[PATCH 3/8] drm/amdgpu: fix user fence CS check

Christian König christian.koenig at amd.com
Fri Sep 16 18:24:15 UTC 2022


Am 16.09.22 um 20:01 schrieb Alex Deucher:
> On Fri, Sep 16, 2022 at 5:09 AM Christian König
> <ckoenig.leichtzumerken at gmail.com> wrote:
>> An offset of zero is valid, check if the BO is present or not.
>>
>> Signed-off-by: Christian König <christian.koenig at amd.com>
> Since this is a bug fix, should we make the first patch in the series?

Good point, going to re-arrange this before pushing.

Christian.

>
> Alex
>
>> ---
>>   drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
>> index 5c37dde97ff2..265ed2118a80 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
>> @@ -313,7 +313,7 @@ static int amdgpu_cs_p2_ib(struct amdgpu_cs_parser *p,
>>
>>          ring = to_amdgpu_ring(entity->rq->sched);
>>          /* MM engine doesn't support user fences */
>> -       if (p->job->uf_addr && ring->funcs->no_user_fence)
>> +       if (p->uf_entry.tv.bo && ring->funcs->no_user_fence)
>>                  return -EINVAL;
>>
>>          if (chunk_ib->ip_type == AMDGPU_HW_IP_GFX &&
>> --
>> 2.25.1
>>



More information about the amd-gfx mailing list