public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@kernel.org>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Anna-Maria Behnsen <anna-maria@linutronix.de>,
	John Stultz <jstultz@google.com>, Stephen Boyd <sboyd@kernel.org>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Juri Lelli <juri.lelli@redhat.com>,
	Vincent Guittot <vincent.guittot@linaro.org>,
	Dietmar Eggemann <dietmar.eggemann@arm.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Ben Segall <bsegall@google.com>, Mel Gorman <mgorman@suse.de>,
	Valentin Schneider <vschneid@redhat.com>,
	x86@kernel.org, Peter Zijlstra <peterz@infradead.org>,
	Frederic Weisbecker <frederic@kernel.org>,
	Eric Dumazet <edumazet@google.com>
Subject: [patch 12/48] tick/sched: Avoid hrtimer_cancel/start() sequence
Date: Tue, 24 Feb 2026 17:36:10 +0100	[thread overview]
Message-ID: <20260224163429.542178086@kernel.org> (raw)
In-Reply-To: 20260224163022.795809588@kernel.org

The sequence of cancel and start is inefficient. It has to do the timer
lock/unlock twice and in the worst case has to reprogram the underlying
clock event device twice.

The reason why it is done this way is the usage of hrtimer_forward_now(),
which requires the timer to be inactive.

But that can be completely avoided as the forward can be done on a variable
and does not need any of the overrun accounting provided by
hrtimer_forward_now().

Implement a trivial forwarding mechanism and replace the cancel/reprogram
sequence with hrtimer_start(..., new_expiry).

For the non high resolution case the timer is not actually armed, but used
for storage so that code checking for expiry times can unconditially look
it up in the timer. So it is safe for that case to set the new expiry time
directly.

Signed-off-by: Thomas Gleixner <tglx@kernel.org
Cc: Frederic Weisbecker <frederic@kernel.org>
---
 kernel/time/tick-sched.c |   27 ++++++++++++++++++++-------
 1 file changed, 20 insertions(+), 7 deletions(-)

--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -864,19 +864,32 @@ u64 get_cpu_iowait_time_us(int cpu, u64
 }
 EXPORT_SYMBOL_GPL(get_cpu_iowait_time_us);
 
+/* Simplified variant of hrtimer_forward_now() */
+static ktime_t tick_forward_now(ktime_t expires, ktime_t now)
+{
+	ktime_t delta = now - expires;
+
+	if (likely(delta < TICK_NSEC))
+		return expires + TICK_NSEC;
+
+	expires += TICK_NSEC * ktime_divns(delta, TICK_NSEC);
+	if (expires > now)
+		return expires;
+	return expires + TICK_NSEC;
+}
+
 static void tick_nohz_restart(struct tick_sched *ts, ktime_t now)
 {
-	hrtimer_cancel(&ts->sched_timer);
-	hrtimer_set_expires(&ts->sched_timer, ts->last_tick);
+	ktime_t expires = ts->last_tick;
 
-	/* Forward the time to expire in the future */
-	hrtimer_forward(&ts->sched_timer, now, TICK_NSEC);
+	if (now >= expires)
+		expires = tick_forward_now(expires, now);
 
 	if (tick_sched_flag_test(ts, TS_FLAG_HIGHRES)) {
-		hrtimer_start_expires(&ts->sched_timer,
-				      HRTIMER_MODE_ABS_PINNED_HARD);
+		hrtimer_start(&ts->sched_timer,	expires, HRTIMER_MODE_ABS_PINNED_HARD);
 	} else {
-		tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
+		hrtimer_set_expires(&ts->sched_timer, expires);
+		tick_program_event(expires, 1);
 	}
 
 	/*


  parent reply	other threads:[~2026-02-24 16:36 UTC|newest]

