* [PATCH] drm/scheduler: Fix hang when sched_entity released
@ 2021-02-16 17:07 Andrey Grodzovsky
2021-02-17 15:53 ` Andrey Grodzovsky
2021-02-17 21:32 ` Christian König
0 siblings, 2 replies; 6+ messages in thread
From: Andrey Grodzovsky @ 2021-02-16 17:07 UTC (permalink / raw)
To: dri-devel, amd-gfx; +Cc: ckoenig.leichtzumerken
Problem: If scheduler is already stopped by the time sched_entity
is released and entity's job_queue not empty I encountred
a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
never becomes false.
Fix: In drm_sched_fini detach all sched_entities from the
scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
Also wakeup all those processes stuck in sched_entity flushing
as the scheduler main thread which wakes them up is stopped by now.
Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
---
drivers/gpu/drm/scheduler/sched_main.c | 31 +++++++++++++++++++++++++++++++
1 file changed, 31 insertions(+)
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 908b0b5..11abf5d 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
*/
void drm_sched_fini(struct drm_gpu_scheduler *sched)
{
+ int i;
+ struct drm_sched_entity *s_entity;
if (sched->thread)
kthread_stop(sched->thread);
+ /* Detach all sched_entites from this scheduler once it's stopped */
+ for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
+ struct drm_sched_rq *rq = &sched->sched_rq[i];
+
+ if (!rq)
+ continue;
+
+ /* Loop this way because rq->lock is taken in drm_sched_rq_remove_entity */
+ spin_lock(&rq->lock);
+ while ((s_entity = list_first_entry_or_null(&rq->entities,
+ struct drm_sched_entity,
+ list))) {
+ spin_unlock(&rq->lock);
+ drm_sched_rq_remove_entity(rq, s_entity);
+
+ /* Mark as stopped to reject adding to any new rq */
+ spin_lock(&s_entity->rq_lock);
+ s_entity->stopped = true;
+ spin_unlock(&s_entity->rq_lock);
+
+ spin_lock(&rq->lock);
+ }
+ spin_unlock(&rq->lock);
+
+ }
+
+ /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
+ wake_up_all(&sched->job_scheduled);
+
/* Confirm no work left behind accessing device structures */
cancel_delayed_work_sync(&sched->work_tdr);
--
2.7.4
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/scheduler: Fix hang when sched_entity released
2021-02-16 17:07 [PATCH] drm/scheduler: Fix hang when sched_entity released Andrey Grodzovsky
@ 2021-02-17 15:53 ` Andrey Grodzovsky
2021-02-17 21:32 ` Christian König
1 sibling, 0 replies; 6+ messages in thread
From: Andrey Grodzovsky @ 2021-02-17 15:53 UTC (permalink / raw)
To: dri-devel, amd-gfx; +Cc: ckoenig.leichtzumerken
Ping
Andrey
On 2/16/21 12:07 PM, Andrey Grodzovsky wrote:
> Problem: If scheduler is already stopped by the time sched_entity
> is released and entity's job_queue not empty I encountred
> a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
> never becomes false.
>
> Fix: In drm_sched_fini detach all sched_entities from the
> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
> Also wakeup all those processes stuck in sched_entity flushing
> as the scheduler main thread which wakes them up is stopped by now.
>
> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 31 +++++++++++++++++++++++++++++++
> 1 file changed, 31 insertions(+)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 908b0b5..11abf5d 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
> */
> void drm_sched_fini(struct drm_gpu_scheduler *sched)
> {
> + int i;
> + struct drm_sched_entity *s_entity;
> if (sched->thread)
> kthread_stop(sched->thread);
>
> + /* Detach all sched_entites from this scheduler once it's stopped */
> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
> + struct drm_sched_rq *rq = &sched->sched_rq[i];
> +
> + if (!rq)
> + continue;
> +
> + /* Loop this way because rq->lock is taken in drm_sched_rq_remove_entity */
> + spin_lock(&rq->lock);
> + while ((s_entity = list_first_entry_or_null(&rq->entities,
> + struct drm_sched_entity,
> + list))) {
> + spin_unlock(&rq->lock);
> + drm_sched_rq_remove_entity(rq, s_entity);
> +
> + /* Mark as stopped to reject adding to any new rq */
> + spin_lock(&s_entity->rq_lock);
> + s_entity->stopped = true;
> + spin_unlock(&s_entity->rq_lock);
> +
> + spin_lock(&rq->lock);
> + }
> + spin_unlock(&rq->lock);
> +
> + }
> +
> + /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
> + wake_up_all(&sched->job_scheduled);
> +
> /* Confirm no work left behind accessing device structures */
> cancel_delayed_work_sync(&sched->work_tdr);
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/scheduler: Fix hang when sched_entity released
2021-02-16 17:07 [PATCH] drm/scheduler: Fix hang when sched_entity released Andrey Grodzovsky
2021-02-17 15:53 ` Andrey Grodzovsky
@ 2021-02-17 21:32 ` Christian König
2021-02-17 21:36 ` Andrey Grodzovsky
1 sibling, 1 reply; 6+ messages in thread
From: Christian König @ 2021-02-17 21:32 UTC (permalink / raw)
To: Andrey Grodzovsky, dri-devel, amd-gfx
Am 16.02.21 um 18:07 schrieb Andrey Grodzovsky:
> Problem: If scheduler is already stopped by the time sched_entity
> is released and entity's job_queue not empty I encountred
> a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
> never becomes false.
