From: Andrew Morton <akpm@linux-foundation.org>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>,
Minchan Kim <minchan.kim@gmail.com>,
Greg Thelen <gthelen@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
containers@lists.osdl.org, Andrea Righi <arighi@develer.com>,
Balbir Singh <balbir@linux.vnet.ibm.com>
Subject: Re: [PATCH v2] memcg: reduce lock time at move charge (Was Re: [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup()
Date: Thu, 7 Oct 2010 16:14:54 -0700 [thread overview]
Message-ID: <20101007161454.84570cf9.akpm@linux-foundation.org> (raw)
In-Reply-To: <20101007170405.27ed964c.kamezawa.hiroyu@jp.fujitsu.com>
On Thu, 7 Oct 2010 17:04:05 +0900
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:
> Now, at task migration among cgroup, memory cgroup scans page table and moving
> account if flags are properly set.
>
> The core code, mem_cgroup_move_charge_pte_range() does
>
> pte_offset_map_lock();
> for all ptes in a page table:
> 1. look into page table, find_and_get a page
> 2. remove it from LRU.
> 3. move charge.
> 4. putback to LRU. put_page()
> pte_offset_map_unlock();
>
> for pte entries on a 3rd level? page table.
>
> This pte_offset_map_lock seems a bit long. This patch modifies a rountine as
>
> for 32 pages: pte_offset_map_lock()
> find_and_get a page
> record it
> pte_offset_map_unlock()
> for all recorded pages
> isolate it from LRU.
> move charge
> putback to LRU
> for all recorded pages
> put_page()
The patch makes the code larger, more complex and slower!
I do think we're owed a more complete description of its benefits than
"seems a bit long". Have problems been observed? Any measurements
taken?
--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2010-10-07 23:15 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-04 6:57 [PATCH 00/10] memcg: per cgroup dirty page accounting Greg Thelen
2010-10-04 6:57 ` [PATCH 01/10] memcg: add page_cgroup flags for dirty page tracking Greg Thelen
2010-10-05 6:20 ` KAMEZAWA Hiroyuki
2010-10-06 0:37 ` Daisuke Nishimura
2010-10-06 11:07 ` Balbir Singh
2010-10-04 6:57 ` [PATCH 02/10] memcg: document cgroup dirty memory interfaces Greg Thelen
2010-10-05 6:48 ` KAMEZAWA Hiroyuki
2010-10-06 0:49 ` Daisuke Nishimura
2010-10-06 11:12 ` Balbir Singh
2010-10-04 6:57 ` [PATCH 03/10] memcg: create extensible page stat update routines Greg Thelen
2010-10-04 13:48 ` Ciju Rajan K
2010-10-04 15:43 ` Greg Thelen
2010-10-04 17:35 ` Ciju Rajan K
2010-10-05 6:51 ` KAMEZAWA Hiroyuki
2010-10-05 7:10 ` Greg Thelen
2010-10-05 15:42 ` Minchan Kim
2010-10-05 19:59 ` Greg Thelen
2010-10-05 23:57 ` Minchan Kim
2010-10-06 0:48 ` Greg Thelen
2010-10-06 16:19 ` Balbir Singh
2010-10-04 6:57 ` [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup() Greg Thelen
2010-10-05 6:54 ` KAMEZAWA Hiroyuki
2010-10-05 7:18 ` Greg Thelen
2010-10-05 16:03 ` Minchan Kim
2010-10-05 23:26 ` Greg Thelen
2010-10-06 0:15 ` Minchan Kim
2010-10-07 0:35 ` KAMEZAWA Hiroyuki
2010-10-07 1:54 ` Daisuke Nishimura
2010-10-07 2:17 ` KAMEZAWA Hiroyuki
2010-10-07 6:21 ` [PATCH] memcg: reduce lock time at move charge (Was " KAMEZAWA Hiroyuki
2010-10-07 6:24 ` [PATCH] memcg: lock-free clear page writeback " KAMEZAWA Hiroyuki
