From 035bca3f017ee9dea3a5a756e77a6f7138cc6eea Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 13 Nov 2025 10:39:24 +0000 Subject: mptcp: fix race condition in mptcp_schedule_work() syzbot reported use-after-free in mptcp_schedule_work() [1] Issue here is that mptcp_schedule_work() schedules a work, then gets a refcount on sk->sk_refcnt if the work was scheduled. This refcount will be released by mptcp_worker(). [A] if (schedule_work(...)) { [B] sock_hold(sk); return true; } Problem is that mptcp_worker() can run immediately and complete before [B] We need instead : sock_hold(sk); if (schedule_work(...)) return true; sock_put(sk); [1] refcount_t: addition on 0; use-after-free. WARNING: CPU: 1 PID: 29 at lib/refcount.c:25 refcount_warn_saturate+0xfa/0x1d0 lib/refcount.c:25 Call Trace: __refcount_add include/linux/refcount.h:-1 [inline] __refcount_inc include/linux/refcount.h:366 [inline] refcount_inc include/linux/refcount.h:383 [inline] sock_hold include/net/sock.h:816 [inline] mptcp_schedule_work+0x164/0x1a0 net/mptcp/protocol.c:943 mptcp_tout_timer+0x21/0xa0 net/mptcp/protocol.c:2316 call_timer_fn+0x17e/0x5f0 kernel/time/timer.c:1747 expire_timers kernel/time/timer.c:1798 [inline] __run_timers kernel/time/timer.c:2372 [inline] __run_timer_base+0x648/0x970 kernel/time/timer.c:2384 run_timer_base kernel/time/timer.c:2393 [inline] run_timer_softirq+0xb7/0x180 kernel/time/timer.c:2403 handle_softirqs+0x22f/0x710 kernel/softirq.c:622 __do_softirq kernel/softirq.c:656 [inline] run_ktimerd+0xcf/0x190 kernel/softirq.c:1138 smpboot_thread_fn+0x542/0xa60 kernel/smpboot.c:160 kthread+0x711/0x8a0 kernel/kthread.c:463 ret_from_fork+0x4bc/0x870 arch/x86/kernel/process.c:158 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245 Cc: stable@vger.kernel.org Fixes: 3b1d6210a957 ("mptcp: implement and use MPTCP-level retransmission") Reported-by: syzbot+355158e7e301548a1424@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6915b46f.050a0220.3565dc.0028.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Reviewed-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251113103924.3737425-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) (limited to 'net/mptcp/protocol.c') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 2d6b8de35c44..e27e0fe2460f 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -935,14 +935,19 @@ static void mptcp_reset_rtx_timer(struct sock *sk) bool mptcp_schedule_work(struct sock *sk) { - if (inet_sk_state_load(sk) != TCP_CLOSE && - schedule_work(&mptcp_sk(sk)->work)) { - /* each subflow already holds a reference to the sk, and the - * workqueue is invoked by a subflow, so sk can't go away here. - */ - sock_hold(sk); + if (inet_sk_state_load(sk) == TCP_CLOSE) + return false; + + /* Get a reference on this socket, mptcp_worker() will release it. + * As mptcp_worker() might complete before us, we can not avoid + * a sock_hold()/sock_put() if schedule_work() returns false. + */ + sock_hold(sk); + + if (schedule_work(&mptcp_sk(sk)->work)) return true; - } + + sock_put(sk); return false; } -- cgit From 17393fa7b7086664be519e7230cb6ed7ec7d9462 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 18 Nov 2025 08:20:21 +0100 Subject: mptcp: fix premature close in case of fallback I'm observing very frequent self-tests failures in case of fallback when running on a CONFIG_PREEMPT kernel. The root cause is that subflow_sched_work_if_closed() closes any subflow as soon as it is half-closed and has no incoming data pending. That works well for regular subflows - MPTCP needs bi-directional connectivity to operate on a given subflow - but for fallback socket is race prone. When TCP peer closes the connection before the MPTCP one, subflow_sched_work_if_closed() will schedule the MPTCP worker to gracefully close the subflow, and shortly after will do another schedule to inject and process a dummy incoming DATA_FIN. On CONFIG_PREEMPT kernel, the MPTCP worker can kick-in and close the fallback subflow before subflow_sched_work_if_closed() is able to create the dummy DATA_FIN, unexpectedly interrupting the transfer. Address the issue explicitly avoiding closing fallback subflows on when the peer is only half-closed. Note that, when the subflow is able to create the DATA_FIN before the worker invocation, the