[PATCH net-next v8 0/3] net: tcp: un-pin tw timer

[PATCH net-next v8 0/3] net: tcp: un-pin tw timer

[Florian Westphal]

Changes since previous iteration:
 - Patch 1: update a comment, I copied Erics v7 RvB tag.
 - Patch 2: move bh off/on into hashdance_schedule and get rid of
   comment mentioning pinned tw timer.
   I did not copy Erics RvB tag over from v7 because of the change.
 - Patch 3 is unchanged, so I kept Erics RvB tag.

This is v8 of the series where the tw_timer is un-pinned to get rid of
interferences in isolated CPUs setups.

First patch makes necessary preparations, existing code relies on
TIMER_PINNED to avoid races.

Second patch un-pins the TW timer. Could be folded into the first one,
but it might help wrt. bisection.

Third patch is a minor cleanup to move a helper from .h to the only
remaining compilation unit.

Tested with iperf3 and stress-ng socket mode.

Florian Westphal (2):
  net: tcp: un-pin the tw_timer
  tcp: move inet_twsk_schedule helper out of header

Valentin Schneider (1):
  net: tcp/dccp: prepare for tw_timer un-pinning

 include/net/inet_timewait_sock.h | 11 ++----
 net/dccp/minisocks.c             |  9 +----
 net/ipv4/inet_timewait_sock.c    | 63 +++++++++++++++++++++++++++-----
 net/ipv4/tcp_ipv4.c              |  2 +-
 net/ipv4/tcp_minisocks.c         |  9 +----
 5 files changed, 61 insertions(+), 33 deletions(-)

-- 
2.44.2

[PATCH net-next v8 1/3] net: tcp/dccp: prepare for tw_timer un-pinning

[Florian Westphal]

From: Valentin Schneider <vschneid@redhat.com>

The TCP timewait timer is proving to be problematic for setups where
scheduler CPU isolation is achieved at runtime via cpusets (as opposed to
statically via isolcpus=domains).

What happens there is a CPU goes through tcp_time_wait(), arming the
time_wait timer, then gets isolated. TCP_TIMEWAIT_LEN later, the timer
fires, causing interference for the now-isolated CPU. This is conceptually
similar to the issue described in commit e02b93124855 ("workqueue: Unbind
kworkers before sending them to exit()")

Move inet_twsk_schedule() to within inet_twsk_hashdance(), with the ehash
lock held. Expand the lock's critical section from inet_twsk_kill() to
inet_twsk_deschedule_put(), serializing the scheduling vs descheduling of
the timer. IOW, this prevents the following race:

			     tcp_time_wait()
			       inet_twsk_hashdance()
  inet_twsk_deschedule_put()
    del_timer_sync()
			       inet_twsk_schedule()

Thanks to Paolo Abeni for suggesting to leverage the ehash lock.

This also restores a comment from commit ec94c2696f0b ("tcp/dccp: avoid
one atomic operation for timewait hashdance") as inet_twsk_hashdance() had
a "Step 1" and "Step 3" comment, but the "Step 2" had gone missing.

inet_twsk_deschedule_put() now acquires the ehash spinlock to synchronize
with inet_twsk_hashdance_schedule().

To ease possible regression search, actual un-pin is done in next patch.

Link: https://lore.kernel.org/all/ZPhpfMjSiHVjQkTk@localhost.localdomain/
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
Co-developed-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 Change since last version:
   - adjust a comment in tcp_twsk_unique().

 include/net/inet_timewait_sock.h |  6 ++--
 net/dccp/minisocks.c             |  3 +-
 net/ipv4/inet_timewait_sock.c    | 52 +++++++++++++++++++++++++++-----
 net/ipv4/tcp_ipv4.c              |  2 +-
 net/ipv4/tcp_minisocks.c         |  3 +-
 5 files changed, 52 insertions(+), 14 deletions(-)

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 2a536eea9424..5b43d220243d 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -93,8 +93,10 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
 					   struct inet_timewait_death_row *dr,
 					   const int state);
 
