linux-kernel - Re: [PATCH 0/2] memcg: improving scalability by reducing lock contention at charge/uncharge

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

Message-Id: <20091013102827.c0280e37.kamezawa.hiroyu@jp.fujitsu.com>
Date:	Tue, 13 Oct 2009 10:28:27 +0900
From:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
To:	Ying Han <yinghan@...gle.com>
Cc:	"linux-mm@...ck.org" <linux-mm@...ck.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
	"balbir@...ux.vnet.ibm.com" <balbir@...ux.vnet.ibm.com>,
	"nishimura@....nes.nec.co.jp" <nishimura@....nes.nec.co.jp>
Subject: Re: [PATCH 0/2] memcg: improving scalability by reducing lock 
 contention at charge/uncharge

On Mon, 12 Oct 2009 18:18:44 -0700
Ying Han <yinghan@...gle.com> wrote:

> Ok. After I am running the task in the child cgroup, I do see performance
> improvement on the page-faults .

Thank you. 
Hmm...I wonder what makes cache-miss dramatically larger.
4 process on 4 cpus ? (my script was for 8cpus.)

-Kame

> [Before]
>  Performance counter stats for './runpause.sh 10' (5 runs):
> 
>   226190.612998  task-clock-msecs         #      3.766 CPUs    ( +-   0.192%
> )
>            4454  context-switches         #      0.000 M/sec   ( +-  14.029%
> )
>              38  CPU-migrations           #      0.000 M/sec   ( +-  15.385%
> )
>        21445296  page-faults              #      0.095 M/sec   ( +-   1.686%
> )
>    498345012124  cycles                   #   2203.208 M/sec   ( +-   0.023%
> )
>    255638483632  instructions             #      0.513 IPC     ( +-   0.196%
> )
>     92240143452  cache-references         #    407.798 M/sec   ( +-   0.673%
> )
>       175412792  cache-misses             #      0.776 M/sec   ( +-   2.188%
> )
> 
>    60.068670564  seconds time elapsed   ( +-   0.014% )
> 
> [After]
>  Performance counter stats for './runpause.sh 10' (5 runs):
> 
>   214409.926571  task-clock-msecs         #      3.570 CPUs    ( +-   0.317%
> )
>            3097  context-switches         #      0.000 M/sec   ( +-  25.726%
> )
>              14  CPU-migrations           #      0.000 M/sec   ( +-  20.000%
> )
>        33977467  page-faults              #      0.158 M/sec   ( +-   4.884%
> )
>    472369769787  cycles                   #   2203.115 M/sec   ( +-   0.024%
> )
>    275624185415  instructions             #      0.583 IPC     ( +-   0.271%
> )
>     98359325470  cache-references         #    458.744 M/sec   ( +-   0.281%
> )
>       941121561  cache-misses             #      4.389 M/sec   ( +-   4.969%
> )
> 
>    60.052748032  seconds time elapsed   ( +-   0.013% )
> 
> --Ying
> 
> 
> On Mon, Oct 12, 2009 at 5:29 PM, KAMEZAWA Hiroyuki <
> kamezawa.hiroyu@...fujitsu.com> wrote:
> 
> > On Sun, 11 Oct 2009 11:34:39 -0700
> > Ying Han <yinghan@...gle.com> wrote:
> >
> > > 2009/10/10 KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
> >
> > > > This patch series is only for "child" cgroup. Sorry, I had to write it
> > > > clearer. No effects to root.
> > > >
> > >
> > > Ok, Thanks for making it clearer. :) So Do you mind post the cgroup+memcg
> > > configuration
> > > while you are running on your host?
> > >
> >
> > #mount -t cgroup /dev/null /cgroups -omemory
> > #mkdir /cgroups/A
> > #echo $$ > /cgroups/A
> >
> > and run test.
> >
> > Thanks,
> > -Kame
> >
> >
> >
> 

--
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/