diff options
29 files changed, 1520 insertions, 877 deletions
diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl index 30b3651d642b..9ddf8c6cb887 100644 --- a/Documentation/DocBook/drm.tmpl +++ b/Documentation/DocBook/drm.tmpl @@ -3982,7 +3982,6 @@ int num_ioctls;</synopsis> <title>Interrupt Handling</title> !Pdrivers/gpu/drm/i915/i915_irq.c interrupt handling !Fdrivers/gpu/drm/i915/i915_irq.c intel_irq_init intel_irq_init_hw intel_hpd_init -!Fdrivers/gpu/drm/i915/i915_irq.c intel_irq_fini !Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_disable_interrupts !Fdrivers/gpu/drm/i915/i915_irq.c intel_runtime_pm_enable_interrupts </sect2> @@ -4199,6 +4198,23 @@ int num_ioctls;</synopsis> !Idrivers/gpu/drm/i915/i915_gem_gtt.c </sect2> <sect2> + <title>GTT Fences and Swizzling</title> +!Idrivers/gpu/drm/i915/i915_gem_fence.c + <sect3> + <title>Global GTT Fence Handling</title> +!Pdrivers/gpu/drm/i915/i915_gem_fence.c fence register handling + </sect3> + <sect3> + <title>Hardware Tiling and Swizzling Details</title> +!Pdrivers/gpu/drm/i915/i915_gem_fence.c tiling swizzling details + </sect3> + </sect2> + <sect2> + <title>Object Tiling IOCTLs</title> +!Idrivers/gpu/drm/i915/i915_gem_tiling.c +!Pdrivers/gpu/drm/i915/i915_gem_tiling.c buffer object tiling + </sect2> + <sect2> <title>Buffer Object Eviction</title> <para> This section documents the interface functions for evicting buffer diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index e52e01251644..41fb8a9c5bef 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -6,12 +6,13 @@ # core driver code i915-y := i915_drv.o \ + i915_irq.o \ i915_params.o \ i915_suspend.o \ i915_sysfs.o \ + intel_csr.o \ intel_pm.o \ - intel_runtime_pm.o \ - intel_csr.o + intel_runtime_pm.o i915-$(CONFIG_COMPAT) += i915_ioc32.o i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o @@ -20,21 +21,20 @@ i915-$(CONFIG_DEBUG_FS) += i915_debugfs.o i915-y += i915_cmd_parser.o \ i915_gem_batch_pool.o \ i915_gem_context.o \ - i915_gem_render_state.o \ i915_gem_debug.o \ i915_gem_dmabuf.o \ i915_gem_evict.o \ i915_gem_execbuffer.o \ + i915_gem_fence.o \ i915_gem_gtt.o \ i915_gem.o \ + i915_gem_render_state.o \ i915_gem_shrinker.o \ i915_gem_stolen.o \ i915_gem_tiling.o \ i915_gem_userptr.o \ i915_gpu_error.o \ - i915_irq.o \ i915_trace_points.o \ - intel_hotplug.o \ intel_lrc.o \ intel_mocs.o \ intel_ringbuffer.o \ @@ -48,11 +48,14 @@ i915-y += intel_renderstate_gen6.o \ # modesetting core code i915-y += intel_audio.o \ + intel_atomic.o \ + intel_atomic_plane.o \ intel_bios.o \ intel_display.o \ intel_fbc.o \ intel_fifo_underrun.o \ intel_frontbuffer.o \ + intel_hotplug.o \ intel_modes.o \ intel_overlay.o \ intel_psr.o \ @@ -68,15 +71,13 @@ i915-y += dvo_ch7017.o \ dvo_ns2501.o \ dvo_sil164.o \ dvo_tfp410.o \ - intel_atomic.o \ - intel_atomic_plane.o \ intel_crt.o \ intel_ddi.o \ - intel_dp.o \ intel_dp_mst.o \ + intel_dp.o \ intel_dsi.o \ - intel_dsi_pll.o \ intel_dsi_panel_vbt.o \ + intel_dsi_pll.o \ intel_dvo.o \ intel_hdmi.o \ intel_i2c.o \ diff --git a/drivers/gpu/drm/i915/i915_cmd_parser.c b/drivers/gpu/drm/i915/i915_cmd_parser.c index 430571b977db..237ff6884a22 100644 --- a/drivers/gpu/drm/i915/i915_cmd_parser.c +++ b/drivers/gpu/drm/i915/i915_cmd_parser.c @@ -151,8 +151,8 @@ static const struct drm_i915_cmd_descriptor render_cmds[] = { CMD( MI_ARB_ON_OFF, SMI, F, 1, R ), CMD( MI_PREDICATE, SMI, F, 1, S ), CMD( MI_TOPOLOGY_FILTER, SMI, F, 1, S ), - CMD( MI_DISPLAY_FLIP, SMI, !F, 0xFF, R ), CMD( MI_SET_APPID, SMI, F, 1, S ), + CMD( MI_DISPLAY_FLIP, SMI, !F, 0xFF, R ), CMD( MI_SET_CONTEXT, SMI, !F, 0xFF, R ), CMD( MI_URB_CLEAR, SMI, !F, 0xFF, S ), CMD( MI_STORE_DWORD_IMM, SMI, !F, 0x3F, B, @@ -564,7 +564,7 @@ static bool validate_cmds_sorted(struct intel_engine_cs *ring, for (j = 0; j < table->count; j++) { const struct drm_i915_cmd_descriptor *desc = - &table->table[i]; + &table->table[j]; u32 curr = desc->cmd.value & desc->cmd.mask; if (curr < previous) { diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 51580bdd587f..23a69307e12e 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -4030,24 +4030,14 @@ static ssize_t i915_displayport_test_active_write(struct file *file, { char *input_buffer; int status = 0; - struct seq_file *m; struct drm_device *dev; struct drm_connector *connector; struct list_head *connector_list; struct intel_dp *intel_dp; int val = 0; - m = file->private_data; - if (!m) { - status = -ENODEV; - return status; - } - dev = m->private; + dev = ((struct seq_file *)file->private_data)->private; - if (!dev) { - status = -ENODEV; - return status; - } connector_list = &dev->mode_config.connector_list; if (len == 0) @@ -4071,9 +4061,7 @@ static ssize_t i915_displayport_test_active_write(struct file *file, DRM_MODE_CONNECTOR_DisplayPort) continue; - if (connector->connector_type == - DRM_MODE_CONNECTOR_DisplayPort && - connector->status == connector_status_connected && + if (connector->status == connector_status_connected && connector->encoder != NULL) { intel_dp = enc_to_intel_dp(connector->encoder); status = kstrtoint(input_buffer, 10, &val); @@ -4105,9 +4093,6 @@ static int i915_displayport_test_active_show(struct seq_file *m, void *data) struct list_head *connector_list = &dev->mode_config.connector_list; struct intel_dp *intel_dp; - if (!dev) - return -ENODEV; - list_for_each_entry(connector, connector_list, head) { if (connector->connector_type != @@ -4152,9 +4137,6 @@ static int i915_displayport_test_data_show(struct seq_file *m, void *data) struct list_head *connector_list = &dev->mode_config.connector_list; struct intel_dp *intel_dp; - if (!dev) - return -ENODEV; - list_for_each_entry(connector, connector_list, head) { if (connector->connector_type != @@ -4194,9 +4176,6 @@ static int i915_displayport_test_type_show(struct seq_file *m, void *data) struct list_head *connector_list = &dev->mode_config.connector_list; struct intel_dp *intel_dp; - if (!dev) - return -ENODEV; - list_for_each_entry(connector, connector_list, head) { if (connector->connector_type != diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index b1f9e5561cf2..ab37d1121be8 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -1274,13 +1274,3 @@ const struct drm_ioctl_desc i915_ioctls[] = { }; int i915_max_ioctl = ARRAY_SIZE(i915_ioctls); - -/* - * This is really ugly: Because old userspace abused the linux agp interface to - * manage the gtt, we need to claim that all intel devices are agp. For - * otherwise the drm core refuses to initialize the agp support code. - */ -int i915_driver_device_is_agp(struct drm_device *dev) -{ - return 1; -} diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 0d6775a3e88c..1d887459e37f 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -935,8 +935,6 @@ static int i915_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (PCI_FUNC(pdev->devfn)) return -ENODEV; - driver.driver_features &= ~(DRIVER_USE_AGP); - return drm_get_pci_dev(pdev, ent, &driver); } @@ -1491,7 +1489,15 @@ static int intel_runtime_suspend(struct device *device) * FIXME: We really should find a document that references the arguments * used below! */ - if (IS_HASWELL(dev)) { + if (IS_BROADWELL(dev)) { + /* + * On Broadwell, if we use PCI_D1 the PCH DDI ports will stop + * being detected, and the call we do at intel_runtime_resume() + * won't be able to restore them. Since PCI_D3hot matches the + * actual specification and appears to be working, use it. + */ + intel_opregion_notify_adapter(dev, PCI_D3hot); + } else { /* * current versions of firmware which depend on this opregion * notification have repurposed the D1 definition to mean @@ -1500,16 +1506,6 @@ static int intel_runtime_suspend(struct device *device) * the suspend path. */ intel_opregion_notify_adapter(dev, PCI_D1); - } else { - /* - * On Broadwell, if we use PCI_D1 the PCH DDI ports will stop - * being detected, and the call we do at intel_runtime_resume() - * won't be able to restore them. Since PCI_D3hot matches the - * actual specification and appears to be working, use it. Let's - * assume the other non-Haswell platforms will stay the same as - * Broadwell. - */ - intel_opregion_notify_adapter(dev, PCI_D3hot); } assert_forcewakes_inactive(dev_priv); @@ -1649,7 +1645,6 @@ static struct drm_driver driver = { * deal with them for Intel hardware. */ .driver_features = - DRIVER_USE_AGP | DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM | DRIVER_PRIME | DRIVER_RENDER, .load = i915_driver_load, @@ -1664,7 +1659,6 @@ static struct drm_driver driver = { .suspend = i915_suspend_legacy, .resume = i915_resume_legacy, - .device_is_agp = i915_driver_device_is_agp, #if defined(CONFIG_DEBUG_FS) .debugfs_init = i915_debugfs_init, .debugfs_cleanup = i915_debugfs_cleanup, diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 23ce125e0298..4e9f7b16a729 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -56,7 +56,7 @@ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20150717" +#define DRIVER_DATE "20150731" #undef WARN_ON /* Many gcc seem to no see through this and fall over :( */ @@ -206,11 +206,11 @@ enum intel_display_power_domain { enum hpd_pin { HPD_NONE = 0, - HPD_PORT_A = HPD_NONE, /* PORT_A is internal */ HPD_TV = HPD_NONE, /* TV is known to be unreliable */ HPD_CRT, HPD_SDVO_B, HPD_SDVO_C, + HPD_PORT_A, HPD_PORT_B, HPD_PORT_C, HPD_PORT_D, @@ -742,7 +742,7 @@ enum csr_state { struct intel_csr { const char *fw_path; - __be32 *dmc_payload; + uint32_t *dmc_payload; uint32_t dmc_fw_size; uint32_t mmio_count; uint32_t mmioaddr[8]; @@ -894,6 +894,7 @@ enum fb_op_origin { ORIGIN_CPU, ORIGIN_CS, ORIGIN_FLIP, + ORIGIN_DIRTYFB, }; struct i915_fbc { @@ -2610,6 +2611,8 @@ struct i915_params { bool reset; bool disable_display; bool disable_vtd_wa; + bool enable_guc_submission; + int guc_log_level; int use_mmio_flip; int mmio_debug; bool verbose_state_checks; @@ -2626,7 +2629,6 @@ extern void i915_driver_preclose(struct drm_device *dev, struct drm_file *file); extern void i915_driver_postclose(struct drm_device *dev, struct drm_file *file); -extern int i915_driver_device_is_agp(struct drm_device * dev); #ifdef CONFIG_COMPAT extern long i915_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); @@ -2646,7 +2648,7 @@ void intel_hpd_irq_handler(struct drm_device *dev, u32 pin_mask, u32 long_mask); void intel_hpd_init(struct drm_i915_private *dev_priv); void intel_hpd_init_work(struct drm_i915_private *dev_priv); void intel_hpd_cancel_work(struct drm_i915_private *dev_priv); -enum port intel_hpd_pin_to_port(enum hpd_pin pin); +bool intel_hpd_pin_to_port(enum hpd_pin pin, enum port *port); /* i915_irq.c */ void i915_queue_hangcheck(struct drm_device *dev); @@ -2758,6 +2760,8 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj, const struct drm_i915_gem_object_ops *ops); struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, size_t size); +struct drm_i915_gem_object *i915_gem_object_create_from_data( + struct drm_device *dev, const void *data, size_t size); void i915_init_vm(struct drm_i915_private *dev_priv, struct i915_address_space *vm); void i915_gem_free_object(struct drm_gem_object *obj); @@ -2864,11 +2868,6 @@ static inline bool i915_gem_request_completed(struct drm_i915_gem_request *req, int __must_check i915_gem_get_seqno(struct drm_device *dev, u32 *seqno); int __must_check i915_gem_set_seqno(struct drm_device *dev, u32 seqno); -int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj); -int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj); - -bool i915_gem_object_pin_fence(struct drm_i915_gem_object *obj); -void i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj); struct drm_i915_gem_request * i915_gem_find_active_request(struct intel_engine_cs *ring); @@ -2966,8 +2965,6 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev, struct dma_buf *i915_gem_prime_export(struct drm_device *dev, struct drm_gem_object *gem_obj, int