-void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
-			 struct inet_hashinfo *hashinfo);
+void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
+				  struct sock *sk,
+				  struct inet_hashinfo *hashinfo,
+				  int timeo);
 
 void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo,
 			  bool rearm);
diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index 251a57cf5822..deb52d7d31b4 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -59,11 +59,10 @@ void dccp_time_wait(struct sock *sk, int state, int timeo)
 		 * we complete the initialization.
 		 */
 		local_bh_disable();
-		inet_twsk_schedule(tw, timeo);
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
-		inet_twsk_hashdance(tw, sk, &dccp_hashinfo);
+		inet_twsk_hashdance_schedule(tw, sk, &dccp_hashinfo, timeo);
 		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index e28075f0006e..628d33a41ce5 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -96,9 +96,13 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
  * Enter the time wait state. This is called with locally disabled BH.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
  * from the SK, and mess with hash chains and list linkage.
+ *
+ * The caller must not access @tw anymore after this function returns.
  */
-void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
-			   struct inet_hashinfo *hashinfo)
+void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
+				  struct sock *sk,
+				  struct inet_hashinfo *hashinfo,
+				  int timeo)
 {
 	const struct inet_sock *inet = inet_sk(sk);
 	const struct inet_connection_sock *icsk = inet_csk(sk);
@@ -129,26 +133,33 @@ void inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 
 	spin_lock(lock);
 
+	/* Step 2: Hash TW into tcp ehash chain */
 	inet_twsk_add_node_rcu(tw, &ehead->chain);
 
 	/* Step 3: Remove SK from hash chain */
 	if (__sk_nulls_del_node_init_rcu(sk))
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
 
-	spin_unlock(lock);
 
+	/* Ensure above writes are committed into memory before updating the
+	 * refcount.
+	 * Provides ordering vs later refcount_inc().
+	 */
+	smp_wmb();
 	/* tw_refcnt is set to 3 because we have :
 	 * - one reference for bhash chain.
 	 * - one reference for ehash chain.
 	 * - one reference for timer.
-	 * We can use atomic_set() because prior spin_lock()/spin_unlock()
-	 * committed into memory all tw fields.
 	 * Also note that after this point, we lost our implicit reference
 	 * so we are not allowed to use tw anymore.
 	 */
 	refcount_set(&tw->tw_refcnt, 3);
+
+	inet_twsk_schedule(tw, timeo);
+
+	spin_unlock(lock);
 }
-EXPORT_SYMBOL_GPL(inet_twsk_hashdance);
+EXPORT_SYMBOL_GPL(inet_twsk_hashdance_schedule);
 
 static void tw_timer_handler(struct timer_list *t)
 {
@@ -217,7 +228,34 @@ EXPORT_SYMBOL_GPL(inet_twsk_alloc);
  */
 void inet_twsk_deschedule_put(struct inet_timewait_sock *tw)
 {
-	if (del_timer_sync(&tw->tw_timer))
+	struct inet_hashinfo *hashinfo = tw->tw_dr->hashinfo;
+	spinlock_t *lock = inet_ehash_lockp(hashinfo, tw->tw_hash);
+
+	/* inet_twsk_purge() walks over all sockets, including tw ones,
+	 * and removes them via inet_twsk_deschedule_put() after a
+	 * refcount_inc_not_zero().
+	 *
+	 * inet_twsk_hashdance_schedule() must (re)init the refcount before
+	 * arming the timer, i.e. inet_twsk_purge can obtain a reference to
+	 * a twsk that did not yet schedule the timer.
+	 *
+	 * The ehash lock synchronizes these two:
+	 * After acquiring the lock, the timer is always scheduled (else
+	 * timer_shutdown returns false), because hashdance_schedule releases
+	 * the ehash lock only after completing the timer initialization.
+	 *
+	 * Without grabbing the ehash lock, we get:
+	 * 1) cpu x sets twsk refcount to 3
+	 * 2) cpu y bumps refcount to 4
+	 * 3) cpu y calls inet_twsk_deschedule_put() and shuts timer down
+	 * 4) cpu x tries to start timer, but mod_timer is a noop post-shutdown
+	 * -> timer refcount is never decremented.
+	 */
+	spin_lock(lock);
+	/*  Makes sure hashdance_schedule() has completed */
+	spin_unlock(lock);
+
+	if (timer_shutdown_sync(&tw->tw_timer))
 		inet_twsk_kill(tw);
 	inet_twsk_put(tw);
 }
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 3613e08ca794..e61c7c974745 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -157,7 +157,7 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
 	if (ts_recent_stamp &&
 	    (!twp || (reuse && time_after32(ktime_get_seconds(),
 					    ts_recent_stamp)))) {
-		/* inet_twsk_hashdance() sets sk_refcnt after putting twsk
+		/* inet_twsk_hashdance_schedule() sets sk_refcnt after putting twsk
 		 * and releasing the bucket lock.
 		 */
 		if (unlikely(!refcount_inc_not_zero(&sktw->sk_refcnt)))
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 4c894e540730..fc9a850ed9bd 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -350,11 +350,10 @@ void tcp_time_wait(struct sock *sk, int state, int timeo)
 		 * we complete the initialization.
 		 */
 		local_bh_disable();
-		inet_twsk_schedule(tw, timeo);
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
-		inet_twsk_hashdance(tw, sk, net->ipv4.tcp_death_row.hashinfo);
+		inet_twsk_hashdance_schedule(tw, sk, net->ipv4.tcp_death_row.hashinfo, timeo);
 		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
-- 
2.44.2

