Merge tag 'drm-fixes-2025-11-15' of https://gitlab.freedesktop.org/drm/kernel

torvalds · torvalds · commit d5c1b4b43249 · 2025-11-14T13:39:15.000-08:00
Pull drm fixes from Dave Airlie: "Weekly fixes, amdgpu and vmwgfx making up the most of it, along with panthor and i915/xe. Seems about right for this time of development, nothing major outstanding. client: - Fix description of module parameter panthor: - Flush writes before mapping buffers vmwgfx: - Improve command validation - Improve ref counting - Fix cursor-plane support amdgpu: - Disallow P2P DMA for GC 12 DCC surfaces - ctx error handling fix - UserQ fixes - VRR fix - ISP fix - JPEG 5.0.1 fix amdkfd: - Save area check fix - Fix GPU mappings for APU after prefetch i915: - Fix PSR's pipe to vblank conversion - Disable Panel Replay on MST links xe: - New HW workarounds affecting PTL and WCL platforms * tag 'drm-fixes-2025-11-15' of https://gitlab.freedesktop.org/drm/kernel: drm/client: fix MODULE_PARM_DESC string for "active" drm/i915/dp_mst: Disable Panel Replay drm/amdkfd: Fix GPU mappings for APU after prefetch drm/amdkfd: relax checks for over allocation of save area drm/amdgpu/jpeg: Add parse_cs for JPEG5_0_1 drm/amd/amdgpu: Ensure isp_kernel_buffer_alloc() creates a new BO drm/amd/display: Allow VRR params change if unsynced with the stream drm/amdgpu: fix lock warning in amdgpu_userq_fence_driver_process drm/amdgpu: jump to the correct label on failure drm/amdgpu: disable peer-to-peer access for DCC-enabled GC12 VRAM surfaces drm/xe/xe3lpg: Extend Wa_15016589081 for xe3lpg drm/xe/xe3: Extend wa_14023061436 drm/xe/xe3: Add WA_14024681466 for Xe3_LPG drm/i915/psr: fix pipe to vblank conversion drm/panthor: Flush shmem writes before mapping buffers CPU-uncached drm/vmwgfx: Restore Guest-Backed only cursor plane support drm/vmwgfx: Use kref in vmw_bo_dirty drm/vmwgfx: Validate command header size against SVGA_CMD_MAX_DATASIZE
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c
@@ -236,7 +236,7 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,
 		r = amdgpu_xcp_select_scheds(adev, hw_ip, hw_prio, fpriv,
 						&num_scheds, &scheds);
 		if (r)
-			goto cleanup_entity;
+			goto error_free_entity;
 	}
 
 	/* disable load balance if the hw engine retains context among dependent jobs */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -82,6 +82,18 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
 	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 
+	/*
+	 * Disable peer-to-peer access for DCC-enabled VRAM surfaces on GFX12+.
+	 * Such buffers cannot be safely accessed over P2P due to device-local
+	 * compression metadata. Fallback to system-memory path instead.
+	 * Device supports GFX12 (GC 12.x or newer)
+	 * BO was created with the AMDGPU_GEM_CREATE_GFX12_DCC flag
+	 *
+	 */
+	if (amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(12, 0, 0) &&
+	    bo->flags & AMDGPU_GEM_CREATE_GFX12_DCC)
+		attach->peer2peer = false;
+
 	if (!amdgpu_dmabuf_is_xgmi_accessible(attach_adev, bo) &&
 	    pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0)
 		attach->peer2peer = false;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c
@@ -280,6 +280,8 @@ int isp_kernel_buffer_alloc(struct device *dev, u64 size,
 	if (ret)
 		return ret;
 
+	/* Ensure *bo is NULL so a new BO will be created */
+	*bo = NULL;
 	ret = amdgpu_bo_create_kernel(adev,
 				      size,
 				      ISP_MC_ADDR_ALIGN,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
@@ -151,15 +151,16 @@ void amdgpu_userq_fence_driver_process(struct amdgpu_userq_fence_driver *fence_d
 {
 	struct amdgpu_userq_fence *userq_fence, *tmp;
 	struct dma_fence *fence;
+	unsigned long flags;
 	u64 rptr;
 	int i;
 
 	if (!fence_drv)
 		return;
 
+	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
 	rptr = amdgpu_userq_fence_read(fence_drv);
 
