mptcp: fix possible list corruption on passive MPJ

At passive MPJ time, if the msk socket lock is held by the user,
the new subflow is appended to the msk->join_list under the msk
data lock.

In mptcp_release_cb()/__mptcp_flush_join_list(), the subflows in
that list are moved from the join_list into the conn_list under the
msk socket lock.

Append and removal could race, possibly corrupting such list.
Address the issue splicing the join list into a temporary one while
still under the msk data lock.

Found by code inspection, the race itself should be almost impossible
to trigger in practice.

Fixes: 3e50149 ("mptcp: cleanup MPJ subflow list handling")
Cc: [email protected]
Signed-off-by: Paolo Abeni <[email protected]>
Reviewed-by: Matthieu Baerts <[email protected]>
Signed-off-by: Matthieu Baerts <[email protected]>
Signed-off-by: Jakub Kicinski <[email protected]>

Author: Paolo Abeni

net/mptcp/protocol.c

@@ -850,12 +850,12 @@ static bool __mptcp_finish_join(struct mptcp_sock *msk, struct sock *ssk)
 	return true;
 }
 
-static void __mptcp_flush_join_list(struct sock *sk)
+static void __mptcp_flush_join_list(struct sock *sk, struct list_head *join_list)
 {
 	struct mptcp_subflow_context *tmp, *subflow;
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
-	list_for_each_entry_safe(subflow, tmp, &msk->join_list, node) {
+	list_for_each_entry_safe(subflow, tmp, join_list, node) {
 		struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 		bool slow = lock_sock_fast(ssk);
 
@@ -3342,9 +3342,14 @@ static void mptcp_release_cb(struct sock *sk)
 	for (;;) {
 		unsigned long flags = (msk->cb_flags & MPTCP_FLAGS_PROCESS_CTX_NEED) |
 				      msk->push_pending;
+		struct list_head join_list;
+
 		if (!flags)
 			break;
 
+		INIT_LIST_HEAD(&join_list);
+		list_splice_init(&msk->join_list, &join_list);
+
 		/* the following actions acquire the subflow socket lock
 		 *
 		 * 1) can't be invoked in atomic scope
@@ -3355,8 +3360,9 @@ static void mptcp_release_cb(struct sock *sk)
 		msk->push_pending = 0;
 		msk->cb_flags &= ~flags;
 		spin_unlock_bh(&sk->sk_lock.slock);
+
 		if (flags & BIT(MPTCP_FLUSH_JOIN_LIST))
-			__mptcp_flush_join_list(sk);
+			__mptcp_flush_join_list(sk, &join_list);
 		if (flags & BIT(MPTCP_PUSH_PENDING))
 			__mptcp_push_pending(sk, 0);
 		if (flags & BIT(MPTCP_RETRANSMIT))