mptcp: wake-up readers only for in sequence data

Currently we rely on the subflow->data_avail field, which is subject to
races:

	ssk1
		skb len = 500 DSS(seq=1, len=1000, off=0)
		# data_avail == MPTCP_SUBFLOW_DATA_AVAIL

	ssk2
		skb len = 500 DSS(seq = 501, len=1000)
		# data_avail == MPTCP_SUBFLOW_DATA_AVAIL

	ssk1
		skb len = 500 DSS(seq = 1, len=1000, off =500)
		# still data_avail == MPTCP_SUBFLOW_DATA_AVAIL,
		# as the skb is covered by a pre-existing map,
		# which was in-sequence at reception time.

Instead we can explicitly check if some has been received in-sequence,
propagating the info from __mptcp_move_skbs_from_subflow().

Additionally add the 'ONCE' annotation to the 'data_avail' memory
access, as msk will read it outside the subflow socket lock.

Fixes: 648ef4b88673 ("mptcp: Implement MPTCP receive path")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 12 insertions, 21 deletions

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 534cf500521d..f6e62a6dc9fb 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -670,15 +670,13 @@ static bool __mptcp_ofo_queue(struct mptcp_sock *msk)
 /* In most cases we will be able to lock the mptcp socket.  If its already
  * owned, we need to defer to the work queue to avoid ABBA deadlock.
  */
-static void move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk)
+static bool move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk)
 {
 	struct sock *sk = (struct sock *)msk;
 	unsigned int moved = 0;
 
 	if (inet_sk_state_load(sk) == TCP_CLOSE)
-		return;
-
-	mptcp_data_lock(sk);
+		return false;
 
 	__mptcp_move_skbs_from_subflow(msk, ssk, &moved);
 	__mptcp_ofo_queue(msk);
@@ -690,7 +688,7 @@ static void move_skbs_to_msk(struct mptcp_sock *msk, struct sock *ssk)
 	 */
 	if (mptcp_pending_data_fin(sk, NULL))
 		mptcp_schedule_work(sk);
-	mptcp_data_unlock(sk);
+	return moved > 0;
 }
 
 void mptcp_data_ready(struct sock *sk, struct sock *ssk)
@@ -698,7 +696,6 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk)
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(ssk);
 	struct mptcp_sock *msk = mptcp_sk(sk);
 	int sk_rbuf, ssk_rbuf;
-	bool wake;
 
 	/* The peer can send data while we are shutting down this
 	 * subflow at msk destruction time, but we must avoid enqueuing
@@ -707,28 +704,22 @@ void mptcp_data_ready(struct sock *sk, struct sock *ssk)
 	if (unlikely(subflow->disposable))
 		return;
 
-	/* move_skbs_to_msk below can legitly clear the data_avail flag,
-	 * but we will need later to properly woke the reader, cache its
-	 * value
-	 */
-	wake = subflow->data_avail == MPTCP_SUBFLOW_DATA_AVAIL;
-	if (wake)
-		set_bit(MPTCP_DATA_READY, &msk->flags);
-
 	ssk_rbuf = READ_ONCE(ssk->sk_rcvbuf);
 	sk_rbuf = READ_ONCE(sk->sk_rcvbuf);
 	if (unlikely(ssk_rbuf > sk_rbuf))
 		sk_rbuf = ssk_rbuf;
 
-	/* over limit? can't append more skbs to msk */
+	/* over limit? can't append more skbs to msk, Also, no need to wake-up*/
 	if (atomic_read(&sk->sk_rmem_alloc) > sk_rbuf)
-		goto wake;
-
-	move_skbs_to_msk(msk, ssk);
+		return;
 
-wake:
-	if (wake)
+	/* Wake-up the reader only for in-sequence data */
+	mptcp_data_lock(sk);
+	if (move_skbs_to_msk(msk, ssk)) {
+		set_bit(MPTCP_DATA_READY, &msk->flags);
 		sk->sk_data_ready(sk);
+	}
+	mptcp_data_unlock(sk);
 }
 
 static bool mptcp_do_flush_join_list(struct mptcp_sock *msk)
@@ -860,7 +851,7 @@ static struct sock *mptcp_subflow_recv_lookup(const struct mptcp_sock *msk)
 	sock_owned_by_me(sk);
 
 	mptcp_for_each_subflow(msk, subflow) {
-		if (subflow->data_avail)
+		if (READ_ONCE(subflow->data_avail))
 			return mptcp_subflow_tcp_sock(subflow);
 	}