Re: [RFC PATCH 6/6] vdpa: remove net cvq migration blocker

[Jason Wang <jasowang@redhat.com>]

On Thu, Jul 20, 2023 at 4:18 PM Eugenio Perez Martin
<eperezma@redhat.com> wrote:
>
> On Mon, Jul 10, 2023 at 9:37 AM Eugenio Perez Martin
> <eperezma@redhat.com> wrote:
> >
> > On Mon, Jul 10, 2023 at 5:54 AM Jason Wang <jasowang@redhat.com> wrote:
> > >
> > > On Fri, Jul 7, 2023 at 3:12 AM Eugenio Pérez <eperezma@redhat.com> wrote:
> > > >
> > > > Now that we have add migration blockers if the device does not support
> > > > all the needed features, remove the general blocker applied to all net
> > > > devices with CVQ.
> > > >
> > > > Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
> > >
> > > I wonder what's the difference compared if we just start cvq first in
> > > vhost_net_start()?
> > >
> >
> > That's interesting. It would complicate the for loop in
> > vhost_net_start, but I think it's less complicated than
> > should_start_op.
> >
> > It would be something like moving from:
> >
> > for (i = 0; i < nvhosts; i++) {
> >     if (i < data_queue_pairs) {
> >         peer = qemu_get_peer(ncs, i);
> >     } else {
> >         peer = qemu_get_peer(ncs, n->max_queue_pairs);
> >     }
> >
> >     ...
> >
> >     r = vhost_net_start_one(get_vhost_net(peer), dev);
> >     if (r < 0) {
> >         goto err_start;
> >     }
> > }
> >
> > To:
> >
> > for (i = 0; i < nvhosts; i++) {
> >     if (i == 0 && cvq) {
> >         peer = qemu_get_peer(ncs, n->max_queue_pairs);
> >     } else {
> >         peer = qemu_get_peer(ncs, i - cvq);
> >     }
> >
> >     ...
> >
> >     r = vhost_net_start_one(get_vhost_net(peer), dev);
> >     if (r < 0) {
> >         goto err_start;
> >     }
> > }
> >
> > Is this what you have in mind?
> >
>

Yes, something like this.

Thanks

> Friendly ping.
>
> Thanks!
>