Re: OOPS: Suspend2+iptables when loading ipw2200

[Nigel Cunningham <ncunningham@cyclades.com>]

Hi.

On Tue, 2005-09-27 at 21:06, Daniel Link wrote:
> Hi,
> 
> my Suspend2-2.6.12-r1 kernel from Gentoo's portage worked almost
> perfectly until I decided to prepare it for using iptables. I've had
> only one system freeze before, which I didn't think much about. After
> adding numerous modules about firewalling to the kernel config and
> recompiling everything I finally rebooted the system. But some seconds
> after loading the ipw2200 module (1.0.6) there was an OOPS. I also tried
> not to use modules but to put the iptables stuff directly into the
> kernel, but that didn't help.
> 
> Here's some output from dmesg. First I thought it had come from the new
> iptables kernel, but that's not the case. It's the one from yesterday's
> kernel without iptables. During normal work with GNOME the computer just
> froze and I had to do a hard reboot.

Suspend2 doesn't touch timer code but I should ask anyway: Can you
reproduce it on a vanilla kernel (without Suspend2)? If so, I would
contact the ipw2200 people. If on the otherhand you become convinced
that Suspend2 is the case, we have lists at suspend2.net and a number of
people using those modules.

Regards,

Nigel

> Sep 26 13:07:54 hermes ------------[ cut here ]------------
> Sep 26 13:07:54 hermes kernel BUG at kernel/workqueue.c:131!
> Sep 26 13:07:54 hermes invalid operand: 0000 [#1]
> Sep 26 13:07:54 hermes PREEMPT
> Sep 26 13:07:54 hermes Modules linked in: ieee80211_crypt_tkip ipw2200
> ieee80211 ieee80211_crypt
> Sep 26 13:07:54 hermes CPU:    0
> Sep 26 13:07:54 hermes EIP:    0060:[<c012f0bb>]    Not tainted VLI
> Sep 26 13:07:54 hermes EFLAGS: 00010286   (2.6.12-suspend2-r6)
> Sep 26 13:07:54 hermes EIP is at queue_delayed_work+0x6b/0x80
> Sep 26 13:07:54 hermes eax: c053f9c0   ebx: 00000000   ecx: c04b13b8
> edx: c04b13a0
> Sep 26 13:07:54 hermes esi: c145e880   edi: 0000000a   ebp: 00000000
> esp: def03f14
> Sep 26 13:07:54 hermes ds: 007b   es: 007b   ss: 0068
> Sep 26 13:07:54 hermes Process events/0 (pid: 3, threadinfo=def02000
> task=c1469020)
> Sep 26 13:07:54 hermes Stack: dea75f00 c04b13a0 c04b13a0 00000202
> c04b13a4 c0329e26 00000000 c012f2e0
> Sep 26 13:07:54 hermes 00000000 def03f68 00000000 def02000 c145e898
> def02000 c145e888 c145e890
> Sep 26 13:07:54 hermes def02000 c0329de0 def02000 ffffffff ffffffff
> 00000001 00000000 c011b140
> Sep 26 13:07:54 hermes Call Trace:
> Sep 26 13:07:54 hermes [<c0329e26>] do_dbs_timer+0x46/0x60
> Sep 26 13:07:54 hermes [<c012f2e0>] worker_thread+0x210/0x300
> Sep 26 13:07:54 hermes [<c0329de0>] do_dbs_timer+0x0/0x60
> Sep 26 13:07:54 hermes [<c011b140>] default_wake_function+0x0/0x20
> Sep 26 13:07:54 hermes [<c011b140>] default_wake_function+0x0/0x20
> Sep 26 13:07:54 hermes [<c0133a07>] kthread+0xd7/0x120
> Sep 26 13:07:54 hermes [<c012f0d0>] worker_thread+0x0/0x300
> Sep 26 13:07:54 hermes [<c0133930>] kthread+0x0/0x120
> Sep 26 13:07:54 hermes [<c010133d>] kernel_thread_helper+0x5/0x18
> Sep 26 13:07:54 hermes Code: 14 c7 41 10 30 f0 12 c0 01 f8 89 41 08 89
> 44 24 04 89 0c 24 e8 b7 85 ff ff 89 d8 8b 74 24 0c 8b 5c 24 08 8b 7c 24
> 10 83 c4 14 c3 <0f> 0b 83  00 36 a7 40 c0 eb b4 0f 0b 84 00 36 a7 40 c0
> eb b2 90
> 
> There was no other OOPS in /var/log/messages. It seems like when using
> the iptables kernel the OOPS comes too fast for logging something, but
> who am I to make any assumptions. Since there is no proof in the logs I
> have made a photo with a digital camera. The first lines look similar to
> the ones above. I haven't compared any more. Just tell me where to send
> the image. I can also send you the working/panic kernel, its config and
> the System.map.
> 
> Bye, Daniel
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
--