tcp/dccp: fix another race at listener dismantle

Ilya reported following lockdep splat:

kernel: =========================
kernel: [ BUG: held lock freed! ]
kernel: 4.5.0-rc1-ceph-00026-g5e0a311 #1 Not tainted
kernel: -------------------------
kernel: swapper/5/0 is freeing memory
ffff880035c9d200-ffff880035c9dbff, with a lock still held there!
kernel: (&(&queue->rskq_lock)->rlock){+.-...}, at:
[<ffffffff816f6a88>] inet_csk_reqsk_queue_add+0x28/0xa0
kernel: 4 locks held by swapper/5/0:
kernel: #0:  (rcu_read_lock){......}, at: [<ffffffff8169ef6b>]
netif_receive_skb_internal+0x4b/0x1f0
kernel: #1:  (rcu_read_lock){......}, at: [<ffffffff816e977f>]
ip_local_deliver_finish+0x3f/0x380
kernel: #2:  (slock-AF_INET){+.-...}, at: [<ffffffff81685ffb>]
sk_clone_lock+0x19b/0x440
kernel: #3:  (&(&queue->rskq_lock)->rlock){+.-...}, at:
[<ffffffff816f6a88>] inet_csk_reqsk_queue_add+0x28/0xa0

To properly fix this issue, inet_csk_reqsk_queue_add() needs
to return to its callers if the child as been queued
into accept queue.

We also need to make sure listener is still there before
calling sk->sk_data_ready(), by holding a reference on it,
since the reference carried by the child can disappear as
soon as the child is put on accept queue.

Reported-by: Ilya Dryomov <idryomov@gmail.com>
Fixes: ebb516af60e1 ("tcp/dccp: fix race at listener dismantle phase")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

2 files changed, 14 insertions, 14 deletions

diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 46b9c887bede..64148914803a 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -789,14 +789,16 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
 	reqsk_put(req);
 }
 
-void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
-			      struct sock *child)
+struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
+				      struct request_sock *req,
+				      struct sock *child)
 {
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 
 	spin_lock(&queue->rskq_lock);
 	if (unlikely(sk->sk_state != TCP_LISTEN)) {
 		inet_child_forget(sk, req, child);
+		child = NULL;
 	} else {
 		req->sk = child;
 		req->dl_next = NULL;
@@ -808,6 +810,7 @@ void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
 		sk_acceptq_added(sk);
 	}
 	spin_unlock(&queue->rskq_lock);
+	return child;
 }
 EXPORT_SYMBOL(inet_csk_reqsk_queue_add);
 
@@ -817,11 +820,8 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
 	if (own_req) {
 		inet_csk_reqsk_queue_drop(sk, req);
 		reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
-		inet_csk_reqsk_queue_add(sk, req, child);
-		/* Warning: caller must not call reqsk_put(req);
-		 * child stole last reference on it.
-		 */
-		return child;
+		if (inet_csk_reqsk_queue_add(sk, req, child))
+			return child;
 	}
 	/* Too bad, another child took ownership of the request, undo. */
 	bh_unlock_sock(child);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index c84477949d3a..487ac67059e2 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1597,30 +1597,30 @@ process:
 
 	if (sk->sk_state == TCP_NEW_SYN_RECV) {
 		struct request_sock *req = inet_reqsk(sk);
-		struct sock *nsk = NULL;
+		struct sock *nsk;
 
 		sk = req->rsk_listener;
 		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb))) {
 			reqsk_put(req);
 			goto discard_it;
 		}
-		if (likely(sk->sk_state == TCP_LISTEN)) {
-			nsk = tcp_check_req(sk, skb, req, false);
-		} else {
+		if (unlikely(sk->sk_state != TCP_LISTEN)) {
 			inet_csk_reqsk_queue_drop_and_put(sk, req);
 			goto lookup;
 		}
+		sock_hold(sk);
+		nsk = tcp_check_req(sk, skb, req, false);
 		if (!nsk) {
 			reqsk_put(req);
-			goto discard_it;
+			goto discard_and_relse;
 		}
 		if (nsk == sk) {
-			sock_hold(sk);
 			reqsk_put(req);
 		} else if (tcp_child_process(sk, nsk, skb)) {
 			tcp_v4_send_reset(nsk, skb);
-			goto discard_it;
+			goto discard_and_relse;
 		} else {
+			sock_put(sk);
 			return 0;
 		}
 	}