[PATCH 1/3] drm/amdgpu: only sync on the dependent read fences for user submission

[Christian König]

On 4/25/25 09:20, Prike Liang wrote:
> The driver doesn't want to sync on the DMA_RESV_USAGE_BOOKKEEP

The driver doesn't want to *implicitely* sync on the...

> usage fences, so here only return and sync the dependent read
> fences.
> 
> Signed-off-by: Prike Liang <Prike.Liang at amd.com>
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c | 5 ++---
>  1 file changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
> index 5576ed0b508f..4e1d30ecb6cc 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
> @@ -249,9 +249,8 @@ int amdgpu_sync_resv(struct amdgpu_device *adev, struct amdgpu_sync *sync,
>  
>  	if (resv == NULL)
>  		return -EINVAL;
> -
> -	/* TODO: Use DMA_RESV_USAGE_READ here */
> -	dma_resv_for_each_fence(&cursor, resv, DMA_RESV_USAGE_BOOKKEEP, f) {
> +	/*Only return and sync the fences of usage <= DMA_RESV_USAGE_READ.*/

That comment just repeats the obvious. We need to comment why the code does something and not what it does.

Write something like /* Implicitly sync only to KERNEL, WRITE and READ fences */

Regards,
Christian.

> +	dma_resv_for_each_fence(&cursor, resv, DMA_RESV_USAGE_READ, f) {
>  		dma_fence_chain_for_each(f, f) {
>  			struct dma_fence *tmp = dma_fence_chain_contained(f);
>