-	spin_lock(&fence_drv->fence_list_lock);
 	list_for_each_entry_safe(userq_fence, tmp, &fence_drv->fences, link) {
 		fence = &userq_fence->base;
 
@@ -174,7 +175,7 @@ void amdgpu_userq_fence_driver_process(struct amdgpu_userq_fence_driver *fence_d
 		list_del(&userq_fence->link);
 		dma_fence_put(fence);
 	}
-	spin_unlock(&fence_drv->fence_list_lock);
+	spin_unlock_irqrestore(&fence_drv->fence_list_lock, flags);
 }
 
 void amdgpu_userq_fence_driver_destroy(struct kref *ref)
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -878,6 +878,7 @@ static const struct amdgpu_ring_funcs jpeg_v5_0_1_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v5_0_1_dec_ring_get_rptr,
 	.get_wptr = jpeg_v5_0_1_dec_ring_get_wptr,
 	.set_wptr = jpeg_v5_0_1_dec_ring_set_wptr,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_queue.c
@@ -297,16 +297,16 @@ int kfd_queue_acquire_buffers(struct kfd_process_device *pdd, struct queue_prope
 		goto out_err_unreserve;
 	}
 
-	if (properties->ctx_save_restore_area_size != topo_dev->node_props.cwsr_size) {
-		pr_debug("queue cwsr size 0x%x not equal to node cwsr size 0x%x\n",
+	if (properties->ctx_save_restore_area_size < topo_dev->node_props.cwsr_size) {
+		pr_debug("queue cwsr size 0x%x not sufficient for node cwsr size 0x%x\n",
 			properties->ctx_save_restore_area_size,
 			topo_dev->node_props.cwsr_size);
 		err = -EINVAL;
 		goto out_err_unreserve;
 	}
 
-	total_cwsr_size = (topo_dev->node_props.cwsr_size + topo_dev->node_props.debug_memory_size)
-			  * NUM_XCC(pdd->dev->xcc_mask);
+	total_cwsr_size = (properties->ctx_save_restore_area_size +
+			   topo_dev->node_props.debug_memory_size) * NUM_XCC(pdd->dev->xcc_mask);
 	total_cwsr_size = ALIGN(total_cwsr_size, PAGE_SIZE);
 
 	err = kfd_queue_buffer_get(vm, (void *)properties->ctx_save_restore_area_address,
@@ -352,8 +352,8 @@ int kfd_queue_release_buffers(struct kfd_process_device *pdd, struct queue_prope
 	topo_dev = kfd_topology_device_by_id(pdd->dev->id);
 	if (!topo_dev)
 		return -EINVAL;
-	total_cwsr_size = (topo_dev->node_props.cwsr_size + topo_dev->node_props.debug_memory_size)
-			  * NUM_XCC(pdd->dev->xcc_mask);
+	total_cwsr_size = (properties->ctx_save_restore_area_size +
+			   topo_dev->node_props.debug_memory_size) * NUM_XCC(pdd->dev->xcc_mask);
 	total_cwsr_size = ALIGN(total_cwsr_size, PAGE_SIZE);
 
 	kfd_queue_buffer_svm_put(pdd, properties->ctx_save_restore_area_address, total_cwsr_size);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -3687,6 +3687,8 @@ svm_range_set_attr(struct kfd_process *p, struct mm_struct *mm,
 		svm_range_apply_attrs(p, prange, nattr, attrs, &update_mapping);
 		/* TODO: unmap ranges from GPU that lost access */
 	}
+	update_mapping |= !p->xnack_enabled && !list_empty(&remap_list);
+
 	list_for_each_entry_safe(prange, next, &remove_list, update_list) {
 		pr_debug("unlink old 0x%p prange 0x%p [0x%lx 0x%lx]\n",
 			 prange->svms, prange, prange->start,
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -1260,6 +1260,17 @@ void mod_freesync_handle_v_update(struct mod_freesync *mod_freesync,
 		update_v_total_for_static_ramp(
 				core_freesync, stream, in_out_vrr);
 	}
+
+	/*
+	 * If VRR is inactive, set vtotal min and max to nominal vtotal
+	 */
+	 if (in_out_vrr->state == VRR_STATE_INACTIVE) {
+		in_out_vrr->adjust.v_total_min =
+			mod_freesync_calc_v_total_from_refresh(stream,
+				in_out_vrr->max_refresh_in_uhz);
+		in_out_vrr->adjust.v_total_max = in_out_vrr->adjust.v_total_min;
+		return;
+	}
 }
 