2010-10-07 9:05 ` KAMEZAWA Hiroyuki
2010-10-07 23:35 ` Minchan Kim
2010-10-08 4:41 ` KAMEZAWA Hiroyuki
2010-10-07 7:28 ` [PATCH] memcg: reduce lock time at move charge " Daisuke Nishimura
2010-10-07 7:42 ` KAMEZAWA Hiroyuki
2010-10-07 8:04 ` [PATCH v2] " KAMEZAWA Hiroyuki
2010-10-07 23:14 ` Andrew Morton [this message]
2010-10-08 1:12 ` Daisuke Nishimura
2010-10-08 4:37 ` KAMEZAWA Hiroyuki
2010-10-08 4:55 ` Andrew Morton
2010-10-08 5:12 ` KAMEZAWA Hiroyuki
2010-10-08 10:41 ` KAMEZAWA Hiroyuki
2010-10-12 3:39 ` Balbir Singh
2010-10-12 3:42 ` KAMEZAWA Hiroyuki
2010-10-12 3:54 ` Balbir Singh
2010-10-12 3:56 ` Daisuke Nishimura
2010-10-12 5:01 ` KAMEZAWA Hiroyuki
2010-10-12 5:48 ` [PATCH v4] memcg: reduce lock time at move charge KAMEZAWA Hiroyuki
2010-10-12 6:23 ` Daisuke Nishimura
2010-10-12 5:39 ` [PATCH 04/10] memcg: disable local interrupts in lock_page_cgroup() Balbir Singh
2010-10-04 6:58 ` [PATCH 05/10] memcg: add dirty page accounting infrastructure Greg Thelen
2010-10-05 7:22 ` KAMEZAWA Hiroyuki
2010-10-05 7:35 ` Greg Thelen
2010-10-05 16:09 ` Minchan Kim
2010-10-05 20:06 ` Greg Thelen
2010-10-04 6:58 ` [PATCH 06/10] memcg: add kernel calls for memcg dirty page stats Greg Thelen
2010-10-05 6:55 ` KAMEZAWA Hiroyuki
2010-10-04 6:58 ` [PATCH 07/10] memcg: add dirty limits to mem_cgroup Greg Thelen
2010-10-05 7:07 ` KAMEZAWA Hiroyuki
2010-10-05 9:43 ` Andrea Righi
2010-10-05 19:00 ` Greg Thelen
2010-10-07 0:13 ` KAMEZAWA Hiroyuki
2010-10-07 0:27 ` Greg Thelen
2010-10-07 0:48 ` KAMEZAWA Hiroyuki
2010-10-12 0:24 ` Greg Thelen
2010-10-12 0:55 ` KAMEZAWA Hiroyuki
2010-10-12 7:32 ` Greg Thelen
2010-10-12 8:38 ` KAMEZAWA Hiroyuki
2010-10-04 6:58 ` [PATCH 08/10] memcg: add cgroupfs interface to memcg dirty limits Greg Thelen
2010-10-05 7:13 ` KAMEZAWA Hiroyuki
2010-10-05 7:33 ` Greg Thelen
2010-10-05 7:31 ` KAMEZAWA Hiroyuki
2010-10-05 9:18 ` Andrea Righi
2010-10-05 18:31 ` David Rientjes
2010-10-06 18:34 ` Greg Thelen
2010-10-06 20:54 ` Andrea Righi
2010-10-06 13:30 ` Balbir Singh
2010-10-06 13:32 ` Balbir Singh
2010-10-06 16:21 ` Greg Thelen
2010-10-06 16:24 ` Balbir Singh
2010-10-07 6:23 ` Ciju Rajan K
2010-10-07 17:46 ` Greg Thelen
2010-10-04 6:58 ` [PATCH 09/10] writeback: make determine_dirtyable_memory() static Greg Thelen
2010-10-05 7:15 ` KAMEZAWA Hiroyuki
2010-10-04 6:58 ` [PATCH 10/10] memcg: check memcg dirty limits in page writeback Greg Thelen
2010-10-05 7:29 ` KAMEZAWA Hiroyuki
2010-10-06 0:32 ` Minchan Kim
2010-10-05 4:20 ` [PATCH 00/10] memcg: per cgroup dirty page accounting Balbir Singh
2010-10-05 4:50 ` Balbir Singh
2010-10-05 5:50 ` Greg Thelen
2010-10-05 8:37 ` Ciju Rajan K
2010-10-05 22:15 ` Andrea Righi
2010-10-06 3:23 ` Balbir Singh
2010-10-18 5:56 ` KAMEZAWA Hiroyuki
2010-10-18 18:09 ` Greg Thelen
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=20101007161454.84570cf9.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=arighi@develer.com \
--cc=balbir@linux.vnet.ibm.com \
--cc=containers@lists.osdl.org \
--cc=gthelen@google.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan.kim@gmail.com \
--cc=nishimura@mxp.nes.nec.co.jp \
/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;
as well as URLs for NNTP newsgroup(s).