[PATCH net 0/2] mptcp: fix warn on bad status

[PATCH net 0/2] mptcp: fix warn on bad status

[Matthieu Baerts (NGI0)]

Two somewhat related fixes addressing different issues found by
syzkaller, and producing the exact same splat: a WARNING in
subflow_data_ready().

- Patch 1: fallback earlier on simultaneous connections to avoid a
  warning. A fix for v5.19.

- Patch 2: ensure context reset on disconnect, also to avoid a similar
  warning. A fix for v6.2.

Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
---
Paolo Abeni (2):
      mptcp: fallback earlier on simult connection
      mptcp: ensure context reset on disconnect()

 net/mptcp/options.c  | 10 ++++++++++
 net/mptcp/protocol.c |  8 +++++---
 net/mptcp/protocol.h |  9 ++++-----
 net/mptcp/subflow.c  |  6 ------
 4 files changed, 19 insertions(+), 14 deletions(-)
---
base-commit: 885bebac9909994050bbbeed0829c727e42bd1b7
change-id: 20251212-net-mptcp-subflow_data_ready-warn-fd8126208c90

Best regards,
-- 
Matthieu Baerts (NGI0) <matttbe@kernel.org>

[PATCH net 1/2] mptcp: fallback earlier on simult connection

[Matthieu Baerts (NGI0)]

From: Paolo Abeni <pabeni@redhat.com>

Syzkaller reports a simult-connect race leading to inconsistent fallback
status:

  WARNING: CPU: 3 PID: 33 at net/mptcp/subflow.c:1515 subflow_data_ready+0x40b/0x7c0 net/mptcp/subflow.c:1515
  Modules linked in:
  CPU: 3 UID: 0 PID: 33 Comm: ksoftirqd/3 Not tainted syzkaller #0 PREEMPT(full)
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
  RIP: 0010:subflow_data_ready+0x40b/0x7c0 net/mptcp/subflow.c:1515
  Code: 89 ee e8 78 61 3c f6 40 84 ed 75 21 e8 8e 66 3c f6 44 89 fe bf 07 00 00 00 e8 c1 61 3c f6 41 83 ff 07 74 09 e8 76 66 3c f6 90 <0f> 0b 90 e8 6d 66 3c f6 48 89 df e8 e5 ad ff ff 31 ff 89 c5 89 c6
  RSP: 0018:ffffc900006cf338 EFLAGS: 00010246
  RAX: 0000000000000000 RBX: ffff888031acd100 RCX: ffffffff8b7f2abf
  RDX: ffff88801e6ea440 RSI: ffffffff8b7f2aca RDI: 0000000000000005
  RBP: 0000000000000000 R08: 0000000000000005 R09: 0000000000000007
  R10: 0000000000000004 R11: 0000000000002c10 R12: ffff88802ba69900
  R13: 1ffff920000d9e67 R14: ffff888046f81800 R15: 0000000000000004
  FS:  0000000000000000(0000) GS:ffff8880d69bc000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000560fc0ca1670 CR3: 0000000032c3a000 CR4: 0000000000352ef0
  Call Trace:
   <TASK>
   tcp_data_queue+0x13b0/0x4f90 net/ipv4/tcp_input.c:5197
   tcp_rcv_state_process+0xfdf/0x4ec0 net/ipv4/tcp_input.c:6922
   tcp_v6_do_rcv+0x492/0x1740 net/ipv6/tcp_ipv6.c:1672
   tcp_v6_rcv+0x2976/0x41e0 net/ipv6/tcp_ipv6.c:1918
   ip6_protocol_deliver_rcu+0x188/0x1520 net/ipv6/ip6_input.c:438
   ip6_input_finish+0x1e4/0x4b0 net/ipv6/ip6_input.c:489
   NF_HOOK include/linux/netfilter.h:318 [inline]
   NF_HOOK include/linux/netfilter.h:312 [inline]
   ip6_input+0x105/0x2f0 net/ipv6/ip6_input.c:500
   dst_input include/net/dst.h:471 [inline]
   ip6_rcv_finish net/ipv6/ip6_input.c:79 [inline]
   NF_HOOK include/linux/netfilter.h:318 [inline]
   NF_HOOK include/linux/netfilter.h:312 [inline]
   ipv6_rcv+0x264/0x650 net/ipv6/ip6_input.c:311
   __netif_receive_skb_one_core+0x12d/0x1e0 net/core/dev.c:5979
   __netif_receive_skb+0x1d/0x160 net/core/dev.c:6092
   process_backlog+0x442/0x15e0 net/core/dev.c:6444
   __napi_poll.constprop.0+0xba/0x550 net/core/dev.c:7494
   napi_poll net/core/dev.c:7557 [inline]
   net_rx_action+0xa9f/0xfe0 net/core/dev.c:7684
   handle_softirqs+0x216/0x8e0 kernel/softirq.c:579
   run_ksoftirqd kernel/softirq.c:968 [inline]
   run_ksoftirqd+0x3a/0x60 kernel/softirq.c:960
   smpboot_thread_fn+0x3f7/0xae0 kernel/smpboot.c:160
   kthread+0x3c2/0x780 kernel/kthread.c:463
   ret_from_fork+0x5d7/0x6f0 arch/x86/kernel/process.c:148
   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
   </TASK>

The TCP subflow can process the simult-connect syn-ack packet after
transitioning to TCP_FIN1 state, bypassing the MPTCP fallback check,
as the sk_state_change() callback is not invoked for * -> FIN_WAIT1
transitions.

