Bug 22632 - Kernel Panic booting 2.6.35 on Toshiba Qosmio X500 PQX33U-01H01M
Summary: Kernel Panic booting 2.6.35 on Toshiba Qosmio X500 PQX33U-01H01M
Status: CLOSED INVALID
Alias: None
Product: ACPI
Classification: Unclassified
Component: ACPICA-Core (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: acpi_acpica-core@kernel-bugs.osdl.org
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-11-10 17:47 UTC by Eli Wapniarski
Modified: 2010-11-24 20:33 UTC (History)
2 users (show)

See Also:
Kernel Version:
Subsystem:
Regression: No
Bisected commit-id:


Attachments

Description Eli Wapniarski 2010-11-10 17:47:28 UTC
Manufacturer: TOSHIBA
        Product Name: Qosmio X500
        Version: PQX33U-01H01M
Comment 1 Eli Wapniarski 2010-11-10 17:54:53 UTC
Sorry about the little blip up top... Must have hit enter by mistake.

Anyway.... Last week I opened up a bug report at:

https://bugzilla.redhat.com/show_bug.cgi?id=649181

When I boot into Kernel 2.6.35.6-48.fc14.x86_64 (Fedora's current build) I get a kernel panic unless I use the boot parameter acpi=off

The details of what information that I'm able to provide are in the above mentioned bug report.

Thank you for your attention on this matter
Comment 2 Zhang Rui 2010-11-15 02:37:11 UTC
please remove the boot option "rhgb quiet" and reboot.
it would be great if you take a screenshot with DC when the system hangs, and attach it here.
Comment 3 Len Brown 2010-11-16 02:33:29 UTC
is it possible to boot kernels built from the upstream kernel tree
rather than the fedora source tree?  (you can use fedora's .config
if you like)

In particular, it would be good to find the latest upstream kernel
that works and the earliest upstream kernel that fails.
Indeed, a git bisect may tell us exactly what commit breaks
this machine.
Comment 4 Eli Wapniarski 2010-11-16 06:03:51 UTC
(In reply to comment #2)
> please remove the boot option "rhgb quiet" and reboot.
> it would be great if you take a screenshot with DC when the system hangs, and
> attach it here.

This would be extremely difficult as the computer crashes on boot. I did try to take a video with a camera that I had in hopes that I could split parts of boot process into stills, but the image is very very bad. Really its imposible to see anything.
Comment 5 Eli Wapniarski 2010-11-16 06:08:53 UTC
I can certainly try... Its been years since I've compiled a kernel from scratch you might need to do some hand holding on that one.

I would be very happy to try to do a bisect, but you will have to hand my hand on this one for sure.
Comment 6 Eli Wapniarski 2010-11-16 06:16:12 UTC
Oh... I'm assuming that you would like to start with 2.6.34?

Please let me know how you would like to proceed.

Thanks
Comment 7 Eli Wapniarski 2010-11-17 20:38:32 UTC
Please let me know if the procedure that I'm following is sound

What I've decided to do is to start with 2.6.35

1) I've downloaded the source rpm from Fedora
2) Installed the source rpm
3) Modified the SPEC file to eliminate any reference to anything
   dealing with ACPI.
4) rebuild the kernel

Same kernel panic at boot.

The next step will be to eliminate any reference to any patches in the SPEC file and rebuild. If that does work, then I will start to try to find what patch has caused the kernel panic. If I continue to get the panic after that I will then start working with the 2.6.34 src rpm in the manner outlined above.

Thanks for any comment you are willing to provide
Comment 8 Eli Wapniarski 2010-11-18 20:28:41 UTC
Yippee!!!!

I found the kernel where the problem was introduced.

Created vanilla kernels for linux-2.6.35.1-4

Upto 2.6.35.3 the computer seemed to have booted up normally. With 2.6.35.4 I got a kernel panic at boot time.


So..... What's the next step?


Thank you so much for you assistance on this.
Comment 9 Eli Wapniarski 2010-11-24 20:33:03 UTC
Well, I finally found the problem and as it turns out it wasn't a kernel problem per se.... I hope that what I'm about to report doesn't get you too annoyed with me.

With your coaching and the coaching over at Fedora, I was finally able to narrow down the problem to a kmod that I used to need... That being the kmod for the omnibook drivers.

It turns out that not only do I not need them anymore, but that they cause a conflict. Once I removed the omnibook drivers all was well.

Thank you so much for your time, effort and coaching


Sincerely

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