Re: [PATCH v3 2/4] vhost-vsock-pci: force virtio version 1

[Cornelia Huck <cohuck@redhat.com>]

On Fri, 18 Sep 2020 09:47:08 +0200
Stefano Garzarella <sgarzare@redhat.com> wrote:

> Commit 9b3a35ec82 ("virtio: verify that legacy support is not
> accidentally on") added a safety check that requires to set
> 'disable-legacy=on' on vhost-vsock-pci device:
> 
>     $ ./qemu-system-x86_64 ... -device vhost-vsock-pci,guest-cid=5
>         qemu-system-x86_64: -device vhost-vsock-pci,guest-cid=5:
>         device is modern-only, use disable-legacy=on
> 
> virtio-vsock was introduced after the release of VIRTIO 1.0
> specifications, so it should be 'modern-only'.
> In addition Cornelia verified that forcing a legacy mode on
> vhost-vsock-pci device using x86-64 host and s390x guest, so with
> different endianness, produces strange behaviours.
> 
> This patch forces virtio version 1 and removes the 'transitional_name'
> property removing the need to specify 'disable-legacy=on' on
> vhost-vsock-pci device.
> 
> To avoid migration issues, we force virtio version 1 only when
> legacy check is enabled in the new machine types (>= 5.1).
> 
> Cc: qemu-stable@nongnu.org
> Reported-by: Qian Cai <caiqian@redhat.com>
> Reported-by: Qinghua Cheng <qcheng@redhat.com>
> Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1868449
> Suggested-by: Cornelia Huck <cohuck@redhat.com>
> Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
> ---
> v3:
>  - forced virtio version 1 only with new machine types
> v2:
>  - fixed commit message [Cornelia]
> ---
>  hw/virtio/vhost-vsock-pci.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/hw/virtio/vhost-vsock-pci.c b/hw/virtio/vhost-vsock-pci.c
> index e56067b427..205da8d1f5 100644
> --- a/hw/virtio/vhost-vsock-pci.c
> +++ b/hw/virtio/vhost-vsock-pci.c
> @@ -44,6 +44,15 @@ static void vhost_vsock_pci_realize(VirtIOPCIProxy *vpci_dev, Error **errp)
>  {
>      VHostVSockPCI *dev = VHOST_VSOCK_PCI(vpci_dev);
>      DeviceState *vdev = DEVICE(&dev->vdev);
> +    VirtIODevice *virtio_dev = VIRTIO_DEVICE(vdev);
> +
> +    /*
> +     * To avoid migration issues, we force virtio version 1 only when
> +     * legacy check is enabled in the new machine types (>= 5.1).
> +     */
> +    if (!virtio_legacy_check_disabled(virtio_dev)) {
> +        virtio_pci_force_virtio_1(vpci_dev);
> +    }
>  
>      qdev_realize(vdev, BUS(&vpci_dev->bus), errp);
>  }
> @@ -73,7 +82,6 @@ static void vhost_vsock_pci_instance_init(Object *obj)
>  static const VirtioPCIDeviceTypeInfo vhost_vsock_pci_info = {
>      .base_name             = TYPE_VHOST_VSOCK_PCI,
>      .generic_name          = "vhost-vsock-pci",
> -    .transitional_name     = "vhost-vsock-pci-transitional",

Hm... this means that vhost-vsock-pci-transitional won't work on compat
machines, which could also lead to migration compatibility issues (I
think?)

Is this way of specifying the device sufficiently uncommon so that we
can ignore that?


>      .non_transitional_name = "vhost-vsock-pci-non-transitional",
>      .instance_size = sizeof(VHostVSockPCI),
>      .instance_init = vhost_vsock_pci_instance_init,