worker will change the msk state before trying to close the subflow and will skip the latter operation as the msk will not match anymore the precondition in __mptcp_close_subflow(). Fixes: f09b0ad55a11 ("mptcp: close subflow when receiving TCP+FIN") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251118-net-mptcp-misc-fixes-6-18-rc6-v1-3-806d3781c95f@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'net/mptcp/protocol.c') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index e27e0fe2460f..e30e9043a694 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2563,7 +2563,8 @@ static void __mptcp_close_subflow(struct sock *sk) if (ssk_state != TCP_CLOSE && (ssk_state != TCP_CLOSE_WAIT || - inet_sk_state_load(sk) != TCP_ESTABLISHED)) + inet_sk_state_load(sk) != TCP_ESTABLISHED || + __mptcp_check_fallback(msk))) continue; /* 'subflow_data_ready' will re-sched once rx queue is empty */ -- cgit From 1bba3f219c5e8c29e63afa3c1fc24f875ebec119 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 18 Nov 2025 08:20:22 +0100 Subject: mptcp: do not fallback when OoO is present In case of DSS corruption, the MPTCP protocol tries to avoid the subflow reset if fallback is possible. Such corruptions happen in the receive path; to ensure fallback is possible the stack additionally needs to check for OoO data, otherwise the fallback will break the data stream. Fixes: e32d262c89e2 ("mptcp: handle consistently DSS corruption") Cc: stable@vger.kernel.org Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/598 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251118-net-mptcp-misc-fixes-6-18-rc6-v1-4-806d3781c95f@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'net/mptcp/protocol.c') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index e30e9043a694..6f0e8f670d83 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -76,6 +76,13 @@ bool __mptcp_try_fallback(struct mptcp_sock *msk, int fb_mib) if (__mptcp_check_fallback(msk)) return true; + /* The caller possibly is not holding the msk socket lock, but + * in the fallback case only the current subflow is touching + * the OoO queue. + */ + if (!RB_EMPTY_ROOT(&msk->out_of_order_queue)) + return false; + spin_lock_bh(&msk->fallback_lock); if (!msk->allow_infinite_fallback) { spin_unlock_bh(&msk->fallback_lock); -- cgit From fff0c87996672816a84c3386797a5e69751c5888 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 18 Nov 2025 08:20:23 +0100 Subject: mptcp: decouple mptcp fastclose from tcp close With the current fastclose implementation, the mptcp_do_fastclose() helper is in charge of two distinct actions: send the fastclose reset and cleanup the subflows. Formally decouple the two steps, ensuring that mptcp explicitly closes all the subflows after the mentioned helper. This will make the upcoming fix simpler, and allows dropping the 2nd argument from mptcp_destroy_common(). The Fixes tag is then the same as in the next commit to help with the backports. Fixes: d21f83485518 ("mptcp: use fastclose on more edge scenarios") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251118-net-mptcp-misc-fixes-6-18-rc6-v1-5-806d3781c95f@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) (limited to 'net/mptcp/protocol.c') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 6f0e8f670d83..c59246c1fde6 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2808,7 +2808,11 @@ static void mptcp_worker(struct work_struct *work) __mptcp_close_subflow(sk); if (mptcp_close_tout_expired(sk)) { + struct mptcp_subflow_context *subflow, *tmp; + mptcp_do_fastclose(sk); + mptcp_for_each_subflow_safe(msk, subflow, tmp) + __mptcp_close_ssk(sk, subflow->tcp_sock, subflow, 0); mptcp_close_wake_up(sk); } @@ -3233,7 +3237,8 @@ static int mptcp_disconnect(struct sock *sk, int flags) /* msk->subflow is still intact, the following will not free the first * subflow */ - mptcp_destroy_common(msk, MPTCP_CF_FASTCLOSE); + mptcp_do_fastclose(sk); + mptcp_destroy_common(msk); /* The first subflow is already in TCP_CLOSE status, the following * can't overlap with a fallback anymore @@ -3412,7 +3417,7 @@ void mptcp_rcv_space_init(struct mptcp_sock *msk, const struct sock *ssk) msk->rcvq_space.space = TCP_INIT_CWND * TCP_MSS_DEFAULT; } -void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags) +void