Re: [Qemu-devel] [PATCH] virtio: validate the existence of handle_output before calling it

[Jason Wang <jasowang@redhat.com>]

On Sat, Feb 14, 2015 at 4:18 AM, Don Koch <dkoch@verizon.com> wrote:
> On Thu, 12 Feb 2015 11:05:17 +0800
> Jason Wang <jasowang@redhat.com> wrote:
> 
>>  We don't validate the existence of handle_output which may let a 
>> buggy
>>  guest to trigger a SIGSEV easily. Fix this by validate its existence
>>  before.
>>  
>>  Cc: qemu-stable@nongnu.org
>>  Cc: Anthony Liguori <aliguori@amazon.com>
>>  Cc: Michael S. Tsirkin <mst@redhat.com>
>>  Signed-off-by: Jason Wang <jasowang@redhat.com>
>>  ---
>>   hw/virtio/virtio.c | 4 ++++
>>   1 file changed, 4 insertions(+)
>>  
>>  diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
>>  index d735343..ffc22e8 100644
>>  --- a/hw/virtio/virtio.c
>>  +++ b/hw/virtio/virtio.c
>>  @@ -761,6 +761,10 @@ void virtio_queue_notify_vq(VirtQueue *vq)
>>   {
>>       if (vq->vring.desc) {
>>           VirtIODevice *vdev = vq->vdev;
>>  +
>>  +        if (!vq->handle_output) {
>>  +            return;
>>  +        }
> 
> Maybe better to just change line 762 to:
>      if (vq->vring.desc && vq->handle_output) {
> 
> -d

Yes, better.
> 
> 
>>           trace_virtio_queue_notify(vdev, vq - vdev->vq, vq);
>>           vq->handle_output(vdev, vq);
>>       }
>>  -- 
>>  1.9.1
>>  
>>