On Thu, May 20, 2010 at 09:58:06AM -0700, David L Stevens wrote:
> [for Michael Tsirkin's vhost development git tree]
> 
> This patch fixes a race between guest and host when
> adding used buffers wraps the ring. Without it, guests
> can see partial packets before num_buffers is set in
> the vnet header.
> 
> Signed-off-by: David L Stevens <dlstev...@us.ibm.com>

Could you please explain what the race is?

> diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
> index 7f2568d..74790ab 100644
> --- a/drivers/vhost/vhost.c
> +++ b/drivers/vhost/vhost.c
> @@ -1065,14 +1065,6 @@ static int __vhost_add_used_n(struct vhost_virtqueue 
> *vq,
>               vq_err(vq, "Failed to write used");
>               return -EFAULT;
>       }
> -     /* Make sure buffer is written before we update index. */
> -     smp_wmb();
> -     if (put_user(vq->last_used_idx + count, &vq->used->idx)) {
> -             vq_err(vq, "Failed to increment used idx");
> -             return -EFAULT;
> -     }
> -     if (unlikely(vq->log_used))
> -             vhost_log_used(vq, used);
>       vq->last_used_idx += count;
>       return 0;
>  }
> @@ -1093,7 +1085,17 @@ int vhost_add_used_n(struct vhost_virtqueue *vq, 
> struct vring_used_elem *heads,
>               heads += n;
>               count -= n;
>       }
> -     return __vhost_add_used_n(vq, heads, count);
> +     r = __vhost_add_used_n(vq, heads, count);
> +
> +     /* Make sure buffer is written before we update index. */
> +     smp_wmb();
> +     if (put_user(vq->last_used_idx, &vq->used->idx)) {
> +             vq_err(vq, "Failed to increment used idx");
> +             return -EFAULT;
> +     }
> +     if (unlikely(vq->log_used))
> +             vhost_log_used(vq, vq->used->ring + start);
> +     return r;
>  }
>  
>  /* This actually signals the guest, using eventfd. */
> 
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to