 unsigned long long mod_freesync_calc_nominal_field_rate(
diff --git a/drivers/gpu/drm/clients/drm_client_setup.c b/drivers/gpu/drm/clients/drm_client_setup.c
@@ -13,8 +13,8 @@
 static char drm_client_default[16] = CONFIG_DRM_CLIENT_DEFAULT;
 module_param_string(active, drm_client_default, sizeof(drm_client_default), 0444);
 MODULE_PARM_DESC(active,
-		 "Choose which drm client to start, default is"
-		 CONFIG_DRM_CLIENT_DEFAULT "]");
+		 "Choose which drm client to start, default is "
+		 CONFIG_DRM_CLIENT_DEFAULT);
 
 /**
  * drm_client_setup() - Setup in-kernel DRM clients
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -585,6 +585,10 @@ static void _panel_replay_init_dpcd(struct intel_dp *intel_dp)
 	struct intel_display *display = to_intel_display(intel_dp);
 	int ret;
 
+	/* TODO: Enable Panel Replay on MST once it's properly implemented. */
+	if (intel_dp->mst_detect == DRM_DP_MST)
+		return;
+
 	ret = drm_dp_dpcd_read_data(&intel_dp->aux, DP_PANEL_REPLAY_CAP_SUPPORT,
 				    &intel_dp->pr_dpcd, sizeof(intel_dp->pr_dpcd));
 	if (ret < 0)
@@ -888,7 +892,8 @@ static bool is_dc5_dc6_blocked(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 	u32 current_dc_state = intel_display_power_get_current_dc_state(display);
-	struct drm_vblank_crtc *vblank = &display->drm->vblank[intel_dp->psr.pipe];
+	struct intel_crtc *crtc = intel_crtc_for_pipe(display, intel_dp->psr.pipe);
+	struct drm_vblank_crtc *vblank = drm_crtc_vblank_crtc(&crtc->base);
 
 	return (current_dc_state != DC_STATE_EN_UPTO_DC5 &&
 		current_dc_state != DC_STATE_EN_UPTO_DC6) ||
diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c
@@ -288,6 +288,23 @@ panthor_gem_create_with_handle(struct drm_file *file,
 
 	panthor_gem_debugfs_set_usage_flags(bo, 0);
 
+	/* If this is a write-combine mapping, we query the sgt to force a CPU
+	 * cache flush (dma_map_sgtable() is called when the sgt is created).
+	 * This ensures the zero-ing is visible to any uncached mapping created
+	 * by vmap/mmap.
+	 * FIXME: Ideally this should be done when pages are allocated, not at
+	 * BO creation time.
+	 */
+	if (shmem->map_wc) {
+		struct sg_table *sgt;
+
+		sgt = drm_gem_shmem_get_pages_sgt(shmem);
+		if (IS_ERR(sgt)) {
+			ret = PTR_ERR(sgt);
+			goto out_put_gem;
+		}
+	}
+
 	/*
 	 * Allocate an id of idr table where the obj is registered
 	 * and handle has the id what user can see.
@@ -296,6 +313,7 @@ panthor_gem_create_with_handle(struct drm_file *file,
 	if (!ret)
 		*size = bo->base.base.size;
 
+out_put_gem:
 	/* drop reference from allocate - handle holds it now. */
 	drm_gem_object_put(&shmem->base);
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.c
@@ -100,8 +100,10 @@ vmw_cursor_update_type(struct vmw_private *vmw, struct vmw_plane_state *vps)
 	if (vmw->has_mob) {
 		if ((vmw->capabilities2 & SVGA_CAP2_CURSOR_MOB) != 0)
 			return VMW_CURSOR_UPDATE_MOB;
+		else
+			return VMW_CURSOR_UPDATE_GB_ONLY;
 	}
-
+	drm_warn_once(&vmw->drm, "Unknown Cursor Type!\n");
 	return VMW_CURSOR_UPDATE_NONE;
 }
 
