drm/sched: Allow using a dedicated workqueue for the timeout/fault tdr

Mali Midgard/Bifrost GPUs have 3 hardware queues but only a global GPU
reset. This leads to extra complexity when we need to synchronize timeout
works with the reset work. One solution to address that is to have an
ordered workqueue at the driver level that will be used by the different
schedulers to queue their timeout work. Thanks to the serialization
provided by the ordered workqueue we are guaranteed that timeout
handlers are executed sequentially, and can thus easily reset the GPU
from the timeout handler without extra synchronization.

v5:
* Add a new paragraph to the timedout_job() method

v3:
* New patch

v4:
* Actually use the timeout_wq to queue the timeout work

Suggested-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Christian König <christian.koenig@amd.com>
Cc: Qiang Yu <yuq825@gmail.com>
Cc: Emma Anholt <emma@anholt.net>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210630062751.2832545-3-boris.brezillon@collabora.com

1 files changed, 22 insertions, 1 deletions

diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index aa90ed1f1b2b..88ae7f331bb1 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -253,6 +253,24 @@ struct drm_sched_backend_ops {
 	 * 5. Restart the scheduler using drm_sched_start(). At that point, new
 	 *    jobs can be queued, and the scheduler thread is unblocked
 	 *
+	 * Note that some GPUs have distinct hardware queues but need to reset
+	 * the GPU globally, which requires extra synchronization between the
+	 * timeout handler of the different &drm_gpu_scheduler. One way to
+	 * achieve this synchronization is to create an ordered workqueue
+	 * (using alloc_ordered_workqueue()) at the driver level, and pass this
+	 * queue to drm_sched_init(), to guarantee that timeout handlers are
+	 * executed sequentially. The above workflow needs to be slightly
+	 * adjusted in that case:
+	 *
+	 * 1. Stop all schedulers impacted by the reset using drm_sched_stop()
+	 * 2. Try to gracefully stop non-faulty jobs on all queues impacted by
+	 *    the reset (optional)
+	 * 3. Issue a GPU reset on all faulty queues (driver-specific)
+	 * 4. Re-submit jobs on all schedulers impacted by the reset using
+	 *    drm_sched_resubmit_jobs()
+	 * 5. Restart all schedulers that were stopped in step #1 using
+	 *    drm_sched_start()
+	 *
 	 * Return DRM_GPU_SCHED_STAT_NOMINAL, when all is normal,
 	 * and the underlying driver has started or completed recovery.
 	 *
@@ -283,6 +301,7 @@ struct drm_sched_backend_ops {
  *                 finished.
  * @hw_rq_count: the number of jobs currently in the hardware queue.
  * @job_id_count: used to assign unique id to the each job.
+ * @timeout_wq: workqueue used to queue @work_tdr
  * @work_tdr: schedules a delayed call to @drm_sched_job_timedout after the
  *            timeout interval is over.
  * @thread: the kthread on which the scheduler which run.
@@ -307,6 +326,7 @@ struct drm_gpu_scheduler {
 	wait_queue_head_t		job_scheduled;
 	atomic_t			hw_rq_count;
 	atomic64_t			job_id_count;
+	struct workqueue_struct		*timeout_wq;
 	struct delayed_work		work_tdr;
 	struct task_struct		*thread;
 	struct list_head		pending_list;
@@ -320,7 +340,8 @@ 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,
+		   uint32_t hw_submission, unsigned hang_limit,
+		   long timeout, struct workqueue_struct *timeout_wq,
 		   atomic_t *score, const char *name);
 
 void drm_sched_fini(struct drm_gpu_scheduler *sched);