flags); -void i915_gem_restore_fences(struct drm_device *dev); - unsigned long i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o, const struct i915_ggtt_view *view); @@ -3062,6 +3059,19 @@ i915_gem_object_ggtt_unpin(struct drm_i915_gem_object *obj) i915_gem_object_ggtt_unpin_view(obj, &i915_ggtt_view_normal); } +/* i915_gem_fence.c */ +int __must_check i915_gem_object_get_fence(struct drm_i915_gem_object *obj); +int __must_check i915_gem_object_put_fence(struct drm_i915_gem_object *obj); + +bool i915_gem_object_pin_fence(struct drm_i915_gem_object *obj); +void i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj); + +void i915_gem_restore_fences(struct drm_device *dev); + +void i915_gem_detect_bit_6_swizzle(struct drm_device *dev); +void i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj); +void i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj); + /* i915_gem_context.c */ int __must_check i915_gem_context_init(struct drm_device *dev); void i915_gem_context_fini(struct drm_device *dev); @@ -3154,10 +3164,6 @@ static inline bool i915_gem_object_needs_bit17_swizzle(struct drm_i915_gem_objec obj->tiling_mode != I915_TILING_NONE; } -void i915_gem_detect_bit_6_swizzle(struct drm_device *dev); -void i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj); -void i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj); - /* i915_gem_debug.c */ #if WATCH_LISTS int i915_verify_lists(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index d9f2701b4593..84f91bcc12f7 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -46,11 +46,6 @@ static void i915_gem_object_retire__write(struct drm_i915_gem_object *obj); static void i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring); -static void i915_gem_write_fence(struct drm_device *dev, int reg, - struct drm_i915_gem_object *obj); -static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, - struct drm_i915_fence_reg *fence, - bool enable); static bool cpu_cache_is_coherent(struct drm_device *dev, enum i915_cache_level level) @@ -66,18 +61,6 @@ static bool cpu_write_needs_clflush(struct drm_i915_gem_object *obj) return obj->pin_display; } -static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj) -{ - if (obj->tiling_mode) - i915_gem_release_mmap(obj); - - /* As we do not have an associated fence register, we will force - * a tiling change if we ever need to acquire one. - */ - obj->fence_dirty = false; - obj->fence_reg = I915_FENCE_REG_NONE; -} - /* some bookkeeping */ static void i915_gem_info_add_obj(struct drm_i915_private *dev_priv, size_t size) @@ -2402,6 +2385,13 @@ i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring) if (obj->active) return; + /* Bump our place on the bound list to keep it roughly in LRU order + * so that we don't steal from recently used but inactive objects + * (unless we are forced to ofc!) + */ + list_move_tail(&obj->global_list, + &to_i915(obj->base.dev)->mm.bound_list); + list_for_each_entry(vma, &obj->vma_list, vma_link) { if (!list_empty(&vma->mm_list)) list_move_tail(&vma->mm_list, &vma->vm->inactive_list); @@ -2793,27 +2783,6 @@ static void i915_gem_reset_ring_cleanup(struct drm_i915_private *dev_priv, } } -void i915_gem_restore_fences(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int i; - - for (i = 0; i < dev_priv->num_fence_regs; i++) { - struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i]; - - /* - * Commit delayed tiling changes if we have an object still - * attached to the fence, otherwise just clear the fence. - */ - if (reg->obj) { - i915_gem_object_update_fence(reg->obj, reg, - reg->obj->tiling_mode); - } else { - i915_gem_write_fence(dev, i, NULL); - } - } -} - void i915_gem_reset(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -3340,343 +3309,6 @@ int i915_gpu_idle(struct drm_device *dev) return 0; } -static void i965_write_fence_reg(struct drm_device *dev, int reg, - struct drm_i915_gem_object *obj) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - int fence_reg; - int fence_pitch_shift; - - if (INTEL_INFO(dev)->gen >= 6) { - fence_reg = FENCE_REG_SANDYBRIDGE_0; - fence_pitch_shift = SANDYBRIDGE_FENCE_PITCH_SHIFT; - } else { - fence_reg = FENCE_REG_965_0; - fence_pitch_shift = I965_FENCE_PITCH_SHIFT; - } - - fence_reg += reg * 8; - - /* To w/a incoherency with non-atomic 64-bit register updates, - * we split the 64-bit update into two 32-bit writes. In order - * for a partial fence not to be evaluated between writes, we - * precede the update with write to turn off the fence register, - * and only enable the fence as the last step. - * - * For extra levels of paranoia, we make sure each step lands - * before applying the next step. - */ - I915_WRITE(fence_reg, 0); - POSTING_READ(fence_reg); - - if (obj) { - u32 size = i915_gem_obj_ggtt_size(obj); - uint64_t val; - - /* Adjust fence size to match tiled area */ - if (obj->tiling_mode != I915_TILING_NONE) { - uint32_t row_size = obj->stride * - (obj->tiling_mode == I915_TILING_Y ? 32 : 8); - size = (size / row_size) * row_size; - } - - val = (uint64_t)((i915_gem_obj_ggtt_offset(obj) + size - 4096) & - 0xfffff000) << 32; - val |= i915_gem_obj_ggtt_offset(obj) & 0xfffff000; - val |= (uint64_t)((obj->stride / 128) - 1) << fence_pitch_shift; - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I965_FENCE_TILING_Y_SHIFT; - val |= I965_FENCE_REG_VALID; - - I915_WRITE(fence_reg + 4, val >> 32); - POSTING_READ(fence_reg + 4); - - I915_WRITE(fence_reg + 0, val); - POSTING_READ(fence_reg); - } else { - I915_WRITE(fence_reg + 4, 0); - POSTING_READ(fence_reg + 4); - } -} - -static void i915_write_fence_reg(struct drm_device *dev, int reg, - struct drm_i915_gem_object *obj) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 val; - - if (obj) { - u32 size = i915_gem_obj_ggtt_size(obj); - int pitch_val; - int tile_width; - - WARN((i915_gem_obj_ggtt_offset(obj) & ~I915_FENCE_START_MASK) || - (size & -size) != size || - (i915_gem_obj_ggtt_offset(obj) & (size - 1)), - "object 0x%08lx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n", - i915_gem_obj_ggtt_offset(obj), obj->map_and_fenceable, size); - - if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) - tile_width = 128; - else - tile_width = 512; - - /* Note: pitch better be a power of two tile widths */ - pitch_val = obj->stride / tile_width; - pitch_val = ffs(pitch_val) - 1; - - val = i915_gem_obj_ggtt_offset(obj); - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I830_FENCE_TILING_Y_SHIFT; - val |= I915_FENCE_SIZE_BITS(size); - val |= pitch_val << I830_FENCE_PITCH_SHIFT; - val |= I830_FENCE_REG_VALID; - } else - val = 0; - - if (reg < 8) - reg = FENCE_REG_830_0 + reg * 4; - else - reg = FENCE_REG_945_8 + (reg - 8) * 4; - - I915_WRITE(reg, val); - POSTING_READ(reg); -} - -static void i830_write_fence_reg(struct drm_device *dev, int reg, - struct drm_i915_gem_object *obj) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t val; - - if (obj) { - u32 size = i915_gem_obj_ggtt_size(obj); - uint32_t pitch_val; - - WARN((i915_gem_obj_ggtt_offset(obj) & ~I830_FENCE_START_MASK) || - (size & -size) != size || - (i915_gem_obj_ggtt_offset(obj) & (size - 1)), - "object 0x%08lx not 512K or pot-size 0x%08x aligned\n", - i915_gem_obj_ggtt_offset(obj), size); - - pitch_val = obj->stride / 128; - pitch_val = ffs(pitch_val) - 1; - - val = i915_gem_obj_ggtt_offset(obj); - if (obj->tiling_mode == I915_TILING_Y) - val |= 1 << I830_FENCE_TILING_Y_SHIFT; - val |= I830_FENCE_SIZE_BITS(size); - val |= pitch_val << I830_FENCE_PITCH_SHIFT; - val |= I830_FENCE_REG_VALID; - } else - val = 0; - - I915_WRITE(FENCE_REG_830_0 + reg * 4, val); - POSTING_READ(FENCE_REG_830_0 + reg * 4); -} - -inline static bool i915_gem_object_needs_mb(struct drm_i915_gem_object *obj) -{ - return obj && obj->base.read_domains & I915_GEM_DOMAIN_GTT; -} - -static void i915_gem_write_fence(struct drm_device *dev, int reg, - struct drm_i915_gem_object *obj) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - /* Ensure that all CPU reads are completed before installing a fence - * and all writes before removing the fence. - */ - if (i915_gem_object_needs_mb(dev_priv->fence_regs[reg].obj)) - mb(); - - WARN(obj && (!obj->stride || !obj->tiling_mode), - "bogus fence setup with stride: 0x%x, tiling mode: %i\n", - obj->stride, obj->tiling_mode); - - if (IS_GEN2(dev)) - i830_write_fence_reg(dev, reg, obj); - else if (IS_GEN3(dev)) - i915_write_fence_reg(dev, reg, obj); - else if (INTEL_INFO(dev)->gen >= 4) - i965_write_fence_reg(dev, reg, obj); - - /* And similarly be paranoid that no direct access to this region - * is reordered to before the fence is installed. - */ - if (i915_gem_object_needs_mb(obj)) - mb(); -} - -static inline int fence_number(struct drm_i915_private *dev_priv, - struct drm_i915_fence_reg *fence) -{ - return fence - dev_priv->fence_regs; -} - -static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, - struct drm_i915_fence_reg *fence, - bool enable) -{ - struct drm_i915_private *dev_priv = obj->base.dev->dev_private; - int reg = fence_number(dev_priv, fence); - - i915_gem_write_fence(obj->base.dev, reg, enable ? obj : NULL); - - if (enable) { - obj->fence_reg = reg; - fence->obj = obj; - list_move_tail(&fence->lru_list, &dev_priv->mm.fence_list); - } else { - obj->fence_reg = I915_FENCE_REG_NONE; - fence->obj = NULL; - list_del_init(&fence->lru_list); - } - obj->fence_dirty = false; -} - -static int -i915_gem_object_wait_fence(struct drm_i915_gem_object *obj) -{ - if (obj->last_fenced_req) { - int ret = i915_wait_request(obj->last_fenced_req); - if (ret) - return ret; - - i915_gem_request_assign(&obj->last_fenced_req, NULL); - } - - return 0; -} - -int -i915_gem_object_put_fence(struct drm_i915_gem_object *obj) -{ - struct drm_i915_private *dev_priv = obj->base.dev->dev_private; - struct drm_i915_fence_reg *fence; - int ret; - - ret = i915_gem_object_wait_fence(obj); - if (ret) - return ret; - - if (obj->fence_reg == I915_FENCE_REG_NONE) - return 0; - - fence = &dev_priv->fence_regs[obj->fence_reg]; - - if (WARN_ON(fence->pin_count)) - return -EBUSY; - - i915_gem_object_fence_lost(obj); - i915_gem_object_update_fence(obj, fence, false); - - return 0; -} - -static struct drm_i915_fence_reg * -i915_find_fence_reg(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - struct drm_i915_fence_reg *reg, *avail; - int i; - - /* First try to find a free reg */ - avail = NULL; - for (i = dev_priv->fence_reg_start; i < dev_priv->num_fence_regs; i++) { - reg = &dev_priv->fence_regs[i]; - if (!reg->obj) - return reg; - - if (!reg->pin_count) - avail = reg; - } - - if (avail == NULL) - goto deadlock; - - /* None available, try to steal one or wait for a user to finish */ - list_for_each_entry(reg, &dev_priv->mm.fence_list, lru_list) { - if (reg->pin_count) - continue; - - return reg; - } - -deadlock: - /* Wait for completion of pending flips which consume fences */ - if (intel_has_pending_fb_unpin(dev)) - return ERR_PTR(-EAGAIN); - - return ERR_PTR(-EDEADLK); -} - -/** - * i915_gem_object_get_fence - set up fencing for an object - * @obj: object to map through a fence reg - * - * When mapping objects through the GTT, userspace wants to be able to write - * to them without having to worry about swizzling if the object is tiled. - * This function walks the fence regs looking for a free one for @obj, - * stealing one if it can't find any. - * - * It then sets up the reg based on the object's properties: address, pitch - * and tiling format. - * - * For an untiled surface, this removes any existing fence. - */ -int -i915_gem_object_get_fence(struct drm_i915_gem_object *obj) -{ - struct drm_device *dev = obj->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; - bool enable = obj->tiling_mode != I915_TILING_NONE; - struct drm_i915_fence_reg *reg; - int ret; - - /* Have we updated the tiling parameters upon the object and so - * will need