Merge tag 'v6.19-rc7' into driver-core-next

We need the driver-core fixes in here as well to build on top of.

Signed-off-by: Danilo Krummrich <dakr@kernel.org>

16 files changed, 72 insertions, 38 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 9f9774f58ce1..b20a06abb65d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -274,6 +274,8 @@ extern int amdgpu_rebar;
 extern int amdgpu_wbrf;
 extern int amdgpu_user_queue;
 
+extern uint amdgpu_hdmi_hpd_debounce_delay_ms;
+
 #define AMDGPU_VM_MAX_NUM_CTX			4096
 #define AMDGPU_SG_THRESHOLD			(256*1024*1024)
 #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index d5c44bd34d45..d2c3885de711 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5063,6 +5063,14 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev)
 
 	amdgpu_ttm_set_buffer_funcs_status(adev, false);
 
+	/*
+	 * device went through surprise hotplug; we need to destroy topology
+	 * before ip_fini_early to prevent kfd locking refcount issues by calling
+	 * amdgpu_amdkfd_suspend()
+	 */
+	if (drm_dev_is_unplugged(adev_to_drm(adev)))
+		amdgpu_amdkfd_device_fini_sw(adev);
+
 	amdgpu_device_ip_fini_early(adev);
 
 	amdgpu_irq_fini_hw(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index b5d34797d606..52bc04452812 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -1880,7 +1880,12 @@ int amdgpu_display_get_scanout_buffer(struct drm_plane *plane,
 				      struct drm_scanout_buffer *sb)
 {
 	struct amdgpu_bo *abo;
-	struct drm_framebuffer *fb = plane->state->fb;
+	struct drm_framebuffer *fb;
+
+	if (drm_drv_uses_atomic_modeset(plane->dev))
+		fb = plane->state->fb;
+	else
+		fb = plane->fb;
 
 	if (!fb)
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index e22cfa7c6d32..c1461317eb29 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -95,18 +95,6 @@ static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
 	    bo->flags & AMDGPU_GEM_CREATE_GFX12_DCC)
 		attach->peer2peer = false;
 
-	/*
-	 * 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_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 848e6b7db482..6ccb80e2d7c9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -247,6 +247,7 @@ int amdgpu_damage_clips = -1; /* auto */
 int amdgpu_umsch_mm_fwlog;
 int amdgpu_rebar = -1; /* auto */
 int amdgpu_user_queue = -1;
+uint amdgpu_hdmi_hpd_debounce_delay_ms;
 
 DECLARE_DYNDBG_CLASSMAP(drm_debug_classes, DD_CLASS_TYPE_DISJOINT_BITS, 0,
 			"DRM_UT_CORE",
@@ -1123,6 +1124,16 @@ module_param_named(rebar, amdgpu_rebar, int, 0444);
 MODULE_PARM_DESC(user_queue, "Enable user queues (-1 = auto (default), 0 = disable, 1 = enable, 2 = enable UQs and disable KQs)");
 module_param_named(user_queue, amdgpu_user_queue, int, 0444);
 
+/*
+ * DOC: hdmi_hpd_debounce_delay_ms (uint)
+ * HDMI HPD disconnect debounce delay in milliseconds.
+ *
+ * Used to filter short disconnect->reconnect HPD toggles some HDMI sinks
+ * generate while entering/leaving power save. Set to 0 to disable by default.
+ */
+MODULE_PARM_DESC(hdmi_hpd_debounce_delay_ms, "HDMI HPD disconnect debounce delay in milliseconds (0 to disable (by default), 1500 is common)");
+module_param_named(hdmi_hpd_debounce_delay_ms, amdgpu_hdmi_hpd_debounce_delay_ms, uint, 0644);
+
 /* These devices are not supported by amdgpu.
  * They are supported by the mach64, r128, radeon drivers
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 06c333b2213b..d78d9e7fb9d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -763,7 +763,7 @@ void amdgpu_fence_save_wptr(struct amdgpu_fence *af)
 }
 
