Merge tag 'drm-fixes-2022-07-22' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Fixes for this week.

  The main one is the i915 firmware fix for the phoronix reported issue.
  I've written some firmware guidelines as a result, should land in
  -next soon. Otherwise a few amdgpu fixes, a scheduler fix, ttm fix and
  two other minor ones.

  scheduler:
   - scheduling while atomic fix

  ttm:
   - locking fix

  edp:
   - variable typo fix

  i915:
   - add back support for v69 firmware on ADL-P

  amdgpu:
   - Drop redundant buffer cleanup that can lead to a segfault
   - Add a bo_list mutex to avoid possible list corruption in CS
   - dmub notification fix

  imx:
   - fix error path"

* tag 'drm-fixes-2022-07-22' of git://anongit.freedesktop.org/drm/drm:
  drm/amdgpu: Protect the amdgpu_bo_list list with a mutex v2
  drm/imx/dcss: Add missing of_node_put() in fail path
  drm/i915/guc: support v69 in parallel to v70
  drm/i915/guc: Support programming the EU priority in the GuC descriptor
  drm/panel-edp: Fix variable typo when saving hpd absent delay from DT
  drm/amdgpu: Remove one duplicated ef removal
  drm/ttm: fix locking in vmap/vunmap TTM GEM helpers
  drm/scheduler: Don't kill jobs in interrupt context
  drm/amd/display: Fix new dmub notification enabling in DM

Author: torvalds

drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c

@@ -1364,16 +1364,10 @@ void amdgpu_amdkfd_gpuvm_destroy_cb(struct amdgpu_device *adev,
 				    struct amdgpu_vm *vm)
 {
 	struct amdkfd_process_info *process_info = vm->process_info;
-	struct amdgpu_bo *pd = vm->root.bo;
 
 	if (!process_info)
 		return;
 
-	/* Release eviction fence from PD */
-	amdgpu_bo_reserve(pd, false);
-	amdgpu_bo_fence(pd, NULL, false);
-	amdgpu_bo_unreserve(pd);
-
 	/* Update process info */
 	mutex_lock(&process_info->lock);
 	process_info->n_vms--;

drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c

@@ -40,7 +40,7 @@ static void amdgpu_bo_list_free_rcu(struct rcu_head *rcu)
 {
 	struct amdgpu_bo_list *list = container_of(rcu, struct amdgpu_bo_list,
 						   rhead);
-
+	mutex_destroy(&list->bo_list_mutex);
 	kvfree(list);
 }
 
@@ -136,6 +136,7 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp,
 
 	trace_amdgpu_cs_bo_status(list->num_entries, total_size);
 
+	mutex_init(&list->bo_list_mutex);
 	*result = list;
 	return 0;
 

drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h

@@ -47,6 +47,10 @@ struct amdgpu_bo_list {
 	struct amdgpu_bo *oa_obj;
 	unsigned first_userptr;
 	unsigned num_entries;
+
+	/* Protect access during command submission.
+	 */
+	struct mutex bo_list_mutex;
 };
 
 int amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id,

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -519,6 +519,8 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 			return r;
 	}
 
+	mutex_lock(&p->bo_list->bo_list_mutex);
+
 	/* One for TTM and one for the CS job */
 	amdgpu_bo_list_for_each_entry(e, p->bo_list)
 		e->tv.num_shared = 2;
@@ -651,6 +653,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 			kvfree(e->user_pages);
 			e->user_pages = NULL;
 		}
+		mutex_unlock(&p->bo_list->bo_list_mutex);
 	}
 	return r;
 }
@@ -690,9 +693,11 @@ static void amdgpu_cs_parser_fini(struct amdgpu_cs_parser *parser, int error,
 {
 	unsigned i;
 
