From: Mel Gorman <mel@csn.ul.ie>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>,
Christoph Lameter <cl@linux-foundation.org>,
Adam Litke <agl@us.ibm.com>, Avi Kivity <avi@redhat.com>,
David Rientjes <rientjes@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH 03/12] Export unusable free space index via /proc/pagetypeinfo
Date: Tue, 16 Feb 2010 08:50:59 +0000 [thread overview]
Message-ID: <20100216085058.GD26086@csn.ul.ie> (raw)
In-Reply-To: <20100216173832.730F.A69D9226@jp.fujitsu.com>
On Tue, Feb 16, 2010 at 05:41:39PM +0900, KOSAKI Motohiro wrote:
> > On Tue, Feb 16, 2010 at 04:03:29PM +0900, KOSAKI Motohiro wrote:
> > > > Unusuable free space index is a measure of external fragmentation that
> > > > takes the allocation size into account. For the most part, the huge page
> > > > size will be the size of interest but not necessarily so it is exported
> > > > on a per-order and per-zone basis via /proc/pagetypeinfo.
> > >
> > > Hmmm..
> > > /proc/pagetype have a machine unfriendly format. perhaps, some user have own ugly
> > > /proc/pagetype parser. It have a little risk to break userland ABI.
> > >
> >
> > It's very low risk. I doubt there are machine parsers of
> > /proc/pagetypeinfo because there are very few machine-orientated actions
> > that can be taken based on the information. It's more informational for
> > a user if they were investigating fragmentation problems.
> >
> > > I have dumb question. Why can't we use another file?
> >
> > I could. What do you suggest?
>
> I agree it's low risk. but personally I hope fragmentation ABI keep very stable because
> I expect some person makes userland compaction daemon. (read fragmentation index
> from /proc and write /proc/compact_memory if necessary).
> then, if possible, I hope fragmentation info have individual /proc file.
>
I'd be somewhat surprised if there was an active userland compaction daemon
because I'd expect them to be depending on direct compaction. Userspace
compaction is more likely to be an all-or-nothing affair and confined to
NUMA nodes if they are being used as containers. If a compaction daemon was
to exist, I'd have expected it to be in-kernel because the triggers from
userspace are so coarse.
Still, I can break out the indices into separate files to cover all the
bases.
--
Mel Gorman
Part-time Phd Student Linux Technology Center
University of Limerick IBM Dublin Software Lab
WARNING: multiple messages have this Message-ID (diff)
From: Mel Gorman <mel@csn.ul.ie>
To: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>,
Christoph Lameter <cl@linux-foundation.org>,
Adam Litke <agl@us.ibm.com>, Avi Kivity <avi@redhat.com>,
David Rientjes <rientjes@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH 03/12] Export unusable free space index via /proc/pagetypeinfo
Date: Tue, 16 Feb 2010 08:50:59 +0000 [thread overview]
Message-ID: <20100216085058.GD26086@csn.ul.ie> (raw)
In-Reply-To: <20100216173832.730F.A69D9226@jp.fujitsu.com>
On Tue, Feb 16, 2010 at 05:41:39PM +0900, KOSAKI Motohiro wrote:
> > On Tue, Feb 16, 2010 at 04:03:29PM +0900, KOSAKI Motohiro wrote:
> > > > Unusuable free space index is a measure of external fragmentation that
> > > > takes the allocation size into account. For the most part, the huge page
> > > > size will be the size of interest but not necessarily so it is exported
> > > > on a per-order and per-zone basis via /proc/pagetypeinfo.
> > >
> > > Hmmm..
> > > /proc/pagetype have a machine unfriendly format. perhaps, some user have own ugly
> > > /proc/pagetype parser. It have a little risk to break userland ABI.
> > >
> >
> > It's very low risk. I doubt there are machine parsers of
> > /proc/pagetypeinfo because there are very few machine-orientated actions
> > that can be taken based on the information. It's more informational for
> > a user if they were investigating fragmentation problems.
> >
> > > I have dumb question. Why can't we use another file?
> >
> > I could. What do you suggest?
>
> I agree it's low risk. but personally I hope fragmentation ABI keep very stable because
> I expect some person makes userland compaction daemon. (read fragmentation index
> from /proc and write /proc/compact_memory if necessary).
> then, if possible, I hope fragmentation info have individual /proc file.
>
I'd be somewhat surprised if there was an active userland compaction daemon
because I'd expect them to be depending on direct compaction. Userspace
compaction is more likely to be an all-or-nothing affair and confined to
NUMA nodes if they are being used as containers. If a compaction daemon was
to exist, I'd have expected it to be in-kernel because the triggers from
userspace are so coarse.
Still, I can break out the indices into separate files to cover all the
bases.
