Bug 2785

Summary:	ACPI page fault during boot in acpi_hw_set_mode()
Product:	ACPI	Reporter:	Bryan Hundven (bhundven)
Component:	Config-Processors	Assignee:	Robert Moore (Robert.Moore)
Status:	REJECTED DUPLICATE
Severity:	blocking	CC:	acpi-bugzilla
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.(6-mm*\|7\|8)	Subsystem:
Regression:	---	Bisected commit-id:

Description Bryan Hundven 2004-05-28 07:19:40 UTC

Distribution:
Gentoo 2004.1 ~x86 (2.6.7-rc1 vanilla kernel)

Hardware Environment:
Intel D875pbzLK - P4 3.2c - 1024Mb Dual-Channel mem.

Software Environment: 
Problem Description:
I get the following from this conifg:
http://nineinchnerds.org/~bman/kernelpanic/config-20040528-4.config

The output is found here:
http://nineinchnerds.org/~bman/kernelpanic/panic-20040528-4.txt

I've seen this same problem from 2.6.6-mm3 to 2.6.7-rc1-mm1.

snippet of death:
===================8< snip 8<==============================================
ACPI: Subsystem revision 20040326
1UUaabe tt haaddle ernel  NgLLg reqntet< de> at rerteal avireus 0003r00s c0
2ric6ing nab:etUnaandle  erndl NULr eo NUeL pernfer nere<1>enael at  irtull
aedres  UL0 poi8
er drrntere eep:
 c0 v94tual *pdr =s 0000000

O pri 0000 [ip]
0PREEMPT <*pSeP  00>
00M0
les linked in:
CPU:    1
EIP:    0060:[<c01194f5>]    Not tainted
EFLAGS: 00010082   (2.6.7-rc1) 
EIP is at do_page_fault+0x180/0x50d
eax: 0000000f   ebx: 0061f000   ecx: c050de74   edx: 00000060
esi: 00000000   edi: c0119375   ebp: f7f9f670   esp: f7f9dc8c
ds: 007b   es: 007b   ss: 0068
Process swapper (pid: 1, threadinfo=f7f9c000 task=f7f9f670)
Stack: c048c276 00030002 00000008 dedfd700 0000001e 00030002 00000003 c1813ca0 
       00030001 f7f9dcec c011c862 c1814660 c1814780 c012984b 00000000 0000090c 
       c18156f4 366bd769 f7f9dcd4 f7f9dcd4 00000001 00000001 c05e8d88 0000000a 
Call Trace:
 [<c011c862>] scheduler_tick+0x11a/0x4dd
 [<c012984b>] run_timer_softirq+0x134/0x1c2
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c0106b39>] error_code+0x2d/0x38
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c012160a>] printk+0x1c6/0x1eb
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c01194cb>] do_page_fault+0x156/0x50d
 [<c0205b4d>] acpi_ns_lookup+0x1d9/0x2e8
 [<c020ded0>] acpi_ut_acquire_mutex+0x5c/0x72
 [<c020df4d>] acpi_ut_release_mutex+0x67/0x6c
 [<c020ded0>] acpi_ut_acquire_mutex+0x5c/0x72
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c0106b39>] error_code+0x2d/0x38
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c0106b39>] error_code+0x2d/0x38
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c020007b>] acpi_ex_opcode_6A_0T_1R+0x127/0x1c4
 [<c0119d63>] fixup_exception+0xb/0x34
 [<c020cf3d>] acpi_ut_release_to_cache+0x31/0x85
 [<c011948e>] do_page_fault+0x119/0x50d
 [<c02067ba>] acpi_ns_delete_namespace_by_owner+0x26/0x9c
 [<c020ded0>] acpi_ut_acquire_mutex+0x5c/0x72
 [<c0207a06>] acpi_ns_get_next_node+0x22/0x4c
 [<c0207a68>] acpi_ns_walk_namespace+0x38/0xf4
 [<c0207b10>] acpi_ns_walk_namespace+0xe0/0xf4
 [<c020df4d>] acpi_ut_release_mutex+0x67/0x6c
 [<c0119375>] do_page_fault+0x0/0x50d
 [<c0106b39>] error_code+0x2d/0x38
 [<c02045ec>] acpi_hw_set_mode+0xc3/0x100
 [<c01fd63e>] acpi_enable+0x22/0x48
 [<c020e2f2>] acpi_enable_subsystem+0x29/0x47
 [<c060058a>] acpi_bus_init+0xa4/0x124
 [<c0600668>] acpi_init+0x5e/0xd9
 [<c05ec909>] do_initcalls+0x28/0xb4
 [<c0131201>] init_workqueues+0x17/0x2e
 [<c0100520>] init+0xf3/0x263
 [<c010042d>] init+0x0/0x263
 [<c0104271>] kernel_thread_helper+0x5/0xb

Code: 8b 42 28 c7 04 24 90 c2 48 c0 89 44 24 04 e8 3c 7f 00 00 8b 
 <1>Unable to handle kernel paging request<0>Kernel panic: Attempted to kill init!
  at virtual address 00002163
 printing eip:
c01ea041
*pde = 00000000
===================8< snip 8<==============================================
The above was provided by my friend minicom...

This problem is _not_ blocking because I can add the arguments:
smpacpi=off pci=noacpi
and it boots without acpi support.

Steps to reproduce:
Build and Reboot.

Comment 1 Bryan Hundven 2004-06-23 20:45:18 UTC

Bug is the same on 2.6.7 on debian testing

Comment 2 Bryan Hundven 2004-07-16 12:10:25 UTC

Has any progress been made on this bug?
I have gotten a lot of responses from other people who are getting this bug, 
but that status hasn't changed (other then being reassigned).

Thanks for any info.

Comment 3 Anil S Keshavamurthy 2004-07-18 13:18:15 UTC

This bug is same as 
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=125841

Comment 4 Bryan Hundven 2004-08-05 16:37:57 UTC

It would seem that the patch supplied:
https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=102078&action=view
on the post above works on both 2.6.8-rc2 and 2.6.8-rc3 for me.

I only bring this up here to echo the results from the redhat bugzilla.

This was tested on the same computer as my original post, but with Debian 3.1
(testing)

Thank you Anil S Keshavamurthy!

Comment 5 Bryan Hundven 2004-09-02 23:20:33 UTC

Continues to fail in 2.6.8 and 2.6.9-rc1.
Applying the above patch still fixes the problem.

Comment 6 Bryan Hundven 2004-09-03 14:59:09 UTC

bug 4921 has fixed this issue. Hence, this is a duplicate.

*** This bug has been marked as a duplicate of 2941 ***

Comment 7 Bryan Hundven 2004-09-03 15:00:27 UTC

pardon my dyslexia, 2941 :)