From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
To: Ingo Molnar <mingo@elte.hu>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>,
Torsten Kaiser <just.for.lkml@googlemail.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Christoph Lameter <clameter@sgi.com>
Subject: Re: Linux 2.6.25-rc2
Date: Tue, 19 Feb 2008 08:02:53 -0500 [thread overview]
Message-ID: <20080219130253.GA29804@Krystal> (raw)
In-Reply-To: <20080219102700.GC14092@elte.hu>
* Ingo Molnar (mingo@elte.hu) wrote:
>
> * Pekka Enberg <penberg@cs.helsinki.fi> wrote:
>
> > Mathieu, Christoph is on vacation and I'm not at all that familiar
> > with this cmpxchg_local() optimization, so if you could take a peek at
> > this bug report to see if you can spot something obviously wrong with
> > it, I would much appreciate that.
>
> hm, it's bad for at least one other reason as well (which is probably
> unrelated to this crash):
>
> /*
> * Currently fastpath is not supported if preemption is enabled.
> */
> #if defined(CONFIG_FAST_CMPXCHG_LOCAL) && !defined(CONFIG_PREEMPT)
> #define SLUB_FASTPATH
> #endif
>
> such !PREEMPT exceptions tend to show "i didnt want to think too hard
> about the preemptible case so just turn it off" thinking.
>
Ingo, a comment in slub.c explains it :
/*
* The SLUB_FASTPATH path is provisional and is currently disabled if the
* kernel is compiled with preemption or if the arch does not support
* fast cmpxchg operations. There are a couple of coming changes that will
* simplify matters and allow preemption. Ultimately we may end up making
* SLUB_FASTPATH the default.
*
* 1. The introduction of the per cpu allocator will avoid array lookups
* through get_cpu_slab(). A special register can be used instead.
*
* 2. The introduction of per cpu atomic operations (cpu_ops) means that
* we can realize the logic here entirely with per cpu atomics. The
* per cpu atomic ops will take care of the preemption issues.
*/
So there is more coming in the preemption area.
> Also, why isnt this "SLUB_FASTPATH" flag done in the Kconfig space?
>
Eventually, I think only CONFIG_FAST_CMPXCHG_LOCAL will be needed (when
the code will support preemption). Therefore, this SLUB_FASTPATH define
seems to be only here temporarily.
I'm looking at the code right now.. more to come.
Mathieu
> Ingo
--
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68
next prev parent reply other threads:[~2008-02-19 13:03 UTC|newest]
Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-02-15 21:23 Linux 2.6.25-rc2 Linus Torvalds
2008-02-16 2:08 ` Rafael J. Wysocki
2008-02-16 5:44 ` [BUG] Linux 2.6.25-rc2 - Kernel Ooops while running dbench Kamalesh Babulal
2008-02-18 12:59 ` Andrew Morton
2008-02-18 12:59 ` Andrew Morton
2008-02-18 14:25 ` Jeff Garzik
2008-02-18 16:11 ` Frans Pop
2008-02-18 16:11 ` Frans Pop
2008-03-03 11:51 ` Pekka Enberg
2008-03-03 11:51 ` Pekka Enberg
2008-03-04 4:03 ` Kamalesh Babulal
2008-03-04 4:03 ` Kamalesh Babulal
2008-02-16 6:10 ` [BUG] Linux 2.6.25-rc2 - Regression from 2.6.24-rc1-git1 softlockup while bootup on powerpc Kamalesh Babulal
2008-02-16 6:10 ` Kamalesh Babulal
2008-02-17 19:29 ` Jens Axboe
2008-02-17 19:29 ` Jens Axboe
2008-02-19 8:04 ` KAMEZAWA Hiroyuki
2008-02-19 8:04 ` KAMEZAWA Hiroyuki
2008-02-19 8:36 ` Jens Axboe
2008-02-19 8:36 ` Jens Axboe
2008-02-19 8:47 ` KAMEZAWA Hiroyuki
