drm/scheduler: provide scheduler score externally

Allow multiple schedulers to share the load balancing score.

This is useful when one engine has different hw rings.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-and-Tested-by: Leo Liu <leo.liu@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210204144405.2737-1-christian.koenig@amd.com

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

@@ -487,7 +487,7 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
 
 		r = drm_sched_init(&ring->sched, &amdgpu_sched_ops,
 				   num_hw_submission, amdgpu_job_hang_limit,
-				   timeout, ring->name);
+				   timeout, NULL, ring->name);
 		if (r) {
 			DRM_ERROR("Failed to create scheduler on ring %s.\n",
 				  ring->name);

drivers/gpu/drm/etnaviv/etnaviv_sched.c

@@ -190,7 +190,7 @@ int etnaviv_sched_init(struct etnaviv_gpu *gpu)
 
 	ret = drm_sched_init(&gpu->sched, &etnaviv_sched_ops,
 			     etnaviv_hw_jobs_limit, etnaviv_job_hang_limit,
-			     msecs_to_jiffies(500), dev_name(gpu->dev));
+			     msecs_to_jiffies(500), NULL, dev_name(gpu->dev));
 	if (ret)
 		return ret;
 

drivers/gpu/drm/lima/lima_sched.c

@@ -509,7 +509,7 @@ int lima_sched_pipe_init(struct lima_sched_pipe *pipe, const char *name)
 
 	return drm_sched_init(&pipe->base, &lima_sched_ops, 1,
 			      lima_job_hang_limit, msecs_to_jiffies(timeout),
-			      name);
+			      NULL, name);
 }
 
 void lima_sched_pipe_fini(struct lima_sched_pipe *pipe)

drivers/gpu/drm/panfrost/panfrost_job.c

@@ -627,7 +627,7 @@ int panfrost_job_init(struct panfrost_device *pfdev)
 		ret = drm_sched_init(&js->queue[j].sched,
 				     &panfrost_sched_ops,
 				     1, 0, msecs_to_jiffies(JOB_TIMEOUT_MS),
-				     "pan_js");
+				     NULL, "pan_js");
 		if (ret) {
 			dev_err(pfdev->dev, "Failed to create scheduler: %d.", ret);
 			goto err_sched;

drivers/gpu/drm/scheduler/sched_entity.c

@@ -489,7 +489,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job,
 	bool first;
 
 	trace_drm_sched_job(sched_job, entity);
-	atomic_inc(&entity->rq->sched->score);
+	atomic_inc(entity->rq->sched->score);
 	WRITE_ONCE(entity->last_user, current->group_leader);
 	first = spsc_queue_push(&entity->job_queue, &sched_job->queue_node);
 

drivers/gpu/drm/scheduler/sched_main.c

@@ -91,7 +91,7 @@ void drm_sched_rq_add_entity(struct drm_sched_rq *rq,
 	if (!list_empty(&entity->list))
 		return;
 	spin_lock(&rq->lock);
-	atomic_inc(&rq->sched->score);
+	atomic_inc(rq->sched->score);
 	list_add_tail(&entity->list, &rq->entities);
 	spin_unlock(&rq->lock);
 }
@@ -110,7 +110,7 @@ void drm_sched_rq_remove_entity(struct drm_sched_rq *rq,
 	if (list_empty(&entity->list))
 		return;
 	spin_lock(&rq->lock);
-	atomic_dec(&rq->sched->score);
+	atomic_dec(rq->sched->score);
 	list_del_init(&entity->list);
 	if (rq->current_entity == entity)
 		rq->current_entity = NULL;
@@ -173,7 +173,7 @@ static void drm_sched_job_done(struct drm_sched_job *s_job)
 	struct drm_gpu_scheduler *sched = s_fence->sched;
 
 	atomic_dec(&sched->hw_rq_count);
-	atomic_dec(&sched->score);
+	atomic_dec(sched->score);
 
 	trace_drm_sched_process_job(s_fence);
 
@@ -732,7 +732,7 @@ drm_sched_pick_best(struct drm_gpu_scheduler **sched_list,
 			continue;
 		}
 
