Bug 5135

Summary:	Boot hang - Asus K8N-DL
Product:	ACPI	Reporter:	Andi Kleen (andi-bz)
Component:	Config-Other	Assignee:	acpi_config-other
Status:	CLOSED CODE_FIX
Severity:	normal	CC:	acpi-bugzilla
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.14rc1	Subsystem:
Regression:	---	Bisected commit-id:
Attachments:	acpidmp

Description Andi Kleen 2005-08-26 14:12:57 UTC

(also tried with acpi-20050729-2.6.13-rc6.diff.bz2 from kernel.org)

Asus K8N-DL hangs at boot. 

I did some instrumentation and the acpi_device_list gets a node that
points to itself, which causes an hang later when the list is walked.

Still debugging, but Len asked for an acpidmp

Comment 1 Andi Kleen 2005-08-26 14:13:39 UTC

Created attachment 5778 [details]
acpidmp

Comment 2 Andi Kleen 2005-08-26 14:15:28 UTC

I must add I cannot supply a boot log because the Serial port seems
to be broken on that board

Comment 3 Andi Kleen 2005-08-26 14:18:40 UTC

Another detail: the device that loops calls itself MEM0

Comment 4 Andi Kleen 2005-09-13 01:13:54 UTC

Problem still happens in 2.6.14rc1.

Comment 5 Len Brown 2005-10-06 22:09:03 UTC

toasted acpi_device_list?
Does the fix here help:?
https://bugzilla.novell.com/show_bug.cgi?id=116763

Comment 6 Andi Kleen 2005-10-10 17:16:21 UTC

With that change the hang goes away yes. 

It still doesn't boot fully without pci=noacpi unfortunately, apparently
the NIC interrupt gets misrouted.

I get (on the pci=noacpi boot, on the one that hangs it is difficult to get
a log because serial is broken and the machine has a nfs root) 

ACPI: Subsystem revision 20050902
    ACPI-0339: *** Error: Looking up [\_SB_.PCI0.LNK0] in namespace, AE_NOT_FOUND
search_node ffff810001e3ca00 start_node ffff810001e3ca00 return_node
0000000000000000
    ACPI-0339: *** Error: Looking up [\_SB_.PCI0.APC0] in namespace, AE_NOT_FOUND
search_node ffff810001e3c9c0 start_node ffff810001e3c9c0 return_node
0000000000000000
ACPI: Interpreter enabled
ACPI: Using PIC for interrupt routing

Comment 7 Len Brown 2005-10-13 10:52:08 UTC

okay, thanks, lets keep 1 failure per 1 bug report.

Comment 8 Len Brown 2007-03-07 23:11:52 UTC

fix found by Thomas in comment #5 shipped in 2.6.16.
closed.