Skip to contentSkip to conversation

FAILED: patch "[PATCH] mptcp: pm: ADD_ADDR rtx: free sk if last" failed to apply to 5.10-stable tree

PATCH In review [email protected] 3 participants 3 messages May 12 – May 19

Subsystems (3)
Networking Stable branch Networking MPTCP
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x b7b9a461569734d33d3259d58d2507adfac107ed
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026051243-blah-wound-3e54@gregkh' --subject-prefix 'PATCH 5.10.y' HEAD^..

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From b7b9a461569734d33d3259d58d2507adfac107ed Mon Sep 17 00:00:00 2001
From: "Matthieu Baerts (NGI0)" <[email protected]>
Date: Tue, 5 May 2026 17:00:53 +0200
Subject: [PATCH] mptcp: pm: ADD_ADDR rtx: free sk if last

When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(), and released at the end.

If at that moment, it was the last reference being held, the sk would not be freed. sock_put() should then be called instead of __sock_put().

But that's not enough: if it is the last reference, sock_put() will call sk_free(), which will end up calling sk_stop_timer_sync() on the same timer, and waiting indefinitely to finish. So it is needed to mark that the timer is done at the end of the timer handler when it has not been rescheduled, not to call sk_stop_timer_sync() on "itself".

Fixes 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout")
Cc [email protected]
Reviewed-by Mat Martineau <[email protected]>
Signed-off-by Matthieu Baerts (NGI0) <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>
net/mptcp/pm.c +19 -11
index 2a01bf1b5bfd..8899327e59a1 100644
@@ -16,6 +16,7 @@ struct mptcp_pm_add_entry {
 	struct list_head	list;
 	struct mptcp_addr_info	addr;
 	u8			retrans_times;
+	bool			timer_done;
 	struct timer_list	add_timer;
 	struct mptcp_sock	*sock;
 	struct rcu_head		rcu;

@@ -327,22 +328,22 @@ static void mptcp_pm_add_timer(struct timer_list *timer)
 							      add_timer);
 	struct mptcp_sock *msk = entry->sock;
 	struct sock *sk = (struct sock *)msk;
-	unsigned int timeout;
+	unsigned int timeout = 0;
 
 	pr_debug("msk=%p\n", msk);
 
-	if (unlikely(inet_sk_state_load(sk) == TCP_CLOSE))
-		goto exit;
-
 	bh_lock_sock(sk);
+	if (unlikely(inet_sk_state_load(sk) == TCP_CLOSE))
+		goto out;
+
 	if (sock_owned_by_user(sk)) {
 		/* Try again later. */
-		sk_reset_timer(sk, timer, jiffies + HZ / 20);
+		timeout = HZ / 20;
 		goto out;
 	}
 
 	if (mptcp_pm_should_add_signal_addr(msk)) {
-		sk_reset_timer(sk, timer, jiffies + TCP_RTO_MAX / 8);
+		timeout = TCP_RTO_MAX / 8;
 		goto out;
 	}
 

@@ -360,8 +361,9 @@ static void mptcp_pm_add_timer(struct timer_list *timer)
 	}
 
 	if (entry->retrans_times < ADD_ADDR_RETRANS_MAX)
-		sk_reset_timer(sk, timer,
-			       jiffies + (timeout << entry->retrans_times));
+		timeout <<= entry->retrans_times;
+	else
+		timeout = 0;
 
 	spin_unlock_bh(&msk->pm.lock);
 

@@ -369,9 +371,13 @@ static void mptcp_pm_add_timer(struct timer_list *timer)
 		mptcp_pm_subflow_established(msk);
 
 out:
+	if (timeout)
+		sk_reset_timer(sk, timer, jiffies + timeout);
+	else
+		/* if sock_put calls sk_free: avoid waiting for this timer */
+		entry->timer_done = true;
 	bh_unlock_sock(sk);
-exit:
-	__sock_put(sk);
+	sock_put(sk);
 }
 
 struct mptcp_pm_add_entry *

@@ -434,6 +440,7 @@ bool mptcp_pm_alloc_anno_list(struct mptcp_sock *msk,
 
 	timer_setup(&add_entry->add_timer, mptcp_pm_add_timer, 0);
 reset_timer:
+	add_entry->timer_done = false;
 	timeout = mptcp_adjust_add_addr_timeout(msk);
 	if (timeout)
 		sk_reset_timer(sk, &add_entry->add_timer, jiffies + timeout);

@@ -454,7 +461,8 @@ static void mptcp_pm_free_anno_list(struct mptcp_sock *msk)
 	spin_unlock_bh(&msk->pm.lock);
 
 	list_for_each_entry_safe(entry, tmp, &free_list, list) {
-		sk_stop_timer_sync(sk, &entry->add_timer);
+		if (!entry->timer_done)
+			sk_stop_timer_sync(sk, &entry->add_timer);
 		kfree_rcu(entry, rcu);
 	}
 }
maintainer
#1 lore raw

PATCH 1/1 mptcp: pm: ADD_ADDR rtx: free sk if last

From: "Matthieu Baerts (NGI0)" <[email protected]>

[ Upstream commit b7b9a461569734d33d3259d58d2507adfac107ed ]

When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(), and released at the end.