-	if (error && backoff)
+	if (error && backoff) {
 		ttm_eu_backoff_reservation(&parser->ticket,
 					   &parser->validated);
+		mutex_unlock(&parser->bo_list->bo_list_mutex);
+	}
 
 	for (i = 0; i < parser->num_post_deps; i++) {
 		drm_syncobj_put(parser->post_deps[i].syncobj);
@@ -832,12 +837,16 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p)
 			continue;
 
 		r = amdgpu_vm_bo_update(adev, bo_va, false);
-		if (r)
+		if (r) {
+			mutex_unlock(&p->bo_list->bo_list_mutex);
 			return r;
+		}
 
 		r = amdgpu_sync_fence(&p->job->sync, bo_va->last_pt_update);
-		if (r)
+		if (r) {
+			mutex_unlock(&p->bo_list->bo_list_mutex);
 			return r;
+		}
 	}
 
 	r = amdgpu_vm_handle_moved(adev, vm);
@@ -1278,6 +1287,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 
 	ttm_eu_fence_buffer_objects(&p->ticket, &p->validated, p->fence);
 	mutex_unlock(&p->adev->notifier_lock);
+	mutex_unlock(&p->bo_list->bo_list_mutex);
 
 	return 0;
 

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

@@ -1653,7 +1653,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
 	adev->dm.crc_rd_wrk = amdgpu_dm_crtc_secure_display_create_work();
 #endif
-	if (dc_enable_dmub_notifications(adev->dm.dc)) {
+	if (dc_is_dmub_outbox_supported(adev->dm.dc)) {
 		init_completion(&adev->dm.dmub_aux_transfer_done);
 		adev->dm.dmub_notify = kzalloc(sizeof(struct dmub_notification), GFP_KERNEL);
 		if (!adev->dm.dmub_notify) {
@@ -1689,6 +1689,13 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 		goto error;
 	}
 
+	/* Enable outbox notification only after IRQ handlers are registered and DMUB is alive.
+	 * It is expected that DMUB will resend any pending notifications at this point, for
+	 * example HPD from DPIA.
+	 */
+	if (dc_is_dmub_outbox_supported(adev->dm.dc))
+		dc_enable_dmub_outbox(adev->dm.dc);
+
 	/* create fake encoders for MST */
 	dm_dp_create_fake_mst_encoders(adev);
 
@@ -2678,9 +2685,6 @@ static int dm_resume(void *handle)
 		 */
 		link_enc_cfg_copy(adev->dm.dc->current_state, dc_state);
 
-		if (dc_enable_dmub_notifications(adev->dm.dc))
-			amdgpu_dm_outbox_init(adev);
-
 		r = dm_dmub_hw_init(adev);
 		if (r)
 			DRM_ERROR("DMUB interface failed to initialize: status=%d\n", r);
@@ -2698,6 +2702,11 @@ static int dm_resume(void *handle)
 			}
 		}
 
+		if (dc_is_dmub_outbox_supported(adev->dm.dc)) {
+			amdgpu_dm_outbox_init(adev);
+			dc_enable_dmub_outbox(adev->dm.dc);
+		}
+
 		WARN_ON(!dc_commit_state(dm->dc, dc_state));
 
 		dm_gpureset_commit_state(dm->cached_dc_state, dm);
@@ -2719,13 +2728,15 @@ static int dm_resume(void *handle)
 	/* TODO: Remove dc_state->dccg, use dc->dccg directly. */
 	dc_resource_state_construct(dm->dc, dm_state->context);
 
-	/* Re-enable outbox interrupts for DPIA. */
-	if (dc_enable_dmub_notifications(adev->dm.dc))
-		amdgpu_dm_outbox_init(adev);
-
 	/* Before powering on DC we need to re-initialize DMUB. */
 	dm_dmub_hw_resume(adev);
 
+	/* Re-enable outbox interrupts for DPIA. */
+	if (dc_is_dmub_outbox_supported(adev->dm.dc)) {
+		amdgpu_dm_outbox_init(adev);
+		dc_enable_dmub_outbox(adev->dm.dc);
+	}
+
 	/* power on hardware */
 	dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0);
 

drivers/gpu/drm/drm_gem_ttm_helper.c

@@ -64,8 +64,13 @@ int drm_gem_ttm_vmap(struct drm_gem_object *gem,
 		     struct iosys_map *map)
 {
 	struct ttm_buffer_object *bo = drm_gem_ttm_of_gem(gem);
+	int ret;
+
+	dma_resv_lock(gem->resv, NULL);
+	ret = ttm_bo_vmap(bo, map);
+	dma_resv_unlock(gem->resv);
 
-	return ttm_bo_vmap(bo, map);
+	return ret;
 }
 EXPORT_SYMBOL(drm_gem_ttm_vmap);
 
@@ -82,7 +87,9 @@ void drm_gem_ttm_vunmap(struct drm_gem_object *gem,
 {
 	struct ttm_buffer_object *bo = drm_gem_ttm_of_gem(gem);
 
