linux-kernel - Re: [PATCH] cpuidle: menu: use nr_running instead of cpuload for calculating perf mult

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: <1338844480.28282.147.camel@twins>
Date:	Mon, 04 Jun 2012 23:14:40 +0200
From:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	Arjan van de Ven <arjan@...ux.intel.com>
Cc:	Vladimir Davydov <vdavydov@...allels.com>,
	Ingo Molnar <mingo@...e.hu>, Len Brown <lenb@...nel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] cpuidle: menu: use nr_running instead of cpuload for
 calculating perf mult

On Mon, 2012-06-04 at 13:45 -0700, Arjan van de Ven wrote:
> On 6/4/2012 1:31 PM, Peter Zijlstra wrote:
> > 
> > And like I said many times now, if you inflate some of the idle periods,
> > the work shifts (it doesn't become less) and a next idle period will be
> > smaller -- since we'll only become idle again once all work is done.
> 
> this is what is not really correct.
> 
> you can be idle for many reasons, not just because you have no work
> left. most common is waiting for a disk IO. the idle period for that
> will not get shorter just because the previous one took more time.

Then you're back to synchronous behaviour and your earlier claim that it
is was not about latency response is false.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/