linux-kernel - Re: Regression in linux 2.6.37: failure on remount / (ext4) rw (was: Re: Regression in linux 2.6.37: failure to boot, caused by commit 37d57443d5 (mm/slub.c))

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20110111130959.GA19003@merz.inka.de>
Date:	Tue, 11 Jan 2011 14:09:59 +0100
From:	Matthias Merz <linux@...z-ka.de>
To:	Pekka Enberg <penberg@...nel.org>
Cc:	Matthias Merz <linux@...z-ka.de>, Christoph Lameter <cl@...ux.com>,
	Matt Mackall <mpm@...enic.com>, linux-mm@...ck.org,
	Tero Roponen <tero.roponen@...il.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...e.hu>
Subject: Re: Regression in linux 2.6.37: failure on remount / (ext4) rw
	(was: Re: Regression in linux 2.6.37: failure to boot, caused by
	commit 37d57443d5 (mm/slub.c))

Hello,

Am Di, 11.01.2011 09:50 schrieb Pekka Enberg
> On Tue, Jan 11, 2011 at 12:31 AM, Matthias Merz <linux@...z-ka.de> wrote:
> > This morning I tried vanilla 2.6.37 on my Desktop system, which failed
> > to boot but continued displaying Debug-Messages too fast to read. Using
> > netconsole I was then able to capture them (see attached file). I was
> > able to trigger this bug even with init=/bin/bash by a simple call of
> > "mount -o remount,rw /" with my / being an ext4 filesystem.
> >
> > Using git bisect I could identify commit 37d57443d5 as "the culprit" -
> > once I reverted that bugfix locally, my system booted happily.

Sorry; this morning the version which I was able to boot yesterday
failed. I assume some "hardware state" influeces triggering of this bug
and my whole bisect-session yesterday was useless because I did mostly
warm restarts and no cold boot when trying :-(


> Now here's an interesting bug! Commit 37d57443d5 shouldn't change
> anything unless you actually access the affected sysfs files. I'm but
> lost with your oops as well so lets CC some scheduler people to see if
> they can help us out:

You're proven to be right; sorry for complicating things by wrong
conclusions - next time I'll hopefully remember to cut the AC-power for
10 seconds when testing revisions...

As I just noticed I don't use slub at all, but slab; I'll attach my
.config for reference.


Keeping the citation of the debug log for inline reference:

> BUG: scheduling while atomic: swapper/0/0x10010000
> Modules linked in: i2c_viapro usbhid snd_via82xx via_ircc
> snd_mpu401_uart parport_pc sata_promise sata_sil tmscsim evdev
> snd_bt87x tda9887 snd_seq_dummy snd_seq_oss snd_seq_midi
> snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss snd_ens1371
> snd_rawmidi snd_seq_device snd_ac97_codec ac97_bus snd_pcm snd_timer
> snd snd_page_alloc parport irtty_sir actisys_sir sir_dev irda
> crc_ccitt tuner_simple tuner_types msp3400 ir_lirc_codec lirc_dev
> ir_sony_decoder bttv ir_jvc_decoder ir_rc6_decoder videobuf_dma_sg
> videobuf_core ir_rc5_decoder btcx_risc ir_nec_decoder ir_common
> ir_core tveeprom tuner v4l2_common videodev v4l1_compat analog
> gameport uhci_hcd ehci_hcd e100
> Modules linked in: i2c_viapro usbhid snd_via82xx via_ircc
> snd_mpu401_uart parport_pc sata_promise sata_sil tmscsim evdev
> snd_bt87x tda9887 snd_seq_dummy snd_seq_oss snd_seq_midi
> snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss snd_ens1371
> snd_rawmidi snd_seq_device snd_ac97_codec ac97_bus snd_pcm snd_timer
> snd snd_page_alloc parport irtty_sir actisys_sir sir_dev irda
> crc_ccitt tuner_simple tuner_types msp3400 ir_lirc_codec lirc_dev
> ir_sony_decoder bttv ir_jvc_decoder ir_rc6_decoder videobuf_dma_sg
> videobuf_core ir_rc5_decoder btcx_risc ir_nec_decoder ir_common
> ir_core tveeprom tuner v4l2_common videodev v4l1_compat analog
> gameport uhci_hcd ehci_hcd e100
> 
> Pid: 0, comm: swapper Not tainted 2.6.37-matthias #28 A7V8X/System Name
> EIP: 0060:[<c10088ba>] EFLAGS: 00000246 CPU: 0
> EIP is at default_idle+0x2a/0x40
> EAX: 00000000 EBX: c1596140 ECX: 00000000 EDX: 00000000
> ESI: 0008d800 EDI: c153d000 EBP: c153bfbc ESP: c153bfbc
>  DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
> Process swapper (pid: 0, ti=f6004000 task=c1541300 task.ti=c153a000)
> Stack:
>  c153bfc4 c1001c7c c153bfcc c13e72a2 c153bfe4 c15706cd 000000a0 c15702b9
>  c1596140 00000000 c153bff8 c157006b 01606d60 00000000 c14b0e88 01827003
>  00000000
> Call Trace:
>  [<c1001c7c>] ? cpu_idle+0x2c/0x50
>  [<c13e72a2>] ? rest_init+0x52/0x60
>  [<c15706cd>] ? start_kernel+0x242/0x248
>  [<c15702b9>] ? unknown_bootoption+0x0/0x19c
>  [<c157006b>] ? i386_start_kernel+0x6b/0x6d
> Code: 00 55 8b 0d 18 67 5c c1 89 e5 85 c9 75 2b 80 3d 05 d5 56 c1 00
> 74 22 89 e0 25 00 e0 ff ff 83 60 0c fb 8b 40 08 a8 08 75 15 fb f4 <89>
> e0 25 00 e0 ff ff 83 48 0c 04 c9 c3 90 fb f3 90 c9 c3 fb eb
> 
> Full log available here:
> 
> http://www.spinics.net/lists/linux-mm/msg13451.html

Thanks for your effort,
Yours
Matthias Merz

-- 
In einem geschlossenen System (Kinderzimmer) wächst die Entropie
(=Unordnung) so lange, bis jemand (=Eltern) dem System Energie zuführt
(=die Türe öffnet und reinruft: "Jetzt räumt Ihr aber auf!")
               "Walter P. Zaehl" in <gervns$m90$1@...cx010.eed.ericsson.se>

View attachment "config-2.6.37-matthias+" of type "text/plain" (79279 bytes)