Merge tag 'amd-drm-next-5.13-2021-04-12' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-5.13-2021-04-12:

amdgpu:
- Re-enable GPU reset on VanGogh
- Enable DPM flags for SMART_SUSPEND and MAY_SKIP_RESUME
- Disentangle HG from vga_switcheroo
- S0ix fixes
- W=1 fixes
- Resource iterator fixes
- DMCUB updates
- UBSAN fixes
- More PM API cleanup
- Aldebaran updates
- Modifier fixes
- Enable VCN load balancing with asymmetric engines
- Rework BO structs
- Aldebaran reset support
- Initial LTTPR display work
- Display MALL fixes
- Fall back to YCbCr420 when YCbCr444 fails
- SR-IOV fixes
- RAS updates
- Misc cleanups and fixes

radeon:
- Typo fixes
- Fix error handling for firmware on r6xx
- Fix a missing check in DP MST handling

Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210412220732.3845-1-alexander.deucher@amd.com

1 files changed, 71 insertions, 32 deletions

diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index d82a7ebf6099..92d8de24d0a1 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -361,40 +361,16 @@ static void drm_sched_job_timedout(struct work_struct *work)
   */
 void drm_sched_increase_karma(struct drm_sched_job *bad)
 {
-	int i;
-	struct drm_sched_entity *tmp;
-	struct drm_sched_entity *entity;
-	struct drm_gpu_scheduler *sched = bad->sched;
-
-	/* don't increase @bad's karma if it's from KERNEL RQ,
-	 * because sometimes GPU hang would cause kernel jobs (like VM updating jobs)
-	 * corrupt but keep in mind that kernel jobs always considered good.
-	 */
-	if (bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) {
-		atomic_inc(&bad->karma);
-		for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_KERNEL;
-		     i++) {
-			struct drm_sched_rq *rq = &sched->sched_rq[i];
-
-			spin_lock(&rq->lock);
-			list_for_each_entry_safe(entity, tmp, &rq->entities, list) {
-				if (bad->s_fence->scheduled.context ==
-				    entity->fence_context) {
-					if (atomic_read(&bad->karma) >
-					    bad->sched->hang_limit)
-						if (entity->guilty)
-							atomic_set(entity->guilty, 1);
-					break;
-				}
-			}
-			spin_unlock(&rq->lock);
-			if (&entity->list != &rq->entities)
-				break;
-		}
-	}
+	drm_sched_increase_karma_ext(bad, 1);
 }
 EXPORT_SYMBOL(drm_sched_increase_karma);
 
+void drm_sched_reset_karma(struct drm_sched_job *bad)
+{
+	drm_sched_increase_karma_ext(bad, 0);
+}
+EXPORT_SYMBOL(drm_sched_reset_karma);
+
 /**
  * drm_sched_stop - stop the scheduler
  *
@@ -534,14 +510,31 @@ EXPORT_SYMBOL(drm_sched_start);
  */
 void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched)
 {
+	drm_sched_resubmit_jobs_ext(sched, INT_MAX);
+}
+EXPORT_SYMBOL(drm_sched_resubmit_jobs);
+
+/**
+ * drm_sched_resubmit_jobs_ext - helper to relunch certain number of jobs from mirror ring list
+ *
+ * @sched: scheduler instance
+ * @max: job numbers to relaunch
+ *
+ */
+void drm_sched_resubmit_jobs_ext(struct drm_gpu_scheduler *sched, int max)
+{
 	struct drm_sched_job *s_job, *tmp;
 	uint64_t guilty_context;
 	bool found_guilty = false;
 	struct dma_fence *fence;
+	int i = 0;
 
 	list_for_each_entry_safe(s_job, tmp, &sched->pending_list, list) {
 		struct drm_sched_fence *s_fence = s_job->s_fence;
 
+		if (i >= max)
+			break;
+
 		if (!found_guilty && atomic_read(&s_job->karma) > sched->hang_limit) {
 			found_guilty = true;
 			guilty_context = s_job->s_fence->scheduled.context;
@@ -552,6 +545,7 @@ void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched)
 
 		dma_fence_put(s_job->s_fence->parent);
 		fence = sched->ops->run_job(s_job);
+		i++;
 
 		if (IS_ERR_OR_NULL(fence)) {
 			if (IS_ERR(fence))
@@ -563,7 +557,7 @@ void drm_sched_resubmit_jobs(struct drm_gpu_scheduler *sched)
 		}
 	}
 }
-EXPORT_SYMBOL(drm_sched_resubmit_jobs);
+EXPORT_SYMBOL(drm_sched_resubmit_jobs_ext);
 
 /**
  * drm_sched_job_init - init a scheduler job
@@ -903,3 +897,48 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched)
 	sched->ready = false;
 }
 EXPORT_SYMBOL(drm_sched_fini);
+
+/**
+ * drm_sched_increase_karma_ext - Update sched_entity guilty flag
+ *
+ * @bad: The job guilty of time out
+ * @type: type for increase/reset karma
+ *
+ */
+void drm_sched_increase_karma_ext(struct drm_sched_job *bad, int type)
+{
+	int i;
+	struct drm_sched_entity *tmp;
+	struct drm_sched_entity *entity;
+	struct drm_gpu_scheduler *sched = bad->sched;
+
+	/* don't change @bad's karma if it's from KERNEL RQ,
+	 * because sometimes GPU hang would cause kernel jobs (like VM updating jobs)
+	 * corrupt but keep in mind that kernel jobs always considered good.
+	 */
+	if (bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) {
+		if (type == 0)
+			atomic_set(&bad->karma, 0);
+		else if (type == 1)
+			atomic_inc(&bad->karma);
+
+		for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_KERNEL;
+		     i++) {
+			struct drm_sched_rq *rq = &sched->sched_rq[i];
+
+			spin_lock(&rq->lock);
+			list_for_each_entry_safe(entity, tmp, &rq->entities, list) {
+				if (bad->s_fence->scheduled.context ==
+				    entity->fence_context) {
+					if (entity->guilty)
+						atomic_set(entity->guilty, type);
+					break;
+				}
+			}
+			spin_unlock(&rq->lock);
+			if (&entity->list != &rq->entities)
+				break;
+		}
+	}
+}
+EXPORT_SYMBOL(drm_sched_increase_karma_ext);