[tip: timers/core] tick: Move tick cancellation up to CPUHP_AP_TICK_DYING

From: tip-bot2 for Frederic Weisbecker
Date: Mon Feb 26 2024 - 17:26:00 EST


The following commit has been merged into the timers/core branch of tip:

Commit-ID: f04e51220ad5cf35540f67f3ca15c8617c1f0bef
Gitweb: https://git.kernel.org/tip/f04e51220ad5cf35540f67f3ca15c8617c1f0bef
Author: Frederic Weisbecker <frederic@xxxxxxxxxx>
AuthorDate: Sun, 25 Feb 2024 23:55:00 +01:00
Committer: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
CommitterDate: Mon, 26 Feb 2024 11:37:31 +01:00

tick: Move tick cancellation up to CPUHP_AP_TICK_DYING

The tick hrtimer is cancelled right before hrtimers are migrated. This
is done from the hrtimer subsystem even though it shouldn't know about
its actual users.

Move instead the tick hrtimer cancellation to the relevant CPU hotplug
state that aims at centralizing high level tick shutdown operations so
that the related flow is easy to follow.

Signed-off-by: Frederic Weisbecker <frederic@xxxxxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Reviewed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20240225225508.11587-9-frederic@xxxxxxxxxx

---
kernel/time/hrtimer.c | 2 --
kernel/time/tick-common.c | 2 ++
2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 95f1f35..3e95474 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -2229,8 +2229,6 @@ int hrtimers_cpu_dying(unsigned int dying_cpu)
int i, ncpu = cpumask_any_and(cpu_active_mask, housekeeping_cpumask(HK_TYPE_TIMER));
struct hrtimer_cpu_base *old_base, *new_base;

- tick_cancel_sched_timer(dying_cpu);
-
old_base = this_cpu_ptr(&hrtimer_bases);
new_base = &per_cpu(hrtimer_bases, ncpu);

diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index a89ef45..b4af8c7 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -410,6 +410,8 @@ int tick_cpu_dying(unsigned int dying_cpu)
if (tick_do_timer_cpu == dying_cpu)
tick_do_timer_cpu = cpumask_first(cpu_online_mask);

+ tick_cancel_sched_timer(dying_cpu);
+
return 0;
}