[PATCH 6.6.y 1/2] tcp: introduce icsk->icsk_keepalive_timer

[Sasha Levin <sashal@kernel.org>]

From: Eric Dumazet <edumazet@google.com>

[ Upstream commit 08dfe370239e53494453cee1e2ded2cdaa1efd12 ]

sk->sk_timer has been used for TCP keepalives.

Keepalive timers are not in fast path, we want to use sk->sk_timer
storage for retransmit timers, for better cache locality.

Create icsk->icsk_keepalive_timer and change keepalive
code to no longer use sk->sk_timer.

Added space is reclaimed in the following patch.

This includes changes to MPTCP, which was also using sk_timer.

Alias icsk->mptcp_tout_timer and icsk->icsk_keepalive_timer
for inet_sk_diag_fill() sake.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>
Link: https://patch.msgid.link/20251124175013.1473655-4-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Backport adaptation for Linux 6.6 and target 85c580b0d859:

Keep icsk_timeout and the existing inet_csk keepalive helpers, updating
those helpers rather than introducing the newer TCP-specific functions.
Use from_timer() and update the DCCP callback to match the shared timer
initializer. Drop the cacheline documentation hunk: that file does not
exist in this branch.

Provide the target's sk->mptcp_retransmit_timer alias and move all MPTCP
retransmit timer operations to that storage, now freed by moving the
timeout timer. TCP retransmit timers retain their stable-tree layout.
Initialize MPTCP's icsk_rto_min to TCP_RTO_MIN so the target's disconnect
reset has a valid value.

Move the existing disconnect close-event notification into the existing
mptcp_pm_connection_closed() function and call it from disconnect, matching
the target's patch context without changing event delivery. No functions
are added. These adaptations allow the target patch to apply unchanged.

[ sashal: Reduced backport -- upstream 08dfe370239e5 touches 11 file(s), this
  backport carries 13. Not backported here:
  Documentation/networking/net_cachelines/inet_connection_sock.rst
  This note is generated from the file lists only; see the resolution record
  for the reasoning. ]

Stable-dep-of: 85c580b0d859 ("mptcp: prevent race between disconnect() and rtx")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/net/inet_connection_sock.h            | 11 +++++--
 include/net/sock.h                            |  6 +++-
 net/dccp/timer.c                              |  4 ++-
 net/ipv4/inet_connection_sock.c               | 10 +++---
 net/ipv4/inet_diag.c                          |  4 +--
 net/ipv4/tcp_ipv4.c                           |  4 +--
 net/ipv4/tcp_timer.c                          |  5 +--
 net/ipv6/tcp_ipv6.c                           |  4 +--
 net/mptcp/pm.c                                |  3 ++
 net/mptcp/protocol.c                          | 32 ++++++++-----------
 net/mptcp/protocol.h                          |  2 +-
 .../selftests/bpf/progs/bpf_iter_tcp4.c       |  4 +--
 .../selftests/bpf/progs/bpf_iter_tcp6.c       |  4 +--
 13 files changed, 53 insertions(+), 40 deletions(-)

diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 4ed883cad65a0..0a33e7cdcc8ba 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -63,6 +63,9 @@ struct inet_connection_sock_af_ops {
  * @icsk_bind2_hash:	   Bind node in the bhash2 table
  * @icsk_timeout:	   Timeout
  * @icsk_retransmit_timer: Resend (no ack)
+ * @icsk_delack_timer:     Delayed ACK timer
+ * @icsk_keepalive_timer:  Keepalive timer
+ * @mptcp_tout_timer:      MPTCP timeout timer
  * @icsk_rto:		   Retransmit timeout
  * @icsk_pmtu_cookie	   Last pmtu seen by socket
  * @icsk_ca_ops		   Pluggable congestion control hook
@@ -89,8 +92,12 @@ struct inet_connection_sock {
 	struct inet_bind_bucket	  *icsk_bind_hash;
 	struct inet_bind2_bucket  *icsk_bind2_hash;
 	unsigned long		  icsk_timeout;
- 	struct timer_list	  icsk_retransmit_timer;
- 	struct timer_list	  icsk_delack_timer;
+	struct timer_list	  icsk_retransmit_timer;
+	struct timer_list	  icsk_delack_timer;
+	union {
+		struct timer_list icsk_keepalive_timer;
+		struct timer_list mptcp_tout_timer;
+	};
 	__u32			  icsk_rto;
 	__u32                     icsk_rto_min;
 	__u32                     icsk_delack_max;
diff --git a/include/net/sock.h b/include/net/sock.h
index c1f129a532b20..f1a5c61d6acf9 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -315,6 +315,7 @@ struct sk_filter;
   *	@sk_txrehash: enable TX hash rethink
   *	@sk_filter: socket filtering instructions
   *	@sk_timer: sock cleanup timer
+  *	@mptcp_retransmit_timer: MPTCP retransmit timer
   *	@sk_stamp: time stamp of last packet received
   *	@sk_stamp_seq: lock for accessing sk_stamp on 32 bit architectures only
   *	@sk_tsflags: SO_TIMESTAMPING flags
@@ -462,7 +463,10 @@ struct sock {
 	__u32			sk_dst_pending_confirm;
 	u32			sk_pacing_status; /* see enum sk_pacing */
 	long			sk_sndtimeo;
-	struct timer_list	sk_timer;
+	union {
+		struct timer_list sk_timer;
+		struct timer_list mptcp_retransmit_timer;
+	};
 	__u32			sk_priority;
 	__u32			sk_mark;
 	unsigned long		sk_pacing_rate; /* bytes per second */
diff --git a/net/dccp/timer.c b/net/dccp/timer.c
index b3255e87cc7e1..b4fb33eb02fb0 100644
--- a/net/dccp/timer.c
+++ b/net/dccp/timer.c
@@ -160,7 +160,9 @@ static void dccp_write_timer(struct timer_list *t)
 
 static void dccp_keepalive_timer(struct timer_list *t)
 {
-	struct sock *sk = from_timer(sk, t, sk_timer);
+	struct inet_connection_sock *icsk =
+		from_timer(icsk, t, icsk_keepalive_timer);
+	struct sock *sk = &icsk->icsk_inet.sk;
 
 	pr_err("dccp should not use a keepalive timer !\n");
 	sock_put(sk);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 530d495ff7d28..c2883a4f25fc9 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -752,7 +752,7 @@ void inet_csk_init_xmit_timers(struct sock *sk,
 
 	timer_setup(&icsk->icsk_retransmit_timer, retransmit_handler, 0);
 	timer_setup(&icsk->icsk_delack_timer, delack_handler, 0);
-	timer_setup(&sk->sk_timer, keepalive_handler, 0);
+	timer_setup(&icsk->icsk_keepalive_timer, keepalive_handler, 0);
 	icsk->icsk_pending = icsk->icsk_ack.pending = 0;
 }
 EXPORT_SYMBOL(inet_csk_init_xmit_timers);
@@ -765,7 +765,7 @@ void inet_csk_clear_xmit_timers(struct sock *sk)
 
 	sk_stop_timer(sk, &icsk->icsk_retransmit_timer);
 	sk_stop_timer(sk, &icsk->icsk_delack_timer);
-	sk_stop_timer(sk, &sk->sk_timer);
+	sk_stop_timer(sk, &icsk->icsk_keepalive_timer);
 }
 EXPORT_SYMBOL(inet_csk_clear_xmit_timers);
 