@@ -139,6 +141,7 @@ static u32 vmw_cursor_mob_size(enum vmw_cursor_update_type update_type,
 {
 	switch (update_type) {
 	case VMW_CURSOR_UPDATE_LEGACY:
+	case VMW_CURSOR_UPDATE_GB_ONLY:
 	case VMW_CURSOR_UPDATE_NONE:
 		return 0;
 	case VMW_CURSOR_UPDATE_MOB:
@@ -623,6 +626,7 @@ int vmw_cursor_plane_prepare_fb(struct drm_plane *plane,
 		if (!surface || vps->cursor.legacy.id == surface->snooper.id)
 			vps->cursor.update_type = VMW_CURSOR_UPDATE_NONE;
 		break;
+	case VMW_CURSOR_UPDATE_GB_ONLY:
 	case VMW_CURSOR_UPDATE_MOB: {
 		bo = vmw_user_object_buffer(&vps->uo);
 		if (bo) {
@@ -737,6 +741,7 @@ void
 vmw_cursor_plane_atomic_update(struct drm_plane *plane,
 			       struct drm_atomic_state *state)
 {
+	struct vmw_bo *bo;
 	struct drm_plane_state *new_state =
 		drm_atomic_get_new_plane_state(state, plane);
 	struct drm_plane_state *old_state =
@@ -762,6 +767,15 @@ vmw_cursor_plane_atomic_update(struct drm_plane *plane,
 	case VMW_CURSOR_UPDATE_MOB:
 		vmw_cursor_update_mob(dev_priv, vps);
 		break;
+	case VMW_CURSOR_UPDATE_GB_ONLY:
+		bo = vmw_user_object_buffer(&vps->uo);
+		if (bo)
+			vmw_send_define_cursor_cmd(dev_priv, bo->map.virtual,
+						   vps->base.crtc_w,
+						   vps->base.crtc_h,
+						   vps->base.hotspot_x,
+						   vps->base.hotspot_y);
+		break;
 	case VMW_CURSOR_UPDATE_NONE:
 		/* do nothing */
 		break;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h b/drivers/gpu/drm/vmwgfx/vmwgfx_cursor_plane.h
@@ -33,6 +33,7 @@ static const u32 __maybe_unused vmw_cursor_plane_formats[] = {
 enum vmw_cursor_update_type {
 	VMW_CURSOR_UPDATE_NONE = 0,
 	VMW_CURSOR_UPDATE_LEGACY,
+	VMW_CURSOR_UPDATE_GB_ONLY,
 	VMW_CURSOR_UPDATE_MOB,
 };
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -3668,6 +3668,11 @@ static int vmw_cmd_check(struct vmw_private *dev_priv,
 
 
 	cmd_id = header->id;
+	if (header->size > SVGA_CMD_MAX_DATASIZE) {
+		VMW_DEBUG_USER("SVGA3D command: %d is too big.\n",
+			       cmd_id + SVGA_3D_CMD_BASE);
+		return -E2BIG;
+	}
 	*size = header->size + sizeof(SVGA3dCmdHeader);
 
 	cmd_id -= SVGA_3D_CMD_BASE;
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c b/drivers/gpu/drm/vmwgfx/vmwgfx_page_dirty.c
@@ -32,22 +32,22 @@ enum vmw_bo_dirty_method {
 
 /**
  * struct vmw_bo_dirty - Dirty information for buffer objects
+ * @ref_count: Reference count for this structure. Must be first member!
  * @start: First currently dirty bit
  * @end: Last currently dirty bit + 1
  * @method: The currently used dirty method
  * @change_count: Number of consecutive method change triggers
- * @ref_count: Reference count for this structure
  * @bitmap_size: The size of the bitmap in bits. Typically equal to the
  * nuber of pages in the bo.
  * @bitmap: A bitmap where each bit represents a page. A set bit means a
  * dirty page.
  */
 struct vmw_bo_dirty {
+	struct   kref ref_count;
 	unsigned long start;
 	unsigned long end;
 	enum vmw_bo_dirty_method method;
 	unsigned int change_count;
-	unsigned int ref_count;
 	unsigned long bitmap_size;
 	unsigned long bitmap[];
 };
@@ -221,7 +221,7 @@ int vmw_bo_dirty_add(struct vmw_bo *vbo)
 	int ret;
 
 	if (dirty) {
-		dirty->ref_count++;
+		kref_get(&dirty->ref_count);
 		return 0;
 	}
 
