From: Johannes Weiner <hannes@cmpxchg.org>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
Minchan Kim <minchan.kim@gmail.com>,
Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Andrew Morton <akpm@linux-foundation.org>,
"Rafael J. Wysocki" <rjw@sisk.pl>, Rik van Riel <riel@redhat.com>,
Balbir Singh <balbir@in.ibm.com>
Subject: Re: [PATCH] vmscan: memcg needs may_swap (Re: [patch] vmscan: rename sc.may_swap to may_unmap)
Date: Wed, 1 Apr 2009 11:49:55 +0200 [thread overview]
Message-ID: <20090401094955.GA1656@cmpxchg.org> (raw)
In-Reply-To: <20090401180445.80b11d90.kamezawa.hiroyu@jp.fujitsu.com>
On Wed, Apr 01, 2009 at 06:04:45PM +0900, KAMEZAWA Hiroyuki wrote:
> On Wed, 1 Apr 2009 06:09:51 +0200
> Johannes Weiner <hannes@cmpxchg.org> wrote:
>
> > On Tue, Mar 31, 2009 at 10:48:32AM +0900, KOSAKI Motohiro wrote:
> > > > > Sorry for too late response.
> > > > > I don't know memcg well.
> > > > >
> > > > > The memcg managed to use may_swap well with global page reclaim until now.
> > > > > I think that was because may_swap can represent both meaning.
> > > > > Do we need each variables really ?
> > > > >
> > > > > How about using union variable ?
> > > >
> > > > or Just removing one of them ?
> > >
> > > I hope all may_unmap user convert to using may_swap.
> > > may_swap is more efficient and cleaner meaning.
> >
> > How about making may_swap mean the following:
> >
> > @@ -642,6 +639,8 @@ static unsigned long shrink_page_list(st
> > * Try to allocate it some swap space here.
> > */
> > if (PageAnon(page) && !PageSwapCache(page)) {
> > + if (!sc->map_swap)
> > + goto keep_locked;
> > if (!(sc->gfp_mask & __GFP_IO))
> > goto keep_locked;
> > if (!add_to_swap(page))
> >
> > try_to_free_pages() always sets it.
> >
> What is the advantage than _not_ scanning ANON LRU at all ?
I thought we could collect anon pages that don't need swap io.
> > try_to_free_mem_cgroup_pages() sets it depending on whether it really
> > wants swapping, and only swapping, right? But the above would still
> > reclaim already swapped anon pages and I don't know the memory
> > controller.
> >
> memory cgroup has 2 calls to this shrink_zone.
> 1. memory usage hits the limit.
> 2. mem+swap usage hits the limit.
>
> At "2", swap-out doesn't decrease the usage of mem+swap, then set may_swap=0.
> So, we want to kick out only file caches.
> But, we can reclaim file cache and "unmap file cache and reclaim it!" is
> necessary even if may_swap=0.
Yes.
> Then, scanning only FILE LRU makes sense at may_swap=0 *if* memcg is
> the only user of may_swap=0.
>
> Let's see others.
>
> - __zone_reclaim sets may_unmap to be 0 when they don't want swap-out.
> .....can be replaced with may_swap.
>
> - shrink_all_memory sets may_swap to be 0. Is this called by hibernation ?
> If you don't want to unmap file caches while hibernation, adding may_unmap
> as *new* paramter makes sense, I think.
Yep, that was my idea too. At least for now and then reevaluate
whether it shouldn't just reclaim in lru order without this flag...
> The change you proposed is for dropping unused SwapCache pages. Right ?
> But this will be dropped by kswapd if necessary.
>
> As far as memcg concerns, scanning ANON LRU even when may_swap=0 is just
> a waste of cpu time.
Okay.
next prev parent reply other threads:[~2009-04-01 9:51 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-02-06 3:11 [PATCH 0/3] [PATCH 0/3] swsusp: shrink file cache first Johannes Weiner
2009-02-06 3:11 ` [PATCH 1/3] swsusp: clean up shrink_all_zones() Johannes Weiner
2009-02-06 3:20 ` KOSAKI Motohiro
2009-02-06 3:11 ` [PATCH 2/3] swsusp: dont fiddle with swappiness Johannes Weiner
2009-02-06 3:21 ` KOSAKI Motohiro
2009-02-06 3:11 ` [PATCH 3/3][RFC] swsusp: shrink file cache first Johannes Weiner
2009-02-06 3:39 ` KOSAKI Motohiro
2009-02-06 4:49 ` Johannes Weiner
2009-02-06 5:59 ` KOSAKI Motohiro
2009-02-06 12:24 ` Johannes Weiner
2009-02-06 13:35 ` MinChan Kim
2009-02-06 17:15 ` MinChan Kim
2009-02-06 23:37 ` Johannes Weiner
2009-02-09 19:43 ` [patch] vmscan: rename sc.may_swap to may_unmap Johannes Weiner
2009-02-09 23:02 ` MinChan Kim
2009-02-10 10:00 ` KOSAKI Motohiro
2009-03-27 6:19 ` [PATCH] vmscan: memcg needs may_swap (Re: [patch] vmscan: rename sc.may_swap to may_unmap) Daisuke Nishimura
2009-03-27 6:30 ` KAMEZAWA Hiroyuki
2009-03-29 23:45 ` KOSAKI Motohiro
2009-03-31 0:18 ` Daisuke Nishimura
2009-03-31 1:26 ` Minchan Kim
2009-03-31 1:42 ` KAMEZAWA Hiroyuki
2009-03-31 1:48 ` KOSAKI Motohiro
2009-04-01 4:09 ` Johannes Weiner
2009-04-01 5:08 ` Daisuke Nishimura
2009-04-01 9:04 ` KAMEZAWA Hiroyuki
2009-04-01 9:11 ` KOSAKI Motohiro
2009-04-01 9:49 ` Johannes Weiner [this message]
2009-04-01 9:55 ` KOSAKI Motohiro
2009-04-01 16:03 ` Johannes Weiner
2009-03-31 1:52 ` Daisuke Nishimura
2009-02-06 21:00 ` [PATCH 3/3][RFC] swsusp: shrink file cache first Andrew Morton
2009-02-06 23:27 ` Johannes Weiner
2009-02-07 17:23 ` Rafael J. Wysocki
2009-02-08 20:56 ` Johannes Weiner
2009-02-07 4:41 ` Nigel Cunningham
2009-02-07 16:51 ` KOSAKI Motohiro
2009-02-07 21:20 ` Nigel Cunningham
2009-02-27 13:27 ` Pavel Machek
2009-03-01 10:37 ` KOSAKI Motohiro
2009-02-06 8:03 ` MinChan Kim
2009-02-06 10:06 ` MinChan Kim
2009-02-06 11:50 ` Johannes Weiner
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=20090401094955.GA1656@cmpxchg.org \
--to=hannes@cmpxchg.org \
--cc=akpm@linux-foundation.org \
--cc=balbir@in.ibm.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kosaki.motohiro@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 \
--cc=riel@redhat.com \
--cc=rjw@sisk.pl \
/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).