Bug 481455

Summary:	ppc32 can't boot with newer rawhide kernels
Product:	[Fedora] Fedora	Reporter:	Kevin Fenzi <kevin>
Component:	kernel	Assignee:	Kernel Maintainer List <kernel-maint>
Status:	CLOSED UPSTREAM	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	low
Version:	rawhide	CC:	jarod, kernel-maint, quintela, tony
Target Milestone:	---
Target Release:	---
Hardware:	powerpc
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2009-02-12 22:12:13 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Kevin Fenzi 2009-01-25 03:30:21 UTC

rawhide kernels including kernel-2.6.29-0.28.rc1.fc11.ppc and after don't seem to allow a normal boot on my ppc32 g4 machine.

The kernel loads and boots, but when it gets to: 

Setting up Logical Volume Management:

it locks completely. 
Adding in a strace I see it's stuck in a read of /dev/cdrom1
However, this machine only has a single cdrom, which on working kernels is just /dev/cdrom. 

Can anyone duplicate this?
Is there further info I can provide?

Smolt profile: http://www.smolts.org/client/show/pub_f3216f28-fbb5-4888-af2e-0d3f0eb3a55c

Comment 1 Jarod Wilson 2009-01-29 19:59:44 UTC

So... kernel-2.6.29-0.69.rc3.fc11.ppc boots just fine on my powerbook, which has a fully updated F10 + the aforementioned rawhide kernel on it. Could be some bad interaction between the rawhide kernel and rawhide lvm though, guess I should go raw-er on my powerbook...

Comment 2 Jarod Wilson 2009-01-29 21:32:26 UTC

Whoops, meant to add myself to the cc list... I've got a sacrificial G4 tower in the office I'll try to reproduce on, hopefully tomorrow...

Comment 3 Kevin Fenzi 2009-02-12 04:00:13 UTC

Strange. 

Just rebooted into 2.6.29-0.99.rc4.git1.fc11.ppc and it worked. 

All I can figure is that it was some unspecified rawhide breakage that got fixed. ;( 

Should we try and track it down more? or just be thankfull and move on?
Either is fine with me.

Comment 4 Jarod Wilson 2009-02-12 22:01:21 UTC

Turns out there was a ppc32-smp-specific bug biting me, in addition to some apparent video hardware related issues tripping up plymouth... But I finally have a rawhide kernel booting on my dual 800 G4 tower, after adding a patch jwb pointed me at:

http://patchwork.ozlabs.org/patch/22887/

Going to tack that onto rawhide in a sec, but I suspect it'll get into linus' tree RSN.

Comment 5 Jarod Wilson 2009-02-12 22:03:00 UTC

Never mind, its already in linus' tree, we'll just pick it up from the next git snap rebase.

Comment 6 Jarod Wilson 2009-02-12 22:12:13 UTC

So yeah, lets just chalk this up to development kernel churn and close it for now...