linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: "linux-mm@kvack.org" <linux-mm@kvack.org>
Cc: Ying Han <yinghan@google.com>,
	"hugh.dickins@tiscali.co.uk" <hugh.dickins@tiscali.co.uk>,
	"hannes@cmpxchg.org" <hannes@cmpxchg.org>,
	Michal Hocko <mhocko@suse.cz>,
	cgroups@vger.kernel.org,
	"bsingharora@gmail.com" <bsingharora@gmail.com>
Subject: [RFC] [PATCH 0/7 v2] memcg: page_cgroup diet
Date: Fri, 13 Jan 2012 17:30:01 +0900	[thread overview]
Message-ID: <20120113173001.ee5260ca.kamezawa.hiroyu@jp.fujitsu.com> (raw)

This is just an RFC for dumping my queue to share and get better idea.
Patch order may not be clean. Advice is welcomed.

Now, struct page_cgroup is defined as
==
struct page_cgroup {
        unsigned long flags;
        struct mem_cgroup *mem_cgroup;
};
==

We want to remove ->flags to shrink the size (and integrate into 'struct page').
To do that, we need to remove some flags.

Now, flag is defined as
==
        PCG_LOCK,  /* Lock for pc->mem_cgroup and following bits. */
        PCG_CACHE, /* charged as cache */
        PCG_USED, /* this object is in use. */
        PCG_MIGRATION, /* under page migration */
        /* flags for mem_cgroup and file and I/O status */
        PCG_MOVE_LOCK, /* For race between move_account v.s. following bits */
        PCG_FILE_MAPPED, /* page is accounted as "mapped" */
==
We have 6bits now.

This patch series removes PCG_CACHE, PCG_MOVE_LOCK, PCG_FILE_MAPPED.
Then, if we use low 3bits of ->mem_cgroup for PCG_LOCK, PCG_USED, PCG_MIGRATION,
we can remove pc->flags, I guess.

To remove flags, this patch modifes page-stat accounting. After this set, 
per-memcg page stat accounting will be

	mem_cgroup_begin_update_page_stat() --(A)
	modify page status
	mem_cgroup_update_page_stat()
	mem_cgroup_end_update_page_stat()   --(B)

Between (A) and (B), it's guaranteed the page's pc->mem_cgroup will not be moved.
By this change, move_account() can make use of page's information rather than
page_cgroup's flag and we don't have to duplicate flags in page_cgroup.
I think this is saner and allow us to add more per-memcg vmstat without any
new flags.

I'm now testing but don't see additional overheads.

Thanks,
-Kame




--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

             reply	other threads:[~2012-01-13  8:31 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-13  8:30 KAMEZAWA Hiroyuki [this message]
2012-01-13  8:32 ` [RFC] [PATCH 1/7 v2] memcg: remove unnecessary check in mem_cgroup_update_page_stat() KAMEZAWA Hiroyuki
2012-01-17 15:16   ` Michal Hocko
2012-01-17 23:55     ` KAMEZAWA Hiroyuki
2012-01-18 13:01       ` Michal Hocko
2012-01-19  2:18         ` KAMEZAWA Hiroyuki
2012-01-19 20:07         ` Ying Han
2012-01-20  0:48           ` KAMEZAWA Hiroyuki
2012-01-13  8:33 ` [RFC] [PATCH 2/7 v2] memcg: add memory barrier for checking account move KAMEZAWA Hiroyuki
2012-01-17 15:26   ` Michal Hocko
2012-01-18  0:06     ` KAMEZAWA Hiroyuki
2012-01-18 12:37       ` Michal Hocko
2012-01-19  2:17         ` KAMEZAWA Hiroyuki
2012-01-19  9:28           ` Michal Hocko
2012-01-19 23:57             ` KAMEZAWA Hiroyuki
2012-01-20 18:08           ` Ying Han
2012-01-23  9:04             ` Michal Hocko
2012-01-24  3:21               ` KAMEZAWA Hiroyuki
2012-01-24  8:49                 ` Michal Hocko
2012-01-24 19:04               ` Ying Han
2012-01-25 11:07                 ` Michal Hocko
2012-01-13  8:40 ` [RFC] [PATCH 3/7 v2] memcg: remove PCG_MOVE_LOCK flag from pc->flags KAMEZAWA Hiroyuki
2012-01-16 12:55   ` Kirill A. Shutemov
2012-01-17  0:22     ` KAMEZAWA Hiroyuki
2012-01-17 16:46   ` Michal Hocko
2012-01-18  0:12     ` KAMEZAWA Hiroyuki
2012-01-18 10:47       ` Michal Hocko
2012-01-18 23:53         ` KAMEZAWA Hiroyuki
2012-01-23 22:05           ` Ying Han
2012-01-24  4:59             ` KAMEZAWA Hiroyuki
2012-01-24  8:43             ` Michal Hocko
2012-01-25 23:07               ` Ying Han
2012-01-26  9:16                 ` Michal Hocko
2012-01-23 22:02   ` Ying Han
2012-01-24  4:47     ` KAMEZAWA Hiroyuki
2012-01-25 22:48       ` Ying Han
2012-01-13  8:41 ` [RFC] [PATCH 4/7 v2] memcg: new scheme to update per-memcg page stat accounting KAMEZAWA Hiroyuki
2012-01-18 16:45   ` Michal Hocko
2012-01-18 23:58     ` KAMEZAWA Hiroyuki
2012-01-26 19:01   ` Ying Han
2012-01-13  8:42 ` [RFC] [PATCH 5/7 v2] memcg: remove PCG_FILE_MAPPED KAMEZAWA Hiroyuki
2012-01-19 14:07   ` Michal Hocko
2012-01-26 19:10     ` Ying Han
2012-01-13  8:43 ` [RFC] [PATCH 6/7 v2] memcg: remove PCG_CACHE KAMEZAWA Hiroyuki
2012-01-13  8:45 ` [RFC] [PATCH 7/7 v2] memcg: make mem_cgroup_begin_update_stat to use global pcpu KAMEZAWA Hiroyuki
2012-01-19 14:47   ` Michal Hocko
2012-01-20  2:19     ` KAMEZAWA Hiroyuki
2012-01-20  8:38       ` Michal Hocko
2012-01-20  8:40   ` Greg Thelen
2012-01-24  3:18     ` KAMEZAWA Hiroyuki

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=20120113173001.ee5260ca.kamezawa.hiroyu@jp.fujitsu.com \
    --to=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=bsingharora@gmail.com \
    --cc=cgroups@vger.kernel.org \
    --cc=hannes@cmpxchg.org \
    --cc=hugh.dickins@tiscali.co.uk \
    --cc=linux-mm@kvack.org \
    --cc=mhocko@suse.cz \
    --cc=yinghan@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