vhost_net: fix OoB on sendmsg() failure.

If the sendmsg() call in vhost_tx_batch() fails, both the 'batched_xdp'
and 'done_idx' indexes are left unchanged. If such failure happens
when batched_xdp == VHOST_NET_BATCH, the next call to
vhost_net_build_xdp() will access and write memory outside the xdp
buffers area.

Since sendmsg() can only error with EBADFD, this change addresses the
issue explicitly freeing the XDP buffers batch on error.

Fixes: 6a22c7284d78 ("vhost_net: batch submitting XDP buffers to underlayer sockets")
Suggested-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 3a249ee7e14401d0f7459f8c401e2f9aebdce201..28ef323882fb29c66c04b6128021041e70b1f49a 100644 (file)
--- a/drivers/vhost/net.c
+++ b/drivers/vhost/net.c
@@ -467,7 +467,7 @@ static void vhost_tx_batch(struct vhost_net *net,
                .num = nvq->batched_xdp,
                .ptr = nvq->xdp,
        };
-       int err;
+       int i, err;
 
        if (nvq->batched_xdp == 0)
                goto signal_used;
@@ -476,6 +476,15 @@ static void vhost_tx_batch(struct vhost_net *net,
        err = sock->ops->sendmsg(sock, msghdr, 0);
        if (unlikely(err < 0)) {
                vq_err(&nvq->vq, "Fail to batch sending packets\n");
+
+               /* free pages owned by XDP; since this is an unlikely error path,
+                * keep it simple and avoid more complex bulk update for the
+                * used pages
+                */
+               for (i = 0; i < nvq->batched_xdp; ++i)
+                       put_page(virt_to_head_page(nvq->xdp[i].data));
+               nvq->batched_xdp = 0;
+               nvq->done_idx = 0;
                return;
        }