Skip to contentSkip to conversation

drm/sched: Fix deadlock in drm_sched_entity_kill_jobs_cb

PATCH Superseded Pierre-Eric Pelloux-Prayer 3 participants 4 messages Oct 31, 2025

Subsystems (4)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 drm/sched: Fix deadlock in drm_sched_entity_kill_jobs_cb

The Mesa issue referenced below pointed out a possible deadlock:

[ 1231.611031]  Possible interrupt unsafe locking scenario:
[ 1231.611033]        CPU0                    CPU1
[ 1231.611034]        ----                    ----
[ 1231.611035]   lock(&xa->xa_lock#17);
[ 1231.611038]                                local_irq_disable();
[ 1231.611039]                                lock(&fence->lock);
[ 1231.611041]                                lock(&xa->xa_lock#17);
[ 1231.611044]   <Interrupt>
[ 1231.611045]     lock(&fence->lock);
[ 1231.611047]
                *** DEADLOCK ***

In this example, CPU0 would be any function accessing job->dependencies through the xa_* functions that doesn't disable interrupts (eg:
drm_sched_job_add_dependency, drm_sched_entity_kill_jobs_cb).

CPU1 is executing drm_sched_entity_kill_jobs_cb as a fence signalling callback so in an interrupt context. It will deadlock when trying to grab the xa_lock which is already held by CPU0.

Replacing all xa_* usage by their xa_*_irq counterparts would fix this issue, but Christian pointed out another issue: dma_fence_signal takes fence.lock and so does dma_fence_add_callback.

dma_fence_signal() // locks f1.lock
-> drm_sched_entity_kill_jobs_cb()
-> foreach dependencies

     -> dma_fence_add_callback() // locks f2.lock

This will deadlock if f1 and f2 share the same spinlock.

To fix both issues, the code iterating on dependencies and re-arming them is moved out to drm_sched_entity_kill_jobs_work.

---
 drivers/gpu/drm/scheduler/sched_entity.c | 34 +++++++++++++-----------
 1 file changed, 19 insertions(+), 15 deletions(-)
drivers/gpu/drm/scheduler/sched_entity.c +19 -15
index c8e949f4a568..fe174a4857be 100644
@@ -173,26 +173,15 @@ int drm_sched_entity_error(struct drm_sched_entity *entity)
 }
 EXPORT_SYMBOL(drm_sched_entity_error);
 
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+					  struct dma_fence_cb *cb);
+
 static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
 {
 	struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
-
-	drm_sched_fence_scheduled(job->s_fence, NULL);
-	drm_sched_fence_finished(job->s_fence, -ESRCH);
-	WARN_ON(job->s_fence->parent);
-	job->sched->ops->free_job(job);
-}
-
-/* Signal the scheduler finished fence when the entity in question is killed. */
-static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
-					  struct dma_fence_cb *cb)
-{
-	struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
-						 finish_cb);
+	struct dma_fence *f;
 	unsigned long index;
 
-	dma_fence_put(f);
-
 	/* Wait for all dependencies to avoid data corruptions */
 	xa_for_each(&job->dependencies, index, f) {
 		struct drm_sched_fence *s_fence = to_drm_sched_fence(f);
@@ -220,6 +209,21 @@ static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
 		dma_fence_put(f);
 	}
 
+	drm_sched_fence_scheduled(job->s_fence, NULL);
+	drm_sched_fence_finished(job->s_fence, -ESRCH);
+	WARN_ON(job->s_fence->parent);
+	job->sched->ops->free_job(job);
+}
+
+/* Signal the scheduler finished fence when the entity in question is killed. */
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+					  struct dma_fence_cb *cb)
+{
+	struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
+						 finish_cb);
+
+	dma_fence_put(f);
+
 	INIT_WORK(&job->work, drm_sched_entity_kill_jobs_work);
 	schedule_work(&job->work);
 }
Signature · 1 line
2.43.0
#1 lore raw
Quoted 27 lines · Pierre-Eric Pelloux-Prayer

The Mesa issue referenced below pointed out a possible deadlock:

