Re: [PATCH] drm/virtio: add definition for venus capset

[Huang Rui <ray.huang@amd.com>]

On Sat, Dec 16, 2023 at 12:45:32AM +0800, Dmitry Osipenko wrote:
> On 11/19/23 06:46, Dmitry Osipenko wrote:
> > On 9/21/23 00:16, Dmitry Osipenko wrote:
> >> On 9/15/23 13:59, Huang Rui wrote:
> >>> This definition is used fro qemu, and qemu imports this marco in the
> >>> headers to enable venus for virtio gpu. So it should add it even kernel
> >>> doesn't use this.
> >>>
> >>> Signed-off-by: Huang Rui <ray.huang@amd.com>
> >>> ---
> >>>
> >>> Hi all,
> >>>
> >>> We would like to add a new definition for venus capset, it will be used for
> >>> qemu. Please see details on below discussion:
> >>>
> >>> https://lore.kernel.org/qemu-devel/b82982aa-5b9e-481e-9491-b9313877bcaa@daynix.com/
> >>>
> >>> Thanks,
> >>> Ray
> >>>
> >>>  include/uapi/linux/virtio_gpu.h | 2 ++
> >>>  1 file changed, 2 insertions(+)
> >>>
> >>> diff --git a/include/uapi/linux/virtio_gpu.h b/include/uapi/linux/virtio_gpu.h
> >>> index f556fde07b76..0e21f3998108 100644
> >>> --- a/include/uapi/linux/virtio_gpu.h
> >>> +++ b/include/uapi/linux/virtio_gpu.h
> >>> @@ -309,6 +309,8 @@ struct virtio_gpu_cmd_submit {
> >>>  
> >>>  #define VIRTIO_GPU_CAPSET_VIRGL 1
> >>>  #define VIRTIO_GPU_CAPSET_VIRGL2 2
> >>> +/* 3 is reserved for gfxstream */
> >>> +#define VIRTIO_GPU_CAPSET_VENUS 4
> >>
> >> Could you please add all other capsets, so we won't needed to do it
> >> again in the future
> > 
> > I've opened request to update virtio-spec with the corrected/updated
> > capsets https://github.com/oasis-tcs/virtio-spec/issues/182. I'm
> > expecting that it will take some time until spec change will be merged
> > and now leaning to apply this v1 patch to not hold the Venus work.
> > 
> > Gerd, do you have objections? R-b/ack?
> 
> Applied patch to misc-next with edited commit message. Updating spec
> taking much time, not worth to hold this change longer. We'll add the
> rest of capsets later on. Thanks, Rui!
> 

Thank you, Dmitry! I will update corresponding patch in qemu, and send v6
qemu patches.

Best Regards,
Ray