[PATCH v7 13/16] drm/scheduler: Fix hang when sched_entity released
Andrey Grodzovsky
andrey.grodzovsky at amd.com
Tue May 18 20:56:56 UTC 2021
On 2021-05-18 2:48 p.m., Andrey Grodzovsky wrote:
>
>
> On 2021-05-18 2:13 p.m., Christian König wrote:
>>
>> Am 18.05.21 um 20:09 schrieb Andrey Grodzovsky:
>>> On 2021-05-18 2:02 p.m., Christian König wrote:
>>>> Am 18.05.21 um 19:43 schrieb Andrey Grodzovsky:
>>>>> On 2021-05-18 12:33 p.m., Christian König wrote:
>>>>>> Am 18.05.21 um 18:17 schrieb Andrey Grodzovsky:
>>>>>>>
>>>>>>>
>>>>>>> On 2021-05-18 11:15 a.m., Christian König wrote:
>>>>>>>> Am 18.05.21 um 17:03 schrieb Andrey Grodzovsky:
>>>>>>>>>
>>>>>>>>> On 2021-05-18 10:07 a.m., Christian König wrote:
>>>>>>>>>> In a separate discussion with Daniel we once more iterated
>>>>>>>>>> over the dma_resv requirements and I came to the conclusion
>>>>>>>>>> that this approach here won't work reliable.
>>>>>>>>>>
>>>>>>>>>> The problem is as following:
>>>>>>>>>> 1. device A schedules some rendering with into a buffer and
>>>>>>>>>> exports it as DMA-buf.
>>>>>>>>>> 2. device B imports the DMA-buf and wants to consume the
>>>>>>>>>> rendering, for the the fence of device A is replaced with a
>>>>>>>>>> new operation.
>>>>>>>>>> 3. device B is hot plugged and the new operation
>>>>>>>>>> canceled/newer scheduled.
>>>>>>>>>>
>>>>>>>>>> The problem is now that we can't do this since the operation
>>>>>>>>>> of device A is still running and by signaling our fences we
>>>>>>>>>> run into the problem of potential memory corruption.
>>>>>>>
>>>>>>> By signaling s_fence->finished of the canceled operation from the
>>>>>>> removed device B we in fact cause memory corruption for the
>>>>>>> uncompleted
>>>>>>> job still running on device A ? Because there is someone waiting to
>>>>>>> read write from the imported buffer on device B and he only waits
>>>>>>> for
>>>>>>> the s_fence->finished of device B we signaled
>>>>>>> in drm_sched_entity_kill_jobs ?
>>>>>>
>>>>>> Exactly that, yes.
>>>>>>
>>>>>> In other words when you have a dependency chain like A->B->C then
>>>>>> memory management only waits for C before freeing up the memory
>>>>>> for example.
>>>>>>
>>>>>> When C now signaled because the device is hot-plugged before A or
>>>>>> B are finished they are essentially accessing freed up memory.
>>>>>
>>>>> But didn't C imported the BO form B or A in this case ? Why would
>>>>> he be
>>>>> the one releasing that memory ? He would be just dropping his
>>>>> reference
>>>>> to the BO, no ?
>>>>
>>>> Well freeing the memory was just an example. The BO could also move
>>>> back to VRAM because of the dropped reference.
>>>>
>>>>> Also in the general case,
>>>>> drm_sched_entity_fini->drm_sched_entity_kill_jobs which is
>>>>> the one who signals the 'C' fence with error code are as far
>>>>> as I looked called from when the user of that BO is stopping
>>>>> the usage anyway (e.g. drm_driver.postclose callback for when use
>>>>> process closes his device file) who would then access and corrupt
>>>>> the exported memory on device A where the job hasn't completed yet ?
>>>>
>>>> Key point is that memory management only waits for the last added
>>>> fence, that is the design of the dma_resv object. How that happens
>>>> is irrelevant.
>>>>
>>>> Because of this we at least need to wait for all dependencies of the
>>>> job before signaling the fence, even if we cancel the job for some
>>>> reason.
>>>>
>>>> Christian.
>>>
>>> Would this be the right way to do it ?
>>
>> Yes, it is at least a start. Question is if we can wait blocking here
>> or not.
>>
>> We install a callback a bit lower to avoid blocking, so I'm pretty
>> sure that won't work as expected.
>>
>> Christian.
