Comment 4 for bug 7304

Andres Salomon wrote:

>On Wed, 04 Aug 2004 10:46:00 +0100, Chris Jones wrote:
>
>
>Is there a backtrace?
>
>
As follows,

Code: 8b 42 28 c7 04 24 fc eb 2a c0 89 44 24 04 e8 77 89 00 00 8b
<1> oops: 0002 [#2]
Kernel panic : fatal exception in interrupt
In interrupt handler - not syncing
PREEMPT SMP
modules linked in
CPU : 1
EIP : 0060 : [<f7abdf74>] not tainted
EFLAG : 00010246 (2.6.7-1-686-smp)
EIP is at 0xf7abdf74
eax: 00000000 ebx: 00000001 ecx: 00000008 edx: 000000b2
esi: 00000bb8 edi: 00000001 ebp: 00000008 esp: f7abdf70
ds:007b es: 007b ss: 0068
process swaper (pid:1, threadinfo=f7abc000 task= f7abf670
Stack
00000001, 00000008, c01cc8b8, 00000000, 00000001, 00000000, c01c58f6,
00000001
00000000, c01d66f2, c03690e4, c030d680, 00000000, 000000f4, c0393540,
c035b742
c030d680, c03469eb, 00000008, c0132bc5, c02a88ce, c0374dc0, 0000001e,
c0100574
call trace: [<c01cc8b8>] acpi_hw_set_mode+0xc3/0x100

>>I have also recived oopses, random asci strings (but no error) followed
>>by hang, and messages about garbage stacks
>>
>>The non smp verson of this package boots, i include the output of dmesg
>>for this kenel
>>
>>
>
>
>Not all that interested in the non-oopsing kernel boot log; can you send
>the boot log of the oopsing kernel?
>
Not unless there is an easy way to capture it, as the kernel either
panics or hangs.

>Also, you can try passing acpi=off to the kernel to keep it from oopsing.
>
This works,

>Alternatively, if it's just the pci apci stuff that's causing problems (i've had problems w/ this
>code), try pci=noacpi.
>
This does not

Additionally i had no problems with 2.6.6-1-686-smp

----
Chris Jones
http://brasskipper.org.uk