to serialise the write to the associated fence register? - */ - if (obj->fence_dirty) { - ret = i915_gem_object_wait_fence(obj); - if (ret) - return ret; - } - - /* Just update our place in the LRU if our fence is getting reused. */ - if (obj->fence_reg != I915_FENCE_REG_NONE) { - reg = &dev_priv->fence_regs[obj->fence_reg]; - if (!obj->fence_dirty) { - list_move_tail(®->lru_list, - &dev_priv->mm.fence_list); - return 0; - } - } else if (enable) { - if (WARN_ON(!obj->map_and_fenceable)) - return -EINVAL; - - reg = i915_find_fence_reg(dev); - if (IS_ERR(reg)) - return PTR_ERR(reg); - - if (reg->obj) { - struct drm_i915_gem_object *old = reg->obj; - - ret = i915_gem_object_wait_fence(old); - if (ret) - return ret; - - i915_gem_object_fence_lost(old); - } - } else - return 0; - - i915_gem_object_update_fence(obj, reg, enable); - - return 0; -} - static bool i915_gem_valid_gtt_space(struct i915_vma *vma, unsigned long cache_level) { @@ -4476,32 +4108,6 @@ i915_gem_object_ggtt_unpin_view(struct drm_i915_gem_object *obj, --vma->pin_count; } -bool -i915_gem_object_pin_fence(struct drm_i915_gem_object *obj) -{ - if (obj->fence_reg != I915_FENCE_REG_NONE) { - struct drm_i915_private *dev_priv = obj->base.dev->dev_private; - struct i915_vma *ggtt_vma = i915_gem_obj_to_ggtt(obj); - - WARN_ON(!ggtt_vma || - dev_priv->fence_regs[obj->fence_reg].pin_count > - ggtt_vma->pin_count); - dev_priv->fence_regs[obj->fence_reg].pin_count++; - return true; - } else - return false; -} - -void -i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj) -{ - if (obj->fence_reg != I915_FENCE_REG_NONE) { - struct drm_i915_private *dev_priv = obj->base.dev->dev_private; - WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0); - dev_priv->fence_regs[obj->fence_reg].pin_count--; - } -} - int i915_gem_busy_ioctl(struct drm_device *dev, void *data, struct drm_file *file) @@ -5477,3 +5083,43 @@ bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj) return false; } + +/* Allocate a new GEM object and fill it with the supplied data */ +struct drm_i915_gem_object * +i915_gem_object_create_from_data(struct drm_device *dev, + const void *data, size_t size) +{ + struct drm_i915_gem_object *obj; + struct sg_table *sg; + size_t bytes; + int ret; + + obj = i915_gem_alloc_object(dev, round_up(size, PAGE_SIZE)); + if (IS_ERR_OR_NULL(obj)) + return obj; + + ret = i915_gem_object_set_to_cpu_domain(obj, true); + if (ret) + goto fail; + + ret = i915_gem_object_get_pages(obj); + if (ret) + goto fail; + + i915_gem_object_pin_pages(obj); + sg = obj->pages; + bytes = sg_copy_from_buffer(sg->sgl, sg->nents, (void *)data, size); + i915_gem_object_unpin_pages(obj); + + if (WARN_ON(bytes != size)) { + DRM_ERROR("Incomplete copy, wrote %zu of %zu", bytes, size); + ret = -EFAULT; + goto fail; + } + + return obj; + +fail: + drm_gem_object_unreference(&obj->base); + return ERR_PTR(ret); +} diff --git a/drivers/gpu/drm/i915/i915_gem_fence.c b/drivers/gpu/drm/i915/i915_gem_fence.c new file mode 100644 index 000000000000..af1f8c461060 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_gem_fence.c @@ -0,0 +1,787 @@ +/* + * Copyright © 2008-2015 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include <drm/drmP.h> +#include <drm/i915_drm.h> +#include "i915_drv.h" + +/** + * DOC: fence register handling + * + * Important to avoid confusions: "fences" in the i915 driver are not execution + * fences used to track command completion but hardware detiler objects which + * wrap a given range of the global GTT. Each platform has only a fairly limited + * set of these objects. + * + * Fences are used to detile GTT memory mappings. They're also connected to the + * hardware frontbuffer render tracking and hence interract with frontbuffer + * conmpression. Furthermore on older platforms fences are required for tiled + * objects used by the display engine. They can also be used by the render + * engine - they're required for blitter commands and are optional for render + * commands. But on gen4+ both display (with the exception of fbc) and rendering + * have their own tiling state bits and don't need fences. + * + * Also note that fences only support X and Y tiling and hence can't be used for + * the fancier new tiling formats like W, Ys and Yf. + * + * Finally note that because fences are such a restricted resource they're + * dynamically associated with objects. Furthermore fence state is committed to + * the hardware lazily to avoid unecessary stalls on gen2/3. Therefore code must + * explictly call i915_gem_object_get_fence() to synchronize fencing status + * for cpu access. Also note that some code wants an unfenced view, for those + * cases the fence can be removed forcefully with i915_gem_object_put_fence(). + * + * Internally these functions will synchronize with userspace access by removing + * CPU ptes into GTT mmaps (not the GTT ptes themselves) as needed. + */ + +static void i965_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int fence_reg; + int fence_pitch_shift; + + if (INTEL_INFO(dev)->gen >= 6) { + fence_reg = FENCE_REG_SANDYBRIDGE_0; + fence_pitch_shift = SANDYBRIDGE_FENCE_PITCH_SHIFT; + } else { + fence_reg = FENCE_REG_965_0; + fence_pitch_shift = I965_FENCE_PITCH_SHIFT; + } + + fence_reg += reg * 8; + + /* To w/a incoherency with non-atomic 64-bit register updates, + * we split the 64-bit update into two 32-bit writes. In order + * for a partial fence not to be evaluated between writes, we + * precede the update with write to turn off the fence register, + * and only enable the fence as the last step. + * + * For extra levels of paranoia, we make sure each step lands + * before applying the next step. + */ + I915_WRITE(fence_reg, 0); + POSTING_READ(fence_reg); + + if (obj) { + u32 size = i915_gem_obj_ggtt_size(obj); + uint64_t val; + + /* Adjust fence size to match tiled area */ + if (obj->tiling_mode != I915_TILING_NONE) { + uint32_t row_size = obj->stride * + (obj->tiling_mode == I915_TILING_Y ? 32 : 8); + size = (size / row_size) * row_size; + } + + val = (uint64_t)((i915_gem_obj_ggtt_offset(obj) + size - 4096) & + 0xfffff000) << 32; + val |= i915_gem_obj_ggtt_offset(obj) & 0xfffff000; + val |= (uint64_t)((obj->stride / 128) - 1) << fence_pitch_shift; + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I965_FENCE_TILING_Y_SHIFT; + val |= I965_FENCE_REG_VALID; + + I915_WRITE(fence_reg + 4, val >> 32); + POSTING_READ(fence_reg + 4); + + I915_WRITE(fence_reg + 0, val); + POSTING_READ(fence_reg); + } else { + I915_WRITE(fence_reg + 4, 0); + POSTING_READ(fence_reg + 4); + } +} + +static void i915_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 val; + + if (obj) { + u32 size = i915_gem_obj_ggtt_size(obj); + int pitch_val; + int tile_width; + + WARN((i915_gem_obj_ggtt_offset(obj) & ~I915_FENCE_START_MASK) || + (size & -size) != size || + (i915_gem_obj_ggtt_offset(obj) & (size - 1)), + "object 0x%08lx [fenceable? %d] not 1M or pot-size (0x%08x) aligned\n", + i915_gem_obj_ggtt_offset(obj), obj->map_and_fenceable, size); + + if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) + tile_width = 128; + else + tile_width = 512; + + /* Note: pitch better be a power of two tile widths */ + pitch_val = obj->stride / tile_width; + pitch_val = ffs(pitch_val) - 1; + + val = i915_gem_obj_ggtt_offset(obj); + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I830_FENCE_TILING_Y_SHIFT; + val |= I915_FENCE_SIZE_BITS(size); + val |= pitch_val << I830_FENCE_PITCH_SHIFT; + val |= I830_FENCE_REG_VALID; + } else + val = 0; + + if (reg < 8) + reg = FENCE_REG_830_0 + reg * 4; + else + reg = FENCE_REG_945_8 + (reg - 8) * 4; + + I915_WRITE(reg, val); + POSTING_READ(reg); +} + +static void i830_write_fence_reg(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t val; + + if (obj) { + u32 size = i915_gem_obj_ggtt_size(obj); + uint32_t pitch_val; + + WARN((i915_gem_obj_ggtt_offset(obj) & ~I830_FENCE_START_MASK) || + (size & -size) != size || + (i915_gem_obj_ggtt_offset(obj) & (size - 1)), + "object 0x%08lx not 512K or pot-size 0x%08x aligned\n", + i915_gem_obj_ggtt_offset(obj), size); + + pitch_val = obj->stride / 128; + pitch_val = ffs(pitch_val) - 1; + + val = i915_gem_obj_ggtt_offset(obj); + if (obj->tiling_mode == I915_TILING_Y) + val |= 1 << I830_FENCE_TILING_Y_SHIFT; + val |= I830_FENCE_SIZE_BITS(size); + val |= pitch_val << I830_FENCE_PITCH_SHIFT; + val |= I830_FENCE_REG_VALID; + } else + val = 0; + + I915_WRITE(FENCE_REG_830_0 + reg * 4, val); + POSTING_READ(FENCE_REG_830_0 + reg * 4); +} + +inline static bool i915_gem_object_needs_mb(struct drm_i915_gem_object *obj) +{ + return obj && obj->base.read_domains & I915_GEM_DOMAIN_GTT; +} + +static void i915_gem_write_fence(struct drm_device *dev, int reg, + struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + + /* Ensure that all CPU reads are completed before installing a fence + * and all writes before removing the fence. + */ + if (i915_gem_object_needs_mb(dev_priv->fence_regs[reg].obj)) + mb(); + + WARN(obj && (!obj->stride || !obj->tiling_mode), + "bogus fence setup with stride: 0x%x, tiling mode: %i\n", + obj->stride, obj->tiling_mode); + + if (IS_GEN2(dev)) + i830_write_fence_reg(dev, reg, obj); + else if (IS_GEN3(dev)) + i915_write_fence_reg(dev, reg, obj); + else if (INTEL_INFO(dev)->gen >= 4) + i965_write_fence_reg(dev, reg, obj); + + /* And similarly be paranoid that no direct access to this region + * is reordered to before the fence is installed. + */ + if (i915_gem_object_needs_mb(obj)) + mb(); +} + +static inline int fence_number(struct drm_i915_private *dev_priv, + struct drm_i915_fence_reg *fence) +{ + return fence - dev_priv->fence_regs; +} + +static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, + struct drm_i915_fence_reg *fence, + bool enable) +{ + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + int reg = fence_number(dev_priv, fence); + + i915_gem_write_fence(obj->base.dev, reg, enable ? obj : NULL); + + if (enable) { + obj->fence_reg = reg; + fence->obj = obj; + list_move_tail(&fence->lru_list, &dev_priv->mm.fence_list); + } else { + obj->fence_reg = I915_FENCE_REG_NONE; + fence->obj = NULL; + list_del_init(&fence->lru_list); + } + obj->fence_dirty = false; +} + +static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj) +{ + if (obj->tiling_mode) + i915_gem_release_mmap(obj); + + /* As we do not have an associated fence register, we will force + * a tiling change if we ever need to acquire one. + */ + obj->fence_dirty = false; + obj->fence_reg = I915_FENCE_REG_NONE; +} + +static int +i915_gem_object_wait_fence(struct drm_i915_gem_object *obj) +{ + if (obj->last_fenced_req) { + int ret = i915_wait_request(obj->last_fenced_req); + if (ret) + return ret; + + i915_gem_request_assign(&obj->last_fenced_req, NULL); + } + + return 0; +} + +/** + * i915_gem_object_put_fence - force-remove fence for an object + * @obj: object to map through a fence reg + * + * This function force-removes any fence from the given object, which is useful + * if the kernel wants to do untiled GTT access. + * + * Returns: + * + * 0 on success, negative error code on failure. + */ +int +i915_gem_object_put_fence(struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + struct drm_i915_fence_reg *fence; + int ret; + + ret = i915_gem_object_wait_fence(obj); + if (ret) + return ret; + + if (obj->fence_reg == I915_FENCE_REG_NONE) + return 0; + + fence = &dev_priv->fence_regs[obj->fence_reg]; + + if (WARN_ON(fence->pin_count)) + return -EBUSY; + + i915_gem_object_fence_lost(obj); + i915_gem_object_update_fence(obj, fence, false); + + return 0; +} + +static struct drm_i915_fence_reg * +i915_find_fence_reg(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_fence_reg *reg, *avail; + int i; + + /* First try to find a free reg */ + avail = NULL; + for (i = dev_priv->fence_reg_start; i < dev_priv->num_fence_regs; i++) { + reg = &dev_priv->fence_regs[i]; + if (!reg->obj) + return reg; + + if (!reg->pin_count) + avail = reg; + } + + if (avail == NULL) + goto deadlock; + + /* None available, try to steal one or wait for a user to finish */ + list_for_each_entry(reg, &dev_priv->mm.fence_list, lru_list) { + if (reg->pin_count) + continue; + + return reg; + } + +deadlock: + /* Wait for completion of pending flips which consume fences */ + if (intel_has_pending_fb_unpin(dev)) + return ERR_PTR(-EAGAIN); + + return ERR_PTR(-EDEADLK); +} + +/** + * i915_gem_object_get_fence - set up fencing for an object + * @obj: object to map through a fence reg + * + * When mapping objects through the GTT, userspace wants to be able to write + * to them without having to worry about swizzling if the object is tiled. + * This function walks the fence regs looking for a free one for @obj, + * stealing one if it can't find any. + * + * It then sets up the reg based on the object's properties: address, pitch + * and tiling format. + * + * For an untiled surface, this removes any existing fence. + * + * Returns: + * + * 0 on success, negative error code on failure. + */ +int +i915_gem_object_get_fence(struct drm_i915_gem_object *obj) +{ + struct drm_device *dev = obj->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + bool enable = obj->tiling_mode != I915_TILING_NONE; + struct drm_i915_fence_reg *reg; + int ret; + + /* Have we updated the tiling parameters upon the object and so + * will need to serialise the write to the associated fence register? + */ + if (obj->fence_dirty) { + ret = i915_gem_object_wait_fence(obj); + if (ret) + return ret; + } + + /* Just update our place in the LRU if our fence is getting reused. */ + if (obj->fence_reg != I915_FENCE_REG_NONE) { + reg = &dev_priv->fence_regs[obj->fence_reg]; + if (!obj->fence_dirty) { + list_move_tail(®->lru_list, + &dev_priv->mm.fence_list); + return 0; + } + } else if (enable) { + if (WARN_ON(!obj->map_and_fenceable)) + return -EINVAL; + + reg = i915_find_fence_reg(dev); + if (IS_ERR(reg)) + return PTR_ERR(reg); + + if (reg->obj) { + struct drm_i915_gem_object *old = reg->obj; + + ret = i915_gem_object_wait_fence(old); + if (ret) + return ret; + + i915_gem_object_fence_lost(old); + } + } else + return 0; + + i915_gem_object_update_fence(obj, reg, enable); + + return 0; +} + +/** + * i915_gem_object_pin_fence - pin fencing state + * @obj: object to pin fencing for + * + * This pins the fencing state (whether tiled or untiled) to make sure the + * object is ready to be used as a scanout target. Fencing status must be + * synchronize first by calling i915_gem_object_get_fence(): + * + * The resulting fence pin reference must be released again with + * i915_gem_object_unpin_fence(). + * + * Returns: + * + * True if the object has a fence, false otherwise. + */ +bool +i915_gem_object_pin_fence(struct drm_i915_gem_object *obj) +{ + if (obj->fence_reg != I915_FENCE_REG_NONE) { + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + struct i915_vma *ggtt_vma = i915_gem_obj_to_ggtt(obj); + + WARN_ON(!ggtt_vma || + dev_priv->fence_regs[obj->fence_reg].pin_count > + ggtt_vma->pin_count); + dev_priv->fence_regs[obj->fence_reg].pin_count++; + return true; + } else + return false; +} + +/** + * i915_gem_object_unpin_fence - unpin fencing state + * @obj: object to unpin fencing for + * + * This releases the fence pin reference acquired through + * i915_gem_object_pin_fence. It will handle both objects with and without an + * attached fence correctly, callers do not need to distinguish this. + */ +void +i915_gem_object_unpin_fence(struct drm_i915_gem_object *obj) +{ + if (obj->fence_reg != I915_FENCE_REG_NONE) { + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + WARN_ON(dev_priv->fence_regs[obj->fence_reg].pin_count <= 0); + dev_priv->fence_regs[obj->fence_reg].pin_count--; + } +} + +/** + * i915_gem_restore_fences - restore fence state + * @dev: DRM device + * + * Restore the hw fence state to match the software tracking again, to be called + * after a gpu reset and on resume. + */ +void i915_gem_restore_fences(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + int i; + + for (i = 0; i < dev_priv->num_fence_regs; i++) { + struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i]; + + /* + * Commit delayed tiling changes if we have an object still + * attached to the fence, otherwise just clear the fence. + */ + if (reg->obj) { + i915_gem_object_update_fence(reg->obj, reg, + reg->obj->tiling_mode); + } else { + i915_gem_write_fence(dev, i, NULL); + } + } +} + +/** + * DOC: tiling swizzling details + * + * The idea behind tiling is to increase cache hit rates by rearranging + * pixel data so that a group of pixel accesses are in the same cacheline. + * Performance improvement from doing this on the back/depth buffer are on + * the order of 30%. + * + * Intel architectures make this somewhat more complicated, though, by + * adjustments made to addressing of data when the memory is in interleaved + * mode (matched pairs of DIMMS) to improve memory bandwidth. + * For interleaved memory, the CPU sends every sequential 64 bytes + * to an alternate memory channel so it can get the bandwidth from both. + * + * The GPU also rearranges its accesses for increased bandwidth to interleaved + * memory, and it matches what the CPU does for non-tiled. However, when tiled + * it does it a little differently, since one walks addresses not just in the + * X direction but also Y. So, along with alternating channels when bit + * 6 of the address flips, it also alternates when other bits flip -- Bits 9 + * (every 512 bytes, an X tile scanline) and 10 (every two X tile scanlines) + * are common to both the 915 and 965-class hardware. + * + * The CPU also sometimes XORs in higher bits as well, to improve + * bandwidth doing strided access like we do so frequently in graphics. This + * is called "Channel XOR Randomization" in the MCH documentation. The result + * is that the CPU is XORing in either bit 11 or bit 17 to bit 6 of its address + * decode. + * + * All of this bit 6 XORing has an effect on our memory management, + * as we need to make sure that the 3d driver can correctly address object + * contents. + * + * If we don't have interleaved memory, all tiling is safe and no swizzling is + * required. + * + * When bit 17 is XORed in, we simply refuse to tile at all. Bit + * 17 is not just a page offset, so as we page an objet out and back in, + * individual pages in it will have different bit 17 addresses, resulting in + * each 64 bytes being swapped with its neighbor! + * + * Otherwise, if interleaved, we have to tell the 3d driver what the address + * swizzling it needs to do is, since it's writing with the CPU to the pages + * (bit 6 and potentially bit 11 XORed in), and the GPU is reading from the + * pages (bit 6, 9, and 10 XORed in), resulting in a cumulative bit swizzling + * required by the CPU of XORing in bit 6, 9, 10, and potentially 11, in order + * to match what the GPU expects. + */ + +/** + * i915_gem_detect_bit_6_swizzle - detect bit 6 swizzling pattern + * @dev: DRM device + * + * Detects bit 6 swizzling of address lookup between IGD access and CPU + * access through main memory. + */ +void +i915_gem_detect_bit_6_swizzle(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; + uint32_t swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; + + if (INTEL_INFO(dev)->gen >= 8 || IS_VALLEYVIEW(dev)) { + /* + * On BDW+, swizzling is not used. We leave the CPU memory + * controller in charge of optimizing memory accesses without + * the extra address manipulation GPU side. + * + * VLV and CHV don't have GPU swizzling. + */ + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } else if (INTEL_INFO(dev)->gen >= 6) { + if (dev_priv->preserve_bios_swizzle) { + if (I915_READ(DISP_ARB_CTL) & + DISP_TILE_SURFACE_SWIZZLING) { + swizzle_x = I915_BIT_6_SWIZZLE_9_10; + swizzle_y = I915_BIT_6_SWIZZLE_9; + } else { + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } + } else { + uint32_t dimm_c0, dimm_c1; + dimm_c0 = I915_READ(MAD_DIMM_C0); + dimm_c1 = I915_READ(MAD_DIMM_C1); + dimm_c0 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK; + dimm_c1 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK; + /* Enable swizzling when the channels are populated + * with identically sized dimms. We don't need to check + * the 3rd channel because no cpu with gpu attached + * ships in that configuration. Also, swizzling only + * makes sense for 2 channels anyway. */ + if (dimm_c0 == dimm_c1) { + swizzle_x = I915_BIT_6_SWIZZLE_9_10; + swizzle_y = I915_BIT_6_SWIZZLE_9; + } else { + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } + } + } else if (IS_GEN5(dev)) { + /* On Ironlake whatever DRAM config, GPU always do + * same swizzling setup. + */ + swizzle_x = I915_BIT_6_SWIZZLE_9_10; + swizzle_y = I915_BIT_6_SWIZZLE_9; + } else if (IS_GEN2(dev)) { + /* As far as we know, the 865 doesn't have these bit 6 + * swizzling issues. + */ + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } else if (IS_MOBILE(dev) || (IS_GEN3(dev) && !IS_G33(dev))) { + uint32_t dcc; + + /* On 9xx chipsets, channel interleave by the CPU is + * determined by DCC. For single-channel, neither the CPU + * nor the GPU do swizzling. For dual channel interleaved, + * the GPU's interleave is bit 9 and 10 for X tiled, and bit + * 9 for Y tiled. The CPU's interleave is independent, and + * can be based on either bit 11 (haven't seen this yet) or + * bit 17 (common). + */ + dcc = I915_READ(DCC); + switch (dcc & DCC_ADDRESSING_MODE_MASK) { + case DCC_ADDRESSING_MODE_SINGLE_CHANNEL: + case DCC_ADDRESSING_MODE_DUAL_CHANNEL_ASYMMETRIC: + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + break; + case DCC_ADDRESSING_MODE_DUAL_CHANNEL_INTERLEAVED: + if (dcc & DCC_CHANNEL_XOR_DISABLE) { + /* This is the base swizzling by the GPU for + * tiled buffers. + */ + swizzle_x = I915_BIT_6_SWIZZLE_9_10; + swizzle_y = I915_BIT_6_SWIZZLE_9; + } else if ((dcc & DCC_CHANNEL_XOR_BIT_17) == 0) { + /* Bit 11 swizzling by the CPU in addition. */ + swizzle_x = I915_BIT_6_SWIZZLE_9_10_11; + swizzle_y = I915_BIT_6_SWIZZLE_9_11; + } else { + /* Bit 17 swizzling by the CPU in addition. */ + swizzle_x = I915_BIT_6_SWIZZLE_9_10_17; + swizzle_y = I915_BIT_6_SWIZZLE_9_17; + } + break; + } + + /* check for L-shaped memory aka modified enhanced addressing */ + if (IS_GEN4(dev)) { + uint32_t ddc2 = I915_READ(DCC2); + + if (!(ddc2 & DCC2_MODIFIED_ENHANCED_DISABLE)) + dev_priv->quirks |= QUIRK_PIN_SWIZZLED_PAGES; + } + + if (dcc == 0xffffffff) { + DRM_ERROR("Couldn't read from MCHBAR. " + "Disabling tiling.