Re: [PATCH 09/18] drm/amdgpu: fix KFD eviction fence enable_signaling path

["Christian König" <christian.koenig@amd.com>]

On 11/27/25 11:57, Philipp Stanner wrote:
> On Thu, 2025-11-13 at 15:51 +0100, Christian König wrote:
>> Calling dma_fence_is_signaled() here is illegal!
> 
> OK, but why is that patch in this series?

Because the next patch depends on it, otherwise the series won't compile.

My plan is to push the amdgpu patches through amd-staging-drm-next as soon as Alex rebased that branch on drm-next during the next cycle.

Regards,
Christian.

> 
> P.
> 
>>
>> Signed-off-by: Christian König <christian.koenig@amd.com>
>> ---
>>  drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c | 6 ------
>>  1 file changed, 6 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
>> index 1ef758ac5076..09c919f72b6c 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_fence.c
>> @@ -120,12 +120,6 @@ static bool amdkfd_fence_enable_signaling(struct dma_fence *f)
>>  {
>>  	struct amdgpu_amdkfd_fence *fence = to_amdgpu_amdkfd_fence(f);
>>  
>> -	if (!fence)
>> -		return false;
>> -
>> -	if (dma_fence_is_signaled(f))
>> -		return true;
>> -
>>  	if (!fence->svm_bo) {
>>  		if (!kgd2kfd_schedule_evict_and_restore_process(fence->mm, f))
>>  			return true;
>