drm/scheduler: Add stopped flag to drm_sched_entity

The flag will prevent another thread from same process to
reinsert the entity queue into scheduler's rq after it was already
removewd from there by another thread during drm_sched_entity_flush.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed-by: Chunming Zhou <david1.zhou@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

1 files changed, 11 insertions, 1 deletions

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 1416edb2642a..812e3530ea25 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -177,8 +177,12 @@ long drm_sched_entity_flush(struct drm_sched_entity *entity, long timeout)
         /* For killed process disable any more IBs enqueue right now */
         last_user = cmpxchg(&entity->last_user, current->group_leader, NULL);
         if ((!last_user || last_user == current->group_leader) &&
-            (current->flags & PF_EXITING) && (current->exit_code == SIGKILL))
+            (current->flags & PF_EXITING) && (current->exit_code == SIGKILL)) {
+                spin_lock(&entity->rq_lock);
+                entity->stopped = true;
                 drm_sched_rq_remove_entity(entity->rq, entity);
+                spin_unlock(&entity->rq_lock);
+        }
 
         return ret;
 }
@@ -504,6 +508,12 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job,
         if (first) {
                 /* Add the entity to the run queue */
                 spin_lock(&entity->rq_lock);
+                if (entity->stopped) {
+                        spin_unlock(&entity->rq_lock);
+
+                        DRM_ERROR("Trying to push to a killed entity\n");
+                        return;
+                }
                 drm_sched_rq_add_entity(entity->rq, entity);
                 spin_unlock(&entity->rq_lock);
                 drm_sched_wakeup(entity->rq->sched);