\n"); + swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; + swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; + } + } else { + /* The 965, G33, and newer, have a very flexible memory + * configuration. It will enable dual-channel mode + * (interleaving) on as much memory as it can, and the GPU + * will additionally sometimes enable different bit 6 + * swizzling for tiled objects from the CPU. + * + * Here's what I found on the G965: + * slot fill memory size swizzling + * 0A 0B 1A 1B 1-ch 2-ch + * 512 0 0 0 512 0 O + * 512 0 512 0 16 1008 X + * 512 0 0 512 16 1008 X + * 0 512 0 512 16 1008 X + * 1024 1024 1024 0 2048 1024 O + * + * We could probably detect this based on either the DRB + * matching, which was the case for the swizzling required in + * the table above, or from the 1-ch value being less than + * the minimum size of a rank. + */ + if (I915_READ16(C0DRB3) != I915_READ16(C1DRB3)) { + swizzle_x = I915_BIT_6_SWIZZLE_NONE; + swizzle_y = I915_BIT_6_SWIZZLE_NONE; + } else { + swizzle_x = I915_BIT_6_SWIZZLE_9_10; + swizzle_y = I915_BIT_6_SWIZZLE_9; + } + } + + dev_priv->mm.bit_6_swizzle_x = swizzle_x; + dev_priv->mm.bit_6_swizzle_y = swizzle_y; +} + +/* + * Swap every 64 bytes of this page around, to account for it having a new + * bit 17 of its physical address and therefore being interpreted differently + * by the GPU. + */ +static void +i915_gem_swizzle_page(struct page *page) +{ + char temp[64]; + char *vaddr; + int i; + + vaddr = kmap(page); + + for (i = 0; i < PAGE_SIZE; i += 128) { + memcpy(temp, &vaddr[i], 64); + memcpy(&vaddr[i], &vaddr[i + 64], 64); + memcpy(&vaddr[i + 64], temp, 64); + } + + kunmap(page); +} + +/** + * i915_gem_object_do_bit_17_swizzle - fixup bit 17 swizzling + * @obj: i915 GEM buffer object + * + * This function fixes up the swizzling in case any page frame number for this + * object has changed in bit 17 since that state has been saved with + * i915_gem_object_save_bit_17_swizzle(). + * + * This is called when pinning backing storage again, since the kernel is free + * to move unpinned backing storage around (either by directly moving pages or + * by swapping them out and back in again). + */ +void +i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj) +{ + struct sg_page_iter sg_iter; + int i; + + if (obj->bit_17 == NULL) + return; + + i = 0; + for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) { + struct page *page = sg_page_iter_page(&sg_iter); + char new_bit_17 = page_to_phys(page) >> 17; + if ((new_bit_17 & 0x1) != + (test_bit(i, obj->bit_17) != 0)) { + i915_gem_swizzle_page(page); + set_page_dirty(page); + } + i++; + } +} + +/** + * i915_gem_object_save_bit_17_swizzle - save bit 17 swizzling + * @obj: i915 GEM buffer object + * + * This function saves the bit 17 of each page frame number so that swizzling + * can be fixed up later on with i915_gem_object_do_bit_17_swizzle(). This must + * be called before the backing storage can be unpinned. + */ +void +i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj) +{ + struct sg_page_iter sg_iter; + int page_count = obj->base.size >> PAGE_SHIFT; + int i; + + if (obj->bit_17 == NULL) { + obj->bit_17 = kcalloc(BITS_TO_LONGS(page_count), + sizeof(long), GFP_KERNEL); + if (obj->bit_17 == NULL) { + DRM_ERROR("Failed to allocate memory for bit 17 " + "record\n"); + return; + } + } + + i = 0; + for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) { + if (page_to_phys(sg_page_iter_page(&sg_iter)) & (1 << 17)) + __set_bit(i, obj->bit_17); + else + __clear_bit(i, obj->bit_17); + i++; + } +} diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.c b/drivers/gpu/drm/i915/i915_gem_render_state.c index a0201fc94d25..5026a6267a88 100644 --- a/drivers/gpu/drm/i915/i915_gem_render_state.c +++ b/drivers/gpu/drm/i915/i915_gem_render_state.c @@ -73,6 +73,24 @@ free_gem: return ret; } +/* + * Macro to add commands to auxiliary batch. + * This macro only checks for page overflow before inserting the commands, + * this is sufficient as the null state generator makes the final batch + * with two passes to build command and state separately. At this point + * the size of both are known and it compacts them by relocating the state + * right after the commands taking care of aligment so we should sufficient + * space below them for adding new commands. + */ +#define OUT_BATCH(batch, i, val) \ + do { \ + if (WARN_ON((i) >= PAGE_SIZE / sizeof(u32))) { \ + ret = -ENOSPC; \ + goto err_out; \ + } \ + (batch)[(i)++] = (val); \ + } while(0) + static int render_state_setup(struct render_state *so) { const struct intel_renderstate_rodata *rodata = so->rodata; @@ -96,8 +114,10 @@ static int render_state_setup(struct render_state *so) s = lower_32_bits(r); if (so->gen >= 8) { if (i + 1 >= rodata->batch_items || - rodata->batch[i + 1] != 0) - return -EINVAL; + rodata->batch[i + 1] != 0) { + ret = -EINVAL; + goto err_out; + } d[i++] = s; s = upper_32_bits(r); @@ -108,6 +128,21 @@ static int render_state_setup(struct render_state *so) d[i++] = s; } + + while (i % CACHELINE_DWORDS) + OUT_BATCH(d, i, MI_NOOP); + + so->aux_batch_offset = i * sizeof(u32); + + OUT_BATCH(d, i, MI_BATCH_BUFFER_END); + so->aux_batch_size = (i * sizeof(u32)) - so->aux_batch_offset; + + /* + * Since we are sending length, we need to strictly conform to + * all requirements. For Gen2 this must be a multiple of 8. + */ + so->aux_batch_size = ALIGN(so->aux_batch_size, 8); + kunmap(page); ret = i915_gem_object_set_to_gtt_domain(so->obj, false); @@ -120,8 +155,14 @@ static int render_state_setup(struct render_state *so) } return 0; + +err_out: + kunmap(page); + return ret; } +#undef OUT_BATCH + void i915_gem_render_state_fini(struct render_state *so) { i915_gem_object_ggtt_unpin(so->obj); @@ -170,6 +211,16 @@ int i915_gem_render_state_init(struct drm_i915_gem_request *req) if (ret) goto out; + if (so.aux_batch_size > 8) { + ret = req->ring->dispatch_execbuffer(req, + (so.ggtt_offset + + so.aux_batch_offset), + so.aux_batch_size, + I915_DISPATCH_SECURE); + if (ret) + goto out; + } + i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req); out: diff --git a/drivers/gpu/drm/i915/i915_gem_render_state.h b/drivers/gpu/drm/i915/i915_gem_render_state.h index 7aa73728178a..e641bb093a90 100644 --- a/drivers/gpu/drm/i915/i915_gem_render_state.h +++ b/drivers/gpu/drm/i915/i915_gem_render_state.h @@ -37,6 +37,8 @@ struct render_state { struct drm_i915_gem_object *obj; u64 ggtt_offset; int gen; + u32 aux_batch_size; + u32 aux_batch_offset; }; int i915_gem_render_state_init(struct drm_i915_gem_request *req); diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 633bd1fcab69..ac3eb566c9d2 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -31,201 +31,32 @@ #include <drm/i915_drm.h> #include "i915_drv.h" -/** @file i915_gem_tiling.c - * - * Support for managing tiling state of buffer objects. - * - * The idea behind tiling is to increase cache hit rates by rearranging - * pixel data so that a group of pixel accesses are in the same cacheline. - * Performance improvement from doing this on the back/depth buffer are on - * the order of 30%. - * - * Intel architectures make this somewhat more complicated, though, by - * adjustments made to addressing of data when the memory is in interleaved - * mode (matched pairs of DIMMS) to improve memory bandwidth. - * For interleaved memory, the CPU sends every sequential 64 bytes - * to an alternate memory channel so it can get the bandwidth from both. - * - * The GPU also rearranges its accesses for increased bandwidth to interleaved - * memory, and it matches what the CPU does for non-tiled. However, when tiled - * it does it a little differently, since one walks addresses not just in the - * X direction but also Y. So, along with alternating channels when bit - * 6 of the address flips, it also alternates when other bits flip -- Bits 9 - * (every 512 bytes, an X tile scanline) and 10 (every two X tile scanlines) - * are common to both the 915 and 965-class hardware. - * - * The CPU also sometimes XORs in higher bits as well, to improve - * bandwidth doing strided access like we do so frequently in graphics. This - * is called "Channel XOR Randomization" in the MCH documentation. The result - * is that the CPU is XORing in either bit 11 or bit 17 to bit 6 of its address - * decode. +/** + * DOC: buffer object tiling * - * All of this bit 6 XORing has an effect on our memory management, - * as we need to make sure that the 3d driver can correctly address object - * contents. + * i915_gem_set_tiling() and i915_gem_get_tiling() is the userspace interface to + * declare fence register requirements. * - * If we don't have interleaved memory, all tiling is safe and no swizzling is - * required. + * In principle GEM doesn't care at all about the internal data layout of an + * object, and hence it also doesn't care about tiling or swizzling. There's two + * exceptions: * - * When bit 17 is XORed in, we simply refuse to tile at all. Bit - * 17 is not just a page offset, so as we page an objet out and back in, - * individual pages in it will have different bit 17 addresses, resulting in - * each 64 bytes being swapped with its neighbor! + * - For X and Y tiling the hardware provides detilers for CPU access, so called + * fences. Since there's only a limited amount of them the kernel must manage + * these, and therefore userspace must tell the kernel the object tiling if it + * wants to use fences for detiling. + * - On gen3 and gen4 platforms have a swizzling pattern for tiled objects which + * depends upon the physical page frame number. When swapping such objects the + * page frame number might change and the kernel must be able to fix this up + * and hence now the tiling. Note that on a subset of platforms with + * asymmetric memory channel population the swizzling pattern changes in an + * unknown way, and for those the kernel simply forbids swapping completely. * - * Otherwise, if interleaved, we have to tell the 3d driver what the address - * swizzling it needs to do is, since it's writing with the CPU to the pages - * (bit 6 and potentially bit 11 XORed in), and the GPU is reading from the - * pages (bit 6, 9, and 10 XORed in), resulting in a cumulative bit swizzling - * required by the CPU of XORing in bit 6, 9, 10, and potentially 11, in order - * to match what the GPU expects. - */ - -/** - * Detects bit 6 swizzling of address lookup between IGD access and CPU - * access through main memory. + * Since neither of this applies for new tiling layouts on modern platforms like + * W, Ys and Yf tiling GEM only allows object tiling to be set to X or Y tiled. + * Anything else can be handled in userspace entirely without the kernel's + * invovlement. */ -void -i915_gem_detect_bit_6_swizzle(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; - uint32_t swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; - - if (INTEL_INFO(dev)->gen >= 8 || IS_VALLEYVIEW(dev)) { - /* - * On BDW+, swizzling is not used. We leave the CPU memory - * controller in charge of optimizing memory accesses without - * the extra address manipulation GPU side. - * - * VLV and CHV don't have GPU swizzling. - */ - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } else if (INTEL_INFO(dev)->gen >= 6) { - if (dev_priv->preserve_bios_swizzle) { - if (I915_READ(DISP_ARB_CTL) & - DISP_TILE_SURFACE_SWIZZLING) { - swizzle_x = I915_BIT_6_SWIZZLE_9_10; - swizzle_y = I915_BIT_6_SWIZZLE_9; - } else { - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } - } else { - uint32_t dimm_c0, dimm_c1; - dimm_c0 = I915_READ(MAD_DIMM_C0); - dimm_c1 = I915_READ(MAD_DIMM_C1); - dimm_c0 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK; - dimm_c1 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK; - /* Enable swizzling when the channels are populated - * with identically sized dimms. We don't need to check - * the 3rd channel because no cpu with gpu attached - * ships in that configuration. Also, swizzling only - * makes sense for 2 channels anyway. */ - if (dimm_c0 == dimm_c1) { - swizzle_x = I915_BIT_6_SWIZZLE_9_10; - swizzle_y = I915_BIT_6_SWIZZLE_9; - } else { - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } - } - } else if (IS_GEN5(dev)) { - /* On Ironlake whatever DRAM config, GPU always do - * same swizzling setup. - */ - swizzle_x = I915_BIT_6_SWIZZLE_9_10; - swizzle_y = I915_BIT_6_SWIZZLE_9; - } else if (IS_GEN2(dev)) { - /* As far as we know, the 865 doesn't have these bit 6 - * swizzling issues. - */ - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } else if (IS_MOBILE(dev) || (IS_GEN3(dev) && !IS_G33(dev))) { - uint32_t dcc; - - /* On 9xx chipsets, channel interleave by the CPU is - * determined by DCC. For single-channel, neither the CPU - * nor the GPU do swizzling. For dual channel interleaved, - * the GPU's interleave is bit 9 and 10 for X tiled, and bit - * 9 for Y tiled. The CPU's interleave is independent, and - * can be based on either bit 11 (haven't seen this yet) or - * bit 17 (common). - */ - dcc = I915_READ(DCC); - switch (dcc & DCC_ADDRESSING_MODE_MASK) { - case DCC_ADDRESSING_MODE_SINGLE_CHANNEL: - case DCC_ADDRESSING_MODE_DUAL_CHANNEL_ASYMMETRIC: - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - break; - case DCC_ADDRESSING_MODE_DUAL_CHANNEL_INTERLEAVED: - if (dcc & DCC_CHANNEL_XOR_DISABLE) { - /* This is the base swizzling by the GPU for - * tiled buffers. - */ - swizzle_x = I915_BIT_6_SWIZZLE_9_10; - swizzle_y = I915_BIT_6_SWIZZLE_9; - } else if ((dcc & DCC_CHANNEL_XOR_BIT_17) == 0) { - /* Bit 11 swizzling by the CPU in addition. */ - swizzle_x = I915_BIT_6_SWIZZLE_9_10_11; - swizzle_y = I915_BIT_6_SWIZZLE_9_11; - } else { - /* Bit 17 swizzling by the CPU in addition. */ - swizzle_x = I915_BIT_6_SWIZZLE_9_10_17; - swizzle_y = I915_BIT_6_SWIZZLE_9_17; - } - break; - } - - /* check for L-shaped memory aka modified enhanced addressing */ - if (IS_GEN4(dev)) { - uint32_t ddc2 = I915_READ(DCC2); - - if (!