linux-kernel - Re: [PATCH 6/6 v4] memcg: fix performance of mem_cgroup_begin_update_page

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]

Message-Id: <20120214200842.df132ab1.kamezawa.hiroyu@jp.fujitsu.com>
Date:	Tue, 14 Feb 2012 20:08:42 +0900
From:	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
To:	Greg Thelen <gthelen@...gle.com>
Cc:	"linux-mm@...ck.org" <linux-mm@...ck.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"hannes@...xchg.org" <hannes@...xchg.org>,
	Michal Hocko <mhocko@...e.cz>,
	"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
	Ying Han <yinghan@...gle.com>, Hugh Dickins <hughd@...gle.com>
Subject: Re: [PATCH 6/6 v4] memcg: fix performance of
 mem_cgroup_begin_update_page_stat()

On Mon, 13 Feb 2012 23:22:50 -0800
Greg Thelen <gthelen@...gle.com> wrote:

> On Mon, Feb 13, 2012 at 7:16 PM, KAMEZAWA Hiroyuki
> <kamezawa.hiroyu@...fujitsu.com> wrote:
> > From 3377fd7b6e23a5d2a368c078eae27e2b49c4f4aa Mon Sep 17 00:00:00 2001
> > From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
> > Date: Mon, 6 Feb 2012 12:14:47 +0900
> > Subject: [PATCH 6/6] memcg: fix performance of mem_cgroup_begin_update_page_stat()
> >
> > mem_cgroup_begin_update_page_stat() should be very fast because
> > it's called very frequently. Now, it needs to look up page_cgroup
> > and its memcg....this is slow.
> >
> > This patch adds a global variable to check "a memcg is moving or not".
> 
> s/a memcg/any memcg/
> 
yes.

> > By this, the caller doesn't need to visit page_cgroup and memcg.
> 
> s/By/With/
> 
ok.

> > Here is a test result. A test program makes page faults onto a file,
> > MAP_SHARED and makes each page's page_mapcount(page) > 1, and free
> > the range by madvise() and page fault again.  This program causes
> > 26214400 times of page fault onto a file(size was 1G.) and shows
> > shows the cost of mem_cgroup_begin_update_page_stat().
> 
> Out of curiosity, what is the performance of the mmap program before
> this series?
> 

Score of 3 runs underlinux-next.
==
[root@...extal test]# time ./mmap 1G

real    0m21.041s
user    0m6.146s
sys     0m14.625s
[root@...extal test]# time ./mmap 1G

real    0m21.063s
user    0m6.019s
sys     0m14.776s
[root@...extal test]# time ./mmap 1G

real    0m21.178s
user    0m6.000s
sys     0m14.849s
==

My program is attached. This program is for checking cost of updating FILE_MAPPED.
I guess sys_time scores's error rate will be 0.2-0.3 sec.

Thanks,
-Kame
==
#include <stdio.h>
#include <stdlib.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <unistd.h>
#include <sys/mman.h>
#include <fcntl.h>

void reader(int fd, int size)
{
	int i, off, x;
	char *addr;

	addr = mmap(NULL, size, PROT_READ, MAP_SHARED, fd, 0);
	for (i = 0; i < 100; i++) {
		for(off = 0; off < size; off += 4096) {
			x += *(addr + off);
		}
		madvise(addr, size, MADV_DONTNEED);
	}
}

int main(int argc, char *argv[])
{
	int fd;
	char *addr, *c;
	unsigned long size;
	struct stat statbuf;

	fd = open(argv[1], O_RDONLY);
	if (fd < 0) {
		perror("cannot open file");
		return 1;
	}

	if (fstat(fd, &statbuf)) {
		perror("fstat failed");
		return 1;
	}
	size = statbuf.st_size;
	/* mmap in 2 place. */
	addr = mmap(NULL, size, PROT_READ, MAP_SHARED, fd, 0);
	mlock(addr, size);
	reader(fd, size);
}

--
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/