From: Scott Wood <swood@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@kernel.org>
Cc: linux-kernel@vger.kernel.org, Scott Wood <swood@redhat.com>
Subject: [PATCH 1/4] tick/sched: Forward timer even in nohz mode
Date: Mon, 16 Dec 2019 18:22:22 -0500 [thread overview]
Message-ID: <1576538545-13274-1-git-send-email-swood@redhat.com> (raw)
Currently when exiting nohz, the expiry will be forwarded as if we
had just run the timer. If we re-enter nohz before this new expiry,
and exit after, this forwarding will happen again. If this load pattern
recurs the tick can be indefinitely postponed.
To avoid this, use last_tick as-is rather than calling hrtimer_forward().
However, in some cases the tick *will* have just run (despite being
"stopped"), and leading to double timer execution.
To avoid that, forward the timer after every tick (regardless of nohz
status) and keep last_tick up-to-date. During nohz, last_tick will
reflect what the expiry would have been if not in nohz mode.
Signed-off-by: Scott Wood <swood@redhat.com>
---
kernel/time/tick-sched.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 8b192e67aabc..8936b604dd6c 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -642,9 +642,6 @@ 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);
- /* Forward the time to expire in the future */
- hrtimer_forward(&ts->sched_timer, now, tick_period);
-
if (ts->nohz_mode == NOHZ_MODE_HIGHRES) {
hrtimer_start_expires(&ts->sched_timer,
HRTIMER_MODE_ABS_PINNED_HARD);
@@ -1207,12 +1204,13 @@ static void tick_nohz_handler(struct clock_event_device *dev)
tick_sched_do_timer(ts, now);
tick_sched_handle(ts, regs);
+ hrtimer_forward(&ts->sched_timer, now, tick_period);
+ ts->last_tick = hrtimer_get_expires(&ts->sched_timer);
/* No need to reprogram if we are running tickless */
if (unlikely(ts->tick_stopped))
return;
- hrtimer_forward(&ts->sched_timer, now, tick_period);
tick_program_event(hrtimer_get_expires(&ts->sched_timer), 1);
}
@@ -1311,12 +1309,13 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
else
ts->next_tick = 0;
+ hrtimer_forward(timer, now, tick_period);
+ ts->last_tick = hrtimer_get_expires(&ts->sched_timer);
+
/* No need to reprogram if we are in idle or full dynticks mode */
if (unlikely(ts->tick_stopped))
return HRTIMER_NORESTART;
- hrtimer_forward(timer, now, tick_period);
-
return HRTIMER_RESTART;
}
--
1.8.3.1
next reply other threads:[~2019-12-16 23:22 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-16 23:22 Scott Wood [this message]
2019-12-16 23:22 ` [PATCH 2/4] tick/sched: Set last_tick in init paths Scott Wood
2019-12-16 23:22 ` [PATCH 3/4] sched/core: Don't skip remote tick for idle cpus Scott Wood
2019-12-16 23:22 ` [PATCH 4/4] timers/nohz: Update nohz load in remote tick Scott Wood
2020-01-07 9:26 ` Peter Zijlstra
2020-01-06 20:12 ` [PATCH 1/4] tick/sched: Forward timer even in nohz mode Scott Wood
2020-01-06 22:18 ` Frederic Weisbecker
2020-01-08 23:18 ` Scott Wood
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=1576538545-13274-1-git-send-email-swood@redhat.com \
--to=swood@redhat.com \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/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
all inboxes | Powered by JetHome®