From 313c8fd33ebc1945f1c875481a7d5b6921265e0d Mon Sep 17 00:00:00 2001 From: Guchun Chen Date: Thu, 13 Feb 2020 15:34:06 +0800 Subject: drm/amdgpu: log on non-zero error conter per IP before GPU reset Once sync flood interrupt is triggered by RAS error, before actual GPU recovery job, it's necessary to log on and print non-zero error counter, this will help user knows where the RAS error source is from quickly. Signed-off-by: Guchun Chen Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 33 +++++++++++++++++++++++++++++++++ 1 file changed, 33 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index cef94e2169fe..6a9a45d6b9e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1319,6 +1319,33 @@ static int amdgpu_ras_interrupt_remove_all(struct amdgpu_device *adev) } /* ih end */ +/* traversal all IPs except NBIO to query error counter */ +static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj; + + if (!con) + return; + + list_for_each_entry(obj, &con->head, node) { + struct ras_query_if info = { + .head = obj->head, + }; + + /* + * PCIE_BIF IP has one different isr by ras controller + * interrupt, the specific ras counter query will be + * done in that isr. So skip such block from common + * sync flood interrupt isr calling. + */ + if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF) + continue; + + amdgpu_ras_error_query(adev, &info); + } +} + /* recovery begin */ /* return 0 on success. @@ -1373,6 +1400,12 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) struct amdgpu_ras *ras = container_of(work, struct amdgpu_ras, recovery_work); + /* + * Query and print non zero error counter per IP block for + * awareness before recovering GPU. + */ + amdgpu_ras_log_on_err_counter(ras->adev); + if (amdgpu_device_should_recover_gpu(ras->adev)) amdgpu_device_gpu_recover(ras->adev, 0); atomic_set(&ras->in_recovery, 0); -- cgit v1.2.3 From 19744f5f2dff50290f0767a68c0dc418c1a59cec Mon Sep 17 00:00:00 2001 From: Hawking Zhang Date: Mon, 24 Feb 2020 15:36:13 +0800 Subject: drm/amdgpu: move get_xgmi_relative_phy_addr to amdgpu_xgmi.c centralize all the xgmi related function to amdgpu_xgmi.c Signed-off-by: Hawking Zhang Acked-by: Evan Quan Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 20 ++++---------------- drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 15 +++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h | 2 ++ 3 files changed, 21 insertions(+), 16 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 6a9a45d6b9e4..930633a0ed64 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -31,6 +31,7 @@ #include "amdgpu.h" #include "amdgpu_ras.h" #include "amdgpu_atomfirmware.h" +#include "amdgpu_xgmi.h" #include "ivsrcid/nbio/irqsrcs_nbif_7_4.h" const char *ras_error_string[] = { @@ -742,20 +743,6 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, return 0; } -uint64_t get_xgmi_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr) -{ - uint32_t df_inst_id; - - if ((!adev->df.funcs) || - (!adev->df.funcs->get_df_inst_id) || - (!adev->df.funcs->get_dram_base_addr)) - return addr; - - df_inst_id = adev->df.funcs->get_df_inst_id(adev); - - return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id); -} - /* wrapper of psp_ras_trigger_error */ int amdgpu_ras_error_inject(struct amdgpu_device *adev, struct ras_inject_if *info) @@ -775,8 +762,9 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, /* Calculate XGMI relative offset */ if (adev->gmc.xgmi.num_physical_nodes > 1) { - block_info.address = get_xgmi_relative_phy_addr(adev, - block_info.address); + block_info.address = + amdgpu_xgmi_get_relative_phy_addr(adev, + block_info.address); } switch (info->head.block) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 490f57d6704c..b58f725cb2f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -530,3 +530,18 @@ void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev) kfree(ras_if); } } + +uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, + uint64_t addr) +{ + uint32_t df_inst_id; + + if ((!adev->df.funcs) || + (!adev->df.funcs->get_df_inst_id) || + (!adev->df.funcs->get_dram_base_addr)) + return addr; + + df_inst_id = adev->df.funcs->get_df_inst_id(adev); + + return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index c62a4acf4c14..2aa61adee459 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -46,6 +46,8 @@ int amdgpu_xgmi_get_hops_count(struct amdgpu_device *adev, struct amdgpu_device *peer_adev); int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev); void amdgpu_xgmi_ras_fini(struct amdgpu_device *adev); +uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, + uint64_t addr); static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, struct amdgpu_device *bo_adev) -- cgit v1.2.3 From ec01fe2dbf8cf2b2a25c6ac73eac357988f1bf7d Mon Sep 17 00:00:00 2001 From: Hawking Zhang Date: Fri, 21 Feb 2020 21:37:15 +0800 Subject: drm/amdgpu: enable PCS error report on VG20 Now driver will report XGMI/WAFL PCS error through sysfs xgmi_wafl_err_count node on Vega20 Signed-off-by: Hawking Zhang Reviewed-by: Guchun Chen Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 930633a0ed64..e1e8fd4b2b89 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -721,6 +721,9 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev, if (adev->nbio.funcs->query_ras_error_count) adev->nbio.funcs->query_ras_error_count(adev, &err_data); break; + case AMDGPU_RAS_BLOCK__XGMI_WAFL: + amdgpu_xgmi_query_ras_error_count(adev, &err_data); + break; default: break; } -- cgit v1.2.3 From f9317014ea517e13a59b9d437948c5ab17ce4ed7 Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Fri, 6 Mar 2020 11:59:16 +0800 Subject: drm/amdgpu: add function to creat all ras debugfs node centralize all debugfs creation in one place for ras this is required to fix ras when the driver does not use the drm load and unload callbacks due to ordering issues with the drm device node. Signed-off-by: Tao Zhou Signed-off-by: Stanley.Yang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 29 +++++++++++++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 2 ++ 2 files changed, 31 insertions(+) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index e1e8fd4b2b89..7f29e704936e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1113,6 +1113,35 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, &amdgpu_ras_debugfs_ops); } +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + struct ras_manager *obj, *tmp; + struct ras_fs_if fs_info; + + /* + * it won't be called in resume path, no need to check + * suspend and gpu reset status + */ + if (!con) + return; + + amdgpu_ras_debugfs_create_ctrl_node(adev); + + list_for_each_entry_safe(obj, tmp, &con->head, node) { + if (!obj) + continue; + + if (amdgpu_ras_is_supported(adev, obj->head.block) && + (obj->attr_inuse == 1)) { + sprintf(fs_info.debugfs_name, "%s_err_inject", + ras_block_str(obj->head.block)); + fs_info.head = obj->head; + amdgpu_ras_debugfs_create(adev, &fs_info); + } + } +} + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index a5fe29a9373e..55c3eceb390d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -592,6 +592,8 @@ int amdgpu_ras_sysfs_remove(struct amdgpu_device *adev, void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, struct ras_fs_if *head); +void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev); + void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev, struct ras_common_if *head); -- cgit v1.2.3 From 204eaac625d57d32d8b4c42b57271a359b76db5a Mon Sep 17 00:00:00 2001 From: Tao Zhou Date: Fri, 6 Mar 2020 12:24:06 +0800 Subject: drm/amdgpu: call ras_debugfs_create_all in debugfs_init and remove each ras IP's own debugfs creation this is required to fix ras when the driver does not use the drm load and unload callbacks due to ordering issues with the drm device node. Signed-off-by: Tao Zhou Signed-off-by: Stanley.Yang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 3 +++ drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 5 ----- drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c | 1 - drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c | 1 - 8 files changed, 3 insertions(+), 11 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 831f70dc8b59..c573edf02afc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1294,6 +1294,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_ib_preempt, NULL, DEFINE_SIMPLE_ATTRIBUTE(fops_sclk_set, NULL, amdgpu_debugfs_sclk_set, "%llu\n"); +extern void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev); int amdgpu_debugfs_init(struct amdgpu_device *adev) { int r, i; @@ -1366,6 +1367,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) } } + amdgpu_ras_debugfs_create_all(adev); + return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list, ARRAY_SIZE(amdgpu_debugfs_list)); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 952725e7243c..6b9c9193cdfa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -573,7 +573,6 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "gfx_err_count", - .debugfs_name = "gfx_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_gfx_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c index 676c48c02d77..ead3dc572ec5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mmhub.c @@ -32,7 +32,6 @@ int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "mmhub_err_count", - .debugfs_name = "mmhub_err_inject", }; if (!adev->mmhub.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c index 7d5c3a9de9ea..6201a5f4b4fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c @@ -30,7 +30,6 @@ int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "pcie_bif_err_count", - .debugfs_name = "pcie_bif_err_inject", }; if (!adev->nbio.ras_if) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 7f29e704936e..ce8548d5fbf3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1174,7 +1174,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev) static int amdgpu_ras_fs_init(struct amdgpu_device *adev) { amdgpu_ras_sysfs_create_feature_node(adev); - amdgpu_ras_debugfs_create_ctrl_node(adev); return 0; } @@ -1878,8 +1877,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, goto interrupt; } - amdgpu_ras_debugfs_create(adev, fs_info); - r = amdgpu_ras_sysfs_create(adev, fs_info); if (r) goto sysfs; @@ -1888,7 +1885,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev, cleanup: amdgpu_ras_sysfs_remove(adev, ras_block); sysfs: - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); interrupt: @@ -1905,7 +1901,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev, return; amdgpu_ras_sysfs_remove(adev, ras_block); - amdgpu_ras_debugfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); amdgpu_ras_feature_enable(adev, ras_block, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c index 7854c053e85d..250a309e4dee 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c @@ -93,7 +93,6 @@ int amdgpu_sdma_ras_late_init(struct amdgpu_device *adev, struct ras_ih_if *ih_info = (struct ras_ih_if *)ras_ih_info; struct ras_fs_if fs_info = { .sysfs_name = "sdma_err_count", - .debugfs_name = "sdma_err_inject", }; if (!ih_info) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c index f4d40855147b..9dd51f0d2c11 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c @@ -28,7 +28,6 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev) int r; struct ras_fs_if fs_info = { .sysfs_name = "umc_err_count", - .debugfs_name = "umc_err_inject", }; struct ras_ih_if ih_info = { .cb = amdgpu_umc_process_ras_data_cb, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 59d171510ddc..95b3327168ac 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -598,7 +598,6 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev) }; struct ras_fs_if fs_info = { .sysfs_name = "xgmi_wafl_err_count", - .debugfs_name = "xgmi_wafl_err_inject", }; if (!adev->gmc.xgmi.supported || -- cgit v1.2.3 From 88474ccad5f85ee934785e2919689282ba6c6767 Mon Sep 17 00:00:00 2001 From: Guchun Chen Date: Tue, 10 Mar 2020 18:27:08 +0800 Subject: drm/amdgpu: update ras capability's query based on mem ecc configuration RAS support capability needs to be updated on top of different memeory ECC enablement, and