virtio: avoid buffer overrun on incoming migration
authorMichael Roth <mdroth@linux.vnet.ibm.com>
Thu, 3 Apr 2014 16:51:46 +0000 (19:51 +0300)
committerJuan Quintela <quintela@redhat.com>
Mon, 5 May 2014 20:15:02 +0000 (22:15 +0200)
CVE-2013-6399

vdev->queue_sel is read from the wire, and later used in the
emulation code as an index into vdev->vq[]. If the value of
vdev->queue_sel exceeds the length of vdev->vq[], currently
allocated to be VIRTIO_PCI_QUEUE_MAX elements, subsequent PIO
operations such as VIRTIO_PCI_QUEUE_PFN can be used to overrun
the buffer with arbitrary data originating from the source.

Fix this by failing migration if the value from the wire exceeds
VIRTIO_PCI_QUEUE_MAX.

Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Juan Quintela <quintela@redhat.com>

hw/virtio/virtio.c

index 05f05e7..0072542 100644 (file)
@@ -907,6 +907,9 @@ int virtio_load(VirtIODevice *vdev, QEMUFile *f)
     qemu_get_8s(f, &vdev->status);
     qemu_get_8s(f, &vdev->isr);
     qemu_get_be16s(f, &vdev->queue_sel);
+    if (vdev->queue_sel >= VIRTIO_PCI_QUEUE_MAX) {
+        return -1;
+    }
     qemu_get_be32s(f, &features);
 
     if (virtio_set_features(vdev, features) < 0) {