(ddc2 & DCC2_MODIFIED_ENHANCED_DISABLE)) - dev_priv->quirks |= QUIRK_PIN_SWIZZLED_PAGES; - } - - if (dcc == 0xffffffff) { - DRM_ERROR("Couldn't read from MCHBAR. " - "Disabling tiling.\n"); - swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN; - swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN; - } - } else { - /* The 965, G33, and newer, have a very flexible memory - * configuration. It will enable dual-channel mode - * (interleaving) on as much memory as it can, and the GPU - * will additionally sometimes enable different bit 6 - * swizzling for tiled objects from the CPU. - * - * Here's what I found on the G965: - * slot fill memory size swizzling - * 0A 0B 1A 1B 1-ch 2-ch - * 512 0 0 0 512 0 O - * 512 0 512 0 16 1008 X - * 512 0 0 512 16 1008 X - * 0 512 0 512 16 1008 X - * 1024 1024 1024 0 2048 1024 O - * - * We could probably detect this based on either the DRB - * matching, which was the case for the swizzling required in - * the table above, or from the 1-ch value being less than - * the minimum size of a rank. - */ - if (I915_READ16(C0DRB3) != I915_READ16(C1DRB3)) { - swizzle_x = I915_BIT_6_SWIZZLE_NONE; - swizzle_y = I915_BIT_6_SWIZZLE_NONE; - } else { - swizzle_x = I915_BIT_6_SWIZZLE_9_10; - swizzle_y = I915_BIT_6_SWIZZLE_9; - } - } - - dev_priv->mm.bit_6_swizzle_x = swizzle_x; - dev_priv->mm.bit_6_swizzle_y = swizzle_y; -} /* Check pitch constriants for all chips & tiling formats */ static bool @@ -313,8 +144,18 @@ i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode) } /** + * i915_gem_set_tiling - IOCTL handler to set tiling mode + * @dev: DRM device + * @data: data pointer for the ioctl + * @file: DRM file for the ioctl call + * * Sets the tiling mode of an object, returning the required swizzling of * bit 6 of addresses in the object. + * + * Called by the user via ioctl. + * + * Returns: + * Zero on success, negative errno on failure. */ int i915_gem_set_tiling(struct drm_device *dev, void *data, @@ -432,7 +273,17 @@ err: } /** + * i915_gem_get_tiling - IOCTL handler to get tiling mode + * @dev: DRM device + * @data: data pointer for the ioctl + * @file: DRM file for the ioctl call + * * Returns the current tiling mode and required bit 6 swizzling for the object. + * + * Called by the user via ioctl. + * + * Returns: + * Zero on success, negative errno on failure. */ int i915_gem_get_tiling(struct drm_device *dev, void *data, @@ -475,75 +326,3 @@ i915_gem_get_tiling(struct drm_device *dev, void *data, return 0; } - -/** - * Swap every 64 bytes of this page around, to account for it having a new - * bit 17 of its physical address and therefore being interpreted differently - * by the GPU. - */ -static void -i915_gem_swizzle_page(struct page *page) -{ - char temp[64]; - char *vaddr; - int i; - - vaddr = kmap(page); - - for (i = 0; i < PAGE_SIZE; i += 128) { - memcpy(temp, &vaddr[i], 64); - memcpy(&vaddr[i], &vaddr[i + 64], 64); - memcpy(&vaddr[i + 64], temp, 64); - } - - kunmap(page); -} - -void -i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj) -{ - struct sg_page_iter sg_iter; - int i; - - if (obj->bit_17 == NULL) - return; - - i = 0; - for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) { - struct page *page = sg_page_iter_page(&sg_iter); - char new_bit_17 = page_to_phys(page) >> 17; - if ((new_bit_17 & 0x1) != - (test_bit(i, obj->bit_17) != 0)) { - i915_gem_swizzle_page(page); - set_page_dirty(page); - } - i++; - } -} - -void -i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj) -{ - struct sg_page_iter sg_iter; - int page_count = obj->base.size >> PAGE_SHIFT; - int i; - - if (obj->bit_17 == NULL) { - obj->bit_17 = kcalloc(BITS_TO_LONGS(page_count), - sizeof(long), GFP_KERNEL); - if (obj->bit_17 == NULL) { - DRM_ERROR("Failed to allocate memory for bit 17 " - "record\n"); - return; - } - } - - i = 0; - for_each_sg_page(obj->pages->sgl, &sg_iter, obj->pages->nents, 0) { - if (page_to_phys(sg_page_iter_page(&sg_iter)) & (1 << 17)) - __set_bit(i, obj->bit_17); - else - __clear_bit(i, obj->bit_17); - i++; - } -} diff --git a/drivers/gpu/drm/i915/i915_guc_reg.h b/drivers/gpu/drm/i915/i915_guc_reg.h new file mode 100644 index 000000000000..ccdc6c8ac20b --- /dev/null +++ b/drivers/gpu/drm/i915/i915_guc_reg.h @@ -0,0 +1,102 @@ +/* + * Copyright © 2014 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ +#ifndef _I915_GUC_REG_H_ +#define _I915_GUC_REG_H_ + +/* Definitions of GuC H/W registers, bits, etc */ + +#define GUC_STATUS 0xc000 +#define GS_BOOTROM_SHIFT 1 +#define GS_BOOTROM_MASK (0x7F << GS_BOOTROM_SHIFT) +#define GS_BOOTROM_RSA_FAILED (0x50 << GS_BOOTROM_SHIFT) +#define GS_UKERNEL_SHIFT 8 +#define GS_UKERNEL_MASK (0xFF << GS_UKERNEL_SHIFT) +#define GS_UKERNEL_LAPIC_DONE (0x30 << GS_UKERNEL_SHIFT) +#define GS_UKERNEL_DPC_ERROR (0x60 << GS_UKERNEL_SHIFT) +#define GS_UKERNEL_READY (0xF0 << GS_UKERNEL_SHIFT) +#define GS_MIA_SHIFT 16 +#define GS_MIA_MASK (0x07 << GS_MIA_SHIFT) + +#define GUC_WOPCM_SIZE 0xc050 +#define GUC_WOPCM_SIZE_VALUE (0x80 << 12) /* 512KB */ +#define GUC_WOPCM_OFFSET 0x80000 /* 512KB */ + +#define SOFT_SCRATCH(n) (0xc180 + ((n) * 4)) + +#define UOS_RSA_SCRATCH_0 0xc200 +#define DMA_ADDR_0_LOW 0xc300 +#define DMA_ADDR_0_HIGH 0xc304 +#define DMA_ADDR_1_LOW 0xc308 +#define DMA_ADDR_1_HIGH 0xc30c +#define DMA_ADDRESS_SPACE_WOPCM (7 << 16) +#define DMA_ADDRESS_SPACE_GTT (8 << 16) +#define DMA_COPY_SIZE 0xc310 +#define DMA_CTRL 0xc314 +#define UOS_MOVE (1<<4) +#define START_DMA (1<<0) +#define DMA_GUC_WOPCM_OFFSET 0xc340 + +#define GEN8_GT_PM_CONFIG 0x138140 +#define GEN9_GT_PM_CONFIG 0x13816c +#define GEN8_GT_DOORBELL_ENABLE (1<<0) + +#define GEN8_GTCR 0x4274 +#define GEN8_GTCR_INVALIDATE (1<<0) + +#define GUC_ARAT_C6DIS 0xA178 + +#define GUC_SHIM_CONTROL 0xc064 +#define GUC_DISABLE_SRAM_INIT_TO_ZEROES (1<<0) +#define GUC_ENABLE_READ_CACHE_LOGIC (1<<1) +#define GUC_ENABLE_MIA_CACHING (1<<2) +#define GUC_GEN10_MSGCH_ENABLE (1<<4) +#define GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA (1<<9) +#define GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA (1<<10) +#define GUC_ENABLE_MIA_CLOCK_GATING (1<<15) +#define GUC_GEN10_SHIM_WC_ENABLE (1<<21) + +#define GUC_SHIM_CONTROL_VALUE (GUC_DISABLE_SRAM_INIT_TO_ZEROES | \ + GUC_ENABLE_READ_CACHE_LOGIC | \ + GUC_ENABLE_MIA_CACHING | \ + GUC_ENABLE_READ_CACHE_FOR_SRAM_DATA | \ + GUC_ENABLE_READ_CACHE_FOR_WOPCM_DATA) + +#define HOST2GUC_INTERRUPT 0xc4c8 +#define HOST2GUC_TRIGGER (1<<0) + +#define DRBMISC1 0x1984 +#define DOORBELL_ENABLE (1<<0) + +#define GEN8_DRBREGL(x) (0x1000 + (x) * 8) +#define GEN8_DRB_VALID (1<<0) +#define GEN8_DRBREGU(x) (GEN8_DRBREGL(x) + 4) + +#define DE_GUCRMR 0x44054 + +#define GUC_BCS_RCS_IER 0xC550 +#define GUC_VCS2_VCS1_IER 0xC554 +#define GUC_WD_VECS_IER 0xC558 +#define GUC_PM_P24C_IER 0xC55C + +#endif diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index d87f173a0179..1118c39281f9 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1227,6 +1227,22 @@ static irqreturn_t gen8_gt_irq_handler(struct drm_i915_private *dev_priv, return ret; } +static bool bxt_port_hotplug_long_detect(enum port port, u32 val) +{ + switch (port) { + case PORT_A: + return val & BXT_PORTA_HOTPLUG_LONG_DETECT; + case PORT_B: + return val & PORTB_HOTPLUG_LONG_DETECT; + case PORT_C: + return val & PORTC_HOTPLUG_LONG_DETECT; + case PORT_D: + return val & PORTD_HOTPLUG_LONG_DETECT; + default: + return false; + } +} + static bool pch_port_hotplug_long_detect(enum port port, u32 val) { switch (port) { @@ -1256,9 +1272,10 @@ static bool i9xx_port_hotplug_long_detect(enum port port, u32 val) } /* Get a bit mask of pins that have triggered, and which ones may be long. */ -static void pch_get_hpd_pins(u32 *pin_mask, u32 *long_mask, +static void intel_get_hpd_pins(u32 *pin_mask, u32 *long_mask, u32 hotplug_trigger, u32 dig_hotplug_reg, - const u32 hpd[HPD_NUM_PINS]) + const u32 hpd[HPD_NUM_PINS], + bool long_pulse_detect(enum port port, u32 val)) { enum port port; int i; @@ -1272,8 +1289,10 @@ static void pch_get_hpd_pins(u32 *pin_mask, u32 *long_mask, *pin_mask |= BIT(i); - port = intel_hpd_pin_to_port(i); - if (pch_port_hotplug_long_detect(port, dig_hotplug_reg)) + if (!intel_hpd_pin_to_port(i, &port)) + continue; + + if (long_pulse_detect(port, dig_hotplug_reg)) *long_mask |= BIT(i); } @@ -1282,34 +1301,6 @@ static void pch_get_hpd_pins(u32 *pin_mask, u32 *long_mask, } -/* Get a bit mask of pins that have triggered, and which ones may be long. */ -static void i9xx_get_hpd_pins(u32 *pin_mask, u32 *long_mask, - u32 hotplug_trigger, const u32 hpd[HPD_NUM_PINS]) -{ - enum port port; - int i; - - *pin_mask = 0; - *long_mask = 0; - - if (!hotplug_trigger) - return; - - for_each_hpd_pin(i) { - if ((hpd[i] & hotplug_trigger) == 0) - continue; - - *pin_mask |= BIT(i); - - port = intel_hpd_pin_to_port(i); - if (i9xx_port_hotplug_long_detect(port, hotplug_trigger)) - *long_mask |= BIT(i); - } - - DRM_DEBUG_DRIVER("hotplug event received, stat 0x%08x, pins 0x%08x\n", - hotplug_trigger, *pin_mask); -} - static void gmbus_irq_handler(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -1547,7 +1538,9 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev) if (IS_G4X(dev) || IS_VALLEYVIEW(dev)) { u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X; - i9xx_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, hpd_status_g4x); + intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + hotplug_trigger, hpd_status_g4x, + i9xx_port_hotplug_long_detect); intel_hpd_irq_handler(dev, pin_mask, long_mask); if (hotplug_status & DP_AUX_CHANNEL_MASK_INT_STATUS_G4X) @@ -1555,7 +1548,9 @@ static void i9xx_hpd_irq_handler(struct drm_device *dev) } else { u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915; - i9xx_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, hpd_status_i915); + intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + hotplug_trigger, hpd_status_g4x, + i9xx_port_hotplug_long_detect); intel_hpd_irq_handler(dev, pin_mask, long_mask); } } @@ -1662,8 +1657,9 @@ static void ibx_irq_handler(struct drm_device *dev, u32 pch_iir) dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG); I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg); - pch_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, - dig_hotplug_reg, hpd_ibx); + intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + dig_hotplug_reg, hpd_ibx, + pch_port_hotplug_long_detect); intel_hpd_irq_handler(dev, pin_mask, long_mask); } @@ -1763,8 +1759,10 @@ static void cpt_irq_handler(struct drm_device *dev, u32 pch_iir) dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG); I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg); - pch_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, - dig_hotplug_reg, hpd_cpt); + + intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + dig_hotplug_reg, hpd_cpt, + pch_port_hotplug_long_detect); intel_hpd_irq_handler(dev, pin_mask, long_mask); } @@ -1981,7 +1979,8 @@ static void bxt_hpd_handler(struct drm_device *dev, uint32_t iir_status) /* Clear sticky bits in hpd status */ I915_WRITE(BXT_HOTPLUG_CTL, hp_control); - pch_get_hpd_pins(&pin_mask, &long_mask, hp_trigger, hp_control, hpd_bxt); + intel_get_hpd_pins(&pin_mask, &long_mask, hp_trigger, hp_control, + hpd_bxt, bxt_port_hotplug_long_detect); intel_hpd_irq_handler(dev, pin_mask, long_mask); } diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index 5f4e7295295f..5ae4b0aba564 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers/gpu/drm/i915/i915_params.c @@ -52,6 +52,8 @@ struct i915_params i915 __read_mostly = { .mmio_debug = 0, .verbose_state_checks = 1, .edp_vswing = 0, + .enable_guc_submission = false, + .guc_log_level = -1, }; module_param_named(modeset, i915.modeset, int, 0400); @@ -181,3 +183,10 @@ MODULE_PARM_DESC(edp_vswing, "Ignore/Override vswing pre-emph table selection from VBT " "(0=use value from vbt [default], 1=low power swing(200mV)," "2=default swing(400mV))"); + +module_param_named_unsafe(enable_guc_submission, i915.enable_guc_submission, bool, 0400); +MODULE_PARM_DESC(enable_guc_submission, "Enable GuC submission (default:false)"); + +module_param_named(guc_log_level, i915.guc_log_level, int, 0400); +MODULE_PARM_DESC(guc_log_level, + "GuC firmware logging level (-1:disabled (default), 0-3:enabled)"); diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index e9a95df639f0..0f67f3da0762 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -5985,6 +5985,11 @@ enum skl_disp_power_wells { /* digital port hotplug */ #define PCH_PORT_HOTPLUG 0xc4030 /* SHOTPLUG_CTL */ +#define BXT_PORTA_HOTPLUG_ENABLE (1 << 28) +#define BXT_PORTA_HOTPLUG_STATUS_MASK (0x3 << 24) +#define BXT_PORTA_HOTPLUG_NO_DETECT (0 << 24) +#define BXT_PORTA_HOTPLUG_SHORT_DETECT (1 << 24) +#define BXT_PORTA_HOTPLUG_LONG_DETECT (2 << 24) #define PORTD_HOTPLUG_ENABLE (1 << 20) #define PORTD_PULSE_DURATION_2ms (0) #define PORTD_PULSE_DURATION_4_5ms (1 << 18) @@ -6846,6 +6851,9 @@ enum skl_disp_power_wells { #define GEN7_MISCCPCTL (0x9424) #define GEN7_DOP_CLOCK_GATE_ENABLE (1<<0) +#define GEN8_GARBCNTL 0xB004 +#define GEN9_GAPS_TSV_CREDIT_DISABLE (1<<7) + /* IVYBRIDGE DPF */ #define GEN7_L3CDERRST1 0xB008 /* L3CD Error Status 1 */ #define HSW_L3CDERRST11 0xB208 /* L3CD Error Status register 1 slice 1 */ diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 6d8a7bf06dfc..ba1ae031e6fd 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -244,7 +244,7 @@ void intel_csr_load_status_set(struct drm_i915_private *dev_priv, void intel_csr_load_program(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; - __be32 *payload = dev_priv->csr.dmc_payload; + u32 *payload = dev_priv->csr.dmc_payload; uint32_t i, fw_size; if (!IS_GEN9(dev)) { @@ -256,7 +256,7 @@ void intel_csr_load_program(struct drm_device *dev) fw_size = dev_priv->csr.dmc_fw_size; for (i = 0; i < fw_size; i++) I915_WRITE(CSR_PROGRAM_BASE + i * 4, - (u32 __force)payload[i]); + payload[i]); for (i = 0; i < dev_priv->csr.mmio_count; i++) { I915_WRITE(dev_priv->csr.mmioaddr[i], @@ -279,7 +279,7 @@ static void finish_csr_load(const struct firmware *fw, void *context) char substepping = intel_get_substepping(dev); uint32_t dmc_offset = CSR_DEFAULT_FW_OFFSET, readcount = 0, nbytes; uint32_t i; - __be32 *dmc_payload; + uint32_t *dmc_payload; bool fw_loaded = false; if (!fw) { @@ -375,15 +375,7 @@ static void finish_csr_load(const struct firmware *fw, void *context) } dmc_payload = csr->dmc_payload; - for (i = 0; i < dmc_header->fw_size; i++) { - uint32_t *tmp = (u32 *)&fw->data[readcount + i * 4]; - /* - * The firmware payload is an array of 32 bit words stored in - * little-endian format in the firmware image and programmed - * as 32 bit big-endian format to memory. - */ - dmc_payload[i] = cpu_to_be32(*tmp); - } + memcpy(dmc_payload, &fw->data[readcount], nbytes); /* load csr program during system boot, as needed for DC states */ intel_csr_load_program(dev); diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 43b0f17ad1fa..a76751633af6 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1936,7 +1936,9 @@ static void intel_disable_shared_dpll(struct intel_crtc *crtc) struct intel_shared_dpll *pll = intel_crtc_to_shared_dpll(crtc); /* PCH only available on ILK+ */ - BUG_ON(INTEL_INFO(dev)->gen < 5); + if (INTEL_INFO(dev)->gen < 5) + return; + if (pll == NULL) return; @@ -10764,15 +10766,12 @@ static void intel_unpin_work_fn(struct work_struct *__work) container_of(__work, struct intel_unpin_work, work); struct intel_crtc *crtc = to_intel_crtc(work->crtc); struct drm_device *dev = crtc->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; struct drm_plane *primary = crtc->base.primary; mutex_lock(&dev->struct_mutex); intel_unpin_fb_obj(work->old_fb, primary->state); drm_gem_object_unreference(&work->pending_flip_obj->base); - intel_fbc_update(dev_priv); - if (work->flip_queued_req) i915_gem_request_assign(&work->flip_queued_req, NULL); mutex_unlock(&dev->struct_mutex); @@ -11544,7 +11543,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, to_intel_plane(primary)->frontbuffer_bit); mutex_unlock(&dev->struct_mutex); - intel_fbc_disable(dev_priv); + intel_fbc_disable_crtc(intel_crtc); intel_frontbuffer_flip_prepare(dev, to_intel_plane(primary)->frontbuffer_bit); @@ -14271,7 +14270,7 @@ static int intel_user_framebuffer_dirty(struct drm_framebuffer *fb, struct drm_i915_gem_object *obj = intel_fb->obj; mutex_lock(&dev->struct_mutex); - intel_fb_obj_flush(obj, false, ORIGIN_GTT); + intel_fb_obj_flush(obj, false, ORIGIN_DIRTYFB); mutex_unlock(&dev->struct_mutex); return 0; diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index f1b9f939b435..df7e2cfef38d 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -1409,7 +1409,10 @@ intel_dp_compute_config(struct intel_encoder *encoder, * bpc in between. */ bpp = pipe_config->pipe_bpp; if (is_edp(intel_dp)) { - if (dev_priv->vbt.edp_bpp && dev_priv->vbt.edp_bpp < bpp) { + + /* Get bpp from vbt only for panels that dont have bpp in edid */ + if (intel_connector->base.display_info.bpc == 0 && + (dev_priv->vbt.edp_bpp && dev_priv->vbt.edp_bpp < bpp)) { DRM_DEBUG_KMS("clamping bpp for eDP panel to BIOS-provided %i\n", dev_priv->vbt.edp_bpp); bpp = dev_priv->vbt.edp_bpp; @@ -3958,43 +3961,67 @@ intel_dp_probe_mst(struct intel_dp *intel_dp) return intel_dp->is_mst; } -int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc) +static void intel_dp_sink_crc_stop(struct intel_dp *intel_dp) { - struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp); - struct drm_device *dev = intel_dig_port->base.base.dev; - struct intel_crtc *intel_crtc = - to_intel_crtc(intel_dig_port->base.base.crtc); + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc); u8 buf; - int test_crc_count; - int attempts = 6; - int ret = 0; - hsw_disable_ips(intel_crtc); - - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) { - ret = -EIO; - goto out; + if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) { + DRM_DEBUG_KMS("Sink CRC couldn't be stopped properly\n"); + return; } - if (!(buf & DP_TEST_CRC_SUPPORTED)) { - ret = -ENOTTY; - goto out; - } + if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK, + buf & ~DP_TEST_SINK_START) < 0) + DRM_DEBUG_KMS("Sink CRC couldn't be stopped properly\n"); - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) { - ret = -EIO; - goto out; - } + hsw_enable_ips(intel_crtc); +} + +static int intel_dp_sink_crc_start(struct intel_dp *intel_dp) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc); + u8 buf; + + if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) + return -EIO; + + if (!(buf & DP_TEST_CRC_SUPPORTED)) + return -ENOTTY; + + if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) + return -EIO; + + hsw_disable_ips(intel_crtc); if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK, - buf | DP_TEST_SINK_START) < 0) { - ret = -EIO; - goto out; + buf | DP_TEST_SINK_START) < 0) { + hsw_enable_ips(intel_crtc); + return -EIO; } + return 0; +} + +int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct drm_device *dev = dig_port->base.base.dev; + struct intel_crtc *intel_crtc = to_intel_crtc(dig_port->base.base.crtc); + u8 buf; + int test_crc_count; + int attempts = 6; + int ret; + + ret = intel_dp_sink_crc_start(intel_dp); + if (ret) + return ret; + if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) { ret = -EIO; - goto out; + goto stop; } test_crc_count = buf & DP_TEST_COUNT_MASK; @@ -4003,7 +4030,7 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc) if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK_MISC, &buf) < 0) { ret = -EIO; - goto out; + goto stop; } intel_wait_for_vblank(dev, intel_crtc->pipe); } while (--attempts && (buf & DP_TEST_COUNT_MASK) == test_crc_count); @@ -4011,25 +4038,13 @@ int intel_dp_sink_crc(struct intel_dp *intel_dp, u8 *crc) if (attempts == 0) { DRM_DEBUG_KMS("Panel is unable to calculate CRC after 6 vblanks\n"); ret = -ETIMEDOUT; - goto out; - } - - if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0) { - ret = -EIO; - goto out; + goto stop; } - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_TEST_SINK, &buf) < 0) { + if (drm_dp_dpcd_read(&intel_dp->aux, DP_TEST_CRC_R_CR, crc, 6) < 0) ret = -EIO; - goto out; - } - if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_TEST_SINK, - buf & ~DP_TEST_SINK_START) < 0) { - ret = -EIO; - goto out; - } -out: - hsw_enable_ips(intel_crtc); +stop: + intel_dp_sink_crc_stop(intel_dp); return ret; } diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 320c9e6bd848..34ad042f18b5 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -1243,7 +1243,7 @@ void intel_fbc_invalidate(struct drm_i915_private *dev_priv, unsigned int frontbuffer_bits, enum fb_op_origin origin); void intel_fbc_flush(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits); + unsigned int frontbuffer_bits, enum fb_op_origin origin); const char *intel_no_fbc_reason_str(enum no_fbc_reason reason); void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c index c271af767981..1f97fb548c2a 100644 --- a/drivers/gpu/drm/i915/intel_fbc.c +++ b/drivers/gpu/drm/i915/intel_fbc.c @@ -884,22 +884,23 @@ void intel_fbc_invalidate(struct drm_i915_private *dev_priv, } void intel_fbc_flush(struct drm_i915_private *dev_priv, - unsigned int frontbuffer_bits) + unsigned int frontbuffer_bits, enum fb_op_origin origin) { if (!dev_priv->fbc.enable_fbc) return; - mutex_lock(&dev_priv->fbc.lock); + if (origin == ORIGIN_GTT) + return; - if (!dev_priv->fbc.busy_bits) - goto out; + mutex_lock(&dev_priv->fbc.lock); dev_priv->fbc.busy_bits &= ~frontbuffer_bits; - if (!dev_priv->fbc.busy_bits) + if (!dev_priv->fbc.busy_bits) { + __intel_fbc_disable(dev_priv); __intel_fbc_update(dev_priv); + } -out: mutex_unlock(&dev_priv->fbc.lock); } diff --git a/drivers/gpu/drm/i915/intel_frontbuffer.c b/drivers/gpu/drm/i915/intel_frontbuffer.c index 777b1d3ccd41..ac85357010b4 100644 --- a/drivers/gpu/drm/i915/intel_frontbuffer.c +++ b/drivers/gpu/drm/i915/intel_frontbuffer.c @@ -129,7 +129,7 @@ static void intel_frontbuffer_flush(struct drm_device *dev, intel_edp_drrs_flush(dev, frontbuffer_bits); intel_psr_flush(dev, frontbuffer_bits, origin); - intel_fbc_flush(dev_priv, frontbuffer_bits); + intel_fbc_flush(dev_priv, frontbuffer_bits, origin); } /** diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h new file mode 100644 index 000000000000..18d7f20936c8 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_guc_fwif.h @@ -0,0 +1,245 @@ +/* + * Copyright © 2014 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ +#ifndef _INTEL_GUC_FWIF_H +#define _INTEL_GUC_FWIF_H + +/* + * This file is partially autogenerated, although currently with some manual + * fixups afterwards. In future, it should be entirely autogenerated, in order + * to ensure that the definitions herein remain in sync with those used by the + * GuC's own firmware. + * + * EDITING THIS FILE IS THEREFORE NOT RECOMMENDED - YOUR CHANGES MAY BE LOST. + */ + +#define