* FAILED: patch "[PATCH] tick/nohz_full: Don't abuse smp_call_function_single() in" failed to apply to 6.6-stable tree
@ 2024-06-17 17:54 gregkh
2024-06-17 20:35 ` [PATCH 6.6.y and earlier] tick/nohz_full: Don't abuse smp_call_function_single() in tick_setup_device() Thomas Gleixner
0 siblings, 1 reply; 3+ messages in thread
From: gregkh @ 2024-06-17 17:54 UTC (permalink / raw)
To: oleg, tglx; +Cc: stable
The patch below does not apply to the 6.6-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 <stable@vger.kernel.org>.
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-6.6.y
git checkout FETCH_HEAD
git cherry-pick -x 07c54cc5988f19c9642fd463c2dbdac7fc52f777
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '2024061706-smudgy-gumball-93c0@gregkh' --subject-prefix 'PATCH 6.6.y' HEAD^..
Possible dependencies:
07c54cc5988f ("tick/nohz_full: Don't abuse smp_call_function_single() in tick_setup_device()")
f87cbcb345d0 ("timekeeping: Use READ/WRITE_ONCE() for tick_do_timer_cpu")
a478ffb2ae23 ("tick: Move individual bit features to debuggable mask accesses")
3ce74f1a8566 ("tick: Move got_idle_tick away from common flags")
3ad6eb0683a1 ("tick: Start centralizing tick related CPU hotplug operations")
3650f49bfb95 ("tick/sched: Rename tick_nohz_stop_sched_tick() to tick_nohz_full_stop_tick()")
27dc08096ce4 ("tick: Use IS_ENABLED() whenever possible")
37263ba0c44b ("tick/nohz: Remove duplicate between lowres and highres handlers")
ffb7e01c4e65 ("tick/nohz: Remove duplicate between tick_nohz_switch_to_nohz() and tick_setup_sched_timer()")
4c532939aa2e ("tick/sched: Split out jiffies update helper function")
73129cf4b69c ("timers: Optimization for timer_base_try_to_set_idle()")
e2e1d724e948 ("timers: Move marking timer bases idle into tick_nohz_stop_tick()")
39ed699fb660 ("timers: Split out get next timer interrupt")
bebed6649e85 ("timers: Restructure get_next_timer_interrupt()")
f365d0550615 ("tick/sched: Add function description for tick_nohz_next_event()")
da65f29dada7 ("timers: Fix nextevt calculation when no timers are pending")
bb8caad5083f ("timers: Rework idle logic")
7a39a5080ef0 ("timers: Use already existing function for forwarding timer base")
b5e6f59888c7 ("timers: Move store of next event into __next_timer_interrupt()")
b573c73101d8 ("tracing/timers: Add tracepoint for tracking timer base is_idle flag")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 07c54cc5988f19c9642fd463c2dbdac7fc52f777 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@redhat.com>
Date: Tue, 28 May 2024 14:20:19 +0200
Subject: [PATCH] tick/nohz_full: Don't abuse smp_call_function_single() in
tick_setup_device()
After the recent commit 5097cbcb38e6 ("sched/isolation: Prevent boot crash
when the boot CPU is nohz_full") the kernel no longer crashes, but there is
another problem.
In this case tick_setup_device() calls tick_take_do_timer_from_boot() to
update tick_do_timer_cpu and this triggers the WARN_ON_ONCE(irqs_disabled)
in smp_call_function_single().
Kill tick_take_do_timer_from_boot() and just use WRITE_ONCE(), the new
comment explains why this is safe (thanks Thomas!).
Fixes: 08ae95f4fd3b ("nohz_full: Allow the boot CPU to be nohz_full")
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20240528122019.GA28794@redhat.com
Link: https://lore.kernel.org/all/20240522151742.GA10400@redhat.com
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index d88b13076b79..a47bcf71defc 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -178,26 +178,6 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
}
}
-#ifdef CONFIG_NO_HZ_FULL
-static void giveup_do_timer(void *info)
-{
- int cpu = *(unsigned int *)info;
-
- WARN_ON(tick_do_timer_cpu != smp_processor_id());
-
- tick_do_timer_cpu = cpu;
-}
-
-static void tick_take_do_timer_from_boot(void)
-{
- int cpu = smp_processor_id();
- int from = tick_do_timer_boot_cpu;
-
- if (from >= 0 && from != cpu)
- smp_call_function_single(from, giveup_do_timer, &cpu, 1);
-}
-#endif
-
/*
* Setup the tick device
*/
@@ -221,19 +201,25 @@ static void tick_setup_device(struct tick_device *td,
tick_next_period = ktime_get();
#ifdef CONFIG_NO_HZ_FULL
/*
- * The boot CPU may be nohz_full, in which case set
- * tick_do_timer_boot_cpu so the first housekeeping
- * secondary that comes up will take do_timer from
- * us.
+ * The boot CPU may be nohz_full, in which case the
+ * first housekeeping secondary will take do_timer()
+ * from it.
