* Re: [PATCH] hrtimer: Use hard expiry when updating timers on the same base
2026-09-07 21:11 [PATCH] hrtimer: Use hard expiry when updating timers on the same base Andrea Parri
@ 2026-09-08 11:09 ` Peter Zijlstra
2026-09-08 13:32 ` Andrea Parri
0 siblings, 1 reply; 3+ messages in thread
From: Peter Zijlstra @ 2026-09-08 11:09 UTC (permalink / raw)
To: Andrea Parri
Cc: Anna-Maria Behnsen, Frederic Weisbecker, Thomas Gleixner,
linux-kernel, stable
On Mon, Sep 07, 2026 at 11:11:33PM +0200, Andrea Parri wrote:
> Rearming a queued timer with nonzero slack can leave the timerqueue out
> of order. remove_and_enqueue_same_base() checks the new soft expiry
> against its neighbours' hard expiries, then stores the new hard expiry
> in the node without requeueing it.
>
> For example, with A at 10 and B at 20, rearming A at 11 with slack 30
> passes the neighbour check but leaves A's hard expiry of 41 before B's
> 20. The same function also caches the soft expiry in base->expires_next
> when updating or inserting the first timer, giving next-event selection
> an earlier deadline than the queue head's hard expiry.
>
> Compute the hard expiry with the same saturating addition used by
> hrtimer_set_expires_range_ns(). Use it for the in-place ordering check
> and both updates to base->expires_next.
>
> Fixes: eddffab8282e3 ("hrtimer: Keep track of first expiring timer per clock base")
> Fixes: 343f2f4dc5425 ("hrtimer: Try to modify timers in place")
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Andrea Parri <parri.andrea@gmail.com>
Urgh!
> ---
> kernel/time/hrtimer.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
> index 530d61257b9a0..c5940323746db 100644
> --- a/kernel/time/hrtimer.c
> +++ b/kernel/time/hrtimer.c
> @@ -1261,6 +1261,7 @@ static inline bool
> remove_and_enqueue_same_base(struct hrtimer *timer, struct hrtimer_clock_base *base,
> const enum hrtimer_mode mode, ktime_t expires, u64 delta_ns)
> {
> + ktime_t hard_expires = ktime_add_safe(expires, ns_to_ktime(delta_ns));
> bool was_first = false;
>
> /* Remove it from the timer queue if active */
> @@ -1268,11 +1269,11 @@ remove_and_enqueue_same_base(struct hrtimer *timer, struct hrtimer_clock_base *b
> was_first = !timerqueue_linked_prev(&timer->node);
>
> /* Try to update in place to avoid the de/enqueue dance */
> - if (hrtimer_can_update_in_place(timer, base, expires)) {
> + if (hrtimer_can_update_in_place(timer, base, hard_expires)) {
> hrtimer_set_expires_range_ns(timer, expires, delta_ns);
> trace_hrtimer_start(timer, mode, true);
> if (was_first)
> - base->expires_next = expires;
> + base->expires_next = hard_expires;
> return was_first;
> }
>
> @@ -1291,7 +1292,7 @@ remove_and_enqueue_same_base(struct hrtimer *timer, struct hrtimer_clock_base *b
>
> /* If it's the first expiring timer now or again, update base */
> if (timerqueue_linked_add(&base->active, &timer->node)) {
> - base->expires_next = expires;
> + base->expires_next = hard_expires;
> return true;
> }
Can't we do something like so?
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 530d61257b9a..8c157c4fa699 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -1263,13 +1263,16 @@ remove_and_enqueue_same_base(struct hrtimer *timer, struct hrtimer_clock_base *b
{
bool was_first = false;
+ /* Set the new expiry time */
+ hrtimer_set_expires_range_ns(timer, expires, delta_ns);
+ expires = timer->node.expires;
+
/* Remove it from the timer queue if active */
if (timer->is_queued) {
was_first = !timerqueue_linked_prev(&timer->node);
/* Try to update in place to avoid the de/enqueue dance */
if (hrtimer_can_update_in_place(timer, base, expires)) {
- hrtimer_set_expires_range_ns(timer, expires, delta_ns);
trace_hrtimer_start(timer, mode, true);
if (was_first)
base->expires_next = expires;
@@ -1280,9 +1283,6 @@ remove_and_enqueue_same_base(struct hrtimer *timer, struct hrtimer_clock_base *b
timerqueue_linked_del(&base->active, &timer->node);
}
- /* Set the new expiry time */
- hrtimer_set_expires_range_ns(timer, expires, delta_ns);
-
debug_activate(timer, mode, timer->is_queued);
base->cpu_base->active_bases |= 1 << base->index;
^ permalink raw reply [flat|nested] 3+ messages in thread