From: Frederic Weisbecker <frederic@kernel.org>
To: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Cc: He Zhe <zhe.he@windriver.com>,
anna-maria@linutronix.de, linux-kernel@vger.kernel.org,
tglx@linutronix.de
Subject: Re: [PATCH] timers: Fix get_next_timer_interrupt() with no timers pending
Date: Sat, 10 Jul 2021 11:05:35 +0200 [thread overview]
Message-ID: <20210710090535.GA28305@lothringen> (raw)
In-Reply-To: <4409fa71931446d9cabd849431ee0098c9b31292.camel@redhat.com>
On Fri, Jul 09, 2021 at 04:13:25PM +0200, Nicolas Saenz Julienne wrote:
> 31cd0e119d50 ("timers: Recalculate next timer interrupt only when
> necessary") subtly altered get_next_timer_interrupt()'s behaviour. The
> function no longer consistently returns KTIME_MAX with no timers
> pending.
>
> In order to decide if there are any timers pending we check whether the
> next expiry will happen NEXT_TIMER_MAX_DELTA jiffies from now.
> Unfortunately, the next expiry time and the timer base clock are no
> longer updated in unison. The former changes upon certain timer
> operations (enqueue, expire, detach), whereas the latter keeps track of
> jiffies as they move forward. Ultimately breaking the logic above.
>
> A simplified example:
>
> - Upon entering get_next_timer_interrupt() with:
>
> jiffies = 1
> base->clk = 0;
> base->next_expiry = NEXT_TIMER_MAX_DELTA;
>
> 'base->next_expiry == base->clk + NEXT_TIMER_MAX_DELTA', the function
> returns KTIME_MAX.
>
> - 'base->clk' is updated to the jiffies value.
>
> - The next time we enter get_next_timer_interrupt(), taking into account
> no timer operations happened:
>
> base->clk = 1;
> base->next_expiry = NEXT_TIMER_MAX_DELTA;
>
> 'base->next_expiry != base->clk + NEXT_TIMER_MAX_DELTA', the function
> returns a valid expire time, which is incorrect.
>
> This ultimately might unnecessarily rearm sched's timer on nohz_full
> setups, and add latency to the system[1].
>
> So, introduce 'base->timers_pending'[2], update it every time
> 'base->next_expiry' changes, and use it in get_next_timer_interrupt().
>
> [1] See tick_nohz_stop_tick().
> [2] A quick pahole check on x86_64 and arm64 shows it doesn't make
> 'struct timer_base' any bigger.
>
> Fixes: 31cd0e119d50 ("timers: Recalculate next timer interrupt only when necessary")
> Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
He Zhe, does it fix your issue?
Thanks.
next prev parent reply other threads:[~2021-07-10 9:05 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-23 15:16 [PATCH v2] timers: Recalculate next timer interrupt only when necessary Frederic Weisbecker
2020-07-24 10:59 ` [tip: timers/core] " tip-bot2 for Frederic Weisbecker
2021-07-08 6:43 ` [PATCH v2] " He Zhe
2021-07-08 11:35 ` Frederic Weisbecker
2021-07-08 15:36 ` Frederic Weisbecker
2021-07-09 5:37 ` He Zhe
2021-07-09 8:43 ` Frederic Weisbecker
2021-07-09 9:25 ` He Zhe
2021-07-09 14:06 ` Nicolas Saenz Julienne
2021-07-09 14:13 ` [PATCH] timers: Fix get_next_timer_interrupt() with no timers pending Nicolas Saenz Julienne
2021-07-10 0:52 ` Frederic Weisbecker
2021-07-12 10:19 ` Nicolas Saenz Julienne
2021-07-16 16:38 ` Nicolas Saenz Julienne
2021-07-19 13:54 ` Frederic Weisbecker
2021-07-10 9:05 ` Frederic Weisbecker [this message]
2021-07-12 6:04 ` He Zhe
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=20210710090535.GA28305@lothringen \
--to=frederic@kernel.org \
--cc=anna-maria@linutronix.de \
--cc=linux-kernel@vger.kernel.org \
--cc=nsaenzju@redhat.com \
--cc=tglx@linutronix.de \
--cc=zhe.he@windriver.com \
/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