@@ -780,18 +780,18 @@ void inet_csk_clear_xmit_timers_sync(struct sock *sk)
 
 	sk_stop_timer_sync(sk, &icsk->icsk_retransmit_timer);
 	sk_stop_timer_sync(sk, &icsk->icsk_delack_timer);
-	sk_stop_timer_sync(sk, &sk->sk_timer);
+	sk_stop_timer_sync(sk, &icsk->icsk_keepalive_timer);
 }
 
 void inet_csk_delete_keepalive_timer(struct sock *sk)
 {
-	sk_stop_timer(sk, &sk->sk_timer);
+	sk_stop_timer(sk, &inet_csk(sk)->icsk_keepalive_timer);
 }
 EXPORT_SYMBOL(inet_csk_delete_keepalive_timer);
 
 void inet_csk_reset_keepalive_timer(struct sock *sk, unsigned long len)
 {
-	sk_reset_timer(sk, &sk->sk_timer, jiffies + len);
+	sk_reset_timer(sk, &inet_csk(sk)->icsk_keepalive_timer, jiffies + len);
 }
 EXPORT_SYMBOL(inet_csk_reset_keepalive_timer);
 
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index 5d09ab3ed735e..2128f28e8248d 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -310,11 +310,11 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		r->idiag_retrans = icsk->icsk_probes_out;
 		r->idiag_expires =
 			jiffies_delta_to_msecs(icsk->icsk_timeout - jiffies);
-	} else if (timer_pending(&sk->sk_timer)) {
+	} else if (timer_pending(&icsk->icsk_keepalive_timer)) {
 		r->idiag_timer = 2;
 		r->idiag_retrans = icsk->icsk_probes_out;
 		r->idiag_expires =
-			jiffies_delta_to_msecs(sk->sk_timer.expires - jiffies);
+			jiffies_delta_to_msecs(icsk->icsk_keepalive_timer.expires - jiffies);
 	}
 
 	if ((ext & (1 << (INET_DIAG_INFO - 1))) && handler->idiag_info_size) {
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 638303d6691eb..b85db4825178b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2713,9 +2713,9 @@ static void get_tcp4_sock(struct sock *sk, struct seq_file *f, int i)
 	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
 		timer_active	= 4;
 		timer_expires	= icsk->icsk_timeout;
-	} else if (timer_pending(&sk->sk_timer)) {
+	} else if (timer_pending(&icsk->icsk_keepalive_timer)) {
 		timer_active	= 2;
-		timer_expires	= sk->sk_timer.expires;
+		timer_expires	= icsk->icsk_keepalive_timer.expires;
 	} else {
 		timer_active	= 0;
 		timer_expires = jiffies;
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index b65cd417b0f7c..1205fcb138af5 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -739,8 +739,9 @@ EXPORT_SYMBOL_GPL(tcp_set_keepalive);
 
 static void tcp_keepalive_timer (struct timer_list *t)
 {
-	struct sock *sk = from_timer(sk, t, sk_timer);
-	struct inet_connection_sock *icsk = inet_csk(sk);
+	struct inet_connection_sock *icsk =
+		from_timer(icsk, t, icsk_keepalive_timer);
+	struct sock *sk = &icsk->icsk_inet.sk;
 	struct tcp_sock *tp = tcp_sk(sk);
 	u32 elapsed;
 
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index f13fdd1214aaa..8b772983e47b9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1990,9 +1990,9 @@ static void get_tcp6_sock(struct seq_file *seq, struct sock *sp, int i)
 	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
 		timer_active	= 4;
 		timer_expires	= icsk->icsk_timeout;
-	} else if (timer_pending(&sp->sk_timer)) {
+	} else if (timer_pending(&icsk->icsk_keepalive_timer)) {
 		timer_active	= 2;
-		timer_expires	= sp->sk_timer.expires;
+		timer_expires	= icsk->icsk_keepalive_timer.expires;
 	} else {
 		timer_active	= 0;
 		timer_expires = jiffies;
diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c
index 0d21f4e693908..ee2ef0171969c 100644
--- a/net/mptcp/pm.c
+++ b/net/mptcp/pm.c
@@ -155,6 +155,9 @@ void mptcp_pm_fully_established(struct mptcp_sock *msk, const struct sock *ssk)
 void mptcp_pm_connection_closed(struct mptcp_sock *msk)
 {
 	pr_debug("msk=%p\n", msk);
+
+	if (msk->token)
+		mptcp_event(MPTCP_EVENT_CLOSED, msk, NULL, GFP_KERNEL);
 }
 