--
Mel Gorman
Part-time Phd Student Linux Technology Center
University of Limerick IBM Dublin Software Lab
--
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:[~2010-02-16 8:51 UTC|newest]
Thread overview: 99+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-12 12:00 [PATCH 0/12] Memory Compaction v2r12 Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-12 12:00 ` [PATCH 01/12] mm: Document /proc/pagetypeinfo Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-12 15:54 ` Christoph Lameter
2010-02-12 15:54 ` Christoph Lameter
2010-02-16 7:05 ` KOSAKI Motohiro
2010-02-16 7:05 ` KOSAKI Motohiro
2010-02-12 12:00 ` [PATCH 02/12] Allow CONFIG_MIGRATION to be set without CONFIG_NUMA or memory hot-remove Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-16 17:43 ` Rik van Riel
2010-02-16 17:43 ` Rik van Riel
2010-02-12 12:00 ` [PATCH 03/12] Export unusable free space index via /proc/pagetypeinfo Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-16 7:03 ` KOSAKI Motohiro
2010-02-16 7:03 ` KOSAKI Motohiro
2010-02-16 8:36 ` Mel Gorman
2010-02-16 8:36 ` Mel Gorman
2010-02-16 8:41 ` KOSAKI Motohiro
2010-02-16 8:41 ` KOSAKI Motohiro
2010-02-16 8:50 ` Mel Gorman [this message]
2010-02-16 8:50 ` Mel Gorman
2010-02-16 18:28 ` Rik van Riel
2010-02-16 18:28 ` Rik van Riel
2010-02-18 15:23 ` Minchan Kim
2010-02-18 15:23 ` Minchan Kim
2010-02-18 15:32 ` Mel Gorman
2010-02-18 15:32 ` Mel Gorman
2010-02-12 12:00 ` [PATCH 04/12] Export fragmentation " Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-16 7:59 ` KOSAKI Motohiro
2010-02-16 7:59 ` KOSAKI Motohiro
2010-02-16 8:41 ` Mel Gorman
2010-02-16 8:41 ` Mel Gorman
2010-02-16 8:49 ` KOSAKI Motohiro
2010-02-16 8:49 ` KOSAKI Motohiro
2010-02-17 1:44 ` Rik van Riel
2010-02-17 1:44 ` Rik van Riel
2010-02-18 15:37 ` Minchan Kim
2010-02-18 15:37 ` Minchan Kim
2010-02-12 12:00 ` [PATCH 05/12] Memory compaction core Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-16 8:31 ` KOSAKI Motohiro
2010-02-16 8:31 ` KOSAKI Motohiro
2010-02-16 8:48 ` Mel Gorman
2010-02-16 8:48 ` Mel Gorman
2010-02-16 14:55 ` Christoph Lameter
2010-02-16 14:55 ` Christoph Lameter
2010-02-16 14:59 ` Mel Gorman
2010-02-16 14:59 ` Mel Gorman
2010-02-18 19:37 ` Christoph Lameter
2010-02-18 19:37 ` Christoph Lameter
2010-02-18 21:35 ` Mel Gorman
2010-02-18 21:35 ` Mel Gorman
2010-02-19 0:04 ` KAMEZAWA Hiroyuki
2010-02-19 0:04 ` KAMEZAWA Hiroyuki
2010-02-17 13:29 ` Mel Gorman
2010-02-17 13:29 ` Mel Gorman
2010-02-17 15:45 ` Rik van Riel
2010-02-17 15:45 ` Rik van Riel
2010-02-18 16:58 ` Minchan Kim
2010-02-18 16:58 ` Minchan Kim
2010-02-18 17:34 ` Mel Gorman
2010-02-18 17:34 ` Mel Gorman
2010-02-19 1:21 ` Minchan Kim
2010-02-19 1:21 ` Minchan Kim
2010-02-19 14:33 ` Mel Gorman
2010-02-19 14:33 ` Mel Gorman
2010-02-12 12:00 ` [PATCH 06/12] Add /proc trigger for memory compaction Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-12 18:34 ` Valdis.Kletnieks
2010-02-12 18:38 ` Mel Gorman
2010-02-12 18:38 ` Mel Gorman
2010-02-17 16:30 ` Rik van Riel
2010-02-17 16:30 ` Rik van Riel
2010-02-18 19:51 ` Christoph Lameter
2010-02-18 19:51 ` Christoph Lameter
2010-02-19 1:56 ` Minchan Kim
2010-02-19 1:56 ` Minchan Kim
2010-02-12 12:00 ` [PATCH 07/12] Add /sys trigger for per-node " Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-17 16:30 ` Rik van Riel
2010-02-17 16:30 ` Rik van Riel
2010-02-12 12:00 ` [PATCH 08/12] Direct compact when a high-order allocation fails Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-18 3:57 ` Rik van Riel
2010-02-18 3:57 ` Rik van Riel
2010-02-12 12:00 ` [PATCH 09/12] Do not compact within a preferred zone after a compaction failure Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-18 4:09 ` Rik van Riel
2010-02-18 4:09 ` Rik van Riel
2010-02-12 12:00 ` [PATCH 10/12] mm: Check for an empty VMA list in rmap_walk_anon Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-17 18:22 ` Mel Gorman
2010-02-17 18:22 ` Mel Gorman
2010-02-12 12:00 ` [PATCH 11/12] mm: Take the RCU read lock " Mel Gorman
2010-02-12 12:00 ` Mel Gorman
2010-02-12 12:00 ` [PATCH 12/12] mm: Check the anon_vma is still valid in rmap_walk_anon() Mel Gorman
2010-02-12 12:00 ` 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=20100216085058.GD26086@csn.ul.ie \
--to=mel@csn.ul.ie \
--cc=aarcange@redhat.com \
--cc=agl@us.ibm.com \
--cc=avi@redhat.com \
--cc=cl@linux-foundation.org \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=rientjes@google.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 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.