Bug 117452

Summary:	system freeze during installation without acpi=off
Product:	[openSUSE] SUSE LINUX 10.0	Reporter:	Matthias Boettger <matthias.boettger>
Component:	Kernel	Assignee:	Thomas Renninger <trenn>
Status:	RESOLVED DUPLICATE	QA Contact:	E-mail List <qa-bugs>
Severity:	Normal
Priority:	P5 - None
Version:	RC 4
Target Milestone:	---
Hardware:	i686
OS:	All
Whiteboard:
Found By:	IS&T	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---

Description Matthias Boettger 2005-09-16 13:26:26 UTC

I tried to install 10.0 on a testsystem but the system freeze during package
installation. Only workaround is acpi=off.
After installation the system crashs on bootup with several stack traces, but I
can't scroll up to see more. I see only a stack trace of ata_piix.

The system is a FSC W300 with P4 2.6Ghz + HT and 512MB RAM with ATA (not SATA) disk.

I can give you access to this machine if would help.

Comment 1 Hubert Mantel 2005-09-19 15:36:23 UTC

Since disabling ACPI is a perfectly acceptable workaround, this is not a BLOCKER.

Comment 2 Thomas Renninger 2005-09-19 16:18:45 UTC

Puhhh, the fans are off all the time? Do you think it could be because the
machine relies on OS to manage fan control?
Is this machine in the house?

Comment 3 Matthias Boettger 2005-09-20 18:59:14 UTC

This is not a notebook. The fan is running. I think the fan control is managed 
by BIOS. 
I can give you one of the machines for a few days if you want ...

Comment 4 Thomas Renninger 2005-10-06 13:02:29 UTC

Sorry for the delay. Yes I can have a look at it...

Comment 5 Thomas Renninger 2005-10-14 09:16:40 UTC

Is already in kernel CVS, not sure if there is already a YOU update kernel and
whether the change is already included...

*** This bug has been marked as a duplicate of 116763 ***

Comment 6 Thomas Renninger 2005-12-08 16:53:21 UTC

I worked on default instead of smp kernel by accident, right?
Seems this is a ht problem.
acpi=ht probably works, but it's nearly the same as acpi=off.
This is not a duplicate of #116763 ...

Would be interesting whether recent ACPI fixes help (newest OpenSuse Alpha kernel/installation). Digging in old code wouldn't be too clever that short before SLES10.

Comment 7 Thomas Renninger 2005-12-22 20:23:09 UTC

I found a patch and added and it to 10.0 branch that went mainline in 2.6.15-rc5.
That one could solve your issue.
You could either send me acpidmp and I can have a look if it is the same bug or, even better, just try the next kotd from:
ftp.suse.com/pub/people/mantel/kotd/10.0-i386/SL100_BRANCH/kernel-smp-2.6.13*.rpm
It may take a while until it pops up there.

You also might want to have a look at #117177 which I expect to be a duplicate of this one. Hmmm, I just mark it as a duplicate ... still enough other bugs ... please reopen again if it does not work. The changelog (rpm -qp --changelog kernel-smp*.rpm |less) is:

Thu Dec 22 21:09:09 CET 2005 - trenn@suse.de
- patches.fixes/acpi_P4HT_same_processor_id_fix: Catches processor
  declarations with same acpi id (P4HT) (117177).

*** This bug has been marked as a duplicate of 117177 ***