That will move the msk socket to an inconsistent status and the next
incoming data will hit the reported splat.

Close the race moving the simult-fallback check at the earliest possible
stage - that is at syn-ack generation time.

About the fixes tags: [2] was supposed to also fix this issue introduced
by [3]. [1] is required as a dependence: it was not explicitly marked as
a fix, but it is one and it has already been backported before [3]. In
other words, this commit should be backported up to [3], including [2]
and [1] if that's not already there.

Fixes: 23e89e8ee7be ("tcp: Don't drop SYN+ACK for simultaneous connect().") [1]
Fixes: 4fd19a307016 ("mptcp: fix inconsistent state on fastopen race") [2]
Fixes: 1e777f39b4d7 ("mptcp: add MSG_FASTOPEN sendmsg flag support") [3]
Cc: stable@vger.kernel.org
Reported-by: syzbot+0ff6b771b4f7a5bce83b@syzkaller.appspotmail.com
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/586
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
---
 net/mptcp/options.c  | 10 ++++++++++
 net/mptcp/protocol.h |  6 ++----
 net/mptcp/subflow.c  |  6 ------
 3 files changed, 12 insertions(+), 10 deletions(-)

diff --git a/net/mptcp/options.c b/net/mptcp/options.c
index f24ae7d40e88..43df4293f58b 100644
--- a/net/mptcp/options.c
+++ b/net/mptcp/options.c
@@ -408,6 +408,16 @@ bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
 	 */
 	subflow->snd_isn = TCP_SKB_CB(skb)->end_seq;
 	if (subflow->request_mptcp) {
+		if (unlikely(subflow_simultaneous_connect(sk))) {
+			WARN_ON_ONCE(!mptcp_try_fallback(sk, MPTCP_MIB_SIMULTCONNFALLBACK));
+
+			/* Ensure mptcp_finish_connect() will not process the
+			 * MPC handshake.
+			 */
+			subflow->request_mptcp = 0;
+			return false;
+		}
+
 		opts->suboptions = OPTION_MPTCP_MPC_SYN;
 		opts->csum_reqd = mptcp_is_checksum_enabled(sock_net(sk));
 		opts->allow_join_id0 = mptcp_allow_join_id0(sock_net(sk));
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 9c0d17876b22..bed0c9aa28b6 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -1337,10 +1337,8 @@ static inline bool subflow_simultaneous_connect(struct sock *sk)
 {
 	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
 
-	return (1 << sk->sk_state) &
-	       (TCPF_ESTABLISHED | TCPF_FIN_WAIT1 | TCPF_FIN_WAIT2 | TCPF_CLOSING) &&
-	       is_active_ssk(subflow) &&
-	       !subflow->conn_finished;
+	/* Note that the sk state implies !subflow->conn_finished. */
+	return sk->sk_state == TCP_SYN_RECV && is_active_ssk(subflow);
 }
 
 #ifdef CONFIG_SYN_COOKIES
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 86ce58ae533d..96d54cb2cd93 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -1878,12 +1878,6 @@ static void subflow_state_change(struct sock *sk)
 
 	__subflow_state_change(sk);
 
-	if (subflow_simultaneous_connect(sk)) {
-		WARN_ON_ONCE(!mptcp_try_fallback(sk, MPTCP_MIB_SIMULTCONNFALLBACK));
-		subflow->conn_finished = 1;
-		mptcp_propagate_state(parent, sk, subflow, NULL);
-	}
-
 	/* as recvmsg() does not acquire the subflow socket for ssk selection
 	 * a fin packet carrying a DSS can be unnoticed if we don't trigger
 	 * the data available machinery here.

-- 
2.51.0

[PATCH net 2/2] mptcp: ensure context reset on disconnect()

[Matthieu Baerts (NGI0)]

From: Paolo Abeni <pabeni@redhat.com>

After the blamed commit below, if the MPC subflow is already in TCP_CLOSE
status or has fallback to TCP at mptcp_disconnect() time,
mptcp_do_fastclose() skips setting the `send_fastclose flag` and the later
__mptcp_close_ssk() does not reset anymore the related subflow context.

