On Tue, Apr 09, 2019 at 12:10:25PM +0800, Jason Wang wrote:
> We used to accept zero size iova range which will lead a infinite loop
> in translate_desc(). Fixing this by failing the request in this case.
>
> Reported-by: syzbot+d21e6e297322a900c128@xxxxxxxxxxxxxxxxxxxxxxxxx
> Fixes: 6b1e6cc7 ("vhost: new device IOTLB API")
> Signed-off-by: Jason Wang <jasowang@xxxxxxxxxx>
Acked-by: Michael S. Tsirkin <mst@xxxxxxxxxx>
Seems appropriate for stable.
> ---
> drivers/vhost/vhost.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
> index 5ace833de746..351af88231ad 100644
> --- a/drivers/vhost/vhost.c
> +++ b/drivers/vhost/vhost.c
> @@ -911,8 +911,12 @@ static int vhost_new_umem_range(struct vhost_umem *umem,
> u64 start, u64 size, u64 end,
> u64 userspace_addr, int perm)
> {
> - struct vhost_umem_node *tmp, *node = kmalloc(sizeof(*node), GFP_ATOMIC);
> + struct vhost_umem_node *tmp, *node;
>
> + if (!size)
> + return -EFAULT;
> +
> + node = kmalloc(sizeof(*node), GFP_ATOMIC);
> if (!node)
> return -ENOMEM;
>
> --
> 2.19.1