[PATCH net-next v8 2/3] net: tcp: un-pin the tw_timer

[Florian Westphal]

After previous patch, even if timer fires immediately on another CPU,
context that schedules the timer now holds the ehash spinlock, so timer
cannot reap tw socket until ehash lock is released.

BH disable is moved into hashdance_schedule.

Signed-off-by: Florian Westphal <fw@strlen.de>
---
 Changes since last version:
 Move local_bh_enable/disable into inet_twsk_hashdance_schedule
 and remove the comments mentioning timer_pinned in the two callers.

 net/dccp/minisocks.c          | 6 ------
 net/ipv4/inet_timewait_sock.c | 6 ++++--
 net/ipv4/tcp_minisocks.c      | 6 ------
 3 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c
index deb52d7d31b4..fecc8190064f 100644
--- a/net/dccp/minisocks.c
+++ b/net/dccp/minisocks.c
@@ -54,16 +54,10 @@ void dccp_time_wait(struct sock *sk, int state, int timeo)
 		if (state == DCCP_TIME_WAIT)
 			timeo = DCCP_TIMEWAIT_LEN;
 
-		/* tw_timer is pinned, so we need to make sure BH are disabled
-		 * in following section, otherwise timer handler could run before
-		 * we complete the initialization.
-		 */
-		local_bh_disable();
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
 		inet_twsk_hashdance_schedule(tw, sk, &dccp_hashinfo, timeo);
-		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
 		 * socket up.  We've got bigger problems than
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 628d33a41ce5..b2d97c816c99 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -93,7 +93,7 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
 }
 
 /*
- * Enter the time wait state. This is called with locally disabled BH.
+ * Enter the time wait state.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
  * from the SK, and mess with hash chains and list linkage.
  *
@@ -118,6 +118,7 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 			hashinfo->bhash_size)];
 	bhead2 = inet_bhashfn_portaddr(hashinfo, sk, twsk_net(tw), inet->inet_num);
 
+	local_bh_disable();
 	spin_lock(&bhead->lock);
 	spin_lock(&bhead2->lock);
 
@@ -158,6 +159,7 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 	inet_twsk_schedule(tw, timeo);
 
 	spin_unlock(lock);
+	local_bh_enable();
 }
 EXPORT_SYMBOL_GPL(inet_twsk_hashdance_schedule);
 
@@ -203,7 +205,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
 		tw->tw_prot	    = sk->sk_prot_creator;
 		atomic64_set(&tw->tw_cookie, atomic64_read(&sk->sk_cookie));
 		twsk_net_set(tw, sock_net(sk));
-		timer_setup(&tw->tw_timer, tw_timer_handler, TIMER_PINNED);
+		timer_setup(&tw->tw_timer, tw_timer_handler, 0);
 		/*
 		 * Because we use RCU lookups, we should not set tw_refcnt
 		 * to a non null value before everything is setup for this
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index fc9a850ed9bd..bc67f6b9efae 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -345,16 +345,10 @@ void tcp_time_wait(struct sock *sk, int state, int timeo)
 		if (state == TCP_TIME_WAIT)
 			timeo = TCP_TIMEWAIT_LEN;
 