 static void amdgpu_ring_backup_unprocessed_command(struct amdgpu_ring *ring,
-						   u64 start_wptr, u32 end_wptr)
+						   u64 start_wptr, u64 end_wptr)
 {
 	unsigned int first_idx = start_wptr & ring->buf_mask;
 	unsigned int last_idx = end_wptr & ring->buf_mask;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index d2237ce9da70..1485f4789440 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -375,7 +375,7 @@ void amdgpu_gart_map(struct amdgpu_device *adev, uint64_t offset,
  * @start_page: first page to map in the GART aperture
  * @num_pages: number of pages to be mapped
  * @flags: page table entry flags
- * @dst: CPU address of the GART table
+ * @dst: valid CPU address of GART table, cannot be null
  *
  * Binds a BO that is allocated in VRAM to the GART page table
  * (all ASICs).
@@ -396,7 +396,7 @@ void amdgpu_gart_map_vram_range(struct amdgpu_device *adev, uint64_t pa,
 		return;
 
 	for (i = 0; i < num_pages; ++i) {
-		amdgpu_gmc_set_pte_pde(adev, adev->gart.ptr,
+		amdgpu_gmc_set_pte_pde(adev, dst,
 			start_page + i, pa + AMDGPU_GPU_PAGE_SIZE * i, flags);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 869bceb0fe2c..7e623f91f2d7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -732,6 +732,12 @@ int amdgpu_gmc_flush_gpu_tlb_pasid(struct amdgpu_device *adev, uint16_t pasid,
 		return 0;
 
 	if (!adev->gmc.flush_pasid_uses_kiq || !ring->sched.ready) {
+
+		if (!adev->gmc.gmc_funcs->flush_gpu_tlb_pasid) {
+			r = 0;
+			goto error_unlock_reset;
+		}
+
 		if (adev->gmc.flush_tlb_needs_extra_type_2)
 			adev->gmc.gmc_funcs->flush_gpu_tlb_pasid(adev, pasid,
 								 2, all_hub,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 586a58facca1..72ec455fa932 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -302,7 +302,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned int num_ibs,
 		if (job && job->vmid)
 			amdgpu_vmid_reset(adev, ring->vm_hub, job->vmid);
 		amdgpu_ring_undo(ring);
-		return r;
+		goto free_fence;
 	}
 	*f = &af->base;
 	/* get a ref for the job */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 0a0dcbf0798d..7ccb724b2488 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -217,8 +217,11 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	if (!entity)
 		return 0;
 
-	return drm_sched_job_init(&(*job)->base, entity, 1, owner,
-				  drm_client_id);
+	r = drm_sched_job_init(&(*job)->base, entity, 1, owner, drm_client_id);
+	if (!r)
+		return 0;
+
+	kfree((*job)->hw_vm_fence);
 
 err_fence:
 	kfree((*job)->hw_fence);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 9a969175900e..58b26c78b642 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -885,12 +885,28 @@ static int amdgpu_userq_input_args_validate(struct drm_device *dev,
 	return 0;
 }
 
+bool amdgpu_userq_enabled(struct drm_device *dev)
+{
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int i;
+
+	for (i = 0; i < AMDGPU_HW_IP_NUM; i++) {
+		if (adev->userq_funcs[i])
+			return true;
+	}
+
+	return false;
+}
+
 int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *filp)
 {
 	union drm_amdgpu_userq *args = data;
 	int r;
 
+	if (!amdgpu_userq_enabled(dev))
+		return -ENOTSUPP;
+
 	if (amdgpu_userq_input_args_validate(dev, args, filp) < 0)
 		return -EINVAL;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
index c37444427a14..b48b3bc293fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -141,6 +141,7 @@ uint64_t amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr,
 					     struct drm_file *filp);
 
 u32 amdgpu_userq_get_supported_ip_mask(struct amdgpu_device *adev);
+bool amdgpu_userq_enabled(struct drm_device *dev);
 
 int amdgpu_userq_suspend(struct amdgpu_device *adev);
 int amdgpu_userq_resume(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
index eba9fb359047..85e9edc1cb6f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
@@ -141,6 +141,8 @@ static void amdgpu_userq_walk_and_drop_fence_drv(struct xarray *xa)
 void
 amdgpu_userq_fence_driver_free(struct amdgpu_usermode_queue *userq)
 {
+	dma_fence_put(userq->last_fence);
+
 	amdgpu_userq_walk_and_drop_fence_drv(&userq->fence_drv_xa);
 	xa_destroy(&userq->fence_drv_xa);
 	/* Drop the fence_drv reference held by user queue */
@@ -471,6 +473,9 @@ int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data,
 	struct drm_exec exec;
 	u64 wptr;
 
