Re: [Qemu-devel] [2.5 issue] virtio-1 in virtio-net and old vhost

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

On Wed, Dec 02, 2015 at 11:11:28AM +0100, Cornelia Huck wrote:
> On Wed, 2 Dec 2015 13:54:09 +0800
> Jason Wang <jasowang@redhat.com> wrote:
> 
> > I wonder instead of rolling back in post_plugged(), maybe we could just
> > delay the region setups to post_plugged(). 
> 
> If this is the saner thing to do for pci, sure.
> 
> > Or just call transport
> > specific device_plugged() after get_features() call in
> > virtio_bus_device_plugged(). 
> 
> The problem is that the VERSION_1 bit is only added in the
> ->device_plugged() callbacks by the transport, so ->get_features() can
> only be called after that. We have a dependency in both directions :(
> 
> > And I'm not sure we need to handle
> > migration compatibility in this case.
> 
> The thing we would need to care about is basically the host kernel on
> the target supporting less than the host kernel on the source.

This is a fundamental problem, we have it with other
features like transport offloads. We don't have a solution for
this now. kvm has the same issue.

> Do we
> care about that in other contexts right now?