From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
To: Tao Ma <tm@tao.ma>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
David Rientjes <rientjes@google.com>,
Minchan Kim <minchan.kim@gmail.com>, Mel Gorman <mel@csn.ul.ie>,
Johannes Weiner <jweiner@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH] mm: do not drain pagevecs for mlock
Date: Fri, 30 Dec 2011 04:31:23 -0500 [thread overview]
Message-ID: <CAHGf_=pODc6fLGJAEZWzQtUd6fj6v=fV9n6UTwysqRR1SwY++A@mail.gmail.com> (raw)
In-Reply-To: <4EFD7AE3.8020403@tao.ma>
2011/12/30 Tao Ma <tm@tao.ma>:
> On 12/30/2011 04:11 PM, KOSAKI Motohiro wrote:
>> 2011/12/30 Tao Ma <tm@tao.ma>:
>>> In our test of mlock, we have found some severe performance regression
>>> in it. Some more investigations show that mlocked is blocked heavily
>>> by lur_add_drain_all which calls schedule_on_each_cpu and flush the work
>>> queue which is very slower if we have several cpus.
>>>
>>> So we have tried 2 ways to solve it:
>>> 1. Add a per cpu counter for all the pagevecs so that we don't schedule
>>> and flush the lru_drain work if the cpu doesn't have any pagevecs(I
>>> have finished the codes already).
>>> 2. Remove the lru_add_drain_all.
>>>
>>> The first one has some problems since in our product system, all the cpus
>>> are busy, so I guess there is very little chance for a cpu to have 0 pagevecs
>>> except that you run several consecutive mlocks.
>>>
>>> From the commit log which added this function(8891d6da), it seems that we
>>> don't have to call it. So the 2nd one seems to be both easy and workable and
>>> comes this patch.
>>
>> Could you please show us your system environment and benchmark programs?
>> Usually lru_drain_** is very fast than mlock() body because it makes
>> plenty memset(page).
> The system environment is: 16 core Xeon E5620. 24G memory.
>
> I have attached the program. It is very simple and just uses mlock/munlock.
Because your test program is too artificial. 20sec/100000times =
200usec. And your
program repeat mlock and munlock the exact same address. so, yes, if
lru_add_drain_all() is removed, it become near no-op. but it's
worthless comparision.
none of any practical program does such strange mlock usage.
But, 200usec is much than I measured before. I'll dig it a bit more.
next prev parent reply other threads:[~2011-12-30 9:31 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-12-30 6:36 [PATCH] mm: do not drain pagevecs for mlock Tao Ma
2011-12-30 8:11 ` KOSAKI Motohiro
2011-12-30 8:48 ` Tao Ma
2011-12-30 9:31 ` KOSAKI Motohiro [this message]
2011-12-30 9:45 ` Tao Ma
2011-12-30 10:07 ` KOSAKI Motohiro
2012-01-01 7:30 ` [PATCH 1/2] mm,mlock: drain pagevecs asynchronously kosaki.motohiro
2012-01-04 1:17 ` Minchan Kim
2012-01-04 2:38 ` KOSAKI Motohiro
2012-01-10 8:53 ` Tao Ma
2012-01-04 2:56 ` Hugh Dickins
2012-01-04 22:05 ` Andrew Morton
2012-01-04 23:33 ` KOSAKI Motohiro
2012-01-05 0:19 ` Hugh Dickins
2012-01-01 7:30 ` [PATCH 2/2] sysvshm: SHM_LOCK use lru_add_drain_all_async() kosaki.motohiro
2012-01-04 1:51 ` Hugh Dickins
2012-01-04 2:19 ` KOSAKI Motohiro
2012-01-04 5:17 ` Hugh Dickins
2012-01-04 8:34 ` KOSAKI Motohiro
2012-01-06 6:13 ` [PATCH] mm: do not drain pagevecs for mlock Tao Ma
2012-01-06 6:18 ` KOSAKI Motohiro
2012-01-06 6:30 ` Tao Ma
2012-01-06 6:33 ` KOSAKI Motohiro
2012-01-06 6:46 ` Tao Ma
2012-01-09 23:58 ` KOSAKI Motohiro
2012-01-10 2:08 ` Tao Ma
2012-01-09 7:25 ` Tao Ma
2011-12-30 10:14 ` KOSAKI Motohiro
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='CAHGf_=pODc6fLGJAEZWzQtUd6fj6v=fV9n6UTwysqRR1SwY++A@mail.gmail.com' \
--to=kosaki.motohiro@jp.fujitsu.com \
--cc=akpm@linux-foundation.org \
--cc=jweiner@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=minchan.kim@gmail.com \
--cc=rientjes@google.com \
--cc=tm@tao.ma \
/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).