mptcp: Check for orphaned subflow before handling MP_FAIL timer

MP_FAIL timeout (waiting for a peer to respond to an MP_FAIL with
another MP_FAIL) is implemented using the MPTCP socket's sk_timer. That
timer is also used at MPTCP socket close, so it's important to not have
the two timer users interfere with each other.

At MPTCP socket close, all subflows are orphaned before sk_timer is
manipulated. By checking the SOCK_DEAD flag on the subflows, each
subflow can determine if the timer is safe to alter without acquiring
any MPTCP-level lock. This replaces code that was using the
mptcp_data_lock and MPTCP-level socket state checks that did not
correctly protect the timer.

Fixes: 49fa1919d6bc ("mptcp: reset subflow when MP_FAIL doesn't respond")
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 8ba51120f35b38ae7970f048b2094255de5fae9a..59a85220edc9ced1b40a5b21ccc3826029ab43c2 100644 (file)
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -312,13 +312,10 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq)
                subflow->send_mp_fail = 1;
                MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPFAILTX);
                subflow->send_infinite_map = 1;
-       } else if (s && inet_sk_state_load(s) != TCP_CLOSE) {
+       } else if (!sock_flag(sk, SOCK_DEAD)) {
                pr_debug("MP_FAIL response received");
 
-               mptcp_data_lock(s);
-               if (inet_sk_state_load(s) != TCP_CLOSE)
-                       sk_stop_timer(s, &s->sk_timer);
-               mptcp_data_unlock(s);
+               sk_stop_timer(s, &s->sk_timer);
        }
 }
 

diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 27273cf091dbeda4e481a00a5ac0716963f7f06d..8841e8cd9ad8dbde4f9853e0dd16d6ace244f09e 100644 (file)
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1016,12 +1016,9 @@ static enum mapping_status get_mapping_status(struct sock *ssk,
                pr_debug("infinite mapping received");
                MPTCP_INC_STATS(sock_net(ssk), MPTCP_MIB_INFINITEMAPRX);
                subflow->map_data_len = 0;
-               if (sk && inet_sk_state_load(sk) != TCP_CLOSE) {
-                       mptcp_data_lock(sk);
-                       if (inet_sk_state_load(sk) != TCP_CLOSE)
-                               sk_stop_timer(sk, &sk->sk_timer);
-                       mptcp_data_unlock(sk);
-               }
+               if (!sock_flag(ssk, SOCK_DEAD))
+                       sk_stop_timer(sk, &sk->sk_timer);
+
                return MAPPING_INVALID;
        }
 
@@ -1241,9 +1238,8 @@ fallback:
                                tcp_send_active_reset(ssk, GFP_ATOMIC);
                                while ((skb = skb_peek(&ssk->sk_receive_queue)))
                                        sk_eat_skb(ssk, skb);
-                       } else {
+                       } else if (!sock_flag(ssk, SOCK_DEAD)) {
                                WRITE_ONCE(subflow->mp_fail_response_expect, true);
-                               /* The data lock is acquired in __mptcp_move_skbs() */
                                sk_reset_timer((struct sock *)msk,
                                               &((struct sock *)msk)->sk_timer,
                                               jiffies + TCP_RTO_MAX);