+	dma_resv_lock(gem->resv, NULL);
 	ttm_bo_vunmap(bo, map);
+	dma_resv_unlock(gem->resv);
 }
 EXPORT_SYMBOL(drm_gem_ttm_vunmap);
 

drivers/gpu/drm/i915/gt/intel_context_types.h

@@ -273,10 +273,17 @@ struct intel_context {
 		u8 child_index;
 		/** @guc: GuC specific members for parallel submission */
 		struct {
-			/** @wqi_head: head pointer in work queue */
+			/** @wqi_head: cached head pointer in work queue */
 			u16 wqi_head;
-			/** @wqi_tail: tail pointer in work queue */
+			/** @wqi_tail: cached tail pointer in work queue */
 			u16 wqi_tail;
+			/** @wq_head: pointer to the actual head in work queue */
+			u32 *wq_head;
+			/** @wq_tail: pointer to the actual head in work queue */
+			u32 *wq_tail;
+			/** @wq_status: pointer to the status in work queue */
+			u32 *wq_status;
+
 			/**
 			 * @parent_page: page in context state (ce->state) used
 			 * by parent for work queue, process descriptor

drivers/gpu/drm/i915/gt/intel_execlists_submission.c

@@ -661,6 +661,16 @@ static inline void execlists_schedule_out(struct i915_request *rq)
 	i915_request_put(rq);
 }
 
+static u32 map_i915_prio_to_lrc_desc_prio(int prio)
+{
+	if (prio > I915_PRIORITY_NORMAL)
+		return GEN12_CTX_PRIORITY_HIGH;
+	else if (prio < I915_PRIORITY_NORMAL)
+		return GEN12_CTX_PRIORITY_LOW;
+	else
+		return GEN12_CTX_PRIORITY_NORMAL;
+}
+
 static u64 execlists_update_context(struct i915_request *rq)
 {
 	struct intel_context *ce = rq->context;
@@ -669,7 +679,7 @@ static u64 execlists_update_context(struct i915_request *rq)
 
 	desc = ce->lrc.desc;
 	if (rq->engine->flags & I915_ENGINE_HAS_EU_PRIORITY)
-		desc |= lrc_desc_priority(rq_prio(rq));
+		desc |= map_i915_prio_to_lrc_desc_prio(rq_prio(rq));
 
 	/*
 	 * WaIdleLiteRestore:bdw,skl

drivers/gpu/drm/i915/gt/intel_lrc.h

@@ -111,16 +111,6 @@ enum {
 #define XEHP_SW_COUNTER_SHIFT			58
 #define XEHP_SW_COUNTER_WIDTH			6
 
-static inline u32 lrc_desc_priority(int prio)
-{
-	if (prio > I915_PRIORITY_NORMAL)
-		return GEN12_CTX_PRIORITY_HIGH;
-	else if (prio < I915_PRIORITY_NORMAL)
-		return GEN12_CTX_PRIORITY_LOW;
-	else
-		return GEN12_CTX_PRIORITY_NORMAL;
-}
-
 static inline void lrc_runtime_start(struct intel_context *ce)
 {
 	struct intel_context_stats *stats = &ce->stats;

drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h

@@ -122,6 +122,9 @@ enum intel_guc_action {
 	INTEL_GUC_ACTION_SCHED_CONTEXT_MODE_DONE = 0x1002,
 	INTEL_GUC_ACTION_SCHED_ENGINE_MODE_SET = 0x1003,
 	INTEL_GUC_ACTION_SCHED_ENGINE_MODE_DONE = 0x1004,
+	INTEL_GUC_ACTION_V69_SET_CONTEXT_PRIORITY = 0x1005,
+	INTEL_GUC_ACTION_V69_SET_CONTEXT_EXECUTION_QUANTUM = 0x1006,
+	INTEL_GUC_ACTION_V69_SET_CONTEXT_PREEMPTION_TIMEOUT = 0x1007,
 	INTEL_GUC_ACTION_CONTEXT_RESET_NOTIFICATION = 0x1008,
 	INTEL_GUC_ACTION_ENGINE_FAILURE_NOTIFICATION = 0x1009,
 	INTEL_GUC_ACTION_HOST2GUC_UPDATE_CONTEXT_POLICIES = 0x100B,