 void mptcp_pm_subflow_established(struct mptcp_sock *msk)
diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c
index 127438ef120e6..eb5ade30f1a31 100644
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -416,9 +416,7 @@ static bool __mptcp_move_skb(struct sock *sk, struct sk_buff *skb)
 
 static void mptcp_stop_rtx_timer(struct sock *sk)
 {
-	struct inet_connection_sock *icsk = inet_csk(sk);
-
-	sk_stop_timer(sk, &icsk->icsk_retransmit_timer);
+	sk_stop_timer(sk, &sk->mptcp_retransmit_timer);
 	mptcp_sk(sk)->timer_ival = 0;
 }
 
@@ -971,12 +969,11 @@ static void __mptcp_flush_join_list(struct sock *sk, struct list_head *join_list
 
 static bool mptcp_rtx_timer_pending(struct sock *sk)
 {
-	return timer_pending(&inet_csk(sk)->icsk_retransmit_timer);
+	return timer_pending(&sk->mptcp_retransmit_timer);
 }
 
 static void mptcp_reset_rtx_timer(struct sock *sk)
 {
-	struct inet_connection_sock *icsk = inet_csk(sk);
 	unsigned long tout;
 
 	/* prevent rescheduling on close */
@@ -984,7 +981,7 @@ static void mptcp_reset_rtx_timer(struct sock *sk)
 		return;
 
 	tout = mptcp_sk(sk)->timer_ival;
-	sk_reset_timer(sk, &icsk->icsk_retransmit_timer, jiffies + tout);
+	sk_reset_timer(sk, &sk->mptcp_retransmit_timer, jiffies + tout);
 }
 
 bool mptcp_schedule_work(struct sock *sk)
@@ -2353,9 +2350,7 @@ static int mptcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
 
 static void mptcp_retransmit_timer(struct timer_list *t)
 {
-	struct inet_connection_sock *icsk = from_timer(icsk, t,
-						       icsk_retransmit_timer);
-	struct sock *sk = &icsk->icsk_inet.sk;
+	struct sock *sk = from_timer(sk, t, mptcp_retransmit_timer);
 	struct mptcp_sock *msk = mptcp_sk(sk);
 
 	bh_lock_sock(sk);
@@ -2373,7 +2368,9 @@ static void mptcp_retransmit_timer(struct timer_list *t)
 
 static void mptcp_tout_timer(struct timer_list *t)
 {
-	struct sock *sk = from_timer(sk, t, sk_timer);
+	struct inet_connection_sock *icsk =
+		from_timer(icsk, t, mptcp_tout_timer);
+	struct sock *sk = &icsk->icsk_inet.sk;
 
 	mptcp_schedule_work(sk);
 	sock_put(sk);
@@ -2816,7 +2813,7 @@ void mptcp_reset_tout_timer(struct mptcp_sock *msk, unsigned long fail_tout)
 	 */
 	timeout = inet_csk(sk)->icsk_mtup.probe_timestamp ? close_timeout : fail_tout;
 
-	sk_reset_timer(sk, &sk->sk_timer, timeout);
+	sk_reset_timer(sk, &inet_csk(sk)->mptcp_tout_timer, timeout);
 }
 
 static void mptcp_mp_fail_no_response(struct mptcp_sock *msk)
@@ -2931,7 +2928,8 @@ static void __mptcp_init_sock(struct sock *sk)
 	msk->first_pending = NULL;
 	msk->rmem_fwd_alloc = 0;
 	WRITE_ONCE(msk->rmem_released, 0);
-	msk->timer_ival = TCP_RTO_MIN;
+	msk->sk.icsk_rto_min = TCP_RTO_MIN;
+	msk->timer_ival = msk->sk.icsk_rto_min;
 	msk->scaling_ratio = TCP_DEFAULT_SCALING_RATIO;
 
