On 2022-12-12 18:44, Luben Tuikov wrote:
> On 2022-12-12 14:19, Christian König wrote:
>> Am 12.12.22 um 18:48 schrieb Luben Tuikov:
>>> Fix amdgpu_bo_validate_size() to check whether the TTM domain manager for 
>>> the
>>> requested memory exists, and to allow for non-exclusive domain allocations, 
>>> as
>>> there would be if the domain is a mask, e.g. AMDGPU_GEM_DOMAIN_VRAM |
>>> AMDGPU_GEM_DOMAIN_GTT.
>>>
>>> Cc: Alex Deucher <[email protected]>
>>> Cc: Christian König <[email protected]>
>>> Signed-off-by: Luben Tuikov <[email protected]>
>>> ---
>>>   drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 19 +++++++------------
>>>   1 file changed, 7 insertions(+), 12 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c 
>>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>> index fd3ab4b5e5bb1f..e0f103f0ec2178 100644
>>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
>>> @@ -448,31 +448,26 @@ static bool amdgpu_bo_validate_size(struct 
>>> amdgpu_device *adev,
>>>   
>>>     /*
>>>      * If GTT is part of requested domains the check must succeed to
>>> -    * allow fall back to GTT
>>> +    * allow fall back to GTT.
>>> +    *
>>> +    * Note that allocations can request from either domain. For
>>> +    * this reason, check either in non-exclusive way, and if
>>> +    * neither satisfies, fail the validation.
>>
>> That's not correct, the original logic was completely intentional.
>>
>> If both VRAM and GTT are specified it's valid if the size fits only into 
>> GTT.
> 
> Given that this patch fixes a kernel oops, should this patch then fail the 
> validation,
> i.e. return false?
> 
> This would then fail, in amdgpu_ttm_reserve_tmr():
> 
>       ret = amdgpu_bo_create_kernel_at(adev,
>                               adev->gmc.real_vram_size - 
> adev->mman.discovery_tmr_size,
>                               adev->mman.discovery_tmr_size,
>                               AMDGPU_GEM_DOMAIN_VRAM |
>                               AMDGPU_GEM_DOMAIN_GTT,
>                               &adev->mman.discovery_memory,
>                               NULL);
> 

I've rewritten the patch to be minimal and just check that the memory manager
is not NULL. The patch follows this email.

Regards,
Luben


Reply via email to