Thread overview: 128+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-24 16:35 [patch 00/48] hrtimer,sched: General optimizations and hrtick enablement Thomas Gleixner
2026-02-24 16:35 ` [patch 01/48] sched/eevdf: Fix HRTICK duration Thomas Gleixner
2026-02-28 15:37   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-03-20 14:59     ` Shrikanth Hegde
2026-03-20 15:38       ` Peter Zijlstra
2026-03-20 15:40         ` Shrikanth Hegde
2026-02-24 16:35 ` [patch 02/48] sched/fair: Simplify hrtick_update() Thomas Gleixner
2026-02-28 15:37   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra (Intel)
2026-02-24 16:35 ` [patch 03/48] sched/fair: Make hrtick resched hard Thomas Gleixner
2026-02-28 15:37   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra (Intel)
2026-02-24 16:35 ` [patch 04/48] sched: Avoid ktime_get() indirection Thomas Gleixner
2026-02-28 15:37   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:35 ` [patch 05/48] hrtimer: Avoid pointless reprogramming in __hrtimer_start_range_ns() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:35 ` [patch 06/48] hrtimer: Provide a static branch based hrtimer_hres_enabled() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:35 ` [patch 07/48] sched: Use hrtimer_highres_enabled() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:35 ` [patch 08/48] sched: Optimize hrtimer handling Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:35 ` [patch 09/48] sched/hrtick: Avoid tiny hrtick rearms Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 10/48] hrtimer: Provide LAZY_REARM mode Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:36 ` [patch 11/48] sched/hrtick: Mark hrtick timer LAZY_REARM Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:36 ` Thomas Gleixner [this message]
2026-02-28 15:36   ` [tip: sched/hrtick] tick/sched: Avoid hrtimer_cancel/start() sequence tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 13/48] clockevents: Remove redundant CLOCK_EVT_FEAT_KTIME Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 14/48] timekeeping: Allow inlining clocksource::read() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 15/48] x86: Inline TSC reads in timekeeping Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 16/48] x86/apic: Remove pointless fence in lapic_next_deadline() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 17/48] x86/apic: Avoid the PVOPS indirection for the TSC deadline timer Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 18/48] timekeeping: Provide infrastructure for coupled clockevents Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 19/48] clockevents: Provide support for clocksource coupled comparators Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-03-03 18:44   ` [patch 19/48] " Michael Kelley
2026-03-03 19:14     ` Peter Zijlstra
2026-03-23  4:24     ` Michael Kelley
2026-03-23 21:36       ` Thomas Gleixner
2026-03-24  0:22         ` mhklkml
2026-03-24  3:37           ` Michael Kelley
2026-03-24 17:24             ` Thomas Gleixner
2026-03-24 17:34               ` Peter Zijlstra
2026-02-24 16:36 ` [patch 20/48] x86/apic: Enable TSC coupled programming mode Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-03-03  1:29   ` [patch 20/48] " Nathan Chancellor
2026-03-03 14:37     ` Thomas Gleixner
2026-03-03 14:45       ` Thomas Gleixner
2026-03-03 17:38       ` Nathan Chancellor
2026-03-03 20:21         ` Thomas Gleixner
2026-03-03 21:30           ` Nathan Chancellor
2026-03-04 18:40             ` Thomas Gleixner
2026-03-04 18:49               ` [patch 20/48] clocksource: Update clocksource::freq_khz on registration Thomas Gleixner
2026-03-04 19:10                 ` Borislav Petkov
2026-03-04 22:57                 ` Nathan Chancellor
2026-03-05 16:47                 ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-03-03 21:56         ` [PATCH] Subject: timekeeping: Initialize the coupled clocksource conversion completely Thomas Gleixner
2026-03-03 23:16           ` John Stultz
2026-03-05 16:47           ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 21/48] hrtimer: Add debug object init assertion Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:36 ` [patch 22/48] hrtimer: Reduce trace noise in hrtimer_start() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 23/48] hrtimer: Use guards where appropriate Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 24/48] hrtimer: Cleanup coding style and comments Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 25/48] hrtimer: Evaluate timer expiry only once Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 26/48] hrtimer: Replace the bitfield in hrtimer_cpu_base Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 27/48] hrtimer: Convert state and properties to boolean Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 28/48] hrtimer: Optimize for local timers Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 29/48] hrtimer: Use NOHZ information for locality Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 30/48] hrtimer: Separate remove/enqueue handling for local timers Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 31/48] hrtimer: Add hrtimer_rearm tracepoint Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 32/48] hrtimer: Re-arrange hrtimer_interrupt() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:37 ` [patch 33/48] hrtimer: Rename hrtimer_cpu_base::in_hrtirq to deferred_rearm Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:37 ` [patch 34/48] hrtimer: Prepare stubs for deferred rearming Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:38 ` [patch 35/48] entry: Prepare for deferred hrtimer rearming Thomas Gleixner
2026-02-27 15:57   ` Christian Loehle
2026-02-27 16:25     ` Peter Zijlstra
2026-02-27 16:32       ` Christian Loehle
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:38 ` [patch 36/48] softirq: " Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:38 ` [patch 37/48] sched/core: " Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:38 ` [patch 38/48] hrtimer: Push reprogramming timers into the interrupt return path Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-24 16:38 ` [patch 39/48] hrtimer: Avoid re-evaluation when nothing changed Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 40/48] hrtimer: Keep track of first expiring timer per clock base Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 41/48] hrtimer: Rework next event evaluation Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 42/48] hrtimer: Simplify run_hrtimer_queues() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 43/48] hrtimer: Optimize for_each_active_base() Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 44/48] rbtree: Provide rbtree with links Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 45/48] timerqueue: Provide linked timerqueue Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:38 ` [patch 46/48] hrtimer: Use " Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:39 ` [patch 47/48] hrtimer: Try to modify timers in place Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Thomas Gleixner
2026-02-24 16:39 ` [patch 48/48] sched: Default enable HRTICK when deferred rearming is enabled Thomas Gleixner
2026-02-28 15:36   ` [tip: sched/hrtick] " tip-bot2 for Peter Zijlstra
2026-02-25 15:25 ` [patch 00/48] hrtimer,sched: General optimizations and hrtick enablement Peter Zijlstra
2026-02-25 16:02   ` Thomas Gleixner
2026-03-04 15:59 ` Christian Loehle

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260224163429.542178086@kernel.org \
    --to=tglx@kernel.org \
    --cc=anna-maria@linutronix.de \
    --cc=bsegall@google.com \
    --cc=daniel.lezcano@linaro.org \
    --cc=dietmar.eggemann@arm.com \
    --cc=edumazet@google.com \
    --cc=frederic@kernel.org \
    --cc=jstultz@google.com \
    --cc=juri.lelli@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@suse.de \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=sboyd@kernel.org \
    --cc=vincent.guittot@linaro.org \
    --cc=vschneid@redhat.com \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox