Re: [Qemu-devel] [PATCH 1/1] virtio: fix feature bit checks

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

On Wed, Jan 21, 2015 at 01:07:59PM +0100, Thomas Huth wrote:
> On Fri, 12 Dec 2014 10:01:46 +0100
> Cornelia Huck <cornelia.huck@de.ibm.com> wrote:
> 
> > Several places check against the feature bit number instead of against
> > the feature bit. Fix them.
> > 
> > Reported-by: Thomas Huth <thuth@linux.vnet.ibm.com>
> > Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
> > ---
> >  hw/scsi/virtio-scsi.c       | 2 +-
> >  hw/virtio/dataplane/vring.c | 4 ++--
> >  2 files changed, 3 insertions(+), 3 deletions(-)
> > 
> > diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
> > index ef48550..a44c410 100644
> > --- a/hw/scsi/virtio-scsi.c
> > +++ b/hw/scsi/virtio-scsi.c
> > @@ -144,7 +144,7 @@ static int virtio_scsi_parse_req(VirtIOSCSIReq *req,
> >       *
> >       * TODO: always disable this workaround for virtio 1.0 devices.
> >       */
> > -    if ((vdev->guest_features & VIRTIO_F_ANY_LAYOUT) == 0) {
> > +    if ((vdev->guest_features & (1 << VIRTIO_F_ANY_LAYOUT)) == 0) {
> >          req_size = req->elem.out_sg[0].iov_len;
> >          resp_size = req->elem.in_sg[0].iov_len;
> >      }
> > diff --git a/hw/virtio/dataplane/vring.c b/hw/virtio/dataplane/vring.c
> > index 61f6d83..78c6f45 100644
> > --- a/hw/virtio/dataplane/vring.c
> > +++ b/hw/virtio/dataplane/vring.c
> > @@ -133,12 +133,12 @@ bool vring_should_notify(VirtIODevice *vdev, Vring *vring)
> >       * interrupts. */
> >      smp_mb();
> > 
> > -    if ((vdev->guest_features & VIRTIO_F_NOTIFY_ON_EMPTY) &&
> > +    if ((vdev->guest_features & (1 << VIRTIO_F_NOTIFY_ON_EMPTY)) &&
> >          unlikely(vring->vr.avail->idx == vring->last_avail_idx)) {
> >          return true;
> >      }
> > 
> > -    if (!(vdev->guest_features & VIRTIO_RING_F_EVENT_IDX)) {
> > +    if (!(vdev->guest_features & (1 << VIRTIO_RING_F_EVENT_IDX))) {
> >          return !(vring->vr.avail->flags & VRING_AVAIL_F_NO_INTERRUPT);
> >      }
> >      old = vring->signalled_used;
> 
> Ping ... somebody taking care of this? It's a bug fix for current
> code, so IMHO this patch should not wait for the inclusion of the
> other virtio-1 stuff...
> 
>  Thomas

It's in my queue.