>
> Fix: In drm_sched_fini detach all sched_entities from the
> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
> Also wakeup all those processes stuck in sched_entity flushing
> as the scheduler main thread which wakes them up is stopped by now.
>
> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
> ---
> drivers/gpu/drm/scheduler/sched_main.c | 31 +++++++++++++++++++++++++++++++
> 1 file changed, 31 insertions(+)
>
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
> index 908b0b5..11abf5d 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
> */
> void drm_sched_fini(struct drm_gpu_scheduler *sched)
> {
> + int i;
> + struct drm_sched_entity *s_entity;
> if (sched->thread)
> kthread_stop(sched->thread);
>
> + /* Detach all sched_entites from this scheduler once it's stopped */
> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
> + struct drm_sched_rq *rq = &sched->sched_rq[i];
> +
> + if (!rq)
> + continue;
> +
> + /* Loop this way because rq->lock is taken in drm_sched_rq_remove_entity */
> + spin_lock(&rq->lock);
> + while ((s_entity = list_first_entry_or_null(&rq->entities,
> + struct drm_sched_entity,
> + list))) {
> + spin_unlock(&rq->lock);
> + drm_sched_rq_remove_entity(rq, s_entity);
> +
> + /* Mark as stopped to reject adding to any new rq */
> + spin_lock(&s_entity->rq_lock);
> + s_entity->stopped = true;
Why not marking it as stopped and then removing it?
Regards,
Christian.
> + spin_unlock(&s_entity->rq_lock);
> +
> + spin_lock(&rq->lock);
> + }
> + spin_unlock(&rq->lock);
> +
> + }
> +
> + /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
> + wake_up_all(&sched->job_scheduled);
> +
> /* Confirm no work left behind accessing device structures */
> cancel_delayed_work_sync(&sched->work_tdr);
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/scheduler: Fix hang when sched_entity released
2021-02-17 21:32 ` Christian König
@ 2021-02-17 21:36 ` Andrey Grodzovsky
2021-02-17 21:37 ` Christian König
0 siblings, 1 reply; 6+ messages in thread
From: Andrey Grodzovsky @ 2021-02-17 21:36 UTC (permalink / raw)
To: Christian König, dri-devel, amd-gfx
On 2/17/21 4:32 PM, Christian König wrote:
> Am 16.02.21 um 18:07 schrieb Andrey Grodzovsky:
>> Problem: If scheduler is already stopped by the time sched_entity
>> is released and entity's job_queue not empty I encountred
>> a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
>> never becomes false.
>>
>> Fix: In drm_sched_fini detach all sched_entities from the
>> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
>> Also wakeup all those processes stuck in sched_entity flushing
>> as the scheduler main thread which wakes them up is stopped by now.
>>
>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
>> ---
>> drivers/gpu/drm/scheduler/sched_main.c | 31 +++++++++++++++++++++++++++++++
>> 1 file changed, 31 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
>> b/drivers/gpu/drm/scheduler/sched_main.c
>> index 908b0b5..11abf5d 100644
>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>> @@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
>> */
>> void drm_sched_fini(struct drm_gpu_scheduler *sched)
>> {
>> + int i;
>> + struct drm_sched_entity *s_entity;
>> if (sched->thread)
>> kthread_stop(sched->thread);
>> + /* Detach all sched_entites from this scheduler once it's stopped */
>> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN; i--) {
>> + struct drm_sched_rq *rq = &sched->sched_rq[i];
>> +
>> + if (!rq)
>> + continue;
>> +
>> + /* Loop this way because rq->lock is taken in
>> drm_sched_rq_remove_entity */
>> + spin_lock(&rq->lock);
>> + while ((s_entity = list_first_entry_or_null(&rq->entities,
>> + struct drm_sched_entity,
>> + list))) {
>> + spin_unlock(&rq->lock);
>> + drm_sched_rq_remove_entity(rq, s_entity);
>> +
>> + /* Mark as stopped to reject adding to any new rq */
>> + spin_lock(&s_entity->rq_lock);
>> + s_entity->stopped = true;
>
> Why not marking it as stopped and then removing it?
