linux-kernel - Re: [PATCH v2 05/12] sched/fair: Optimize __update_sched

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160503084901.GI3430@twins.programming.kicks-ass.net>
Date:	Tue, 3 May 2016 10:49:01 +0200
From:	Peter Zijlstra <peterz@...radead.org>
To:	Yuyang Du <yuyang.du@...el.com>
Cc:	mingo@...nel.org, linux-kernel@...r.kernel.org, bsegall@...gle.com,
	pjt@...gle.com, morten.rasmussen@....com,
	vincent.guittot@...aro.org, dietmar.eggemann@....com,
	juri.lelli@....com
Subject: Re: [PATCH v2 05/12] sched/fair: Optimize __update_sched_avg()

On Tue, May 03, 2016 at 05:54:31AM +0800, Yuyang Du wrote:
> __update_sched_avg() has these steps:
> 1. add the left of the last incomplete period
> 2. decay old sum
> 3. accumulate new sum since last_update_time
> 4. add the current incomplete period
> 5. update averages
> 
> Previously, we separately computed steps 1, 3, and 4, leading to
> each one of them ugly in codes and costly in overhead. But actually
> they all do the same thing, so we combine them together. The result
> will be much cleaner codes and less CPU cycles.

I would very much like to see an explanation of how we can fold all that
here. Without me having to untangle the code first.

That also helps me to verify if the code does indeed implement what you
meant it to, etc..