[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <jhj1rd45ope.mognet@arm.com>
Date: Thu, 25 Feb 2021 12:14:37 +0000
From: Valentin Schneider <valentin.schneider@....com>
To: Vincent Guittot <vincent.guittot@...aro.org>
Cc: Ingo Molnar <mingo@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Juri Lelli <juri.lelli@...hat.com>,
Dietmar Eggemann <dietmar.eggemann@....com>,
Steven Rostedt <rostedt@...dmis.org>,
Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
Daniel Bristot de Oliveira <bristot@...hat.com>,
linux-kernel <linux-kernel@...r.kernel.org>,
Joel Fernandes <joel@...lfernandes.org>,
Frederic Weisbecker <fweisbec@...il.com>,
Thomas Gleixner <tglx@...utronix.de>,
Qais Yousef <qais.yousef@....com>
Subject: Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance
On 25/02/21 09:05, Vincent Guittot wrote:
>> One last thing for patch 7: mayhaps we could do a tad better to avoid
>> duplicate updates going through a heapful of leaf cfs rqs, see
>>
>> http://lore.kernel.org/r/jhj4kiht7oh.mognet@arm.com
>
> rq->last_blocked_load_update_tick is there only to filter duplicate
> update during _nohz_idle_balance but not for other normal LB.
>
Right, update_nohz_stats() is now only used in _nohz_idle_balance()...
IIUC the pattern being covered here would be a CPUX getting kicked for nohz
stats/balance, then some CPUX-N (so it gets picked by find_new_ilb())
entering idle and getting kick in turn (less than a jiffy apart).
Powered by blists - more mailing lists