>
> Regards,
> Christian.
You mean just reverse the order of operations here to prevent a race where
someone adding it again to rq before marking it as stopped ?
Andrey
>
>> + spin_unlock(&s_entity->rq_lock);
>> +
>> + spin_lock(&rq->lock);
>> + }
>> + spin_unlock(&rq->lock);
>> +
>> + }
>> +
>> + /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
>> + wake_up_all(&sched->job_scheduled);
>> +
>> /* Confirm no work left behind accessing device structures */
>> cancel_delayed_work_sync(&sched->work_tdr);
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/scheduler: Fix hang when sched_entity released
2021-02-17 21:36 ` Andrey Grodzovsky
@ 2021-02-17 21:37 ` Christian König
2021-02-17 21:37 ` Andrey Grodzovsky
0 siblings, 1 reply; 6+ messages in thread
From: Christian König @ 2021-02-17 21:37 UTC (permalink / raw)
To: Andrey Grodzovsky, dri-devel, amd-gfx
Am 17.02.21 um 22:36 schrieb Andrey Grodzovsky:
>
> On 2/17/21 4:32 PM, Christian König wrote:
>> Am 16.02.21 um 18:07 schrieb Andrey Grodzovsky:
>>> Problem: If scheduler is already stopped by the time sched_entity
>>> is released and entity's job_queue not empty I encountred
>>> a hang in drm_sched_entity_flush. This is because
>>> drm_sched_entity_is_idle
>>> never becomes false.
>>>
>>> Fix: In drm_sched_fini detach all sched_entities from the
>>> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
>>> Also wakeup all those processes stuck in sched_entity flushing
>>> as the scheduler main thread which wakes them up is stopped by now.
>>>
>>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
>>> ---
>>> drivers/gpu/drm/scheduler/sched_main.c | 31
>>> +++++++++++++++++++++++++++++++
>>> 1 file changed, 31 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
>>> b/drivers/gpu/drm/scheduler/sched_main.c
>>> index 908b0b5..11abf5d 100644
>>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>>> @@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
>>> */
>>> void drm_sched_fini(struct drm_gpu_scheduler *sched)
>>> {
>>> + int i;
>>> + struct drm_sched_entity *s_entity;
>>> if (sched->thread)
>>> kthread_stop(sched->thread);
>>> + /* Detach all sched_entites from this scheduler once it's
>>> stopped */
>>> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >=
>>> DRM_SCHED_PRIORITY_MIN; i--) {
>>> + struct drm_sched_rq *rq = &sched->sched_rq[i];
>>> +
>>> + if (!rq)
>>> + continue;
>>> +
>>> + /* Loop this way because rq->lock is taken in
>>> drm_sched_rq_remove_entity */
>>> + spin_lock(&rq->lock);
>>> + while ((s_entity = list_first_entry_or_null(&rq->entities,
>>> + struct drm_sched_entity,
>>> + list))) {
>>> + spin_unlock(&rq->lock);
>>> + drm_sched_rq_remove_entity(rq, s_entity);
>>> +
>>> + /* Mark as stopped to reject adding to any new rq */
>>> + spin_lock(&s_entity->rq_lock);
>>> + s_entity->stopped = true;
>>
>> Why not marking it as stopped and then removing it?
>>
>> Regards,
>> Christian.
>
>
> You mean just reverse the order of operations here to prevent a race
> where someone adding it again to rq before marking it as stopped ?
Exactly that, yeah.
Christian.
