Re: [PATCH v3 3/6] virtio: allow finalize_features to fail

From: Cornelia Huck
Date: Tue Dec 09 2014 - 07:56:24 EST


On Tue, 9 Dec 2014 14:07:54 +0200
"Michael S. Tsirkin" <mst@xxxxxxxxxx> wrote:

> On Tue, Dec 09, 2014 at 11:46:59AM +0100, Cornelia Huck wrote:
> > On Mon, 8 Dec 2014 15:05:58 +0200
> > "Michael S. Tsirkin" <mst@xxxxxxxxxx> wrote:
> >
> > > This will make it easy for transports to validate features and return
> > > failure.
> > >
> > > Signed-off-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
> > > ---
> > > include/linux/virtio_config.h | 3 ++-
> > > drivers/lguest/lguest_device.c | 4 +++-
> > > drivers/misc/mic/card/mic_virtio.c | 4 +++-
> > > drivers/remoteproc/remoteproc_virtio.c | 4 +++-
> > > drivers/s390/kvm/kvm_virtio.c | 4 +++-
> > > drivers/s390/kvm/virtio_ccw.c | 6 ++++--
> > > drivers/virtio/virtio.c | 21 ++++++++++++++-------
> > > drivers/virtio/virtio_mmio.c | 4 +++-
> > > drivers/virtio/virtio_pci.c | 4 +++-
> > > 9 files changed, 38 insertions(+), 16 deletions(-)
> > >
> >
> > > diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
> > > index c792b5f..789275f 100644
> > > --- a/drivers/s390/kvm/virtio_ccw.c
> > > +++ b/drivers/s390/kvm/virtio_ccw.c
> > > @@ -752,7 +752,7 @@ out_free:
> > > return rc;
> > > }
> > >
> > > -static void virtio_ccw_finalize_features(struct virtio_device *vdev)
> > > +static int virtio_ccw_finalize_features(struct virtio_device *vdev)
> > > {
> > > struct virtio_ccw_device *vcdev = to_vc_device(vdev);
> > > struct virtio_feature_desc *features;
> > > @@ -760,7 +760,7 @@ static void virtio_ccw_finalize_features(struct virtio_device *vdev)
> > >
> > > ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL);
> > > if (!ccw)
> > > - return;
> > > + return 0;
> >
> > I think we'll want to return an error in this case as well to fail
> > probing. Also for the other places where something can go wrong.
> > Something like this:
>
>
> I guess the fact this previously failed silently is a bug in original
> code?

Not really a bug; it simply was not possible to fail it.

>
> Can you please send your S.O.B so I can add this patch on top?

Sure. Code below is

Signed-off-by: Cornelia Huck <cornelia.huck@xxxxxxxxxx>

>
> We might also need to Cc stable.

Not sure whether this is stable-worthy, as these are largely
pathological cases (and for the one non-pathological case, device going
away, we're already fine).

>
>
> > diff --git a/drivers/s390/kvm/virtio_ccw.c b/drivers/s390/kvm/virtio_ccw.c
> > index 789275f..bc6e671 100644
> > --- a/drivers/s390/kvm/virtio_ccw.c
> > +++ b/drivers/s390/kvm/virtio_ccw.c
> > @@ -757,15 +757,17 @@ static int virtio_ccw_finalize_features(struct virtio_device *vdev)
> > struct virtio_ccw_device *vcdev = to_vc_device(vdev);
> > struct virtio_feature_desc *features;
> > struct ccw1 *ccw;
> > + int ret;
> >
> > ccw = kzalloc(sizeof(*ccw), GFP_DMA | GFP_KERNEL);
> > if (!ccw)
> > - return 0;
> > + return -ENOMEM;
> >
> > features = kzalloc(sizeof(*features), GFP_DMA | GFP_KERNEL);
> > - if (!features)
> > + if (!features) {
> > + ret = -ENOMEM;
> > goto out_free;
> > -
> > + }
> > /* Give virtio_ring a chance to accept features. */
> > vring_transport_features(vdev);
> >
> > @@ -776,7 +778,9 @@ static int virtio_ccw_finalize_features(struct virtio_device *vdev)
> > ccw->flags = 0;
> > ccw->count = sizeof(*features);
> > ccw->cda = (__u32)(unsigned long)features;
> > - ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT);
> > + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT);
> > + if (ret)
> > + goto out_free;
> >
> > if (vcdev->revision == 0)
> > goto out_free;
> > @@ -788,13 +792,13 @@ static int virtio_ccw_finalize_features(struct virtio_device *vdev)
> > ccw->flags = 0;
> > ccw->count = sizeof(*features);
> > ccw->cda = (__u32)(unsigned long)features;
> > - ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT);
> > + ret = ccw_io_helper(vcdev, ccw, VIRTIO_CCW_DOING_WRITE_FEAT);
> >
> > out_free:
> > kfree(features);
> > kfree(ccw);
> >
> > - return 0;
> > + return ret;
> > }
> >
> > static void virtio_ccw_get_config(struct virtio_device *vdev,
>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/