From: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
To: balbir@linux.vnet.ibm.com
Cc: linux-mm <linux-mm@kvack.org>,
Andrew Morton <akpm@linux-foundation.org>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>,
Li Zefan <lizf@cn.fujitsu.com>, Paul Menage <menage@google.com>,
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Subject: Re: [PATCH -mmotm 4/5] memcg: avoid oom during recharge at task move
Date: Fri, 27 Nov 2009 13:58:10 +0900 [thread overview]
Message-ID: <20091127135810.ef5fee0b.nishimura@mxp.nes.nec.co.jp> (raw)
In-Reply-To: <20091124114358.80e0cafe.nishimura@mxp.nes.nec.co.jp>
[-- Attachment #1: Type: text/plain, Size: 1426 bytes --]
> > Sorry, if I missed it, but I did not see any time overhead of moving a
> > task after these changes. Could you please help me understand the cost
> > of moving say a task with 1G anonymous memory to another group and
> > the cost of moving a task with 512MB anonymous and 512 page cache
> > mapped, etc. It would be nice to understand the overall cost.
> >
> O.K.
> I'll test programs with big anonymous pages and measure the time and report.
>
I measured the elapsed time of "echo <pid> > <some path>/tasks" on KVM guest
with 4CPU/4GB(Xeon/3GHz).
- used the attached simple program.
- made 2 directories(00, 01) under root, and enabled recharge_at_immigrate in both.
- measured the elapsed time by "time -p" for moving between:
(1) root -> 00
(2) 00 -> 01
we don't need to call res_counter_uncharge against root, so (1) would be smaller
than (2).
(3) 00(setting mem.limit to half size of total) -> 01
To compare the overhead of anon and swap.
Results:
| 252M | 512M | 1G
-----+--------+--------+--------
(1) | 0.21 | 0.41 | 0.821
-----+--------+--------+--------
(2) | 0.43 | 0.85 | 1.71
-----+--------+--------+--------
(3) | 0.40 | 0.81 | 1.62
-----+--------+--------+--------
hmm, it would be better to add some comments to memory.txt like:
Note: It may take several seconds if you move charges in giga bytes order.
Regards,
Daisuke Nishimura.
[-- Attachment #2: bigmem.c --]
[-- Type: text/x-csrc, Size: 660 bytes --]
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#include <errno.h>
#include <string.h>
#include <sys/mman.h>
#include <unistd.h>
void
usage(void)
{
fprintf(stderr, "bigmem <anon size(MB)>\n");
}
int
main(int argc, char *argv[])
{
void *buf;
size_t size;
pid_t pid;
if (argc != 2) {
usage();
return 1;
}
pid = getpid();
fprintf(stdout, "pid is %d\n", pid);
size = atol(argv[1]) * 1024 * 1024;
buf = mmap(NULL, size, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, 0, 0);
if (buf == MAP_FAILED) {
perror(NULL);
return errno;
}
memset(buf, 0, size);
fprintf(stdout, "allocated %ld bytes anonymous memory\n");
pause();
}
next prev parent reply other threads:[~2009-11-27 5:02 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-11-19 4:27 [PATCH -mmotm 0/5] memcg: recharge at task move (19/Nov) Daisuke Nishimura
2009-11-19 4:28 ` [PATCH -mmotm 1/5] cgroup: introduce cancel_attach() Daisuke Nishimura
2009-11-19 21:42 ` Paul Menage
2009-11-19 23:49 ` Daisuke Nishimura
2009-11-19 4:29 ` [PATCH -mmotm 2/5] memcg: add interface to recharge at task move Daisuke Nishimura
2009-11-20 15:42 ` Balbir Singh
2009-11-23 23:56 ` Daisuke Nishimura
2009-11-19 4:29 ` [PATCH -mmotm 3/5] memcg: recharge charges of anonymous page Daisuke Nishimura
2009-11-19 4:30 ` [PATCH -mmotm 4/5] memcg: avoid oom during recharge at task move Daisuke Nishimura
2009-11-23 5:10 ` Balbir Singh
2009-11-24 2:43 ` Daisuke Nishimura
2009-11-27 4:58 ` Daisuke Nishimura [this message]
2009-12-03 4:58 ` Daisuke Nishimura
2009-12-03 5:22 ` KAMEZAWA Hiroyuki
2009-12-03 6:00 ` Daisuke Nishimura
2009-12-03 7:40 ` KAMEZAWA Hiroyuki
2009-11-19 4:31 ` [PATCH -mmotm 5/5] memcg: recharge charges of anonymous swap Daisuke Nishimura
2009-11-23 6:59 ` Balbir Singh
2009-11-24 7:54 ` Daisuke Nishimura
2009-11-19 19:03 ` [PATCH -mmotm 0/5] memcg: recharge at task move (19/Nov) Balbir Singh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20091127135810.ef5fee0b.nishimura@mxp.nes.nec.co.jp \
--to=nishimura@mxp.nes.nec.co.jp \
--cc=akpm@linux-foundation.org \
--cc=balbir@linux.vnet.ibm.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=lizf@cn.fujitsu.com \
--cc=menage@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox