linux-kernel - Re: A question regarding TG load tracking functions in fair.c

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] [day] [month] [year] [list]

Date:	Fri, 28 Feb 2014 13:29:22 +0100
From:	Peter Zijlstra <peterz@...radead.org>
To:	Matthew Leach <matthew.leach@....com>
Cc:	Paul Turner <pjt@...gle.com>, linux-kernel@...r.kernel.org,
	Dietmar Eggemann <Dietmar.Eggemann@....com>
Subject: Re: A question regarding TG load tracking functions in fair.c

On Fri, Feb 28, 2014 at 12:04:19PM +0000, Matthew Leach wrote:
> Hi Paul,
> 
> I have a question regarding two lines of code in the functions
> __update_cfs_rq_tg_load_contrib and __update_tg_runnable_avg. AFAICS
> these functions update a load-tracking signal for the local RQ and an
> aggregated signal in the TG for all RQs that belong to this TG. The
> update is surrounded by conditions:
> 
>   if (abs(contrib) > cfs_rq->tg_runnable_contrib / 64) {
> 
> for __update_tg_runnable_avg and
> 
>   if (force_update || abs(tg_contrib) > cfs_rq->tg_load_contrib / 8) {
> 
> for __update_cfs_rq_tg_load_contrib. Is this an optimisation to prevent
> the signals being updated for small variations? Also, could you please
> shed some light on the specific numbers used?

Yes, its an optimization; updating global state is expensive. I'll let
Paul explain the specific numbers.
--
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/