>
> I can't see why this would create problems, as long as the dependencies
> complete or force competed if they are from same device (extracted) but
> on a different ring then looks to me it should work. I will give it
> a try.
>
> Andrey
Well, i gave it a try with my usual tests like IGT hot unplug while
rapid command submissions and unplug the card while glxgears runs
with DRI_PRIME=1 and haven't seen issues.
Andrey
>
>>
>>>
>>> diff --git a/drivers/gpu/drm/scheduler/sched_entity.c
>>> b/drivers/gpu/drm/scheduler/sched_entity.c
>>> index 2e93e881b65f..10f784874b63 100644
>>> --- a/drivers/gpu/drm/scheduler/sched_entity.c
>>> +++ b/drivers/gpu/drm/scheduler/sched_entity.c
>>> @@ -223,10 +223,14 @@ static void drm_sched_entity_kill_jobs(struct
>>> drm_sched_entity *entity)
>>> {
>>> struct drm_sched_job *job;
>>> int r;
>>> + struct dma_fence *f;
>>>
>>> while ((job =
>>> to_drm_sched_job(spsc_queue_pop(&entity->job_queue)))) {
>>> struct drm_sched_fence *s_fence = job->s_fence;
>>>
>>> + while (f = sched->ops->dependency(sched_job, entity))
>>> + dma_fence_wait(f);
>>> +
>>> drm_sched_fence_scheduled(s_fence);
>>> dma_fence_set_error(&s_fence->finished, -ESRCH);
>>>
>>> Andrey
>>>
>>>
>>>
>>>>
>>>>>
>>>>> Andrey
>>>>>
>>>>>>
>>>>>> Christian.
>>>>>>
>>>>>>>
>>>>>>> Andrey
>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> I am not sure this problem you describe above is related to
>>>>>>>>> this patch.
>>>>>>>>
>>>>>>>> Well it is kind of related.
>>>>>>>>
>>>>>>>>> Here we purely expand the criteria for when sched_entity is
>>>>>>>>> considered idle in order to prevent a hang on device remove.
>>>>>>>>
>>>>>>>> And exactly that is problematic. See the jobs on the entity need
>>>>>>>> to cleanly wait for their dependencies before they can be
>>>>>>>> completed.
>>>>>>>>
>>>>>>>> drm_sched_entity_kill_jobs() is also not handling that correctly
>>>>>>>> at the moment, we only wait for the last scheduled fence but not
>>>>>>>> for the dependencies of the job.
>>>>>>>>
>>>>>>>>> Were you addressing the patch from yesterday in which you
>>>>>>>>> commented
>>>>>>>>> that you found a problem with how we finish fences ? It was
>>>>>>>>> '[PATCH v7 12/16] drm/amdgpu: Fix hang on device removal.'
>>>>>>>>>
>>>>>>>>> Also, in the patch series as it is now we only signal HW fences
>>>>>>>>> for the
>>>>>>>>> extracted device B, we are not touching any other fences. In
>>>>>>>>> fact as you
>>>>>>>>> may remember, I dropped all new logic to forcing fence
>>>>>>>>> completion in
>>>>>>>>> this patch series and only call
>>>>>>>>> amdgpu_fence_driver_force_completion
>>>>>>>>> for the HW fences of the extracted device as it's done today
>>>>>>>>> anyway.
>>>>>>>>
>>>>>>>> Signaling hardware fences is unproblematic since they are
>>>>>>>> emitted when the software scheduling is already completed.
>>>>>>>>
>>>>>>>> Christian.
>>>>>>>>
>>>>>>>>>
>>>>>>>>> Andrey
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Not sure how to handle that case. One possibility would be to
>>>>>>>>>> wait for all dependencies of unscheduled jobs before signaling
>>>>>>>>>> their fences as canceled.
>>>>>>>>>>
>>>>>>>>>> Christian.
>>>>>>>>>>
>>>>>>>>>> Am 12.05.21 um 16:26 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 0249c7450188..2e93e881b65f 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 8d1211e87101..a2a953693b45 100644
>>>>>>>>>>> --- a/drivers/gpu/drm/scheduler/sched_main.c
>>>>>>>>>>> +++ b/drivers/gpu/drm/scheduler/sched_main.c
>>>>>>>>>>> @@ -898,9 +898,33 @@ 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);
>>>>>>>>>>
>>>>>>>>
>>>>>>
>>>>
>>
More information about the amd-gfx
mailing list