Bug 39032

Summary:	Causes MCE with CPUIDLE driver enabled - Core i7 920
Product:	Power Management	Reporter:	frith.harry
Component:	cpuidle	Assignee:	Len Brown (lenb)
Status:	CLOSED INSUFFICIENT_DATA
Severity:	blocking	CC:	acpi-bugzilla, rui.zhang
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	2.6.39.3	Subsystem:
Regression:	No	Bisected commit-id:

Description frith.harry 2011-07-09 09:54:38 UTC

Hi all,

When the CPUIDLE driver is enabled, my computer fails to boot with the error "Fatal machine check - not syncing", on my Core i7 920. Unfortunately I don't have any technical dumps to hand but I guess I could boot up a live cd and write it all down if neccessary

Thanks a lot

Comment 1 frith.harry 2011-07-11 09:33:16 UTC

Any progress?

Comment 2 Len Brown 2011-07-16 23:16:41 UTC

please enable CONFIG_CPU_IDLE=y again.

boot with both command line params
intel_idle.max_cstate=0
processor.max_cstate=1

and capture the complete dmesg

increase N in processor.max_cstate=N
until failure or until you stop seeing a max_cstate
message in dmesg (probably at 4).

then increase N in intel_idle.max_cstate=N
from 1 onward until no max_cstate message.
capture the dmesg for the last one that works
and report the 1st one to fail.

Comment 3 Len Brown 2011-07-31 21:14:03 UTC

What kernel is this reported against?
Is it still a problem with v3.0?

Comment 4 frith.harry 2011-08-01 09:22:08 UTC

Hi, I'm very sorry I forgot all about this... I'm not entirely sure if it's still a problem with v3.0, my dilemma is, I can't even get a Linux distro to install... I'll burn my distro I made a while ago with Suse sources and try what you said

Comment 5 Zhang Rui 2012-01-18 05:25:47 UTC

It's great that the kernel bugzilla is back.

Can you please verify if the problem still exists in the latest upstream
kernel?

Comment 6 Zhang Rui 2012-05-24 08:01:10 UTC

bug closed as there is no response from the bug reporter.
please feel free to reopen it if the problem still exists in the latest upstream kernel.