[ 1231.611031]  Possible interrupt unsafe locking scenario:
[ 1231.611033]        CPU0                    CPU1
[ 1231.611034]        ----                    ----
[ 1231.611035]   lock(&xa->xa_lock#17);
[ 1231.611038]                                local_irq_disable();
[ 1231.611039]                                lock(&fence->lock);
[ 1231.611041]                                lock(&xa->xa_lock#17);
[ 1231.611044]   <Interrupt>
[ 1231.611045]     lock(&fence->lock);
[ 1231.611047]
                 *** DEADLOCK ***

In this example, CPU0 would be any function accessing job->dependencies through the xa_* functions that doesn't disable interrupts (eg:
drm_sched_job_add_dependency, drm_sched_entity_kill_jobs_cb).

CPU1 is executing drm_sched_entity_kill_jobs_cb as a fence signalling callback so in an interrupt context. It will deadlock when trying to grab the xa_lock which is already held by CPU0.

Replacing all xa_* usage by their xa_*_irq counterparts would fix this issue, but Christian pointed out another issue: dma_fence_signal takes fence.lock and so does dma_fence_add_callback.

Quoted 6 lines
   dma_fence_signal() // locks f1.lock
   -> drm_sched_entity_kill_jobs_cb()
   -> foreach dependencies
      -> dma_fence_add_callback() // locks f2.lock

This will deadlock if f1 and f2 share the same spinlock.

Is it possible to hit this case?

Same lock means same execution timeline, which should mean dependency should have been squashed in drm_sched_job_add_dependency(), no?

Or would sharing the lock but not sharing the entity->fence_context be considered legal? It would be surprising at least.

Also, would anyone have time to add a kunit test? ;)

Regards,

Tvrtko

Quoted 69 lines

To fix both issues, the code iterating on dependencies and re-arming them is moved out to drm_sched_entity_kill_jobs_work.

---
  drivers/gpu/drm/scheduler/sched_entity.c | 34 +++++++++++++-----------
  1 file changed, 19 insertions(+), 15 deletions(-)
drivers/gpu/drm/scheduler/sched_entity.c +19 -15
index c8e949f4a568..fe174a4857be 100644
@@ -173,26 +173,15 @@ int drm_sched_entity_error(struct drm_sched_entity *entity)
  }
  EXPORT_SYMBOL(drm_sched_entity_error);
 
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+					  struct dma_fence_cb *cb);
+
  static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
  {
  	struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
-
-	drm_sched_fence_scheduled(job->s_fence, NULL);
-	drm_sched_fence_finished(job->s_fence, -ESRCH);
-	WARN_ON(job->s_fence->parent);
-	job->sched->ops->free_job(job);
-}
-
-/* Signal the scheduler finished fence when the entity in question is killed. */
-static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
-					  struct dma_fence_cb *cb)
-{
-	struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
-						 finish_cb);
+	struct dma_fence *f;
  	unsigned long index;
 
-	dma_fence_put(f);
-
  	/* Wait for all dependencies to avoid data corruptions */
  	xa_for_each(&job->dependencies, index, f) {
  		struct drm_sched_fence *s_fence = to_drm_sched_fence(f);
@@ -220,6 +209,21 @@ static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
  		dma_fence_put(f);
  	}
 
+	drm_sched_fence_scheduled(job->s_fence, NULL);
+	drm_sched_fence_finished(job->s_fence, -ESRCH);
+	WARN_ON(job->s_fence->parent);
+	job->sched->ops->free_job(job);
+}
+
+/* Signal the scheduler finished fence when the entity in question is killed. */
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+					  struct dma_fence_cb *cb)
+{
+	struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
+						 finish_cb);
+
+	dma_fence_put(f);
+
  	INIT_WORK(&job->work, drm_sched_entity_kill_jobs_work);
  	schedule_work(&job->work);
  }
#2 lore raw
Quoted 34 lines · Tvrtko Ursulin
Quoted 32 lines · Pierre-Eric Pelloux-Prayer

The Mesa issue referenced below pointed out a possible deadlock:

[ 1231.611031]  Possible interrupt unsafe locking scenario:
[ 1231.611033]        CPU0                    CPU1
[ 1231.611034]        ----                    ----
[ 1231.611035]   lock(&xa->xa_lock#17);
[ 1231.611038]                                local_irq_disable();
[ 1231.611039]                                lock(&fence->lock);
[ 1231.611041]                                lock(&xa->xa_lock#17);
[ 1231.611044]   <Interrupt>
[ 1231.611045]     lock(&fence->lock);
[ 1231.611047]

                 *** DEADLOCK ***

In this example, CPU0 would be any function accessing job->dependencies through the xa_* functions that doesn't disable interrupts (eg:
drm_sched_job_add_dependency, drm_sched_entity_kill_jobs_cb).

