Re: [PATCH 2/3] virtio: indirect ring entries (VIRTIO_RING_F_INDIRECT_DESC)

[Mark McLoughlin <markmc@redhat.com>]

Hi Ingo,

On Sat, 2008-12-20 at 12:38 +0100, Ingo Oeser wrote:
> Hi Mark,
> 
> On Thursday 18 December 2008, Mark McLoughlin wrote:
> > diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
> > index 5777196..2330c4b 100644
> > --- a/drivers/virtio/virtio_ring.c
> > +++ b/drivers/virtio/virtio_ring.c
> > @@ -70,6 +73,55 @@ struct vring_virtqueue
> >  
> >  #define to_vvq(_vq) container_of(_vq, struct vring_virtqueue, vq)
> >  
> > +/* Set up an indirect table of descriptors and add it to the queue. */
> > +static int vring_add_indirect(struct vring_virtqueue *vq,
> > +			      struct scatterlist sg[],
> > +			      unsigned int out,
> > +			      unsigned int in)
> > +{
> > +	struct vring_desc *desc;
> > +	unsigned head;
> > +	int i;
> > +
> > +	desc = kmalloc((out + in) * sizeof(struct vring_desc), GFP_ATOMIC);
> 
> kmalloc() returns ZERO_SIZE_PTR, if (out + in) == 0

vring_add_buf() has:

  BUG_ON(out + in == 0)

I should just add that here too before the kmalloc() call.

Thanks,
Mark.