Bug 149530

Summary:	SMP Kernel crashes on boot unless 'noapic' is used
Product:	Gentoo Linux	Reporter:	jonathan e. Snow <jesnow>
Component:	[OLD] Core system	Assignee:	Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel>
Status:	RESOLVED NEEDINFO
Severity:	normal	CC:	jakub, stefhome
Priority:	High
Version:	2006.1
Hardware:	x86
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---

Description jonathan e. Snow 2006-09-29 09:02:04 UTC

While the kernel is initializing the sata driver, I get

BUG: Unable to handle kernel NULL pointer dereference
and then a call stack that points to functions in in libata-core.c

If I pass the kernel param 'noapic=TRUE' this doesn't happen. 

It's an ASUS P5NSLI Motherboard with 1G mem and a Core 2 Duo 
that I just installed into an otherwise healthy gentoo system.

thanks in advance

Interestingly, the 2005.1 install cd was able to boot. Am I missing some
kernel parameters? Or do the install cd's turn off APIC for stability?

Who do I go to to track this problem down? I've already googled quite a 
lot and found nothing. APIC and SMP are old enemies, I know that.

Comment 1 Jakub Moc (RETIRED) gentoo-dev

2006-09-29 09:06:17 UTC

Well, I had this issue as well, and all I can tell you is that Asus BIOSes are broken. :) Do you have the latest one?

Comment 2 Stefano 2006-10-01 06:56:47 UTC

Hello
I had the same problem with an ASUS pundit PI-AH2 with an AMD64 X2 CPU while I was trying to boot from the minimal install CD (and later from the 2.6.17-r8 Kernel I compiled). The boot process stopped when the different sata drivers where checked or it continued when I disabled it with the "nosata" boot option but got later a lot of problems when trying to access the network/lan by loading the  "forcedeth" drivers.
The system works perfectly after using the "noapic" boot option.
I didn't have any such problems with my other machines which use an AMD Sempron 2600 or an Intel P4 extreme HT 3.0GHZ.
Greetings!
Stefano

Comment 3 Daniel Drake (RETIRED) gentoo-dev

2006-10-11 17:57:13 UTC

Stefano: sounds like you have a different problem, at least I have no reason to believe from your description that its a kernel oops like the one described in the original report.

jonathan: Please attach lspci output and a photo of the kernel crash

Comment 4 Daniel Drake (RETIRED) gentoo-dev

2006-10-23 17:01:36 UTC

See comment #3