 	WRITE_ONCE(msk->first, NULL);
@@ -2945,9 +2943,8 @@ static void __mptcp_init_sock(struct sock *sk)
 	mptcp_pm_data_init(msk);
 	spin_lock_init(&msk->fallback_lock);
 
-	/* re-use the csk retrans timer for MPTCP-level retrans */
-	timer_setup(&msk->sk.icsk_retransmit_timer, mptcp_retransmit_timer, 0);
-	timer_setup(&sk->sk_timer, mptcp_tout_timer, 0);
+	timer_setup(&sk->mptcp_retransmit_timer, mptcp_retransmit_timer, 0);
+	timer_setup(&msk->sk.mptcp_tout_timer, mptcp_tout_timer, 0);
 }
 
 static void mptcp_ca_reset(struct sock *sk)
@@ -3148,7 +3145,7 @@ static void __mptcp_destroy_sock(struct sock *sk)
 	might_sleep();
 
 	mptcp_stop_rtx_timer(sk);
-	sk_stop_timer(sk, &sk->sk_timer);
+	sk_stop_timer(sk, &inet_csk(sk)->mptcp_tout_timer);
 	msk->pm.status = 0;
 	mptcp_release_sched(msk);
 
@@ -3320,8 +3317,7 @@ static int mptcp_disconnect(struct sock *sk, int flags)
 	mptcp_stop_rtx_timer(sk);
 	mptcp_stop_tout_timer(sk);
 
-	if (msk->token)
-		mptcp_event(MPTCP_EVENT_CLOSED, msk, NULL, GFP_KERNEL);
+	mptcp_pm_connection_closed(msk);
 
 	/* msk->subflow is still intact, the following will not free the first
 	 * subflow
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index af00d64e61f4b..6d390269da5c5 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -764,7 +764,7 @@ static inline void mptcp_stop_tout_timer(struct sock *sk)
 	if (!inet_csk(sk)->icsk_mtup.probe_timestamp)
 		return;
 
-	sk_stop_timer(sk, &sk->sk_timer);
+	sk_stop_timer(sk, &inet_csk(sk)->mptcp_tout_timer);
 	inet_csk(sk)->icsk_mtup.probe_timestamp = 0;
 }
 
diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c
index 92267abb462fc..be540970fcc1f 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp4.c
@@ -103,9 +103,9 @@ static int dump_tcp_sock(struct seq_file *seq, struct tcp_sock *tp,
 	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
 		timer_active = 4;
 		timer_expires = icsk->icsk_timeout;
-	} else if (timer_pending(&sp->sk_timer)) {
+	} else if (timer_pending(&icsk->icsk_keepalive_timer)) {
 		timer_active = 2;
-		timer_expires = sp->sk_timer.expires;
+		timer_expires = icsk->icsk_keepalive_timer.expires;
 	} else {
 		timer_active = 0;
 		timer_expires = bpf_jiffies64();
diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c
index 943f7bba180e7..9f72398fc3cc3 100644
--- a/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c
+++ b/tools/testing/selftests/bpf/progs/bpf_iter_tcp6.c
@@ -103,9 +103,9 @@ static int dump_tcp6_sock(struct seq_file *seq, struct tcp6_sock *tp,
 	} else if (icsk->icsk_pending == ICSK_TIME_PROBE0) {
 		timer_active = 4;
 		timer_expires = icsk->icsk_timeout;
-	} else if (timer_pending(&sp->sk_timer)) {
+	} else if (timer_pending(&icsk->icsk_keepalive_timer)) {
 		timer_active = 2;
-		timer_expires = sp->sk_timer.expires;
+		timer_expires = icsk->icsk_keepalive_timer.expires;
 	} else {
 		timer_active = 0;
 		timer_expires = bpf_jiffies64();
-- 
2.53.0