remove redundant memory ecc check in gmc module for vega20 and arcturus. v2: check HBM ECC enablement and set ras mask accordingly. v3: avoid to invoke atomfirmware interface to query twice. Suggested-by: Hawking Zhang Signed-off-by: Guchun Chen Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 24 +++++++++++++++------ drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 38 ++++++++++++--------------------- 2 files changed, 32 insertions(+), 30 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index ce8548d5fbf3..c3dca4ca8081 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1765,18 +1765,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev, *hw_supported = 0; *supported = 0; - if (amdgpu_sriov_vf(adev) || + if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw || (adev->asic_type != CHIP_VEGA20 && adev->asic_type != CHIP_ARCTURUS)) return; - if (adev->is_atom_fw && - (amdgpu_atomfirmware_mem_ecc_supported(adev) || - amdgpu_atomfirmware_sram_ecc_supported(adev))) - *hw_supported = AMDGPU_RAS_BLOCK_MASK; + if (amdgpu_atomfirmware_mem_ecc_supported(adev)) { + DRM_INFO("HBM ECC is active.\n"); + *hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("HBM ECC is not presented.\n"); + + if (amdgpu_atomfirmware_sram_ecc_supported(adev)) { + DRM_INFO("SRAM ECC is active.\n"); + *hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC | + 1 << AMDGPU_RAS_BLOCK__DF); + } else + DRM_INFO("SRAM ECC is not presented.\n"); + + /* hw_supported needs to be aligned with RAS block mask. */ + *hw_supported &= AMDGPU_RAS_BLOCK_MASK; *supported = amdgpu_ras_enable == 0 ? - 0 : *hw_supported & amdgpu_ras_mask; + 0 : *hw_supported & amdgpu_ras_mask; } int amdgpu_ras_init(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 6ceaab553130..8606f877478f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -922,30 +922,20 @@ static int gmc_v9_0_late_init(void *handle) if (r) return r; /* Check if ecc is available */ - if (!amdgpu_sriov_vf(adev)) { - switch (adev->asic_type) { - case CHIP_VEGA10: - case CHIP_VEGA20: - case CHIP_ARCTURUS: - r = amdgpu_atomfirmware_mem_ecc_supported(adev); - if (!r) { - DRM_INFO("ECC is not present.\n"); - if (adev->df.funcs->enable_ecc_force_par_wr_rmw) - adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); - } else { - DRM_INFO("ECC is active.\n"); - } - - r = amdgpu_atomfirmware_sram_ecc_supported(adev); - if (!r) { - DRM_INFO("SRAM ECC is not present.\n"); - } else { - DRM_INFO("SRAM ECC is active.\n"); - } - break; - default: - break; - } + if (!amdgpu_sriov_vf(adev) && (adev->asic_type == CHIP_VEGA10)) { + r = amdgpu_atomfirmware_mem_ecc_supported(adev); + if (!r) { + DRM_INFO("ECC is not present.\n"); + if (adev->df.funcs->enable_ecc_force_par_wr_rmw) + adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); + } else + DRM_INFO("ECC is active.\n"); + + r = amdgpu_atomfirmware_sram_ecc_supported(adev); + if (!r) + DRM_INFO("SRAM ECC is not present.\n"); + else + DRM_INFO("SRAM ECC is active.\n"); } if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) -- cgit v1.2.3 From c1509f3f6fa41addf863aee5af7f406e56d110b4 Mon Sep 17 00:00:00 2001 From: "Stanley.Yang" Date: Thu, 12 Mar 2020 18:18:39 +0800 Subject: drm/amdgpu: fix warning in ras_debugfs_create_all() Fix the warning "warn: variable dereferenced before check 'obj' (see line 1131)" by removing unnecessary checks as amdgpu_ras_debugfs_create_all() is only called from amdgpu_debugfs_init() where obj member in con->head list is not NULL. Use list_for_each_entry() instead list_for_each_entry_safe() as obj do not to be freeing or removing from list during this process. Signed-off-by: Stanley.Yang Reviewed-by: Hawking Zhang Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index c3dca4ca8081..43055a01f35e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1116,7 +1116,7 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev, void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); - struct ras_manager *obj, *tmp; + struct ras_manager *obj; struct ras_fs_if fs_info; /* @@ -1128,10 +1128,7 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) amdgpu_ras_debugfs_create_ctrl_node(adev); - list_for_each_entry_safe(obj, tmp, &con->head, node) { - if (!obj) - continue; - + list_for_each_entry(obj, &con->head, node) { if (amdgpu_ras_is_supported(adev, obj->head.block) && (obj->attr_inuse == 1)) { sprintf(fs_info.debugfs_name, "%s_err_inject", -- cgit v1.2.3 From 43c4d57618bef018eecd769c2805ce6f4e849a0d Mon Sep 17 00:00:00 2001 From: John Clements Date: Thu, 19 Mar 2020 14:41:55 +0800 Subject: drm/amdgpu: protect RAS sysfs during GPU reset MMHub EDC becomes dirty after BACO reset EDC registers should be cleared early on in reset phase Reviewed-by: Hawking Zhang Signed-off-by: John Clements Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 12 +++++++++++- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 9 +++++++++ 2 files changed, 20 insertions(+), 1 deletion(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 6f469facabfb..faa3e7102156 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -2742,6 +2742,9 @@ static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) if (adev->asic_reset_res) goto fail; + + if (adev->mmhub.funcs && adev->mmhub.funcs->reset_ras_error_count) + adev->mmhub.funcs->reset_ras_error_count(adev); } else { task_barrier_full(&hive->tb); @@ -3910,8 +3913,15 @@ static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, } } - if (!r && amdgpu_ras_intr_triggered()) + if (!r && amdgpu_ras_intr_triggered()) { + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + if (tmp_adev->mmhub.funcs && + tmp_adev->mmhub.funcs->reset_ras_error_count) + tmp_adev->mmhub.funcs->reset_ras_error_count(tmp_adev); + } + amdgpu_ras_intr_cleared(); + } list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { if (need_full_reset) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 43055a01f35e..3c32a94d2424 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -281,6 +281,11 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user * struct ras_debug_if data; int ret = 0; + if (amdgpu_ras_intr_triggered()) { + DRM_WARN("RAS WARN: error injection currently inaccessible\n"); + return size; + } + ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data); if (ret) return -EINVAL; @@ -394,6 +399,10 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev, .head = obj->head, }; + if (amdgpu_ras_intr_triggered()) + return snprintf(buf, PAGE_SIZE, + "Query currently inaccessible\n"); + if (amdgpu_ras_error_query(obj->adev, &info)) return -EINVAL; -- cgit v1.2.3 From 0b9ebd7eebb7419d18030f0364a9392ffbf1d793 Mon Sep 17 00:00:00 2001 From: John Clements Date: Tue, 7 Apr 2020 15:08:15 +0800 Subject: drm/amdgpu: resolve mGPU RAS query instability upon receiving uncorrectable error, query every GPU node for ras errors Reviewed-by: Hawking Zhang Signed-off-by: John Clements Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c') diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 3c32a94d2424..ab379b44679c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -1424,12 +1424,22 @@ static void amdgpu_ras_do_recovery(struct work_struct *work) { struct amdgpu_ras *ras = container_of(work, struct amdgpu_ras, recovery_work); + struct amdgpu_device *remote_adev = NULL; + struct amdgpu_device *adev = ras->adev; + struct list_head device_list, *device_list_handle = NULL; + struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev, false); + + /* Build list of devices to query RAS related errors */ + if (hive && adev->gmc.xgmi.num_physical_nodes > 1) { + device_list_handle = &hive->device_list; + } else { + list_add_tail(&adev->gmc.xgmi.head, &device_list); + device_list_handle = &device_list; + } - /* - * Query and print non zero error counter per IP block for - * awareness before recovering GPU. - */ - amdgpu_ras_log_on_err_counter(ras->adev); + list_for_each_entry(remote_adev, device_list_handle, gmc.xgmi.head) { + amdgpu_ras_log_on_err_counter(remote_adev); + } if (amdgpu_device_should_recover_gpu(ras->adev)) amdgpu_device_gpu_recover(ras->adev, 0); -- cgit v1.2.3