drm/sched: Barriers are needed for entity->last_scheduled

It might be good enough on x86 with just READ_ONCE, but the write side
should then at least be WRITE_ONCE because x86 has total store order.

It's definitely not enough on arm.

Fix this proplery, which means
- explain the need for the barrier in both places
- point at the other side in each comment

Also pull out the !sched_list case as the first check, so that the
code flow is clearer.

While at it sprinkle some comments around because it was very
non-obvious to me what's actually going on here and why.

Note that we really need full barriers here, at first I thought
store-release and load-acquire on ->last_scheduled would be enough,
but we actually requiring ordering between that and the queue state.

v2: Put smp_rmp() in the right place and fix up comment (Andrey)
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: Steven Price <steven.price@arm.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Cc: Lee Jones <lee.jones@linaro.org>
Cc: Boris Brezillon <boris.brezillon@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210805104705.862416-4-daniel.vetter@ffwll.ch

drivers/gpu/drm/scheduler/sched_entity.c

@@ -439,8 +439,16 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 		dma_fence_set_error(&sched_job->s_fence->finished, -ECANCELED);
 
 	dma_fence_put(entity->last_scheduled);
+
 	entity->last_scheduled = dma_fence_get(&sched_job->s_fence->finished);
 
+	/*
+	 * If the queue is empty we allow drm_sched_entity_select_rq() to
+	 * locklessly access ->last_scheduled. This only works if we set the
+	 * pointer before we dequeue and if we a write barrier here.
+	 */
+	smp_wmb();
+
 	spsc_queue_pop(&entity->job_queue);
 	return sched_job;
 }
@@ -459,10 +467,25 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity)
 	struct drm_gpu_scheduler *sched;
 	struct drm_sched_rq *rq;
 
-	if (spsc_queue_count(&entity->job_queue) || !entity->sched_list)
+	/* single possible engine and already selected */
+	if (!entity->sched_list)
+		return;
+
+	/* queue non-empty, stay on the same engine */
+	if (spsc_queue_count(&entity->job_queue))
 		return;
 
-	fence = READ_ONCE(entity->last_scheduled);
+	/*
+	 * Only when the queue is empty are we guaranteed that the scheduler
+	 * thread cannot change ->last_scheduled. To enforce ordering we need
+	 * a read barrier here. See drm_sched_entity_pop_job() for the other
+	 * side.
+	 */
+	smp_rmb();
+
+	fence = entity->last_scheduled;
+
+	/* stay on the same engine if the previous job hasn't finished */
 	if (fence && !dma_fence_is_signaled(fence))
 		return;