Bug 16331 - lots of unknown symbols with a kernel panic without the nolapic option at boot - Dell Latitude D630
Summary: lots of unknown symbols with a kernel panic without the nolapic option at boo...
Status: REJECTED UNREPRODUCIBLE
Alias: None
Product: ACPI
Classification: Unclassified
Component: Config-Other (show other bugs)
Hardware: All Linux
: P1 high
Assignee: acpi_config-other
URL: http://bugs.archlinux.org/task/18445
Keywords:
Depends on:
Blocks:
 
Reported: 2010-07-03 16:33 UTC by Ludovic Gasc
Modified: 2010-09-27 01:30 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.32 - 2.6.33 - 2.6.34
Subsystem:
Regression: Yes
Bisected commit-id:


Attachments
acpidump + dmesg + dmidecode + interrupts + lspci (53.75 KB, application/x-bzip)
2010-07-03 16:33 UTC, Ludovic Gasc
Details
dmesg from the 2.6.30 kernel (35.11 KB, text/plain)
2010-07-05 12:32 UTC, Ludovic Gasc
Details

Description Ludovic Gasc 2010-07-03 16:33:15 UTC
Created attachment 27008 [details]
acpidump + dmesg + dmidecode + interrupts + lspci

On my Dell Latitude D630, I've several problems with the acpi since some Linux versions:
# HAL crashes at boot, see the logs: http://bbs.archlinux.org/viewtopic.php?pid=713412#p713412
# Xorg crashes randomly without logs.
# Often I've a kernel panic during the boot

With acpi=off, it works correctly, but without the power management and only one CPU :-/

I've posted a bug report on the Arch Linux bug tracker.
After some questions from a developer of Arch Linux, he recommends me to fill a bug on kernel.org.

You can read our discussion here: http://bugs.archlinux.org/task/18445

Since this discussion, I've tested several version of the Linux kernel (2.6.32 - 2.6.33 - 2.6.34), especially with the vanilla version: http://aur.archlinux.org/packages.php?ID=24450

I've the same problem, except with the 2.6.30 version.

After the reading of this web page: http://www.lesswatts.org/projects/acpi/debug.php I've tested all options.

The bug disappears with the "nolapic" option at boot.

Now, I've the status of my battery, but I still have only one CPU, instead of two CPUs.

I can provide some logs or tests.

BIOS version: A17
Linux kernel: 2.6.34

Thanks for your support.
Comment 1 Zhang Rui 2010-07-05 07:15:42 UTC
so the problem doesn't exist in 2.6.30, right?
can you attach the full dmesg output after boot in 2.6.30?
Comment 2 Ludovic Gasc 2010-07-05 12:32:08 UTC
Created attachment 27021 [details]
dmesg from the 2.6.30 kernel
Comment 3 Ludovic Gasc 2010-07-05 12:36:19 UTC
(In reply to comment #1)
> so the problem doesn't exist in 2.6.30, right?

Yes, you understand the situation.

> can you attach the full dmesg output after boot in 2.6.30?

See Comment 2.
I've made the dmesg output with the Arch-Linux USB installer ftp://mirrors.kernel.org/archlinux/iso/2009.08/archlinux-2009.08-netinstall-x86_64.img
This installer has the 2.6.30 version.

Thank you for your answer.
Comment 4 Len Brown 2010-07-13 01:48:39 UTC
Since 2.6.30 works and 2.6.23 and later fail, is it possible
for you to git-bisect what kernel change introduced the failure?
Comment 5 Ludovic Gasc 2010-07-15 16:15:45 UTC
Ok, I'm testing with git-bisect.

I'll update this bug report with my results.
Comment 6 Zhang Rui 2010-08-16 02:13:30 UTC
ping... :)
Comment 7 Len Brown 2010-08-31 02:31:05 UTC
ping
Comment 8 Ludovic Gasc 2010-09-01 12:42:59 UTC
Sorry, I haven't this computer anymore.
I'd made several tests, I'd always the same problem.
Comment 9 Zhang Rui 2010-09-27 01:30:33 UTC
bad news. :(
please re-open this bug report if anyone with this computer can help us debug.

Note You need to log in before you can comment on or make changes to this bug.