[tip: timers/urgent] tick/broadcast: Move per CPU pointer access into the atomic section
From: tip-bot2 for Thomas Gleixner
Date: Wed Jul 31 2024 - 06:42:10 EST
The following commit has been merged into the timers/urgent branch of tip:
Commit-ID: 6881e75237a84093d0986f56223db3724619f26e
Gitweb: https://git.kernel.org/tip/6881e75237a84093d0986f56223db3724619f26e
Author: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
AuthorDate: Wed, 31 Jul 2024 12:23:51 +02:00
Committer: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
CommitterDate: Wed, 31 Jul 2024 12:37:43 +02:00
tick/broadcast: Move per CPU pointer access into the atomic section
The recent fix for making the take over of the broadcast timer more
reliable retrieves a per CPU pointer in preemptible context.
This went unnoticed as compilers hoist the access into the non-preemptible
region where the pointer is actually used. But of course it's valid that
the compiler keeps it at the place where the code puts it which rightfully
triggers:
BUG: using smp_processor_id() in preemptible [00000000] code:
caller is hotplug_cpu__broadcast_tick_pull+0x1c/0xc0
Move it to the actual usage site which is in a non-preemptible region.
Fixes: f7d43dd206e7 ("tick/broadcast: Make takeover of broadcast hrtimer reliable")
Reported-by: David Wang <00107082@xxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Tested-by: Yu Liao <liaoyu15@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Link: https://lore.kernel.org/all/87ttg56ers.ffs@tglx
---
kernel/time/tick-broadcast.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index b484309..ed58eeb 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -1141,7 +1141,6 @@ void tick_broadcast_switch_to_oneshot(void)
#ifdef CONFIG_HOTPLUG_CPU
void hotplug_cpu__broadcast_tick_pull(int deadcpu)
{
- struct tick_device *td = this_cpu_ptr(&tick_cpu_device);
struct clock_event_device *bc;
unsigned long flags;
@@ -1167,6 +1166,8 @@ void hotplug_cpu__broadcast_tick_pull(int deadcpu)
* device to avoid the starvation.
*/
if (tick_check_broadcast_expired()) {
+ struct tick_device *td = this_cpu_ptr(&tick_cpu_device);
+
cpumask_clear_cpu(smp_processor_id(), tick_broadcast_force_mask);
tick_program_event(td->evtdev->next_event, 1);
}