Any later connection will be created with both the `request_mptcp` flag
and the msk-level fallback status off (it is unconditionally cleared at
MPTCP disconnect time), leading to a warning in subflow_data_ready():

  WARNING: CPU: 26 PID: 8996 at net/mptcp/subflow.c:1519 subflow_data_ready (net/mptcp/subflow.c:1519 (discriminator 13))
  Modules linked in:
  CPU: 26 UID: 0 PID: 8996 Comm: syz.22.39 Not tainted 6.18.0-rc7-05427-g11fc074f6c36 #1 PREEMPT(voluntary)
  Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
  RIP: 0010:subflow_data_ready (net/mptcp/subflow.c:1519 (discriminator 13))
  Code: 90 0f 0b 90 90 e9 04 fe ff ff e8 b7 1e f5 fe 89 ee bf 07 00 00 00 e8 db 19 f5 fe 83 fd 07 0f 84 35 ff ff ff e8 9d 1e f5 fe 90 <0f> 0b 90 e9 27 ff ff ff e8 8f 1e f5 fe 4c 89 e7 48 89 de e8 14 09
  RSP: 0018:ffffc9002646fb30 EFLAGS: 00010293
  RAX: 0000000000000000 RBX: ffff88813b218000 RCX: ffffffff825c8435
  RDX: ffff8881300b3580 RSI: ffffffff825c8443 RDI: 0000000000000005
  RBP: 000000000000000b R08: ffffffff825c8435 R09: 000000000000000b
  R10: 0000000000000005 R11: 0000000000000007 R12: ffff888131ac0000
  R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
  FS:  00007f88330af6c0(0000) GS:ffff888a93dd2000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f88330aefe8 CR3: 000000010ff59000 CR4: 0000000000350ef0
  Call Trace:
   <TASK>
   tcp_data_ready (net/ipv4/tcp_input.c:5356)
   tcp_data_queue (net/ipv4/tcp_input.c:5445)
   tcp_rcv_state_process (net/ipv4/tcp_input.c:7165)
   tcp_v4_do_rcv (net/ipv4/tcp_ipv4.c:1955)
   __release_sock (include/net/sock.h:1158 (discriminator 6) net/core/sock.c:3180 (discriminator 6))
   release_sock (net/core/sock.c:3737)
   mptcp_sendmsg (net/mptcp/protocol.c:1763 net/mptcp/protocol.c:1857)
   inet_sendmsg (net/ipv4/af_inet.c:853 (discriminator 7))
   __sys_sendto (net/socket.c:727 (discriminator 15) net/socket.c:742 (discriminator 15) net/socket.c:2244 (discriminator 15))
   __x64_sys_sendto (net/socket.c:2247)
   do_syscall_64 (arch/x86/entry/syscall_64.c:63 (discriminator 1) arch/x86/entry/syscall_64.c:94 (discriminator 1))
   entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)
  RIP: 0033:0x7f883326702d

Address the issue setting an explicit `fastclosing` flag at fastclose
time, and checking such flag after mptcp_do_fastclose().

Fixes: ae155060247b ("mptcp: fix duplicate reset on fastclose")
Cc: stable@vger.kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
---
 net/mptcp/protocol.c | 8 +++++---
 net/mptcp/protocol.h | 3 ++-
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 9b1fafd87cb9..f505b780f713 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -2467,10 +2467,10 @@ bool __mptcp_retransmit_pending_data(struct sock *sk)
  */
 static void __mptcp_subflow_disconnect(struct sock *ssk,
 				       struct mptcp_subflow_context *subflow,
-				       unsigned int flags)
+				       bool fastclosing)
 {
 	if (((1 << ssk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)) ||
-	    subflow->send_fastclose) {
+	    fastclosing) {
 		/* The MPTCP code never wait on the subflow sockets, TCP-level
 		 * disconnect should never fail
 		 */
@@ -2538,7 +2538,7 @@ static void __mptcp_close_ssk(struct sock *sk, struct sock *ssk,
 
 	need_push = (flags & MPTCP_CF_PUSH) && __mptcp_retransmit_pending_data(sk);
 	if (!dispose_it) {
-		__mptcp_subflow_disconnect(ssk, subflow, flags);
+		__mptcp_subflow_disconnect(ssk, subflow, msk->fastclosing);
 		release_sock(ssk);
 
 		goto out;
@@ -2884,6 +2884,7 @@ static void mptcp_do_fastclose(struct sock *sk)
 
 	mptcp_set_state(sk, TCP_CLOSE);
 	mptcp_backlog_purge(sk);
+	msk->fastclosing = 1;
 
 	/* Explicitly send the fastclose reset as need */
 	if (__mptcp_check_fallback(msk))
@@ -3418,6 +3419,7 @@ static int mptcp_disconnect(struct sock *sk, int flags)
 	msk->bytes_sent = 0;
 	msk->bytes_retrans = 0;
 	msk->rcvspace_init = 0;
+	msk->fastclosing = 0;
 
 	/* for fallback's sake */
 	WRITE_ONCE(msk->ack_seq, 0);
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index bed0c9aa28b6..66e973500791 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -320,7 +320,8 @@ struct mptcp_sock {
 			fastopening:1,
 			in_accept_queue:1,
 			free_first:1,
-			rcvspace_init:1;
+			rcvspace_init:1,
+			fastclosing:1;
 	u32		notsent_lowat;
 	int		keepalive_cnt;
 	int		keepalive_idle;

-- 
2.51.0