On Thu, Feb 11, 2021 at 09:59:36AM +0000, Stefan Hajnoczi wrote:
> On Tue, Feb 09, 2021 at 07:02:01PM +0000, Dr. David Alan Gilbert (git) wrote:
> > +static uint64_t vhost_user_slave_handle_vring_host_notifier(
> > + struct vhost_dev *dev,
> > + VhostUserVringArea *area,
> > + int fd)
>
> Indentation looks off. Only worth changing if you respin.
>
> > @@ -1398,7 +1399,8 @@ static void slave_read(void *opaque)
> > struct vhost_user *u = dev->opaque;
> > VhostUserHeader hdr = { 0, };
> > VhostUserPayload payload = { 0, };
> > - int size, ret = 0;
> > + int size;
> > + uint64_t ret = 0;
> > struct iovec iov;
> > struct msghdr msgh;
> > int fd[VHOST_USER_SLAVE_MAX_FDS];
> > @@ -1472,7 +1474,7 @@ static void slave_read(void *opaque)
> > break;
> > default:
> > error_report("Received unexpected msg type: %d.", hdr.request);
> > - ret = -EINVAL;
> > + ret = (uint64_t)-EINVAL;
>
> The !!ret was removed below so it would have previously been true (1).
> Now it has changed value.
>
> If there is no specific reason to change the value, please keep it true
> (1) just in case a vhost-user device backend depends on that value.
Good catch. I guess it will be nice to send -EINVAL back but we probably
can't change it now due to backward compatibility issue. Just in case,
someone is relying on reading back true (instead of -EINVAL).
Vivek
