linux-kernel - Re: 2.6.23-rc6-mm1: atomic counter underflow

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20070919102716.09004317@gondolin.boeblingen.de.ibm.com>
Date:	Wed, 19 Sep 2007 10:27:16 +0200
From:	Cornelia Huck <cornelia.huck@...ibm.com>
To:	Mariusz Kozlowski <m.kozlowski@...land.pl>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org, Greg KH <gregkh@...e.de>
Subject: Re: 2.6.23-rc6-mm1: atomic counter underflow

On Tue, 18 Sep 2007 23:45:13 +0200,
Mariusz Kozlowski <m.kozlowski@...land.pl> wrote:

> Hello,
> 
> 	I wrote a simple script that finds all modules in /lib/modules/`uname -r`
> and performs sth like 'modprobe $x; rmmod $x;' for every of them. The result is the
> output below. This actually happens when rmmod'ing a module.
> 
> I narrowed it down to modules mmc_core and firewire_core. To reproduce that just
> rmmod them.

Hm, does that mean you can reproduce this when you modprobe/rmmod
either of them on a fresh boot? (There could be follow-on errors...)

Could you try with DEBUG_KOBJECT set?

> PS. X doesn't work but that's another story.
> 
> Regards,
> 
> 	Mariusz
> 
> WARNING: at lib/kref.c:33 kref_get()
>  [<c0104a1c>] dump_trace+0x202/0x22b
>  [<c0104a5f>] show_trace_log_lvl+0x1a/0x30
>  [<c0105608>] show_trace+0x12/0x14
>  [<c010572c>] dump_stack+0x15/0x17
>  [<c024c567>] kref_get+0x40/0x42
>  [<c024b81e>] kobject_get+0x12/0x17
>  [<c02d5800>] bus_get+0x11/0x22
>  [<c02d5a89>] bus_remove_file+0xe/0x27
>  [<c02d5ab2>] remove_probe_files+0x10/0x1f
>  [<c02d5afc>] bus_unregister+0x3b/0x66
>  [<dee3c2bd>] sdio_unregister_bus+0xd/0xf [mmc_core]
>  [<dee3d66d>] mmc_exit+0xd/0x23 [mmc_core]
>  [<c013c9d6>] sys_delete_module+0x154/0x202
>  [<c010406a>] sysenter_past_esp+0x5f/0x99
>  [<ffffe410>] 0xffffe410
>  =======================

Is this the first error message you see? This looks like the embedded
kset had its refcount decreased too much. (Or was the sdio bus never
registered? Don't see how this could happen if mmc_init was successful,
though.)

(The next errors seem to have the same cause.)

> modprobing: firewire-core
> WARNING: at lib/kref.c:33 kref_get()
>  [<c0104a1c>] dump_trace+0x202/0x22b
>  [<c0104a5f>] show_trace_log_lvl+0x1a/0x30
>  [<c0105608>] show_trace+0x12/0x14
>  [<c010572c>] dump_stack+0x15/0x17
>  [<c024c567>] kref_get+0x40/0x42
>  [<c024b81e>] kobject_get+0x12/0x17
>  [<c024b97b>] kobject_init+0x43/0x82
>  [<c024b9c8>] kset_init+0xe/0x2e
>  [<c024be4f>] kset_register+0x15/0x38
>  [<c024be7a>] subsystem_register+0x8/0xa
>  [<c02d6008>] bus_register+0x65/0x209
>  [<ded6300d>] fw_core_init+0xd/0x99 [firewire_core]
>  [<c013aceb>] sys_init_module+0x125/0x1612
>  [<c010406a>] sysenter_past_esp+0x5f/0x99
>  [<ffffe410>] 0xffffe410

Strange. Can't see how this can happen except as a follow-on error.

(Next ones are follow-on errors from this one.)

> modprobing: mmc_block

(Following errors don't look like they could have anything to do with
the driver core.)

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/