From: Paolo Abeni <[email protected]>

Sashiko noted that the two event can race, leading to inconsistent
status. Prevent the race using the synchronous timer stop operation.

Cc: [email protected]
Fixes: b29fcfb54cd7 ("mptcp: full disconnect implementation")
Signed-off-by: Paolo Abeni <[email protected]>
Reviewed-by: Matthieu Baerts (NGI0) <[email protected]>
Signed-off-by: Matthieu Baerts (NGI0) <[email protected]>
Link: 
https://patch.msgid.link/20260908-net-mptcp-misc-fixes-7-3-rc1-v2-6-df1de7034...@kernel.org
Signed-off-by: Jakub Kicinski <[email protected]>
(cherry picked from commit 85c580b0d8590520ae00a15c29e9fb9c99427a3e)

Rebase note: this tree lacks commits 9a5e5334adc0 ("tcp: remove
icsk->icsk_retransmit_timer") and e4141ea18a73 ("tcp: introduce
icsk->icsk_keepalive_timer"), so the timers stopped are still
icsk_retransmit_timer and sk_timer. It also lacks commit ea4eb2adb0d3
("mptcp: honour configured min/max RTO in retransmit paths"), which
initialises icsk_rto_min, so timer_ival is reset to TCP_RTO_MIN, as in
__mptcp_init_sock() here and in the 6.12.y backport.

https://virtuozzo.atlassian.net/browse/VSTOR-155322
Feature: fix ms/net

Signed-off-by: Eva Kurchatova <[email protected]>
---
 net/mptcp/protocol.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 40d78fe6e111..1d4754b20cea 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -3188,6 +3188,7 @@ static void mptcp_copy_inaddrs(struct sock *msk, const 
struct sock *ssk)
 
 static int mptcp_disconnect(struct sock *sk, int flags)
 {
+       struct inet_connection_sock *icsk = inet_csk(sk);
        struct mptcp_sock *msk = mptcp_sk(sk);
 
        /* We are on the fastopen error path. We can't call straight into the
@@ -3200,8 +3201,13 @@ static int mptcp_disconnect(struct sock *sk, int flags)
        mptcp_check_listen_stop(sk);
        mptcp_set_state(sk, TCP_CLOSE);
 
-       mptcp_stop_rtx_timer(sk);
-       mptcp_stop_tout_timer(sk);
+       /* The later subflow close can not kick again the tout timer,
+        * as the msk is already in closed status.
+        */
+       msk->timer_ival = TCP_RTO_MIN;
+       sk_stop_timer_sync(sk, &icsk->icsk_retransmit_timer);
+       icsk->icsk_mtup.probe_timestamp = 0;
+       sk_stop_timer_sync(sk, &sk->sk_timer);
 
        mptcp_pm_connection_closed(msk);
 
-- 
2.56.1

_______________________________________________
Devel mailing list
[email protected]
https://lists.openvz.org/mailman/listinfo/devel

Reply via email to