Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 190989 - [2.6.21 regression] ACPI-related freeze during kernel boot
Summary: [2.6.21 regression] ACPI-related freeze during kernel boot
Status: VERIFIED UPSTREAM
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: x86 Linux
: High critical
Assignee: Gentoo Kernel Bug Wranglers and Kernel Maintainers
URL: http://bugzilla.kernel.org/show_bug.c...
Whiteboard: linux-2.6.21-regression linux-2.6.24
Keywords:
Depends on:
Blocks:
 
Reported: 2007-09-01 18:31 UTC by Christian Wiegele
Modified: 2008-12-04 13:58 UTC (History)
0 users

See Also:
Package list:
Runtime testing required: ---


Attachments
Kernel Config (.config,33.98 KB, text/plain)
2007-09-02 14:02 UTC, Christian Wiegele
Details
Error Message wit last kerel.org 2.6.22.6 (IMG_2892.JPG,724.21 KB, image/jpeg)
2007-09-03 09:57 UTC, Christian Wiegele
Details
error message with gentoo-soures-2.6.22-r5 (IMG_2893.JPG,671.48 KB, image/jpeg)
2007-09-03 10:37 UTC, Christian Wiegele
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Christian Wiegele 2007-09-01 18:31:37 UTC
Hello,

i have go a Samsung P55 notebook which is running fine on gentoo with gentoo-sources <=2.6.20-gentoo-r9.
I did an upgrade to 2.6.22-gentoo-r5. Compiling worked find but booting does not work. Loading of the new krenel stops after a few seconds with a black screen for a second. after my screen comes back the system is freezed an i see the message:

pnp: 00:01: iomem range 0xfed1c000-0xfed1ffff could not be reserved (these message comes 5 times with different numbers)

PCI: Failed to allocate mem resource #6:20000@0000000 for 0000:01:00.0 (seemes to be the Nvidia GF 8600M GS)


i tried several other kernels > 2.6.20-r9 and different configurations


non of these did work.

it seemes to me that smoeting has changed between 2.6.20 -> 2.6.21 do not know exactly what en when.



Reproducible: Always

Steps to Reproduce:
1.install gentoo-sources > 2.6.20-r9
2.boot the system


Actual Results:  
system freezed with error message:

pnp: 00:01: iomem range 0xfed1c000-0xfed1ffff could not be reserved (these message comes 5 times with different numbers)

PCI: Failed to allocate mem resource #6:20000@0000000 for 0000:01:00.0



Expected Results:  
boot normaly

maybe its a problem of the Nvidia Geforce 8600M GS ?
Comment 1 Jason Smathers (RETIRED) gentoo-dev 2007-09-02 03:08:31 UTC
Does the same problem exist in a kernel built from recent vanilla sources?
Are you able to capture the full error via serial or net console?
Can you post your kernel config?
Comment 2 Christian Wiegele 2007-09-02 14:00:02 UTC
(In reply to comment #1)
> Does the same problem exist in a kernel built from recent vanilla sources?
> Are you able to capture the full error via serial or net console?
> Can you post your kernel config?
> 

- Yes, its the same problem with the vanilla sources. kernel-config is the default:
run make menuconfig, exit with no changes and save. then run make

- how does capture via serial or net console work?
Comment 3 Christian Wiegele 2007-09-02 14:02:10 UTC
Created attachment 129834 [details]
Kernel Config
Comment 4 Mike Pagano gentoo-dev 2007-09-02 17:48:57 UTC
(In reply to comment #2)
> - how does capture via serial or net console work?

In the Documentation directory of your kernel source is a file called serial-console.txt that provides the details.



Comment 5 Maarten Bressers (RETIRED) gentoo-dev 2007-09-02 19:46:17 UTC
(In reply to comment #2)
> - how does capture via serial or net console work?

And details on using netconsole are in Documentation/networking/netconsole.txt
Comment 6 Christian Wiegele 2007-09-03 08:03:57 UTC
i'm getting netconsole working, but the probem is, that the kernel freezes before it loads "netconsole" so no error messages are transfered.

so i tried to use serial-conole. i bootet the kernel but im not getting any messages on the other side. i used hyperterminal and a linux box with "cat /dev/ttyS0"

any other ideas?
Comment 7 Christian Wiegele 2007-09-03 09:57:15 UTC
Created attachment 129893 [details]
Error Message wit last kerel.org 2.6.22.6

this is the error message
Comment 8 Christian Wiegele 2007-09-03 10:37:31 UTC
Created attachment 129897 [details]
error message with gentoo-soures-2.6.22-r5

same error
Comment 9 Christian Wiegele 2007-09-03 10:54:12 UTC
i have some news:

i disabled acpi. the problem then was me computer powerd off when it startes loading the kernel. so i enabled acpi, but diabled all sub features. now i am able to boot the kernel - without usind acpi.... its not good :-(
Comment 10 Christian Wiegele 2007-09-03 11:33:20 UTC
i compiled some acpi sub-features as a module. when the system is booted and i load such a module, the screen goes dark for a second. when it comes back it shows me my console a few lines earlier. the system is freezed then.
Comment 11 Jason Smathers (RETIRED) gentoo-dev 2007-09-03 16:07:09 UTC
try booting with kernel option pci=routeirq and please let me know the result.
Comment 12 Christian Wiegele 2007-09-04 06:14:00 UTC
pci=routeirq makes no difference. system is still freezing...
Comment 13 Jason Smathers (RETIRED) gentoo-dev 2007-09-05 01:49:47 UTC
It seems we will have to send this one upstream.  Before doing so can you please try once more with a recent development kernel to make sure that this problem has not already been fixed?  vanilla-sources-2.6.23_rc5 is the latest and is in portage.  If this kernel does not resolve the problem, please file a bug on http://bugzilla.kernel.org/ and post the upstream bug number here so we can track it.  Also, you can reference this bug, but please provide the specific problem and error message upstream so they do not have to come here for it.
Comment 14 Christian Wiegele 2007-09-05 06:36:57 UTC
still the same problem with the latest vanilla-sources

http://bugzilla.kernel.org/show_bug.cgi?id=8973
Comment 15 Daniel Drake (RETIRED) gentoo-dev 2007-09-19 10:36:47 UTC
thanks, will work on upstream bug from this point