-		num_score = atomic_read(&sched->score);
+		num_score = atomic_read(sched->score);
 		if (num_score < min_score) {
 			min_score = num_score;
 			picked_sched = sched;
@@ -842,16 +842,15 @@ static int drm_sched_main(void *param)
  * @hw_submission: number of hw submissions that can be in flight
  * @hang_limit: number of times to allow a job to hang before dropping it
  * @timeout: timeout value in jiffies for the scheduler
+ * @score: optional score atomic shared with other schedulers
  * @name: name used for debugging
  *
  * Return 0 on success, otherwise error code.
  */
 int drm_sched_init(struct drm_gpu_scheduler *sched,
 		   const struct drm_sched_backend_ops *ops,
-		   unsigned hw_submission,
-		   unsigned hang_limit,
-		   long timeout,
-		   const char *name)
+		   unsigned hw_submission, unsigned hang_limit, long timeout,
+		   atomic_t *score, const char *name)
 {
 	int i, ret;
 	sched->ops = ops;
@@ -859,6 +858,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
 	sched->name = name;
 	sched->timeout = timeout;
 	sched->hang_limit = hang_limit;
+	sched->score = score ? score : &sched->_score;
 	for (i = DRM_SCHED_PRIORITY_MIN; i < DRM_SCHED_PRIORITY_COUNT; i++)
 		drm_sched_rq_init(sched, &sched->sched_rq[i]);
 
@@ -868,7 +868,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
 	spin_lock_init(&sched->job_list_lock);
 	atomic_set(&sched->hw_rq_count, 0);
 	INIT_DELAYED_WORK(&sched->work_tdr, drm_sched_job_timedout);
-	atomic_set(&sched->score, 0);
+	atomic_set(&sched->_score, 0);
 	atomic64_set(&sched->job_id_count, 0);
 
 	/* Each scheduler will run on a seperate kernel thread */

drivers/gpu/drm/v3d/v3d_sched.c

@@ -403,7 +403,7 @@ v3d_sched_init(struct v3d_dev *v3d)
 			     &v3d_bin_sched_ops,
 			     hw_jobs_limit, job_hang_limit,
 			     msecs_to_jiffies(hang_limit_ms),
-			     "v3d_bin");
+			     NULL, "v3d_bin");
 	if (ret) {
 		dev_err(v3d->drm.dev, "Failed to create bin scheduler: %d.", ret);
 		return ret;
@@ -413,7 +413,7 @@ v3d_sched_init(struct v3d_dev *v3d)
 			     &v3d_render_sched_ops,
 			     hw_jobs_limit, job_hang_limit,
 			     msecs_to_jiffies(hang_limit_ms),
-			     "v3d_render");
+			     NULL, "v3d_render");
 	if (ret) {
 		dev_err(v3d->drm.dev, "Failed to create render scheduler: %d.",
 			ret);
@@ -425,7 +425,7 @@ v3d_sched_init(struct v3d_dev *v3d)
 			     &v3d_tfu_sched_ops,
 			     hw_jobs_limit, job_hang_limit,
 			     msecs_to_jiffies(hang_limit_ms),
-			     "v3d_tfu");
+			     NULL, "v3d_tfu");
 	if (ret) {
 		dev_err(v3d->drm.dev, "Failed to create TFU scheduler: %d.",
 			ret);
@@ -438,7 +438,7 @@ v3d_sched_init(struct v3d_dev *v3d)
 				     &v3d_csd_sched_ops,
 				     hw_jobs_limit, job_hang_limit,
 				     msecs_to_jiffies(hang_limit_ms),
-				     "v3d_csd");
+				     NULL, "v3d_csd");
 		if (ret) {
 			dev_err(v3d->drm.dev, "Failed to create CSD scheduler: %d.",
 				ret);
@@ -450,7 +450,7 @@ v3d_sched_init(struct v3d_dev *v3d)
 				     &v3d_cache_clean_sched_ops,
 				     hw_jobs_limit, job_hang_limit,
 				     msecs_to_jiffies(hang_limit_ms),
-				     "v3d_cache_clean");
+				     NULL, "v3d_cache_clean");
 		if (ret) {
 			dev_err(v3d->drm.dev, "Failed to create CACHE_CLEAN scheduler: %d.",
 				ret);

include/drm/gpu_scheduler.h

@@ -297,7 +297,8 @@ struct drm_gpu_scheduler {
 	struct list_head		pending_list;
 	spinlock_t			job_list_lock;
 	int				hang_limit;
-	atomic_t                        score;
+	atomic_t                        *score;
+	atomic_t                        _score;
 	bool				ready;
 	bool				free_guilty;
 };
@@ -305,7 +306,7 @@ struct drm_gpu_scheduler {
 int drm_sched_init(struct drm_gpu_scheduler *sched,
 		   const struct drm_sched_backend_ops *ops,
 		   uint32_t hw_submission, unsigned hang_limit, long timeout,
-		   const char *name);
+		   atomic_t *score, const char *name);
 
 void drm_sched_fini(struct drm_gpu_scheduler *sched);
 int drm_sched_job_init(struct drm_sched_job *job,