From: Andrew Morton <akpm@linux-foundation.org>
To: Christoph Lameter <clameter@sgi.com>
Cc: linux-mm@kvack.org, Mel Gorman <mel@skynet.ie>,
andi@firstfloor.org, Nick Piggin <npiggin@suse.de>,
Rik van Riel <riel@redhat.com>,
Pekka Enberg <penberg@cs.helsinki.fi>
Subject: Re: [patch 05/18] SLUB: Slab defrag core
Date: Tue, 8 Apr 2008 14:11:35 -0700 [thread overview]
Message-ID: <20080408141135.de5a6350.akpm@linux-foundation.org> (raw)
In-Reply-To: <Pine.LNX.4.64.0804081401350.31230@schroedinger.engr.sgi.com>
On Tue, 8 Apr 2008 14:02:46 -0700 (PDT) Christoph Lameter <clameter@sgi.com> wrote:
> On Mon, 7 Apr 2008, Andrew Morton wrote:
>
> > > Fragmentation is skipped if it was less than a tenth of a second since we
> > > last checked a slab cache. An unsuccessful defrag attempt pauses attempts
> > > for at least one second.
> >
> > Can we not do this? It's a really nasty hack. Wall time has almost no
> > correlation with reclaim and allocation activity.
> >
> > If we really cannot think of anything smarter than just throttling then the
> > decision regarding when to throttle and for how long should at least be
> > driven by something which is vaguely correlated with the present/recent
> > allocation/reclaim activity.
>
> The reclaim interval increases to 1 second if slab reclaim was not
> succcessful.
I know that. It's still an arbitrary-to-the-point-of-uselessness hack.
Reclaim is clocked by scanning rates, allocation rates and disk speed. Not
wall time.
--
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:[~2008-04-08 21:11 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20080404230158.365359425@sgi.com>
[not found] ` <20080404230229.401345769@sgi.com>
2008-04-08 6:13 ` [patch 16/18] FS: Socket inode defragmentation Andrew Morton
[not found] ` <20080404230225.862960359@sgi.com>
[not found] ` <20080407231052.eb37a8fd.akpm@linux-foundation.org>
2008-04-08 18:55 ` [patch 01/18] SLUB: Add defrag_ratio field and sysfs support Pekka J Enberg
[not found] ` <20080404230226.340749825@sgi.com>
[not found] ` <20080407231059.e8c173fa.akpm@linux-foundation.org>
2008-04-08 19:07 ` [patch 03/18] SLUB: Add get() and kick() methods Pekka J Enberg
[not found] ` <20080404230226.577197795@sgi.com>
[not found] ` <20080407231113.855e2ba3.akpm@linux-foundation.org>
2008-04-08 19:18 ` [patch 04/18] SLUB: Sort slab cache list and establish maximum objects for defrag slabs Pekka Enberg
2008-04-08 21:01 ` Christoph Lameter
2008-04-08 21:07 ` Andi Kleen
[not found] ` <84144f020804072317g5b2b9f42yb300cad9a4258a15@mail.gmail.com>
[not found] ` <20080407233001.3e1e5147.akpm@linux-foundation.org>
2008-04-10 16:17 ` Pekka Enberg
[not found] ` <20080404230226.847485429@sgi.com>
[not found] ` <20080407231129.3c044ba1.akpm@linux-foundation.org>
2008-04-08 21:02 ` [patch 05/18] SLUB: Slab defrag core Christoph Lameter
2008-04-08 21:11 ` Andrew Morton [this message]
2008-04-08 21:17 ` Christoph Lameter
2008-04-08 21:25 ` Andrew Morton
2008-04-08 21:47 ` Christoph Lameter
2008-04-10 18:28 ` Christoph Lameter
2008-04-10 19:00 ` Andrew Morton
2008-04-10 20:33 ` Christoph Lameter
2008-04-10 20:49 ` Pekka Enberg
[not found] ` <20080404230227.768964864@sgi.com>
[not found] ` <20080407231137.6e3a38cd.akpm@linux-foundation.org>
2008-04-08 21:05 ` [patch 09/18] SLUB: Trigger defragmentation from memory reclaim Christoph Lameter
[not found] ` <20080404230229.678047976@sgi.com>
[not found] ` <20080407231402.63284bb5.akpm@linux-foundation.org>
2008-04-08 21:09 ` [patch 17/18] dentries: Add constructor Christoph Lameter
[not found] ` <20080404230229.922470579@sgi.com>
[not found] ` <20080407231434.88352977.akpm@linux-foundation.org>
2008-04-08 21:14 ` [patch 18/18] dentries: dentry defragmentation Christoph Lameter
2008-04-08 21:22 ` Andrew Morton
2008-04-08 21:41 ` Christoph Lameter
[not found] ` <20080404230229.169327879@sgi.com>
[not found] ` <20080407231346.8a17d27d.akpm@linux-foundation.org>
2008-04-13 13:39 ` RIP __kmem_cache_shrink (was Re: [patch 15/18] FS: Proc filesystem support for slab defrag) Alexey Dobriyan
2008-04-14 19:41 ` Christoph Lameter
2008-04-14 20:12 ` Alexey Dobriyan
2008-04-14 20:36 ` Pekka Enberg
[not found] ` <20080404230228.523868817@sgi.com>
[not found] ` <20080407231341.ac45cd9d.akpm@linux-foundation.org>
2008-05-08 3:49 ` [patch 12/18] FS: ExtX filesystem defrag Christoph Lameter
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=20080408141135.de5a6350.akpm@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=clameter@sgi.com \
--cc=linux-mm@kvack.org \
--cc=mel@skynet.ie \
--cc=npiggin@suse.de \
--cc=penberg@cs.helsinki.fi \
--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 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.