Bug 43140 - boot fails in 3.0.0.12 (worked up to 2.6.38.8) , Pentium D processor - Intel D865PERL
Summary: boot fails in 3.0.0.12 (worked up to 2.6.38.8) , Pentium D processor - Intel ...
Status: CLOSED INSUFFICIENT_DATA
Alias: None
Product: ACPI
Classification: Unclassified
Component: Config-Other (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: acpi_config-other
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-04-21 04:08 UTC by Bill Murnane
Modified: 2013-06-03 07:02 UTC (History)
3 users (show)

See Also:
Kernel Version:
Subsystem:
Regression: No
Bisected commit-id:

Attachments
picture of log message (296.44 KB, image/jpeg)
2012-04-21 04:08 UTC, Bill Murnane 		Details
dmesg - 64000 (43.23 KB, text/plain)
2012-04-21 04:11 UTC, Bill Murnane 		Details
acpidump (79.86 KB, text/plain)
2012-04-21 04:12 UTC, Bill Murnane 		Details
lspci (11.54 KB, text/plain)
2012-04-21 04:13 UTC, Bill Murnane 		Details
Add an attachment (proposed patch, testcase, etc.)

Description Bill Murnane 2012-04-21 04:08:03 UTC 
Created attachment 73026 [details]
picture of log message

After upgrading to kernel 3.0.0.12 from 2.6.38.8 the kernel failed to start. I reported this to Unbuntu support via Launchpad, reference  number 878031, and there is a lot of diagnostic info there captured by their apport tool. 

This startup failure can be avoided by having acpi=off or nolapic

At the time kernel.bugzilla was unavailable so it was reported via email  on 2.6.38.8)to len.brown@intel.com whi I believed was the ACPI maintainer,
cc acpi-bugzilla@lists.sourceforge.net. with 8 attachments

subject was:
kernel fails to start regression in 3.0.0 unless nolapic specified, still in 3.1.0
Comment 1 Bill Murnane 2012-04-21 04:11:35 UTC 
Created attachment 73027 [details]
dmesg - 64000
Comment 2 Bill Murnane 2012-04-21 04:12:35 UTC 
Created attachment 73028 [details]
acpidump
Comment 3 Bill Murnane 2012-04-21 04:13:19 UTC 
Created attachment 73029 [details]
lspci
Comment 4 Bill Murnane 2012-04-21 04:21:02 UTC 
Still fails in most recent kernel I tried (ubuntu precise beta2)

The effect of acpi=off or nolapic is that only 1 of the 2 processors is detected.

I have only just discovered that kernel.bugzilla.org is back in operation, hence the new attempt to report this. I'm away next week so will reply to any questions or information requests then.
Comment 5 Len Brown 2013-02-08 18:18:17 UTC 
does it boot if you replace the workaround cmdline params
with maxcpus=1 ?

Can you reproduce using an upstream kernel that you built from scratch,
or have all the tests been with the ubuntu binary kernel?
Comment 6 Aaron Lu 2013-03-27 03:26:17 UTC 
Hi Bill,

Can you please try Len's suggestion in comment #5? Thanks.
Note You need to log in before you can comment on or make changes to this bug.