dccp: fix freeing skb too early for IPV6_RECVPKTINFO
authorAndrey Konovalov <andreyknvl@google.com>
Thu, 16 Feb 2017 16:22:46 +0000 (17:22 +0100)
committerBen Hutchings <ben@decadent.org.uk>
Sun, 26 Feb 2017 20:01:32 +0000 (20:01 +0000)
commit 5edabca9d4cff7f1f2b68f0bac55ef99d9798ba4 upstream.

In the current DCCP implementation an skb for a DCCP_PKT_REQUEST packet
is forcibly freed via __kfree_skb in dccp_rcv_state_process if
dccp_v6_conn_request successfully returns.

However, if IPV6_RECVPKTINFO is set on a socket, the address of the skb
is saved to ireq->pktopts and the ref count for skb is incremented in
dccp_v6_conn_request, so skb is still in use. Nevertheless, it gets freed
in dccp_rcv_state_process.

Fix by calling consume_skb instead of doing goto discard and therefore
calling __kfree_skb.

Similar fixes for TCP:

fb7e2399ec17f1004c0e0ccfd17439f8759ede01 [TCP]: skb is unexpectedly freed.
0aea76d35c9651d55bbaf746e7914e5f9ae5a25d tcp: SYN packets are now
simply consumed

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

index 51d5fe5..2305191 100644 (file)
@@ -606,7 +606,8 @@ int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
                        if (inet_csk(sk)->icsk_af_ops->conn_request(sk,
                                                                    skb) < 0)
                                return 1;
-                       goto discard;
+                       consume_skb(skb);
+                       return 0;
                if (dh->dccph_type == DCCP_PKT_RESET)
                        goto discard;