Re: [PATCH 4/5] virtio: Fail if iommu_platform is requested, but unsupported

[Kevin Wolf]

Am 28.04.2021 um 21:24 hat Raphael Norwitz geschrieben:
> On Thu, Apr 22, 2021 at 07:02:20PM +0200, Kevin Wolf wrote:
> > Commit 2943b53f6 (' virtio: force VIRTIO_F_IOMMU_PLATFORM') made sure
> > that vhost can't just reject VIRTIO_F_IOMMU_PLATFORM when it was
> > requested. However, just adding it back to the negotiated flags isn't
> > right either because it promises support to the guest that the device
> > actually doesn't support. One example of a vhost-user device that
> > doesn't have support for the flag is the vhost-user-blk export of QEMU.
> > 
> > Instead of successfully creating a device that doesn't work, just fail
> > to plug the device when it doesn't support the feature, but it was
> > requested. This results in much clearer error messages.
> > 
> > Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1935031
> > Signed-off-by: Kevin Wolf <kwolf@redhat.com>
> > ---
> >  hw/virtio/virtio-bus.c | 5 +++++
> >  1 file changed, 5 insertions(+)
> > 
> > diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
> > index d6332d45c3..859978d248 100644
> > --- a/hw/virtio/virtio-bus.c
> > +++ b/hw/virtio/virtio-bus.c
> > @@ -69,6 +69,11 @@ void virtio_bus_device_plugged(VirtIODevice *vdev, Error 
> > **errp)
> >          return;
> >      }
> >
> 
> Can you explain this check a little more?
> 
> Above we have:
> bool has_iommu = virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);

If I underdstand the code correctly, at this point this is still the
unchanged value of the iommu_platform=on|off qdev property as given by
the user.

> and then we get the host features from the bckend:
> vdev->host_features = vdc->get_features(vdev, vdev->host_features

Yes, and now a flag is only set if the user had requested it and the
backend also supports it.

> So as is this is catching the case where vdev->host_features had
> VIRTIO_F_IOMMU_PLATFORM set before (by default?), but doesn't now that
> the features have been retrieved? 
> 
> Why not just:
>     if (!virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM)) {

We don't want to fail if the user hadn't even requested the feature, but
just if it was requested, but could not be provided.

> > +    if (has_iommu && !virtio_host_has_feature(vdev, 
> > VIRTIO_F_IOMMU_PLATFORM)) {
> > +        error_setg(errp, "iommu_platform=true is not supported by the 
> > device");
> > +        return;
> > +    }
> > +
> >      if (klass->device_plugged != NULL) {
> >          klass->device_plugged(qbus->parent, &local_err);
> >      }

Kevin