Re: [RFC v2 08/13] vdpa: Negotiate _F_SUSPEND feature

[Eugenio Perez Martin]

On Fri, Jan 13, 2023 at 5:39 AM Jason Wang <jasowang@redhat.com> wrote:
>
> On Fri, Jan 13, 2023 at 1:25 AM Eugenio Pérez <eperezma@redhat.com> wrote:
> >
> > This is needed for qemu to know it can suspend the device to retrieve
> > its status and enable SVQ with it, so all the process is transparent to
> > the guest.
> >
> > Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
>
> Acked-by: Jason Wang <jasowang@redhat.com>
>
> We probably need to add the resume in the future to have a quick
> recovery from migration failures.
>

The capability of a resume can be useful here but only in a small
window. During the most time of the migration SVQ is enabled, so in
the event of a migration failure we may need to reset the whole device
to enable passthrough again.

But maybe is it worth giving a quick review and adding some TODOs
where it can be useful in this series?

Thanks!

> Thanks
>
> > ---
> >  hw/virtio/vhost-vdpa.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
> > index 4296427a69..a61a6b2a74 100644
> > --- a/hw/virtio/vhost-vdpa.c
> > +++ b/hw/virtio/vhost-vdpa.c
> > @@ -659,7 +659,8 @@ static int vhost_vdpa_set_backend_cap(struct vhost_dev 
> > *dev)
> >      uint64_t features;
> >      uint64_t f = 0x1ULL << VHOST_BACKEND_F_IOTLB_MSG_V2 |
> >          0x1ULL << VHOST_BACKEND_F_IOTLB_BATCH |
> > -        0x1ULL << VHOST_BACKEND_F_IOTLB_ASID;
> > +        0x1ULL << VHOST_BACKEND_F_IOTLB_ASID |
> > +        0x1ULL << VHOST_BACKEND_F_SUSPEND;
> >      int r;
> >
> >      if (vhost_vdpa_call(dev, VHOST_GET_BACKEND_FEATURES, &features)) {
> > --
> > 2.31.1
> >
>