From: Peter Zijlstra <peterz@infradead.org>
To: Valentin Schneider <valentin.schneider@arm.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>,
Morten Rasmussen <morten.rasmussen@foss.arm.com>,
Ingo Molnar <mingo@kernel.org>,
linux-kernel <linux-kernel@vger.kernel.org>,
Brendan Jackman <brendan.jackman@arm.com>,
Dietmar Eggemann <dietmar.eggemann@arm.com>,
Morten Rasmussen <morten.rasmussen@arm.com>
Subject: Re: [RFC PATCH 2/5] sched: Add NOHZ_STATS_KICK
Date: Thu, 1 Feb 2018 19:16:34 +0100 [thread overview]
Message-ID: <20180201181634.GG2269@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <b1efd041-2faa-76e0-7d9b-3d2917e03012@arm.com>
On Mon, Jan 29, 2018 at 07:31:07PM +0000, Valentin Schneider wrote:
> But for load update via _nohz_idle_balance(), we iterate through all of the
> nohz CPUS and unconditionally call update_blocked_averages(). This could be
> avoided by remembering which CPUs have stale load before going idle.
> Initially I thought that was what nohz.stats_state was for, but it isn't.
> With Vincent's patches it's only ever set to either 0 or 1, but we could use
> it as a CPU mask, and use it to skip nohz CPUs that don't have stale load in
> _nohz_idle_balance() (when NOHZ_STATS_KICK).
Yes, you'd need to allocate a second cpumask, worse you need atomic
bitops to set and clear bits there.
That all _might_ be worth it... dunno.
next prev parent reply other threads:[~2018-02-01 18:16 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-21 10:21 [RFC PATCH 0/5] sched: On remote stats updates Peter Zijlstra
2017-12-21 10:21 ` [RFC PATCH 1/5] sched: Convert nohz_flags to atomic_t Peter Zijlstra
2017-12-21 10:21 ` [RFC PATCH 2/5] sched: Add NOHZ_STATS_KICK Peter Zijlstra
2017-12-21 16:23 ` Vincent Guittot
2017-12-21 16:56 ` Vincent Guittot
2017-12-22 7:59 ` Peter Zijlstra
2017-12-22 8:05 ` Vincent Guittot
2017-12-22 8:29 ` Peter Zijlstra
2017-12-22 9:12 ` Peter Zijlstra
2017-12-22 14:31 ` Peter Zijlstra
2017-12-22 14:34 ` Vincent Guittot
2017-12-22 14:32 ` Vincent Guittot
2017-12-22 18:56 ` Peter Zijlstra
2017-12-22 20:42 ` Peter Zijlstra
2018-01-02 15:44 ` Morten Rasmussen
2018-01-15 9:43 ` Peter Zijlstra
2018-01-18 10:32 ` Morten Rasmussen
2018-01-03 9:16 ` Vincent Guittot
2018-01-15 8:26 ` Vincent Guittot
2018-01-18 10:38 ` Morten Rasmussen
2018-01-24 8:25 ` Vincent Guittot
2018-01-29 18:43 ` Dietmar Eggemann
2018-01-30 8:00 ` Vincent Guittot
2018-01-29 19:31 ` Valentin Schneider
2018-01-30 8:32 ` Vincent Guittot
2018-01-30 11:41 ` Valentin Schneider
2018-01-30 13:05 ` Vincent Guittot
2018-02-05 22:18 ` Valentin Schneider
2018-02-06 9:22 ` Vincent Guittot
2018-02-01 18:16 ` Peter Zijlstra [this message]
2018-02-01 16:57 ` Peter Zijlstra
2018-02-01 17:26 ` Vincent Guittot
2018-02-01 18:10 ` Peter Zijlstra
2018-02-01 19:11 ` Vincent Guittot
2018-02-06 8:32 ` [PATCH 1/3] sched: Stop nohz stats when decayed Vincent Guittot
2018-02-06 8:32 ` [PATCH 2/3] sched: reduce the periodic update duration Vincent Guittot
2018-02-06 8:32 ` [PATCH 3/3] sched: update blocked load when newly idle Vincent Guittot
2018-02-06 14:32 ` Valentin Schneider
2018-02-06 16:17 ` Vincent Guittot
2018-02-06 16:32 ` Valentin Schneider
2018-02-06 8:55 ` [PATCH 1/3] sched: Stop nohz stats when decayed Vincent Guittot
2018-02-06 14:16 ` Valentin Schneider
2018-02-06 14:31 ` Vincent Guittot
2018-02-01 16:55 ` [RFC PATCH 2/5] sched: Add NOHZ_STATS_KICK Peter Zijlstra
2018-01-22 9:40 ` Dietmar Eggemann
2018-01-22 10:23 ` Vincent Guittot
2018-02-01 16:52 ` Peter Zijlstra
2018-02-01 17:25 ` Vincent Guittot
2017-12-22 7:56 ` Peter Zijlstra
2017-12-22 8:04 ` Vincent Guittot
2017-12-21 10:21 ` [RFC PATCH 3/5] sched: Restructure nohz_balance_kick Peter Zijlstra
2017-12-21 10:21 ` [RFC PATCH 4/5] sched: Add nohz stats balancing Peter Zijlstra
2017-12-21 10:21 ` [RFC PATCH 5/5] sched: Update blocked load from NEWIDLE Peter Zijlstra
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=20180201181634.GG2269@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=brendan.jackman@arm.com \
--cc=dietmar.eggemann@arm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=morten.rasmussen@arm.com \
--cc=morten.rasmussen@foss.arm.com \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.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