On Wed, Jun 15, 2016 at 08:32:45AM +0800, Huawei Xie wrote:
> fixes the following coverity issues:
> CID 107118 (#1 of 1): Dereference null return value (NULL_RETURNS)
> CID 119262 (#1 of 1): Dereference null return value (NULL_RETURNS)
>
> Fixes: 8f972312b8f4 ("vhost: support vhost-user")
> Fixes: 77d20126b4c2 ("vhost-user: handle message to enable vring")
>
> Signed-off-by: Huawei Xie <huawei.xie at intel.com>
> ---
> lib/librte_vhost/vhost_user/virtio-net-user.c | 11 +++++++++--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_vhost/vhost_user/virtio-net-user.c
> b/lib/librte_vhost/vhost_user/virtio-net-user.c
> index 5803182..5844a42 100644
> --- a/lib/librte_vhost/vhost_user/virtio-net-user.c
> +++ b/lib/librte_vhost/vhost_user/virtio-net-user.c
> @@ -283,6 +283,9 @@ user_set_vring_kick(struct vhost_device_ctx ctx, struct
> VhostUserMsg *pmsg)
> struct vhost_vring_file file;
> struct virtio_net *dev = get_device(ctx);
>
> + if (!dev)
> + return;
> +
I just found that it's been fixed by commit a03aabf2e8d6 ("vhost: export
vid as the only interface to applications"), "accidentally".
> file.index = pmsg->payload.u64 & VHOST_USER_VRING_IDX_MASK;
> if (pmsg->payload.u64 & VHOST_USER_VRING_NOFD_MASK)
> file.fd = VIRTIO_INVALID_EVENTFD;
> @@ -306,8 +309,9 @@ user_get_vring_base(struct vhost_device_ctx ctx,
> {
> struct virtio_net *dev = get_device(ctx);
>
> - if (dev == NULL)
> + if (!dev)
> return -1;
What's that for? Interestingly, if you look at the DPDK coding style,
you will find that it's encouraged to use "dev == NULL" check instead
of the "!dev" check. I sometimes uses both though.
So, I don't think it's a big deal, and I will leave it as it is.
> +
> /* We have to stop the queue (virtio) if it is running. */
> if (dev->flags & VIRTIO_DEV_RUNNING)
> notify_ops->destroy_device(dev);
> @@ -341,6 +345,9 @@ user_set_vring_enable(struct vhost_device_ctx ctx,
> struct virtio_net *dev = get_device(ctx);
> int enable = (int)state->num;
>
> + if (!dev)
> + return -1;
> +
It's been fixed by commit 590a936287c6 ("vhost: fix null pointer
dereference").
So, sorry, I will NACK this patch.
--yliu
