Re: [PATCH] virtio: remove the excess virtio features check

["Michael S. Tsirkin" <mst@redhat.com>]

On Wed, Nov 09, 2022 at 07:10:21PM +0800, Xuan Zhuo wrote:
> In virtio_queue_enable(), we checked virtio feature VIRTIO_F_VERSION_1.
> 
> This check is not necessary, and conflict with SeaBIOS. The problem
> appeared in SeaBIOS. But we also remove this check.
> 
> Link: https://www.mail-archive.com/qemu-devel@nongnu.org/msg920538.html
> Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
> ---
>  hw/virtio/virtio.c | 5 -----
>  1 file changed, 5 deletions(-)
> 
> diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
> index 9683b2e158..701e23ea6a 100644
> --- a/hw/virtio/virtio.c
> +++ b/hw/virtio/virtio.c
> @@ -2499,11 +2499,6 @@ void virtio_queue_enable(VirtIODevice *vdev, uint32_t queue_index)
>  {
>      VirtioDeviceClass *k = VIRTIO_DEVICE_GET_CLASS(vdev);
>  
> -    if (!virtio_vdev_has_feature(vdev, VIRTIO_F_VERSION_1)) {
> -        error_report("queue_enable is only suppported in devices of virtio "
> -                     "1.0 or later.");
> -    }
> -
>      if (k->queue_enable) {
>          k->queue_enable(vdev, queue_index);
>      }

Well this warning turned out to be actually useful.
Let's see whether we can fix seabios in time for release.
If yes I would just make it LOG_GUEST_ERR and limit to
latest machine type but not drop completely.

> -- 
> 2.32.0.3.g01195cf9f