Re: SKL BOOT FAILURE unless idle=nomwait (was Re: PROBLEM: Cpufreq constantly keeps frequency at maximum on 4.5-rc4)

[Arto Jantunen <viiru@iki.fi>]

Len Brown <lenb@kernel.org> writes:

> On Wed, Mar 2, 2016 at 11:01 AM, Arto Jantunen <viiru@iki.fi> wrote:
>> Len Brown <lenb@kernel.org> writes:
>>> Does it happen if you use default cmdline,
>>> but edit intel_idle.c to comment out c8 and c9 support?
>>
>> That also boots, ts.out attached (as is the patch to do this, to make
>> sure there is no miscommunication).
>
> yes, patch looks good, and thanks for running turbostat.
>
> If your system boots and runs properly with "intel_idle.max_cstate=7",
> then you are suffering this issue:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=109081

That's the one, indeed. Based on the report this is actually a microcode
bug?

This is the version info that I see at boot:

microcode: CPU0 sig=0x506e3, pf=0x20, revision=0x33

The machine is an Asus GL552VW (reported as affected in the bug report),
with BIOS version 216, which is the latest.

>> With this change the original bug disappears as well.
>
> Hmm, so your problems with frequency staying constant,
> as well as the boot issue both went away by deleting c8 and c9?

Yes. I assume that removing idle=nomwait is the fix for the frequency
issue, while removing the deepest idle states fixes the boot issue.

> Still works with intel_idle.max_cstate=7 ?

I'll test this.

-- 
Arto Jantunen