[PATCH 1/2] drm/amdgpu: sync ring type with drm hw_ip type

[Christian König]

Am 31.03.20 um 14:54 schrieb Nirmoy Das:
> Move AMDGPU_RING_TYPE_KIQ at the end of amdgpu_ring_type enum
> to sync amdgpu_ring_type values with AMDGPU_HW_IP_*
>
> Signed-off-by: Nirmoy Das <nirmoy.das at amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> index 080024d21d3e..121041369c42 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
> @@ -52,11 +52,11 @@ enum amdgpu_ring_type {
>   	AMDGPU_RING_TYPE_SDMA,
>   	AMDGPU_RING_TYPE_UVD,
>   	AMDGPU_RING_TYPE_VCE,
> -	AMDGPU_RING_TYPE_KIQ,
>   	AMDGPU_RING_TYPE_UVD_ENC,
>   	AMDGPU_RING_TYPE_VCN_DEC,
>   	AMDGPU_RING_TYPE_VCN_ENC,
> -	AMDGPU_RING_TYPE_VCN_JPEG
> +	AMDGPU_RING_TYPE_VCN_JPEG,
> +	AMDGPU_RING_TYPE_KIQ

That is most likely not sufficient to keep the two enums in sync since 
we are getting even more kernel internal ring types in the future.

We either need some form of static assert or your original approach of 
defining AMDGPU_RING_TYPE_SDMA = AMDGPU_HW_IP_SDMA.....

Christian.

>   };
>
>   struct amdgpu_device;
> --
> 2.25.1
>
> _______________________________________________
> amd-gfx mailing list
> amd-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/amd-gfx