[PATCH] drm/amd/display: Fix deadlock with display during hanged ring recovery.

[Kazlauskas, Nicholas]

On 2/14/19 12:47 PM, Grodzovsky, Andrey wrote:
> 
> On 2/14/19 11:57 AM, Kazlauskas, Nicholas wrote:
>> On 2/14/19 11:47 AM, Grodzovsky, Andrey wrote:
>>> On 2/14/19 11:07 AM, Michel Dänzer wrote:
>>>> On 2019-02-14 4:54 p.m., Kazlauskas, Nicholas wrote:
>>>>> On 2/14/19 10:42 AM, Grodzovsky, Andrey wrote:
>>>>>> On 2/14/19 4:05 AM, Christian König wrote:
>>>>>>> Am 13.02.19 um 19:58 schrieb Andrey Grodzovsky:
>>>>>>>> When ring hang happens amdgpu_dm_commit_planes during flip is holding
>>>>>>>> the BO reserved and then stack waiting for fences to signal in
>>>>>>>> reservation_object_wait_timeout_rcu (which won't signal because there
>>>>>>>> was a hnag). Then when we try to shutdown display block during reset
>>>>>>>> recovery from drm_atomic_helper_suspend we also try to reserve the BO
>>>>>>>> from dm_plane_helper_cleanup_fb ending in deadlock.
>>>>>>>> Also remove useless WARN_ON
>>>>>>> Well it is good that you pointed this out, but there are more problems
>>>>>>> than just waiting wile the BO is reserved here.
>>>>>>>
>>>>>>>> Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky at amd.com>
>>>>>>>> ---
>>>>>>>>        drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 19
>>>>>>>> +++++++++++++------
>>>>>>>>        1 file changed, 13 insertions(+), 6 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>>>>>>>> b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>>>>>>>> index acc4ff8..f8dec36 100644
>>>>>>>> --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>>>>>>>> +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
>>>>>>>> @@ -4802,14 +4802,21 @@ static void amdgpu_dm_commit_planes(struct
>>>>>>>> drm_atomic_state *state,
>>>>>>>>                     */
>>>>>>>>                    abo = gem_to_amdgpu_bo(fb->obj[0]);
>>>>>>>>                    r = amdgpu_bo_reserve(abo, true);
>>>>>>> Well why do we reserve the BO in the first place? As the name
>>>>>>> indicates reservation_object_wait_timeout_rcu() just uses rcu to wait
>>>>>>> for the BO to be idle, no need to actually reserve it at all.
>>>>>> That a good point, I honestly can't remember any more why it's here...
>>>>>> It does look unnecessary given that we are not validating the BO here.
>>>>> I think we reserve it to grab the tiling flags to update the plane
>>>>> address (and some other bits of the plane state with a patch I'm
>>>>> currently working on). I'm guessing that we still need the reservation
>>>>> so nothing else tries to grab it after it becomes idle, but I'm not
>>>>> overly familiar with that bit.
>>>> The BO is already pinned at this point, isn't it? That's enough to
>>>> prevent it from moving, no reservation needed.
>>>
>>> Yes, agree.
>>>
>>> Andrey
>>>
>>>
>>>>
>> I'm not worried about the buffer moving so much as I am about userspace
>> changing the tiling flags by calling the IOCTL. Can that happen when
>> it's pinned?
>>
>> Nicholas Kazlauskas
> 
> In any case this has nothing to do with reserving the BO.
> 
> Andrey
> 
> 

Technically it does? See the function:

void amdgpu_bo_get_tiling_flags(struct amdgpu_bo *bo, u64 *tiling_flags)
{
	lockdep_assert_held(&bo->tbo.resv->lock.base);

	if (tiling_flags)
		*tiling_flags = bo->tiling_flags;
}

Nicholas Kazlauskas