@@ -235,7 +235,7 @@ int vmw_bo_dirty_add(struct vmw_bo *vbo)
 	dirty->bitmap_size = num_pages;
 	dirty->start = dirty->bitmap_size;
 	dirty->end = 0;
-	dirty->ref_count = 1;
+	kref_init(&dirty->ref_count);
 	if (num_pages < PAGE_SIZE / sizeof(pte_t)) {
 		dirty->method = VMW_BO_DIRTY_PAGETABLE;
 	} else {
@@ -274,10 +274,8 @@ void vmw_bo_dirty_release(struct vmw_bo *vbo)
 {
 	struct vmw_bo_dirty *dirty = vbo->dirty;
 
-	if (dirty && --dirty->ref_count == 0) {
-		kvfree(dirty);
+	if (dirty && kref_put(&dirty->ref_count, (void *)kvfree))
 		vbo->dirty = NULL;
-	}
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -168,6 +168,7 @@
 
 #define XEHP_SLICE_COMMON_ECO_CHICKEN1		XE_REG_MCR(0x731c, XE_REG_OPTION_MASKED)
 #define   MSC_MSAA_REODER_BUF_BYPASS_DISABLE	REG_BIT(14)
+#define   FAST_CLEAR_VALIGN_FIX			REG_BIT(13)
 
 #define XE2LPM_CCCHKNREG1			XE_REG(0x82a8)
 
diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
@@ -679,6 +679,8 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	},
 	{ XE_RTP_NAME("14023061436"),
 	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001),
+		       FUNC(xe_rtp_match_first_render_or_compute), OR,
+		       GRAPHICS_VERSION_RANGE(3003, 3005),
 		       FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(TDL_CHICKEN, QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE))
 	},
@@ -916,6 +918,15 @@ static const struct xe_rtp_entry_sr lrc_was[] = {
 	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3003), ENGINE_CLASS(RENDER)),
 	  XE_RTP_ACTIONS(SET(COMMON_SLICE_CHICKEN4, SBE_PUSH_CONSTANT_BEHIND_FIX_ENABLE))
 	},
+	{ XE_RTP_NAME("14024681466"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3005), ENGINE_CLASS(RENDER)),
+	  XE_RTP_ACTIONS(SET(XEHP_SLICE_COMMON_ECO_CHICKEN1, FAST_CLEAR_VALIGN_FIX))
+	},
+	{ XE_RTP_NAME("15016589081"),
+	  XE_RTP_RULES(GRAPHICS_VERSION(3000), GRAPHICS_STEP(A0, B0),
+		       ENGINE_CLASS(RENDER)),
+	  XE_RTP_ACTIONS(SET(CHICKEN_RASTER_1, DIS_CLIP_NEGATIVE_BOUNDING_BOX))
+	},
 };
 
 static __maybe_unused const struct xe_rtp_entry oob_was[] = {

Original file line number	Diff line number	Diff line change
`@@ -236,7 +236,7 @@ static int amdgpu_ctx_init_entity(struct amdgpu_ctx *ctx, u32 hw_ip,`
`236`	`236`	`r = amdgpu_xcp_select_scheds(adev, hw_ip, hw_prio, fpriv,`
`237`	`237`	`&num_scheds, &scheds);`
`238`	`238`	`if (r)`
`239`		`- goto cleanup_entity;`
	`239`	`+ goto error_free_entity;`
`240`	`240`	`}`
`241`	`241`
`242`	`242`	`/* disable load balance if the hw engine retains context among dependent jobs */`
Original file line number	Diff line number	Diff line change
`@@ -3687,6 +3687,8 @@ svm_range_set_attr(struct kfd_process p, struct mm_struct mm,`
`3687`	`3687`	`svm_range_apply_attrs(p, prange, nattr, attrs, &update_mapping);`
`3688`	`3688`	`/* TODO: unmap ranges from GPU that lost access */`
`3689`	`3689`	`}`
	`3690`	`+ update_mapping \|= !p->xnack_enabled && !list_empty(&remap_list);`
	`3691`	`+`
`3690`	`3692`	`list_for_each_entry_safe(prange, next, &remove_list, update_list) {`
`3691`	`3693`	`pr_debug("unlink old 0x%p prange 0x%p [0x%lx 0x%lx]\n",`
`3692`	`3694`	`prange->svms, prange, prange->start,`