GFXCORE_FAMILY_GEN8 11 +#define GFXCORE_FAMILY_GEN9 12 +#define GFXCORE_FAMILY_FORCE_ULONG 0x7fffffff + +#define GUC_CTX_PRIORITY_CRITICAL 0 +#define GUC_CTX_PRIORITY_HIGH 1 +#define GUC_CTX_PRIORITY_NORMAL 2 +#define GUC_CTX_PRIORITY_LOW 3 + +#define GUC_MAX_GPU_CONTEXTS 1024 +#define GUC_INVALID_CTX_ID (GUC_MAX_GPU_CONTEXTS + 1) + +/* Work queue item header definitions */ +#define WQ_STATUS_ACTIVE 1 +#define WQ_STATUS_SUSPENDED 2 +#define WQ_STATUS_CMD_ERROR 3 +#define WQ_STATUS_ENGINE_ID_NOT_USED 4 +#define WQ_STATUS_SUSPENDED_FROM_RESET 5 +#define WQ_TYPE_SHIFT 0 +#define WQ_TYPE_BATCH_BUF (0x1 << WQ_TYPE_SHIFT) +#define WQ_TYPE_PSEUDO (0x2 << WQ_TYPE_SHIFT) +#define WQ_TYPE_INORDER (0x3 << WQ_TYPE_SHIFT) +#define WQ_TARGET_SHIFT 10 +#define WQ_LEN_SHIFT 16 +#define WQ_NO_WCFLUSH_WAIT (1 << 27) +#define WQ_PRESENT_WORKLOAD (1 << 28) +#define WQ_WORKLOAD_SHIFT 29 +#define WQ_WORKLOAD_GENERAL (0 << WQ_WORKLOAD_SHIFT) +#define WQ_WORKLOAD_GPGPU (1 << WQ_WORKLOAD_SHIFT) +#define WQ_WORKLOAD_TOUCH (2 << WQ_WORKLOAD_SHIFT) + +#define WQ_RING_TAIL_SHIFT 20 +#define WQ_RING_TAIL_MASK (0x7FF << WQ_RING_TAIL_SHIFT) + +#define GUC_DOORBELL_ENABLED 1 +#define GUC_DOORBELL_DISABLED 0 + +#define GUC_CTX_DESC_ATTR_ACTIVE (1 << 0) +#define GUC_CTX_DESC_ATTR_PENDING_DB (1 << 1) +#define GUC_CTX_DESC_ATTR_KERNEL (1 << 2) +#define GUC_CTX_DESC_ATTR_PREEMPT (1 << 3) +#define GUC_CTX_DESC_ATTR_RESET (1 << 4) +#define GUC_CTX_DESC_ATTR_WQLOCKED (1 << 5) +#define GUC_CTX_DESC_ATTR_PCH (1 << 6) + +/* The guc control data is 10 DWORDs */ +#define GUC_CTL_CTXINFO 0 +#define GUC_CTL_CTXNUM_IN16_SHIFT 0 +#define GUC_CTL_BASE_ADDR_SHIFT 12 +#define GUC_CTL_ARAT_HIGH 1 +#define GUC_CTL_ARAT_LOW 2 +#define GUC_CTL_DEVICE_INFO 3 +#define GUC_CTL_GTTYPE_SHIFT 0 +#define GUC_CTL_COREFAMILY_SHIFT 7 +#define GUC_CTL_LOG_PARAMS 4 +#define GUC_LOG_VALID (1 << 0) +#define GUC_LOG_NOTIFY_ON_HALF_FULL (1 << 1) +#define GUC_LOG_ALLOC_IN_MEGABYTE (1 << 3) +#define GUC_LOG_CRASH_PAGES 1 +#define GUC_LOG_CRASH_SHIFT 4 +#define GUC_LOG_DPC_PAGES 3 +#define GUC_LOG_DPC_SHIFT 6 +#define GUC_LOG_ISR_PAGES 3 +#define GUC_LOG_ISR_SHIFT 9 +#define GUC_LOG_BUF_ADDR_SHIFT 12 +#define GUC_CTL_PAGE_FAULT_CONTROL 5 +#define GUC_CTL_WA 6 +#define GUC_CTL_WA_UK_BY_DRIVER (1 << 3) +#define GUC_CTL_FEATURE 7 +#define GUC_CTL_VCS2_ENABLED (1 << 0) +#define GUC_CTL_KERNEL_SUBMISSIONS (1 << 1) +#define GUC_CTL_FEATURE2 (1 << 2) +#define GUC_CTL_POWER_GATING (1 << 3) +#define GUC_CTL_DISABLE_SCHEDULER (1 << 4) +#define GUC_CTL_PREEMPTION_LOG (1 << 5) +#define GUC_CTL_ENABLE_SLPC (1 << 7) +#define GUC_CTL_DEBUG 8 +#define GUC_LOG_VERBOSITY_SHIFT 0 +#define GUC_LOG_VERBOSITY_LOW (0 << GUC_LOG_VERBOSITY_SHIFT) +#define GUC_LOG_VERBOSITY_MED (1 << GUC_LOG_VERBOSITY_SHIFT) +#define GUC_LOG_VERBOSITY_HIGH (2 << GUC_LOG_VERBOSITY_SHIFT) +#define GUC_LOG_VERBOSITY_ULTRA (3 << GUC_LOG_VERBOSITY_SHIFT) +/* Verbosity range-check limits, without the shift */ +#define GUC_LOG_VERBOSITY_MIN 0 +#define GUC_LOG_VERBOSITY_MAX 3 + +#define GUC_CTL_MAX_DWORDS (GUC_CTL_DEBUG + 1) + +struct guc_doorbell_info { + u32 db_status; + u32 cookie; + u32 reserved[14]; +} __packed; + +union guc_doorbell_qw { + struct { + u32 db_status; + u32 cookie; + }; + u64 value_qw; +} __packed; + +#define GUC_MAX_DOORBELLS 256 +#define GUC_INVALID_DOORBELL_ID (GUC_MAX_DOORBELLS) + +#define GUC_DB_SIZE (PAGE_SIZE) +#define GUC_WQ_SIZE (PAGE_SIZE * 2) + +/* Work item for submitting workloads into work queue of GuC. */ +struct guc_wq_item { + u32 header; + u32 context_desc; + u32 ring_tail; + u32 fence_id; +} __packed; + +struct guc_process_desc { + u32 context_id; + u64 db_base_addr; + u32 head; + u32 tail; + u32 error_offset; + u64 wq_base_addr; + u32 wq_size_bytes; + u32 wq_status; + u32 engine_presence; + u32 priority; + u32 reserved[30]; +} __packed; + +/* engine id and context id is packed into guc_execlist_context.context_id*/ +#define GUC_ELC_CTXID_OFFSET 0 +#define GUC_ELC_ENGINE_OFFSET 29 + +/* The execlist context including software and HW information */ +struct guc_execlist_context { + u32 context_desc; + u32 context_id; + u32 ring_status; + u32 ring_lcra; + u32 ring_begin; + u32 ring_end; + u32 ring_next_free_location; + u32 ring_current_tail_pointer_value; + u8 engine_state_submit_value; + u8 engine_state_wait_value; + u16 pagefault_count; + u16 engine_submit_queue_count; +} __packed; + +/*Context descriptor for communicating between uKernel and Driver*/ +struct guc_context_desc { + u32 sched_common_area; + u32 context_id; + u32 pas_id; + u8 engines_used; + u64 db_trigger_cpu; + u32 db_trigger_uk; + u64 db_trigger_phy; + u16 db_id; + + struct guc_execlist_context lrc[I915_NUM_RINGS]; + + u8 attribute; + + u32 priority; + + u32 wq_sampled_tail_offset; + u32 wq_total_submit_enqueues; + + u32 process_desc; + u32 wq_addr; + u32 wq_size; + + u32 engine_presence; + + u32 reserved0[1]; + u64 reserved1[1]; + + u64 desc_private; +} __packed; + +/* This Action will be programmed in C180 - SOFT_SCRATCH_O_REG */ +enum host2guc_action { + HOST2GUC_ACTION_DEFAULT = 0x0, + HOST2GUC_ACTION_SAMPLE_FORCEWAKE = 0x6, + HOST2GUC_ACTION_ALLOCATE_DOORBELL = 0x10, + HOST2GUC_ACTION_DEALLOCATE_DOORBELL = 0x20, + HOST2GUC_ACTION_SLPC_REQUEST = 0x3003, + HOST2GUC_ACTION_LIMIT +}; + +/* + * The GuC sends its response to a command by overwriting the + * command in SS0. The response is distinguishable from a command + * by the fact that all the MASK bits are set. The remaining bits + * give more detail. + */ +#define GUC2HOST_RESPONSE_MASK ((u32)0xF0000000) +#define GUC2HOST_IS_RESPONSE(x) ((u32)(x) >= GUC2HOST_RESPONSE_MASK) +#define GUC2HOST_STATUS(x) (GUC2HOST_RESPONSE_MASK | (x)) + +/* GUC will return status back to SOFT_SCRATCH_O_REG */ +enum guc2host_status { + GUC2HOST_STATUS_SUCCESS = GUC2HOST_STATUS(0x0), + GUC2HOST_STATUS_ALLOCATE_DOORBELL_FAIL = GUC2HOST_STATUS(0x10), + GUC2HOST_STATUS_DEALLOCATE_DOORBELL_FAIL = GUC2HOST_STATUS(0x20), + GUC2HOST_STATUS_GENERIC_FAIL = GUC2HOST_STATUS(0x0000F000) +}; + +#endif diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index 3c9171f11531..032a0bf75f3b 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -76,17 +76,23 @@ * it will use i915_hotplug_work_func where this logic is handled. */ -enum port intel_hpd_pin_to_port(enum hpd_pin pin) +bool intel_hpd_pin_to_port(enum hpd_pin pin, enum port *port) { switch (pin) { + case HPD_PORT_A: + *port = PORT_A; + return true; case HPD_PORT_B: - return PORT_B; + *port = PORT_B; + return true; case HPD_PORT_C: - return PORT_C; + *port = PORT_C; + return true; case HPD_PORT_D: - return PORT_D; + *port = PORT_D; + return true; default: - return PORT_A; /* no hpd */ + return false; /* no hpd */ } } @@ -369,8 +375,8 @@ void intel_hpd_irq_handler(struct drm_device *dev, if (!(BIT(i) & pin_mask)) continue; - port = intel_hpd_pin_to_port(i); - is_dig_port = port && dev_priv->hotplug.irq_port[port]; + is_dig_port = intel_hpd_pin_to_port(i, &port) && + dev_priv->hotplug.irq_port[port]; if (is_dig_port) { bool long_hpd = long_mask & BIT(i); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 9faad82c42ec..99bba8ece464 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1740,6 +1740,12 @@ static int intel_lr_context_render_state_init(struct drm_i915_gem_request *req) if (ret) goto out; + ret = req->ring->emit_bb_start(req, + (so.ggtt_offset + so.aux_batch_offset), + I915_DISPATCH_SECURE); + if (ret) + goto out; + i915_vma_move_to_active(i915_gem_obj_to_ggtt(so.obj), req); out: diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 5004c4a46a9e..fff0c22682ee 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -102,6 +102,12 @@ static void skl_init_clock_gating(struct drm_device *dev) /* WaDisableLSQCROPERFforOCL:skl */ I915_WRITE(GEN8_L3SQCREG4, I915_READ(GEN8_L3SQCREG4) | GEN8_LQSC_RO_PERF_DIS); + + /* WaEnableGapsTsvCreditFix:skl */ + if (IS_SKYLAKE(dev) && (INTEL_REVID(dev) >= SKL_REVID_C0)) { + I915_WRITE(GEN8_GARBCNTL, (I915_READ(GEN8_GARBCNTL) | + GEN9_GAPS_TSV_CREDIT_DISABLE)); + } } static void bxt_init_clock_gating(struct drm_device *dev) @@ -4266,7 +4272,7 @@ static void ironlake_enable_drps(struct drm_device *dev) if (wait_for_atomic((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10)) DRM_ERROR("stuck trying to change perf mode\n"); - msleep(1); + mdelay(1); ironlake_set_drps(dev, fstart); @@ -4297,10 +4303,10 @@ static void ironlake_disable_drps(struct drm_device *dev) /* Go back to the starting frequency */ ironlake_set_drps(dev, dev_priv->ips.fstart); - msleep(1); + mdelay(1); rgvswctl |= MEMCTL_CMD_STS; I915_WRITE(MEMSWCTL, rgvswctl); - msleep(1); + mdelay(1); spin_unlock_irq(&mchdev_lock); } diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c index acd8ec859f71..a04b4dc5ed9b 100644 --- a/drivers/gpu/drm/i915/intel_psr.c +++ b/drivers/gpu/drm/i915/intel_psr.c @@ -698,6 +698,7 @@ void intel_psr_flush(struct drm_device *dev, struct drm_i915_private *dev_priv = dev->dev_private; struct drm_crtc *crtc; enum pipe pipe; + int delay_ms = HAS_DDI(dev) ? 100 : 500; mutex_lock(&dev_priv->psr.lock); if (!dev_priv->psr.enabled) { @@ -733,7 +734,7 @@ void intel_psr_flush(struct drm_device *dev, if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits) schedule_delayed_work(&dev_priv->psr.work, - msecs_to_jiffies(100)); + msecs_to_jiffies(delay_ms)); mutex_unlock(&dev_priv->psr.lock); } diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 177f7ed16cf0..1c14233d179f 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -1041,13 +1041,6 @@ static int skl_init_workarounds(struct intel_engine_cs *ring) WA_SET_BIT_MASKED(HIZ_CHICKEN, BDW_HIZ_POWER_COMPILER_CLOCK_GATING_DISABLE); - if (INTEL_REVID(dev) == SKL_REVID_C0 || - INTEL_REVID(dev) == SKL_REVID_D0) - /* WaBarrierPerformanceFixDisable:skl */ - WA_SET_BIT_MASKED(HDC_CHICKEN0, - HDC_FENCE_DEST_SLM_DISABLE | - HDC_BARRIER_PERFORMANCE_DISABLE); - if (INTEL_REVID(dev) <= SKL_REVID_D0) { /* *Use Force Non-Coherent whenever executing a 3D context. This diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 6393b76f87ff..821644d1b544 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -68,6 +68,22 @@ bool intel_display_power_well_is_enabled(struct drm_i915_private *dev_priv, int power_well_id); +static void intel_power_well_enable(struct drm_i915_private *dev_priv, + struct i915_power_well *power_well) +{ + DRM_DEBUG_KMS("enabling %s\n", power_well->name); + power_well->ops->enable(dev_priv, power_well); + power_well->hw_enabled = true; +} + +static void intel_power_well_disable(struct drm_i915_private *dev_priv, + struct i915_power_well *power_well) +{ + DRM_DEBUG_KMS("disabling %s\n", power_well->name); + power_well->hw_enabled = false; + power_well->ops->disable(dev_priv, power_well); +} + /* * We should only use the power well if we explicitly asked the hardware to * enable it, so check if it's enabled and also check if we've requested it to @@ -1104,11 +1120,8 @@ void intel_display_power_get(struct drm_i915_private *dev_priv, mutex_lock(&power_domains->lock); for_each_power_well(i, power_well, BIT(domain), power_domains) { - if (!power_well->count++) { - DRM_DEBUG_KMS("enabling %s\n", power_well->name); - power_well->ops->enable(dev_priv, power_well); - power_well->hw_enabled = true; - } + if (!power_well->count++) + intel_power_well_enable(dev_priv, power_well); } power_domains->domain_use_count[domain]++; @@ -1142,11 +1155,8 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, for_each_power_well_rev(i, power_well, BIT(domain), power_domains) { WARN_ON(!power_well->count); - if (!--power_well->count && i915.disable_power_well) { - DRM_DEBUG_KMS("disabling %s\n", power_well->name); - power_well->hw_enabled = false; - power_well->ops->disable(dev_priv, power_well); - } + if (!--power_well->count && i915.disable_power_well) + intel_power_well_disable(dev_priv, power_well); } mutex_unlock(&power_domains->lock); |