*/
if (tick_nohz_full_cpu(cpu))
tick_do_timer_boot_cpu = cpu;
- } else if (tick_do_timer_boot_cpu != -1 &&
- !tick_nohz_full_cpu(cpu)) {
- tick_take_do_timer_from_boot();
+ } else if (tick_do_timer_boot_cpu != -1 && !tick_nohz_full_cpu(cpu)) {
tick_do_timer_boot_cpu = -1;
- WARN_ON(READ_ONCE(tick_do_timer_cpu) != cpu);
+ /*
+ * The boot CPU will stay in periodic (NOHZ disabled)
+ * mode until clocksource_done_booting() called after
+ * smp_init() selects a high resolution clocksource and
+ * timekeeping_notify() kicks the NOHZ stuff alive.
+ *
+ * So this WRITE_ONCE can only race with the READ_ONCE
+ * check in tick_periodic() but this race is harmless.
+ */
+ WRITE_ONCE(tick_do_timer_cpu, cpu);
#endif
}
^ permalink raw reply related [flat|nested] 3+ messages in thread* [PATCH 6.6.y and earlier] tick/nohz_full: Don't abuse smp_call_function_single() in tick_setup_device()
2024-06-17 17:54 FAILED: patch "[PATCH] tick/nohz_full: Don't abuse smp_call_function_single() in" failed to apply to 6.6-stable tree gregkh
@ 2024-06-17 20:35 ` Thomas Gleixner
2024-06-19 9:28 ` Greg KH
0 siblings, 1 reply; 3+ messages in thread
From: Thomas Gleixner @ 2024-06-17 20:35 UTC (permalink / raw)
To: gregkh, oleg; +Cc: stable
From: Oleg Nesterov <oleg@redhat.com>
commit 07c54cc5988f19c9642fd463c2dbdac7fc52f777 upstream.
After the recent commit 5097cbcb38e6 ("sched/isolation: Prevent boot crash
when the boot CPU is nohz_full") the kernel no longer crashes, but there is
another problem.
In this case tick_setup_device() calls tick_take_do_timer_from_boot() to
update tick_do_timer_cpu and this triggers the WARN_ON_ONCE(irqs_disabled)
in smp_call_function_single().
Kill tick_take_do_timer_from_boot() and just use WRITE_ONCE(), the new
comment explains why this is safe (thanks Thomas!).
Fixes: 08ae95f4fd3b ("nohz_full: Allow the boot CPU to be nohz_full")
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20240528122019.GA28794@redhat.com
Link: https://lore.kernel.org/all/20240522151742.GA10400@redhat.com
---
Backport to v6.6.y and earlier
---
kernel/time/tick-common.c | 42 ++++++++++++++----------------------------
1 file changed, 14 insertions(+), 28 deletions(-)
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -177,26 +177,6 @@ void tick_setup_periodic(struct clock_ev
}
}
-#ifdef CONFIG_NO_HZ_FULL
-static void giveup_do_timer(void *info)
-{
- int cpu = *(unsigned int *)info;
-
- WARN_ON(tick_do_timer_cpu != smp_processor_id());
-
- tick_do_timer_cpu = cpu;
-}
-
-static void tick_take_do_timer_from_boot(void)
-{
- int cpu = smp_processor_id();
- int from = tick_do_timer_boot_cpu;
-
- if (from >= 0 && from != cpu)
- smp_call_function_single(from, giveup_do_timer, &cpu, 1);
-}
-#endif
-
/*
* Setup the tick device
*/
@@ -220,19 +200,25 @@ static void tick_setup_device(struct tic
tick_next_period = ktime_get();
#ifdef CONFIG_NO_HZ_FULL
/*
- * The boot CPU may be nohz_full, in which case set
- * tick_do_timer_boot_cpu so the first housekeeping
- * secondary that comes up will take do_timer from
- * us.
+ * The boot CPU may be nohz_full, in which case the
+ * first housekeeping secondary will take do_timer()
+ * from it.
*/
if (tick_nohz_full_cpu(cpu))
tick_do_timer_boot_cpu = cpu;
- } else if (tick_do_timer_boot_cpu != -1 &&
- !tick_nohz_full_cpu(cpu)) {
- tick_take_do_timer_from_boot();
+ } else if (tick_do_timer_boot_cpu != -1 && !tick_nohz_full_cpu(cpu)) {
tick_do_timer_boot_cpu = -1;
- WARN_ON(tick_do_timer_cpu != cpu);
+ /*
+ * The boot CPU will stay in periodic (NOHZ disabled)
+ * mode until clocksource_done_booting() called after
+ * smp_init() selects a high resolution clocksource and
+ * timekeeping_notify() kicks the NOHZ stuff alive.
+ *
+ * So this WRITE_ONCE can only race with the READ_ONCE
+ * check in tick_periodic() but this race is harmless.
+ */
+ WRITE_ONCE(tick_do_timer_cpu, cpu);
#endif
}
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2024-06-19 9:28 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-06-17 17:54 FAILED: patch "[PATCH] tick/nohz_full: Don't abuse smp_call_function_single() in" failed to apply to 6.6-stable tree gregkh
2024-06-17 20:35 ` [PATCH 6.6.y and earlier] tick/nohz_full: Don't abuse smp_call_function_single() in tick_setup_device() Thomas Gleixner
2024-06-19 9:28 ` Greg KH
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).