If at that moment, it was the last reference being held, the sk would not be freed. sock_put() should then be called instead of __sock_put().

But that's not enough: if it is the last reference, sock_put() will call sk_free(), which will end up calling sk_stop_timer_sync() on the same timer, and waiting indefinitely to finish. So it is needed to mark that the timer is done at the end of the timer handler when it has not been rescheduled, not to call sk_stop_timer_sync() on "itself".

Fixes 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout")
Cc [email protected]
Reviewed-by Mat Martineau <[email protected]>
Signed-off-by Matthieu Baerts (NGI0) <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>
[ applied fix to pm_netlink.c instead of pm.c due to pre-split file layout ]
Signed-off-by Sasha Levin <[email protected]>
---
 net/mptcp/pm_netlink.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
net/mptcp/pm_netlink.c +8 -2
index b88246b8f21c1..1466634a38234 100644
@@ -34,6 +34,7 @@ struct mptcp_pm_add_entry {
 	struct timer_list	add_timer;
 	struct mptcp_sock	*sock;
 	u8			retrans_times;
+	bool			timer_done;
 	struct rcu_head		rcu;
 };
 

@@ -241,11 +242,14 @@ static void mptcp_pm_add_timer(struct timer_list *timer)
 
 	if (entry->retrans_times < ADD_ADDR_RETRANS_MAX)
 		sk_reset_timer(sk, timer, jiffies + TCP_RTO_MAX);
+	else
+		/* if sock_put calls sk_free: avoid waiting for this timer */
+		entry->timer_done = true;
 
 	spin_unlock_bh(&msk->pm.lock);
 
 out:
-	__sock_put(sk);
+	sock_put(sk);
 }
 
 struct mptcp_pm_add_entry *

@@ -296,6 +300,7 @@ static bool mptcp_pm_alloc_anno_list(struct mptcp_sock *msk,
 	add_entry->addr = entry->addr;
 	add_entry->sock = msk;
 	add_entry->retrans_times = 0;
+	add_entry->timer_done = false;
 
 	timer_setup(&add_entry->add_timer, mptcp_pm_add_timer, 0);
 	sk_reset_timer(sk, &add_entry->add_timer, jiffies + TCP_RTO_MAX);

@@ -316,7 +321,8 @@ void mptcp_pm_free_anno_list(struct mptcp_sock *msk)
 	spin_unlock_bh(&msk->pm.lock);
 
 	list_for_each_entry_safe(entry, tmp, &free_list, list) {
-		sk_stop_timer_sync(sk, &entry->add_timer);
+		if (!entry->timer_done)
+			sk_stop_timer_sync(sk, &entry->add_timer);
 		kfree_rcu(entry, rcu);
 	}
 }
Signature · 1 line
2.53.0
maintainer
#2 lore raw

Re: [PATCH 5.10.y] mptcp: pm: ADD_ADDR rtx: free sk if last

Hi Sasha,

Quoted 16 lines · Sasha Levin

From: "Matthieu Baerts (NGI0)" <[email protected]>

[ Upstream commit b7b9a461569734d33d3259d58d2507adfac107ed ]

When an ADD_ADDR is retransmitted, the sk is held in sk_reset_timer(), and released at the end.

If at that moment, it was the last reference being held, the sk would not be freed. sock_put() should then be called instead of __sock_put().

But that's not enough: if it is the last reference, sock_put() will call sk_free(), which will end up calling sk_stop_timer_sync() on the same timer, and waiting indefinitely to finish. So it is needed to mark that the timer is done at the end of the timer handler when it has not been rescheduled, not to call sk_stop_timer_sync() on "itself".

Quoted 8 lines
Fixes 00cfd77b9063 ("mptcp: retransmit ADD_ADDR when timeout")
Cc [email protected]
Reviewed-by Mat Martineau <[email protected]>
Signed-off-by Matthieu Baerts (NGI0) <[email protected]>
Link https://patch.msgid.link/[email protected] open here
Signed-off-by Jakub Kicinski <[email protected]>
[ applied fix to pm_netlink.c instead of pm.c due to pre-split file layout ]
Signed-off-by Sasha Levin <[email protected]>

See my comment for the same patch for 6.12: most conflicts fixed here shouldn't exist if patches you sent earlier are applied first. So a new version is needed.

Cheers,
Matt

Signature · 1 line
Sponsored by the NGI0 Core fund.