CPU1 is executing drm_sched_entity_kill_jobs_cb as a fence signalling callback so in an interrupt context. It will deadlock when trying to grab the xa_lock which is already held by CPU0.

Replacing all xa_* usage by their xa_*_irq counterparts would fix this issue, but Christian pointed out another issue: dma_fence_signal takes fence.lock and so does dma_fence_add_callback.

   dma_fence_signal() // locks f1.lock
   -> drm_sched_entity_kill_jobs_cb()
   -> foreach dependencies
      -> dma_fence_add_callback() // locks f2.lock

Quoted 5 lines
Quoted 1 line

This will deadlock if f1 and f2 share the same spinlock.

Is it possible to hit this case?

Same lock means same execution timeline

Nope, exactly that is incorrect. It's completely up to the implementation what they use this lock for.

Quoted 1 line

, which should mean dependency should have been squashed in drm_sched_job_add_dependency(), no?

This makes it less likely, but not impossible to trigger.

Regards,
Christian.

Quoted 76 lines

Or would sharing the lock but not sharing the entity->fence_context be considered legal? It would be surprising at least.

Also, would anyone have time to add a kunit test? ;)

Regards,

Tvrtko

Quoted 66 lines

To fix both issues, the code iterating on dependencies and re-arming them is moved out to drm_sched_entity_kill_jobs_work.

---
  drivers/gpu/drm/scheduler/sched_entity.c | 34 +++++++++++++-----------

  1 file changed, 19 insertions(+), 15 deletions(-)
drivers/gpu/drm/scheduler/sched_entity.c +17 -14
index c8e949f4a568..fe174a4857be 100644
@@ -173,26 +173,15 @@ int drm_sched_entity_error(struct drm_sched_entity *entity)
   }
   EXPORT_SYMBOL(drm_sched_entity_error);
   +static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+                      struct dma_fence_cb *cb);
+
   static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
   {
       struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
-
-    drm_sched_fence_scheduled(job->s_fence, NULL);
-    drm_sched_fence_finished(job->s_fence, -ESRCH);
-    WARN_ON(job->s_fence->parent);
-    job->sched->ops->free_job(job);
-}
-
-/* Signal the scheduler finished fence when the entity in question is killed. */
-static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
-                      struct dma_fence_cb *cb)
-{
-    struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
-                         finish_cb);
+    struct dma_fence *f;
       unsigned long index;
   -    dma_fence_put(f);
-
       /* Wait for all dependencies to avoid data corruptions */
       xa_for_each(&job->dependencies, index, f) {
           struct drm_sched_fence *s_fence = to_drm_sched_fence(f);
@@ -220,6 +209,21 @@ static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
           dma_fence_put(f);
       }
   +    drm_sched_fence_scheduled(job->s_fence, NULL);
+    drm_sched_fence_finished(job->s_fence, -ESRCH);
+    WARN_ON(job->s_fence->parent);
+    job->sched->ops->free_job(job);
+}
+
+/* Signal the scheduler finished fence when the entity in question is killed. */
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+                      struct dma_fence_cb *cb)
+{
+    struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
+                         finish_cb);
+
+    dma_fence_put(f);
+
       INIT_WORK(&job->work, drm_sched_entity_kill_jobs_work);
       schedule_work(&job->work);
   }
#3 lore raw
Quoted 37 lines · Christian König
Quoted 34 lines · Tvrtko Ursulin
Quoted 32 lines · Pierre-Eric Pelloux-Prayer

The Mesa issue referenced below pointed out a possible deadlock:

[ 1231.611031]  Possible interrupt unsafe locking scenario:
[ 1231.611033]        CPU0                    CPU1
[ 1231.611034]        ----                    ----
[ 1231.611035]   lock(&xa->xa_lock#17);
[ 1231.611038]                                local_irq_disable();
[ 1231.611039]                                lock(&fence->lock);
[ 1231.611041]                                lock(&xa->xa_lock#17);
[ 1231.611044]   <Interrupt>
[ 1231.611045]     lock(&fence->lock);
[ 1231.611047]
                  *** DEADLOCK ***

In this example, CPU0 would be any function accessing job->dependencies through the xa_* functions that doesn't disable interrupts (eg:
drm_sched_job_add_dependency, drm_sched_entity_kill_jobs_cb).

CPU1 is executing drm_sched_entity_kill_jobs_cb as a fence signalling callback so in an interrupt context. It will deadlock when trying to grab the xa_lock which is already held by CPU0.

Replacing all xa_* usage by their xa_*_irq counterparts would fix this issue, but Christian pointed out another issue: dma_fence_signal takes fence.lock and so does dma_fence_add_callback.

   dma_fence_signal() // locks f1.lock
   -> drm_sched_entity_kill_jobs_cb()
   -> foreach dependencies
      -> dma_fence_add_callback() // locks f2.lock

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

This will deadlock if f1 and f2 share the same spinlock.

Is it possible to hit this case?

Same lock means same execution timeline

Nope, exactly that is incorrect. It's completely up to the implementation what they use this lock for.

Yes, sorry, I got confused for a moment. The lock can be per hw scheduler while execution timeline is per entity.

Regards,

Tvrtko

Quoted 85 lines
Quoted 2 lines

, which should mean dependency should have been squashed in drm_sched_job_add_dependency(), no?

This makes it less likely, but not impossible to trigger.

Regards,
Christian.

Quoted 76 lines

Or would sharing the lock but not sharing the entity->fence_context be considered legal? It would be surprising at least.

Also, would anyone have time to add a kunit test? ;)

Regards,

Tvrtko

Quoted 66 lines

To fix both issues, the code iterating on dependencies and re-arming them is moved out to drm_sched_entity_kill_jobs_work.

---
   drivers/gpu/drm/scheduler/sched_entity.c | 34 +++++++++++++-----------
   1 file changed, 19 insertions(+), 15 deletions(-)
drivers/gpu/drm/scheduler/sched_entity.c +17 -14
index c8e949f4a568..fe174a4857be 100644
@@ -173,26 +173,15 @@ int drm_sched_entity_error(struct drm_sched_entity *entity)
   }
   EXPORT_SYMBOL(drm_sched_entity_error);
   +static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+                      struct dma_fence_cb *cb);
+
   static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk)
   {
       struct drm_sched_job *job = container_of(wrk, typeof(*job), work);
-
-    drm_sched_fence_scheduled(job->s_fence, NULL);
-    drm_sched_fence_finished(job->s_fence, -ESRCH);
-    WARN_ON(job->s_fence->parent);
-    job->sched->ops->free_job(job);
-}
-
-/* Signal the scheduler finished fence when the entity in question is killed. */
-static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
-                      struct dma_fence_cb *cb)
-{
-    struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
-                         finish_cb);
+    struct dma_fence *f;
       unsigned long index;
   -    dma_fence_put(f);
-
       /* Wait for all dependencies to avoid data corruptions */
       xa_for_each(&job->dependencies, index, f) {
           struct drm_sched_fence *s_fence = to_drm_sched_fence(f);
@@ -220,6 +209,21 @@ static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
           dma_fence_put(f);
       }
   +    drm_sched_fence_scheduled(job->s_fence, NULL);
+    drm_sched_fence_finished(job->s_fence, -ESRCH);
+    WARN_ON(job->s_fence->parent);
+    job->sched->ops->free_job(job);
+}
+
+/* Signal the scheduler finished fence when the entity in question is killed. */
+static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
+                      struct dma_fence_cb *cb)
+{
+    struct drm_sched_job *job = container_of(cb, struct drm_sched_job,
+                         finish_cb);
+
+    dma_fence_put(f);
+
       INIT_WORK(&job->work, drm_sched_entity_kill_jobs_work);
       schedule_work(&job->work);
   }