Bug 10857 - boot hang unless acpi=off - Asus M51SE
Summary: boot hang unless acpi=off - Asus M51SE
Status: REJECTED INSUFFICIENT_DATA
Alias: None
Product: ACPI
Classification: Unclassified
Component: BIOS (show other bugs)
Hardware: All Linux
: P1 blocking
Assignee: Zhang Rui
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2008-06-04 19:31 UTC by born.into.silence
Modified: 2008-08-14 00:14 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.26-rc4(vanilla) 2.6.25(vanilla) 2.6.25-r4(gentoo)
Subsystem:
Regression: ---
Bisected commit-id:


Attachments
Contains the text files listed above (30.13 KB, application/gzip)
2008-06-04 19:43 UTC, born.into.silence
Details
acpidump output, running 2.6.25.4 with acpi=off (70.22 KB, application/gzip)
2008-06-05 04:05 UTC, born.into.silence
Details
dmesg -s64000 output (after increasing log_buf value) (20.32 KB, application/octet-stream)
2008-06-13 18:18 UTC, born.into.silence
Details
The logs supposed to be found on the pictures from comment # 2 (7.77 KB, application/octet-stream)
2008-06-13 18:21 UTC, born.into.silence
Details

Description born.into.silence 2008-06-04 19:31:37 UTC
Latest working kernel version: None (brand new laptop)
Earliest failing kernel version: 2.6.24
Distribution: gentoo
Hardware Environment: Asus M51SE
Software Environment:
Problem Description: Several errors about ACPI & IRQs occurs boot time. Passing acpi=off to the kernel makes it boot. I've got no binary drivers.
I tried several boot options : noapic, nolapic, pci=noacpi, irqpoll, acpi=noirq, pci=nomsi.

I've got some graphic problems too : system just hangs (black screen + no ssh) when starting xorg with every driver I tried (vesa, fglrx, radeonhd). Could that be related to the irqs problem ?

Steps to reproduce: Boot without acpi=no

Included in the archive :
- pictures of the crash logs
- lspci -vvnn output
- cat /proc/interrupts output
- dmesg output
- uname -a output
- cat /proc/version output
- 2.6.26-rc4 .config (=cfg26) ------> (kernel panic : attempting to kill init)
- 2.6.25.4 .config (=cfg25)------> see pics

That's my first bug report
Comment 1 born.into.silence 2008-06-04 19:43:27 UTC
Created attachment 16400 [details]
Contains the text files listed above
Comment 2 born.into.silence 2008-06-04 19:48:05 UTC
I posted here the picture of the crash I took : http://rapidshare.com/files/120172775/2.6.25.4_crash_pics.tar.gz.html%20target=

I forgot to mention that kernel crashes too when enabling vesa framebuffer.
Comment 3 ykzhao 2008-06-04 20:08:03 UTC
Will you please boot the system with "acpi=off" and attach the output of acpidump?
Thanks.
Comment 4 born.into.silence 2008-06-05 04:05:33 UTC
Created attachment 16403 [details]
acpidump output, running 2.6.25.4 with acpi=off

I tried booting without acpi=off and a custom dsdt table, but didn't worked.
Comment 5 Len Brown 2008-06-12 14:16:56 UTC
I can't read the attachment in comment #2,
and you attach it directly to this bug report?

also, the dmesg in the report is not long enough.
please try dmesg -s64000 and if that doesn't go
back far enough, then increase CONFIG_LOG_BUF_SHIFT

please try booting with pci=nommconf

does an out-of-box live CD, such as fedora core 9 boot?
does windows boot?
Comment 6 born.into.silence 2008-06-13 18:17:24 UTC
Windows xp works, vista works, and fedora 9 fails booting with acpi on.
The attachment of comment #2 are pictures I stored on rapidshare, because they're too big for bugzilla.
Comment 7 born.into.silence 2008-06-13 18:18:12 UTC
Created attachment 16463 [details]
dmesg -s64000 output (after increasing log_buf value)
Comment 8 born.into.silence 2008-06-13 18:21:22 UTC
Created attachment 16464 [details]
The logs supposed to be found on the pictures from comment # 2

I wrote them up from the screen (I checked it twice and respected every single white space)
Kernel hangs like that with fedora 9 too.
(without acpi=off)
Comment 9 ykzhao 2008-06-17 00:42:02 UTC
Will you please try the boot option of "hpet=off" and see whether the problem still exists? 
If exist, please capture the picture of screen when the system hangs and attach it.
Thanks.
Comment 10 Zhang Rui 2008-08-14 00:14:46 UTC
Reject the bug because there is no response from the bug reporter.

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