2008-02-19 8:47 ` KAMEZAWA Hiroyuki
2008-02-19 8:58 ` Jens Axboe
2008-02-19 8:58 ` Jens Axboe
2008-02-19 9:07 ` KAMEZAWA Hiroyuki
2008-02-19 9:07 ` KAMEZAWA Hiroyuki
2008-02-19 9:09 ` Jens Axboe
2008-02-19 9:09 ` Jens Axboe
2008-02-19 9:02 ` KAMEZAWA Hiroyuki
2008-02-19 9:02 ` KAMEZAWA Hiroyuki
2008-02-19 9:01 ` Jens Axboe
2008-02-19 9:01 ` Jens Axboe
2008-02-19 13:19 ` Kamalesh Babulal
2008-02-19 13:19 ` Kamalesh Babulal
2008-02-22 7:24 ` Andrew Morton
2008-02-22 7:24 ` Andrew Morton
2008-02-22 7:40 ` Jens Axboe
2008-02-22 7:40 ` Jens Axboe
2008-02-17 20:08 ` Rafael J. Wysocki
2008-02-17 20:08 ` Rafael J. Wysocki
2008-02-16 16:52 ` Linux 2.6.25-rc2 Jan Engelhardt
2008-02-16 19:14 ` Linux 2.6.25-rc2 regression: LVM cannot find volume group Tilman Schmidt
2008-02-16 20:12 ` Alan Cox
2008-02-16 22:37 ` Jiri Slaby
2008-02-18 0:57 ` Tilman Schmidt
2008-02-18 1:22 ` Jeff Chua
2008-02-18 10:35 ` Tilman Schmidt
2008-02-19 1:53 ` Alasdair G Kergon
2008-02-19 8:56 ` Tilman Schmidt
2008-02-16 21:38 ` Linux 2.6.25-rc2 Torsten Kaiser
2008-02-17 20:25 ` Rafael J. Wysocki
2008-02-17 21:32 ` Torsten Kaiser
2008-02-18 23:54 ` Linus Torvalds
2008-02-19 6:44 ` Torsten Kaiser
2008-02-19 6:11 ` Ingo Molnar
2008-02-19 6:54 ` Torsten Kaiser
2008-02-19 7:21 ` Pekka Enberg
2008-02-19 10:27 ` Ingo Molnar
2008-02-19 10:45 ` Pekka Enberg
2008-02-19 13:02 ` Mathieu Desnoyers [this message]
2008-02-19 14:00 ` Ingo Molnar
2008-02-19 14:02 ` Mathieu Desnoyers
2008-02-19 14:21 ` Pekka Enberg
2008-02-19 14:38 ` Pekka Enberg
2008-02-19 14:55 ` Ingo Molnar
2008-02-19 14:57 ` Ingo Molnar
2008-02-19 15:54 ` Pekka Enberg
2008-02-19 15:52 ` Pekka Enberg
2008-02-20 0:36 ` Zhang, Yanmin
2008-02-20 2:08 ` Zhang, Yanmin
2008-02-20 6:53 ` Zhang, Yanmin
2008-02-20 7:10 ` Pekka Enberg
2008-02-19 16:20 ` Linus Torvalds
2008-02-19 16:45 ` Ingo Molnar
2008-02-19 16:48 ` Ingo Molnar
2008-02-19 19:27 ` Torsten Kaiser
2008-02-19 20:08 ` Mathieu Desnoyers
2008-02-27 23:32 ` Christoph Lameter
2008-02-28 1:57 ` Andrew Morton
2008-02-28 2:43 ` Christoph Lameter
2008-02-28 8:14 ` Ingo Molnar
2008-02-28 11:15 ` Alan Cox
2008-02-28 11:13 ` Jiri Kosina
2008-02-19 16:27 ` Eric Dumazet
2008-02-19 16:38 ` Linus Torvalds
2008-02-19 20:03 ` Mathieu Desnoyers
2008-02-27 23:34 ` Christoph Lameter
2008-02-28 5:55 ` [PATCH] Implement slub fastpath in terms of freebase and freeoffset Mathieu Desnoyers
2008-02-28 19:08 ` Christoph Lameter
2008-02-28 23:25 ` Mathieu Desnoyers
2008-02-29 0:57 ` Christoph Lameter
2008-02-29 1:56 ` Mathieu Desnoyers
2008-02-29 2:12 ` Christoph Lameter
2008-02-29 3:32 ` Mathieu Desnoyers
2008-02-29 5:11 ` Christoph Lameter
2008-02-29 13:03 ` Mathieu Desnoyers
2008-02-29 19:57 ` Christoph Lameter
2008-02-29 13:28 ` [PATCH] Slub Freeoffset check overflow Mathieu Desnoyers
2008-03-04 6:17 ` [PATCH] Slub Freeoffset check overflow (updated) Mathieu Desnoyers
2008-03-04 7:15 ` Christoph Lameter
2008-02-27 23:32 ` Linux 2.6.25-rc2 Christoph Lameter
2008-02-19 18:39 ` Torsten Kaiser
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=20080219130253.GA29804@Krystal \
--to=mathieu.desnoyers@polymtl.ca \
--cc=clameter@sgi.com \
--cc=just.for.lkml@googlemail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=penberg@cs.helsinki.fi \
--cc=torvalds@linux-foundation.org \
/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.