+	if (!amdgpu_userq_enabled(dev))
+		return -ENOTSUPP;
+
 	num_syncobj_handles = args->num_syncobj_handles;
 	syncobj_handles = memdup_user(u64_to_user_ptr(args->syncobj_handles),
 				      size_mul(sizeof(u32), num_syncobj_handles));
@@ -653,6 +658,9 @@ int amdgpu_userq_wait_ioctl(struct drm_device *dev, void *data,
 	int r, i, rentry, wentry, cnt;
 	struct drm_exec exec;
 
+	if (!amdgpu_userq_enabled(dev))
+		return -ENOTSUPP;
+
 	num_read_bo_handles = wait_info->num_bo_read_handles;
 	bo_handles_read = memdup_user(u64_to_user_ptr(wait_info->bo_read_handles),
 				      size_mul(sizeof(u32), num_read_bo_handles));
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index c362d4dfb5bb..a67285118c37 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1069,9 +1069,7 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params,
 	}
 
 	/* Prepare a TLB flush fence to be attached to PTs */
-	if (!params->unlocked &&
-	    /* SI doesn't support pasid or KIQ/MES */
-	    params->adev->family > AMDGPU_FAMILY_SI) {
+	if (!params->unlocked) {
 		amdgpu_vm_tlb_fence_create(params->adev, vm, fence);
 
 		/* Makes sure no PD/PT is freed before the flush */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index d01d2712cf57..b786967022d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -278,7 +278,6 @@ static void gfx_v12_0_select_se_sh(struct amdgpu_device *adev, u32 se_num,
 				   u32 sh_num, u32 instance, int xcc_id);
 static u32 gfx_v12_0_get_wgp_active_bitmap_per_sh(struct amdgpu_device *adev);
 
-static void gfx_v12_0_ring_emit_frame_cntl(struct amdgpu_ring *ring, bool start, bool secure);
 static void gfx_v12_0_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg,
 				     uint32_t val);
 static int gfx_v12_0_wait_for_rlc_autoload_complete(struct amdgpu_device *adev);
@@ -4634,16 +4633,6 @@ static int gfx_v12_0_ring_preempt_ib(struct amdgpu_ring *ring)
 	return r;
 }
 
-static void gfx_v12_0_ring_emit_frame_cntl(struct amdgpu_ring *ring,
-					   bool start,
-					   bool secure)
-{
-	uint32_t v = secure ? FRAME_TMZ : 0;
-
-	amdgpu_ring_write(ring, PACKET3(PACKET3_FRAME_CONTROL, 0));
-	amdgpu_ring_write(ring, v | FRAME_CMD(start ? 0 : 1));
-}
-
 static void gfx_v12_0_ring_emit_rreg(struct amdgpu_ring *ring, uint32_t reg,
 				     uint32_t reg_val_offs)
 {
@@ -5520,7 +5509,6 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_gfx = {
 	.emit_cntxcntl = gfx_v12_0_ring_emit_cntxcntl,
 	.init_cond_exec = gfx_v12_0_ring_emit_init_cond_exec,
 	.preempt_ib = gfx_v12_0_ring_preempt_ib,
-	.emit_frame_cntl = gfx_v12_0_ring_emit_frame_cntl,
 	.emit_wreg = gfx_v12_0_ring_emit_wreg,
 	.emit_reg_wait = gfx_v12_0_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = gfx_v12_0_ring_emit_reg_write_reg_wait,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 8ad7519f7b58..f1ee3921d970 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1235,16 +1235,16 @@ static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev,
 		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_NC);
 		break;
 	case AMDGPU_VM_MTYPE_WC:
-		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC);
+		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC);
 		break;
 	case AMDGPU_VM_MTYPE_RW:
-		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW);
+		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW);
 		break;
 	case AMDGPU_VM_MTYPE_CC:
-		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC);
+		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC);
 		break;
 	case AMDGPU_VM_MTYPE_UC:
-		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC);
+		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC);
 		break;
 	}