Re: Re: [RFC PATCH 13/15] virtio-mem: Handle virtio_device_ready() failure

[Yongji Xie]

On Mon, May 17, 2021 at 5:48 PM David Hildenbrand <david@xxxxxxxxxx> wrote:
>
> On 17.05.21 11:34, Xie Yongji wrote:
> > Now virtio_device_ready() will return error if we get
> > invalid status. Let's handle this case on probe.
> >
> > Signed-off-by: Xie Yongji <xieyongji@xxxxxxxxxxxxx>
> > ---
> >   drivers/virtio/virtio_mem.c | 6 +++++-
> >   1 file changed, 5 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/virtio/virtio_mem.c b/drivers/virtio/virtio_mem.c
> > index 32a8e359a5c3..1148c392ff94 100644
> > --- a/drivers/virtio/virtio_mem.c
> > +++ b/drivers/virtio/virtio_mem.c
> > @@ -2602,13 +2602,17 @@ static int virtio_mem_probe(struct virtio_device *vdev)
> >       if (rc)
> >               goto out_unreg_mem;
> >
> > -     virtio_device_ready(vdev);
> > +     rc = virtio_device_ready(vdev);
> > +     if (rc)
> > +             goto out_unreg_device;
> >
> >       /* trigger a config update to start processing the requested_size */
> >       atomic_set(&vm->config_changed, 1);
> >       queue_work(system_freezable_wq, &vm->wq);
> >
> >       return 0;
> > +out_unreg_device:
> > +     unregister_virtio_mem_device(vm);
> >   out_unreg_mem:
> >       unregister_memory_notifier(&vm->memory_notifier);
> >   out_del_resource:
> >
>
> I assume this will really be a corner case to hit, right?
>

Yes, it should almost never happen if the device is trusted.

Thanks,
Yongji