[PATCH v5 15/27] drm/scheduler: Fix hang when sched_entity released

[Andrey Grodzovsky]

On 2021-04-29 3:18 a.m., Christian König wrote:
> I need to take another look at this part when I don't have a massive 
> headache any more.
> 
> Maybe split the patch set up into different parts, something like:
> 1. Adding general infrastructure.
> 2. Making sure all memory is unpolated.
> 3. Job and fence handling

I am not sure you mean this patch here, maybe another one ?
Also note you already RBed it.

Andrey

> 
> Christian.
> 
> Am 28.04.21 um 17:11 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.
>>
>> v2:
>> Reverse order of drm_sched_rq_remove_entity and marking
>> s_entity as stopped to prevent reinserion back to rq due
>> to race.
>>
>> v3:
>> Drop drm_sched_rq_remove_entity, only modify entity->stopped
>> and check for it in drm_sched_entity_is_idle
>>
>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky at amd.com>
>> Reviewed-by: Christian König <christian.koenig at amd.com>
>> ---
>>   drivers/gpu/drm/scheduler/sched_entity.c |  3 ++-
>>   drivers/gpu/drm/scheduler/sched_main.c   | 24 ++++++++++++++++++++++++
>>   2 files changed, 26 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/scheduler/sched_entity.c 
>> b/drivers/gpu/drm/scheduler/sched_entity.c
>> index f0790e9471d1..cb58f692dad9 100644
>> --- a/drivers/gpu/drm/scheduler/sched_entity.c
>> +++ b/drivers/gpu/drm/scheduler/sched_entity.c
>> @@ -116,7 +116,8 @@ static bool drm_sched_entity_is_idle(struct 
>> drm_sched_entity *entity)
>>       rmb(); /* for list_empty to work without lock */
>>       if (list_empty(&entity->list) ||
>> -        spsc_queue_count(&entity->job_queue) == 0)
>> +        spsc_queue_count(&entity->job_queue) == 0 ||
>> +        entity->stopped)
>>           return true;
>>       return false;
>> diff --git a/drivers/gpu/drm/scheduler/sched_main.c 
>> b/drivers/gpu/drm/scheduler/sched_main.c
>> index 908b0b56032d..ba087354d0a8 100644
>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>> @@ -897,9 +897,33 @@ EXPORT_SYMBOL(drm_sched_init);
>>    */
>>   void drm_sched_fini(struct drm_gpu_scheduler *sched)
>>   {
>> +    struct drm_sched_entity *s_entity;
>> +    int i;
>> +
>>       if (sched->thread)
>>           kthread_stop(sched->thread);
>> +    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;
>> +
>> +        spin_lock(&rq->lock);
>> +        list_for_each_entry(s_entity, &rq->entities, list)
>> +            /*
>> +             * Prevents reinsertion and marks job_queue as idle,
>> +             * it will removed from rq in drm_sched_entity_fini
>> +             * eventually
>> +             */
>> +            s_entity->stopped = true;
>> +        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);
>