Bug 133585 - kernel panic when loading r8169 network driver
Summary: kernel panic when loading r8169 network driver
Status: VERIFIED FIXED
Alias: None
Product: SUSE LINUX 10.0
Classification: openSUSE
Component: Kernel (show other bugs)
Version: unspecified
Hardware: i686 Other
: P5 - None : Critical
Target Milestone: ---
Assignee: Karsten Keil
QA Contact: E-mail List
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-11-12 12:36 UTC by Stefan Sauer
Modified: 2006-11-23 00:11 UTC (History)
1 user (show)

See Also:
Found By: Other
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
screenshot of kernel oops (209.05 KB, image/jpeg)
2005-11-16 20:15 UTC, Stefan Sauer
Details
This patch should fix the oops (394 bytes, patch)
2005-11-17 16:47 UTC, Olaf Kirch
Details | Diff
new patch from netdevel list (1.03 KB, patch)
2005-11-18 22:46 UTC, Karsten Keil
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Stefan Sauer 2005-11-12 12:36:11 UTC
with an online update in suse9.3 around the 26. oct. 2005 the networkdriver has changed. Form this time loading the network driver resulted in a kernel panic.

lspci show the onboard network controller as:
RealTek RTL-8169

The kernel output is something like:
r8169: Gigabit Ethernet driver 2.2Lk loaded
ACPI: PCI interrupt ...
r8169: Cannot find Powermanagement ....
ACPI: ....
Unable to handle kernel NULL ...
<the kernel panic dump>

I then updated to Suse 10 (SUSE-10.0-EvalDVD-i386-GM.iso) and still can't use the network driver. Same thing happends when booting e.g. Gentoo Live 2005.1. So it does no seem to be suse specific.

Question 1: can you help me to find out what exactly has changed, so that I can report it to the kernel developers.

Question 2: any ideas how one can redirect the kernel error message, normal file-redirection does not work. if needed I will tae a picture using my digital camera, upload it to my webserver and post the url here.
Comment 1 Michael Gross 2005-11-14 16:08:31 UTC
On boot time, the messages are stored in /var/log/boot.msg
This file could be broken nthl. Check it for the messages. If they are not there or you're using a life system, just make a picture of the relevant lines and post it here.

To work around this, you can try passing several kernel options at boot time, like pci=routeirq, acpi=off (probably not really required on a normal workstation) or pci=noacpi.
Comment 2 Stefan Sauer 2005-11-16 20:15:09 UTC
Created attachment 57568 [details]
screenshot of kernel oops

I've screenshoted the kernel oops. Using parameters like  ACPI=off and so on didn't helped to work around the problem.
The captured output also did not apprear in the boot.oldmsg.
Comment 3 Michael Gross 2005-11-17 16:37:56 UTC
Olaf: Something for you? If not, please assign it back to us.
Comment 4 Olaf Kirch 2005-11-17 16:47:01 UTC
Created attachment 57662 [details]
This patch should fix the oops
Comment 5 Olaf Kirch 2005-11-17 16:48:38 UTC
But that doesn't explain the underlying problem of what's wrong with
the driver and why it refuses to init the card. What was the last
working kernel you used?
Comment 6 Stefan Sauer 2005-11-18 07:46:33 UTC
I didn't built my own kernel. So I was using the one which was up-to-date in Suse9.3 as of 26th october 2005. I run the online-updates at least twice a week.

It's unlikely that I can find out which updates have been installed with this online-update as I have tried to upgrade to SuSE 10 afterwards. 
Comment 7 Karsten Keil 2005-11-18 10:10:32 UTC
I'll make a test kernel available for you.
Comment 8 Karsten Keil 2005-11-18 22:46:08 UTC
Created attachment 57775 [details]
new patch from netdevel list

This should fix it, I did sent you a email with a precompiled module for your kernel to test.
Comment 10 Stefan Sauer 2005-11-20 16:05:31 UTC
Your wonderful! It works.
Comment 11 Andreas Jaeger 2005-11-21 08:07:16 UTC
Karsten, please add this to our 10.0 kernel for the next update.