From: Andrea Arcangeli <aarcange@redhat.com>
To: Mel Gorman <mel@csn.ul.ie>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
Andrew Morton <akpm@linux-foundation.org>,
Rik van Riel <riel@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 3/3] mm,vmscan: Reclaim order-0 and compact instead of lumpy reclaim when under light pressure
Date: Mon, 15 Nov 2010 16:23:10 +0100 [thread overview]
Message-ID: <20101115152310.GG6809@random.random> (raw)
In-Reply-To: <20101115092256.GE27362@csn.ul.ie>
On Mon, Nov 15, 2010 at 09:22:56AM +0000, Mel Gorman wrote:
> GFP_LUMPY is something else and is only partially related. Transparent Huge
> Pages (THP) does not want to hit lumpy reclaim no matter what the circumstances
> are - It is always better for THP to not use lumpy reclaim. It's debatable
Agreed.
> whether it should even reclaim order-0 pages for compaction so even with
> this series, THP might still introduce GFP_LUMPY.
reclaim of some order 0 page shouldn't do any significant harm as long
as the young bits are not ignored and it's just going "normal" and not
aggressive like lumpy.
Also we it's ok to do some reclaim as we can free some slab that can't
be compacted in case there's excessive amount of slab caches to be
shrunk to have a chance to convert unmovable pageblocks to movable
ones. And we need at least 2M fully available as migration
destination (but I guess that is always available :).
In general interleaving compaction with regular-reclaim (no lumpy)
before failing allocation sounds ok to me.
I guess these days compaction would tend to succeed before lumpy ever
gets invoked so the trouble with lumpy would then only trigger when
compaction starts failing and we enter reclaim to create more movable
pageblocks, but I don't want to risk bad behavior when the amount of
anoymous memory goes very high and not all anonymous memory can be
backed fully by hugepages.
next prev parent reply other threads:[~2010-11-15 15:23 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-11 19:07 [RFC PATCH 0/3] Use compaction to reduce a dependency on lumpy reclaim Mel Gorman
2010-11-11 19:07 ` [PATCH 1/3] mm,vmscan: Convert lumpy_mode into a bitmask Mel Gorman
2010-11-14 5:40 ` KOSAKI Motohiro
2010-11-15 9:16 ` Mel Gorman
2010-11-11 19:07 ` [PATCH 2/3] mm,compaction: Add COMPACTION_BUILD Mel Gorman
2010-11-14 5:45 ` KOSAKI Motohiro
2010-11-15 9:26 ` Mel Gorman
2010-11-11 19:07 ` [PATCH 3/3] mm,vmscan: Reclaim order-0 and compact instead of lumpy reclaim when under light pressure Mel Gorman
2010-11-12 9:37 ` Mel Gorman
2010-11-14 5:43 ` KOSAKI Motohiro
2010-11-15 9:17 ` Mel Gorman
2010-11-14 6:02 ` KOSAKI Motohiro
2010-11-15 9:22 ` Mel Gorman
2010-11-15 15:23 ` Andrea Arcangeli [this message]
2010-11-14 5:59 ` KOSAKI Motohiro
2010-11-15 9:25 ` Mel Gorman
2010-11-14 5:31 ` [RFC PATCH 0/3] Use compaction to reduce a dependency on lumpy reclaim 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=20101115152310.GG6809@random.random \
--to=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@csn.ul.ie \
--cc=riel@redhat.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;
as well as URLs for NNTP newsgroup(s).