>
> Andrey
>
>
>>
>>> + spin_unlock(&s_entity->rq_lock);
>>> +
>>> + spin_lock(&rq->lock);
>>> + }
>>> + spin_unlock(&rq->lock);
>>> +
>>> + }
>>> +
>>> + /* Wakeup everyone stuck in drm_sched_entity_flush for this
>>> scheduler */
>>> + wake_up_all(&sched->job_scheduled);
>>> +
>>> /* Confirm no work left behind accessing device structures */
>>> cancel_delayed_work_sync(&sched->work_tdr);
>>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] drm/scheduler: Fix hang when sched_entity released
2021-02-17 21:37 ` Christian König
@ 2021-02-17 21:37 ` Andrey Grodzovsky
0 siblings, 0 replies; 6+ messages in thread
From: Andrey Grodzovsky @ 2021-02-17 21:37 UTC (permalink / raw)
To: Christian König, dri-devel, amd-gfx
Will do.
Andrey
On 2/17/21 4:37 PM, Christian König wrote:
>
>
> Am 17.02.21 um 22:36 schrieb Andrey Grodzovsky:
>>
>> On 2/17/21 4:32 PM, Christian König wrote:
>>> Am 16.02.21 um 18:07 schrieb Andrey Grodzovsky:
>>>> Problem: If scheduler is already stopped by the time sched_entity
>>>> is released and entity's job_queue not empty I encountred
>>>> a hang in drm_sched_entity_flush. This is because drm_sched_entity_is_idle
>>>> never becomes false.
>>>>
>>>> Fix: In drm_sched_fini detach all sched_entities from the
>>>> scheduler's run queues. This will satisfy drm_sched_entity_is_idle.
>>>> Also wakeup all those processes stuck in sched_entity flushing
>>>> as the scheduler main thread which wakes them up is stopped by now.
>>>>
>>>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
>>>> ---
>>>> drivers/gpu/drm/scheduler/sched_main.c | 31 +++++++++++++++++++++++++++++++
>>>> 1 file changed, 31 insertions(+)
>>>>
>>>> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
>>>> b/drivers/gpu/drm/scheduler/sched_main.c
>>>> index 908b0b5..11abf5d 100644
>>>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>>>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>>>> @@ -897,9 +897,40 @@ EXPORT_SYMBOL(drm_sched_init);
>>>> */
>>>> void drm_sched_fini(struct drm_gpu_scheduler *sched)
>>>> {
>>>> + int i;
>>>> + struct drm_sched_entity *s_entity;
>>>> if (sched->thread)
>>>> kthread_stop(sched->thread);
>>>> + /* Detach all sched_entites from this scheduler once it's stopped */
>>>> + for (i = DRM_SCHED_PRIORITY_COUNT - 1; i >= DRM_SCHED_PRIORITY_MIN;
>>>> i--) {
>>>> + struct drm_sched_rq *rq = &sched->sched_rq[i];
>>>> +
>>>> + if (!rq)
>>>> + continue;
>>>> +
>>>> + /* Loop this way because rq->lock is taken in
>>>> drm_sched_rq_remove_entity */
>>>> + spin_lock(&rq->lock);
>>>> + while ((s_entity = list_first_entry_or_null(&rq->entities,
>>>> + struct drm_sched_entity,
>>>> + list))) {
>>>> + spin_unlock(&rq->lock);
>>>> + drm_sched_rq_remove_entity(rq, s_entity);
>>>> +
>>>> + /* Mark as stopped to reject adding to any new rq */
>>>> + spin_lock(&s_entity->rq_lock);
>>>> + s_entity->stopped = true;
>>>
>>> Why not marking it as stopped and then removing it?
>>>
>>> Regards,
>>> Christian.
>>
>>
>> You mean just reverse the order of operations here to prevent a race where
>> someone adding it again to rq before marking it as stopped ?
>
> Exactly that, yeah.
>
> Christian.
>
>>
>> Andrey
>>
>>
>>>
>>>> + spin_unlock(&s_entity->rq_lock);
>>>> +
>>>> + spin_lock(&rq->lock);
>>>> + }
>>>> + spin_unlock(&rq->lock);
>>>> +
>>>> + }
>>>> +
>>>> + /* Wakeup everyone stuck in drm_sched_entity_flush for this scheduler */
>>>> + wake_up_all(&sched->job_scheduled);
>>>> +
>>>> /* Confirm no work left behind accessing device structures */
>>>> cancel_delayed_work_sync(&sched->work_tdr);
>>>
>
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2021-02-17 21:37 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-16 17:07 [PATCH] drm/scheduler: Fix hang when sched_entity released Andrey Grodzovsky
2021-02-17 15:53 ` Andrey Grodzovsky
2021-02-17 21:32 ` Christian König
2021-02-17 21:36 ` Andrey Grodzovsky
2021-02-17 21:37 ` Christian König
2021-02-17 21:37 ` Andrey Grodzovsky
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).