mptcp_destroy_common(struct mptcp_sock *msk) { struct mptcp_subflow_context *subflow, *tmp; struct sock *sk = (struct sock *)msk; @@ -3421,7 +3426,7 @@ void mptcp_destroy_common(struct mptcp_sock *msk, unsigned int flags) /* join list will be eventually flushed (with rst) at sock lock release time */ mptcp_for_each_subflow_safe(msk, subflow, tmp) - __mptcp_close_ssk(sk, mptcp_subflow_tcp_sock(subflow), subflow, flags); + __mptcp_close_ssk(sk, mptcp_subflow_tcp_sock(subflow), subflow, 0); __skb_queue_purge(&sk->sk_receive_queue); skb_rbtree_purge(&msk->out_of_order_queue); @@ -3439,7 +3444,7 @@ static void mptcp_destroy(struct sock *sk) /* allow the following to close even the initial subflow */ msk->free_first = 1; - mptcp_destroy_common(msk, 0); + mptcp_destroy_common(msk); sk_sockets_allocated_dec(sk); } -- cgit From ae155060247be8dcae3802a95bd1bdf93ab3215d Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 18 Nov 2025 08:20:24 +0100 Subject: mptcp: fix duplicate reset on fastclose The CI reports sporadic failures of the fastclose self-tests. The root cause is a duplicate reset, not carrying the relevant MPTCP option. In the failing scenario the bad reset is received by the peer before the fastclose one, preventing the reception of the latter. Indeed there is window of opportunity at fastclose time for the following race: mptcp_do_fastclose __mptcp_close_ssk __tcp_close() tcp_set_state() [1] tcp_send_active_reset() [2] After [1] the stack will send reset to in-flight data reaching the now closed port. Such reset may race with [2]. Address the issue explicitly sending a single reset on fastclose before explicitly moving the subflow to close status. Fixes: d21f83485518 ("mptcp: use fastclose on more edge scenarios") Cc: stable@vger.kernel.org Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/596 Signed-off-by: Paolo Abeni Reviewed-by: Geliang Tang Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20251118-net-mptcp-misc-fixes-6-18-rc6-v1-6-806d3781c95f@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-) (limited to 'net/mptcp/protocol.c') diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index c59246c1fde6..a70267a74e3c 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2409,7 +2409,6 @@ bool __mptcp_retransmit_pending_data(struct sock *sk) /* flags for __mptcp_close_ssk() */ #define MPTCP_CF_PUSH BIT(1) -#define MPTCP_CF_FASTCLOSE BIT(2) /* be sure to send a reset only if the caller asked for it, also * clean completely the subflow status when the subflow reaches @@ -2420,7 +2419,7 @@ static void __mptcp_subflow_disconnect(struct sock *ssk, unsigned int flags) { if (((1 << ssk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) || - (flags & MPTCP_CF_FASTCLOSE)) { + subflow->send_fastclose) { /* The MPTCP code never wait on the subflow sockets, TCP-level * disconnect should never fail */ @@ -2467,14 +2466,8 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk, lock_sock_nested(ssk, SINGLE_DEPTH_NESTING); - if ((flags & MPTCP_CF_FASTCLOSE) && !__mptcp_check_fallback(msk)) { - /* be sure to force the tcp_close path - * to generate the egress reset - */ - ssk->sk_lingertime = 0; - sock_set_flag(ssk, SOCK_LINGER); - subflow->send_fastclose = 1; - } + if (subflow->send_fastclose && ssk->sk_state != TCP_CLOSE) + tcp_set_state(ssk, TCP_CLOSE); need_push = (flags & MPTCP_CF_PUSH) && __mptcp_retransmit_pending_data(sk); if (!dispose_it) { @@ -2779,9 +2772,26 @@ static void mptcp_do_fastclose(struct sock *sk) struct mptcp_sock *msk = mptcp_sk(sk); mptcp_set_state(sk, TCP_CLOSE); - mptcp_for_each_subflow_safe(msk, subflow, tmp) - __mptcp_close_ssk(sk, mptcp_subflow_tcp_sock(subflow), - subflow, MPTCP_CF_FASTCLOSE); + + /* Explicitly send the fastclose reset as need */ + if (__mptcp_check_fallback(msk)) + return; + + mptcp_for_each_subflow_safe(msk, subflow, tmp) { + struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + + lock_sock(ssk); + + /* Some subflow socket states don't allow/need a reset.*/ + if ((1 << ssk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE)) + goto unlock; + + subflow->send_fastclose = 1; + tcp_send_active_reset(ssk, ssk->sk_allocation, + SK_RST_REASON_TCP_ABORT_ON_CLOSE); +unlock: + release_sock(ssk); + } } static void mptcp_worker(struct work_struct *work) -- cgit