From: Andi Kleen <ak@suse.de>
To: Mel Gorman <mel@skynet.ie>
Cc: akpm@linux-foundation.org, Lee.Schermerhorn@hp.com,
clameter@sgi.com, kamezawa.hiroyu@jp.fujitsu.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Apply memory policies to top two highest zones when highest zone is ZONE_MOVABLE
Date: Sat, 4 Aug 2007 10:51:13 +0200 [thread overview]
Message-ID: <200708041051.14324.ak@suse.de> (raw)
In-Reply-To: <20070804002354.GA2841@skynet.ie>
> It only affects hot paths in the NUMA case so non-NUMA users will not care.
For x86-64 most distribution kernels are NUMA these days.
> For NUMA users, I have posted patches that eliminate multiple zonelists
> altogether which will reduce cache footprint (something like 7K per node on
> x86_64)
How do you get to 7k? We got worst case 3 zones node (normally less);
that's three pointers per GFP level.
> and make things like MPOL_BIND behave in a consistent manner. That
> would cost on CPU but save on cache which would (hopefully) result in a net
> gain in most cases.
That might be a good tradeoff, but without seeing the patch
the 7k number sounds very dubious.
> I would like to go with this patch for now just for policies but for
> 2.6.23, we could leave it as "policies only apply to ZONE_MOVABLE when it
> is used" if you really insisted on it. It's less than ideal though for
> sure.
Or disable ZONE_MOVABLE. It seems to be clearly not well thought
out well yet. Perhaps make it dependent on !CONFIG_NUMA.
-Andi
WARNING: multiple messages have this Message-ID (diff)
From: Andi Kleen <ak@suse.de>
To: Mel Gorman <mel@skynet.ie>
Cc: akpm@linux-foundation.org, Lee.Schermerhorn@hp.com,
clameter@sgi.com, kamezawa.hiroyu@jp.fujitsu.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Apply memory policies to top two highest zones when highest zone is ZONE_MOVABLE
Date: Sat, 4 Aug 2007 10:51:13 +0200 [thread overview]
Message-ID: <200708041051.14324.ak@suse.de> (raw)
In-Reply-To: <20070804002354.GA2841@skynet.ie>
> It only affects hot paths in the NUMA case so non-NUMA users will not care.
For x86-64 most distribution kernels are NUMA these days.
> For NUMA users, I have posted patches that eliminate multiple zonelists
> altogether which will reduce cache footprint (something like 7K per node on
> x86_64)
How do you get to 7k? We got worst case 3 zones node (normally less);
that's three pointers per GFP level.
> and make things like MPOL_BIND behave in a consistent manner. That
> would cost on CPU but save on cache which would (hopefully) result in a net
> gain in most cases.
That might be a good tradeoff, but without seeing the patch
the 7k number sounds very dubious.
> I would like to go with this patch for now just for policies but for
> 2.6.23, we could leave it as "policies only apply to ZONE_MOVABLE when it
> is used" if you really insisted on it. It's less than ideal though for
> sure.
Or disable ZONE_MOVABLE. It seems to be clearly not well thought
out well yet. Perhaps make it dependent on !CONFIG_NUMA.
-Andi
--
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:[~2007-08-04 8:53 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-08-02 17:21 [PATCH] Apply memory policies to top two highest zones when highest zone is ZONE_MOVABLE Mel Gorman
2007-08-02 17:21 ` Mel Gorman
2007-08-02 19:41 ` Lee Schermerhorn
2007-08-02 19:41 ` Lee Schermerhorn
2007-08-02 20:45 ` Christoph Lameter
2007-08-02 20:45 ` Christoph Lameter
2007-08-06 19:44 ` Andrew Morton
2007-08-06 19:44 ` Andrew Morton
2007-08-06 20:13 ` Christoph Lameter
2007-08-06 20:13 ` Christoph Lameter
2007-08-06 21:56 ` Paul Jackson
2007-08-06 21:56 ` Paul Jackson
2007-08-03 22:02 ` Andi Kleen
2007-08-03 22:02 ` Andi Kleen
2007-08-04 0:23 ` Mel Gorman
2007-08-04 0:23 ` Mel Gorman
2007-08-04 8:51 ` Andi Kleen [this message]
2007-08-04 8:51 ` Andi Kleen
2007-08-04 16:39 ` Mel Gorman
2007-08-04 16:39 ` Mel Gorman
2007-08-06 19:15 ` Andrew Morton
2007-08-06 19:15 ` Andrew Morton
2007-08-06 19:18 ` Christoph Lameter
2007-08-06 19:18 ` Christoph Lameter
2007-08-06 20:31 ` Andi Kleen
2007-08-06 20:31 ` Andi Kleen
2007-08-06 21:55 ` Mel Gorman
2007-08-06 21:55 ` Mel Gorman
2007-08-07 5:12 ` Andrew Morton
2007-08-07 5:12 ` Andrew Morton
2007-08-07 16:55 ` Mel Gorman
2007-08-07 16:55 ` Mel Gorman
2007-08-07 18:14 ` Andrew Morton
2007-08-07 18:14 ` Andrew Morton
2007-08-07 20:37 ` Christoph Lameter
2007-08-07 20:37 ` Christoph Lameter
2007-08-08 16:49 ` Mel Gorman
2007-08-08 16:49 ` Mel Gorman
2007-08-08 17:03 ` Christoph Lameter
2007-08-08 17:03 ` Christoph Lameter
2007-08-06 21:48 ` Mel Gorman
2007-08-06 21:48 ` Mel Gorman
2007-08-06 22:31 ` Christoph Lameter
2007-08-06 22:31 ` Christoph Lameter
2007-08-06 22:57 ` Mel Gorman
2007-08-06 22:57 ` Mel Gorman
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=200708041051.14324.ak@suse.de \
--to=ak@suse.de \
--cc=Lee.Schermerhorn@hp.com \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mel@skynet.ie \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.