On 6/16/26 13:30, SHANMUGAM, SRINIVASAN wrote:
> AMD General
>
> Hi Christian,
>
>> -----Original Message-----
>> From: Koenig, Christian <[email protected]>
>> Sent: Tuesday, June 16, 2026 2:28 PM
>> To: SHANMUGAM, SRINIVASAN <[email protected]>;
>> Deucher, Alexander <[email protected]>
>> Cc: [email protected]
>> Subject: Re: [PATCH 2/5] drm/amdgpu: Signal QUEUE_RESET EVENTFD
>> notifications
>>
>>
>>
>> On 6/12/26 07:52, Srinivasan Shanmugam wrote:
>>> Signal QUEUE_RESET EVENTFD subscriptions when hung user queues are
>>> detected.
>>>
>>> The queue reset path already identifies the affected user queues. Use
>>> those queue objects to wake up the matching EVENTFD subscribers.
>>>
>>> EVENTFD remains notification-only.
>>>
>>> Cc: Alex Deucher <[email protected]>
>>> Cc: Christian König <[email protected]>
>>> Signed-off-by: Srinivasan Shanmugam <[email protected]>
>>> ---
>>> drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 9 +++++++++
>>> 1 file changed, 9 insertions(+)
>>>
>>> diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
>>> b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
>>> index e9189f07c6dc..ee8dbd70e1c8 100644
>>> --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
>>> +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
>>> @@ -238,7 +238,16 @@ static int mes_userq_detect_and_reset(struct
>> amdgpu_device *adev,
>>> if (queue->queue_type == queue_type) {
>>> for (i = 0; i < hung_db_num; i++) {
>>> if (queue->doorbell_index ==
>>> db_array[i]) {
>>
>> That code is extremely questionable to begin with.
>>
>> My suggestion is to clean that up first and add a function in amdgpu_userq.c
>> which
>> gets the doorbell index and does the necessary handling.
>>
>> E.g. actually lock the doorbell XA, lookup the queue and then do the right
>> handling.
>
> Just one clarification pls regarding the QUEUE_RESET feedback.
>
> Would you prefer the new helper in amdgpu_userq.c to only perform the
> doorbell_index -> queue lookup and return the queue, with the caller handling
> the reset/eventfd logic?
No the handling of the reset logic should be full in amdgpu_userq.c
>
> Or should the helper encapsulate the full handling for a hung doorbell (queue
> state updates, force completion, EVENTFD signaling, etc.)?
Yes, exactly that.
Thanks,
Christian.
>
> Thanks,
> Srini
>
>>
>> Regards,
>> Christian.
>>
>