-		/* tw_timer is pinned, so we need to make sure BH are disabled
-		 * in following section, otherwise timer handler could run before
-		 * we complete the initialization.
-		 */
-		local_bh_disable();
 		/* Linkage updates.
 		 * Note that access to tw after this point is illegal.
 		 */
 		inet_twsk_hashdance_schedule(tw, sk, net->ipv4.tcp_death_row.hashinfo, timeo);
-		local_bh_enable();
 	} else {
 		/* Sorry, if we're out of memory, just CLOSE this
 		 * socket up.  We've got bigger problems than
-- 
2.44.2

Re: [PATCH net-next v8 2/3] net: tcp: un-pin the tw_timer

[Eric Dumazet]

On Thu, Jun 6, 2024 at 5:18 PM Florian Westphal <fw@strlen.de> wrote:
>
> After previous patch, even if timer fires immediately on another CPU,
> context that schedules the timer now holds the ehash spinlock, so timer
> cannot reap tw socket until ehash lock is released.
>
> BH disable is moved into hashdance_schedule.
>
> Signed-off-by: Florian Westphal <fw@strlen.de>
> ---
>

Reviewed-by: Eric Dumazet <edumazet@google.com>

[PATCH net-next v8 3/3] tcp: move inet_twsk_schedule helper out of header

[Florian Westphal]

Its no longer used outside inet_timewait_sock.c, so move it there.

Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 No changes since last version.

 include/net/inet_timewait_sock.h | 5 -----
 net/ipv4/inet_timewait_sock.c    | 5 +++++
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h
index 5b43d220243d..f88b68269012 100644
--- a/include/net/inet_timewait_sock.h
+++ b/include/net/inet_timewait_sock.h
@@ -101,11 +101,6 @@ void inet_twsk_hashdance_schedule(struct inet_timewait_sock *tw,
 void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo,
 			  bool rearm);
 
-static inline void inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo)
-{
-	__inet_twsk_schedule(tw, timeo, false);
-}
-
 static inline void inet_twsk_reschedule(struct inet_timewait_sock *tw, int timeo)
 {
 	__inet_twsk_schedule(tw, timeo, true);
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index b2d97c816c99..337390ba85b4 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -92,6 +92,11 @@ static void inet_twsk_add_node_rcu(struct inet_timewait_sock *tw,
 	hlist_nulls_add_head_rcu(&tw->tw_node, list);
 }
 
+static void inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo)
+{
+	__inet_twsk_schedule(tw, timeo, false);
+}
+
 /*
  * Enter the time wait state.
  * Essentially we whip up a timewait bucket, copy the relevant info into it
-- 
2.44.2

Re: [PATCH net-next v8 0/3] net: tcp: un-pin tw timer

[Sebastian Andrzej Siewior]

On 2024-06-06 17:11:36 [+0200], Florian Westphal wrote:
> Changes since previous iteration:

Reviewed-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>

Sebastian

Re: [PATCH net-next v8 0/3] net: tcp: un-pin tw timer

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by David S. Miller <davem@davemloft.net>:

On Thu,  6 Jun 2024 17:11:36 +0200 you wrote:
> Changes since previous iteration:
>  - Patch 1: update a comment, I copied Erics v7 RvB tag.
>  - Patch 2: move bh off/on into hashdance_schedule and get rid of
>    comment mentioning pinned tw timer.
>    I did not copy Erics RvB tag over from v7 because of the change.
>  - Patch 3 is unchanged, so I kept Erics RvB tag.
> 
> [...]

Here is the summary with links:
  - [net-next,v8,1/3] net: tcp/dccp: prepare for tw_timer un-pinning
    https://git.kernel.org/netdev/net-next/c/b334b924c9b7
  - [net-next,v8,2/3] net: tcp: un-pin the tw_timer
    https://git.kernel.org/netdev/net-next/c/c75ad7c759cc
  - [net-next,v8,3/3] tcp: move inet_twsk_schedule helper out of header
    https://git.kernel.org/netdev/net-next/c/f81d0dd2fde3

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html