cpufreq and suspend

cpufreq and suspend

[lucke]

Dear cpufreq gurus,

I would be grateful if you could have a look at:
http://lists.suspend2.net/lurker/message/20060827.234345.486617dc.en.html

Forgive me if this issue has been already mentioned, I can't seem to find a 
way to search through archives.

Regards,

Luke

RE: cpufreq and suspend

[Langsdorf, Mark]

> I would be grateful if you could have a look at:
>
http://lists.suspend2.net/lurker/message/20060827.234345.486617dc.en.htm
l
> 
> Forgive me if this issue has been already mentioned, I can't 
> seem to find a way to search through archives.

I'll go take a look at this.

-Mark Langsdorf
AMD, Inc.

RE: cpufreq and suspend

[lucke]

>> I would be grateful if you could have a look at:
>>
>>http://lists.suspend2.net/lurker/message/20060827.234345.486617dc.en.html
>> 
>> Forgive me if this issue has been already mentioned, I can't 
>> seem to find a way to search through archives.

>I'll go take a look at this.
>
>-Mark Langsdorf
>AMD, Inc. 

Is there any hope for a solution to this problem, Mark et al?

Regards,

Luke

RE: cpufreq and suspend

[Langsdorf, Mark]

> >>http://lists.suspend2.net/lurker/message/20060827.234345.486
> 617dc.en.html
> >> 
> >> Forgive me if this issue has been already mentioned, I can't 
> >> seem to find a way to search through archives.
> 
> >I'll go take a look at this.
> 
> Is there any hope for a solution to this problem, Mark et al?

It's been a busy week, and I just secured the system necessary to
do the testing.

Currently, I cannot reproduce with OpenSUSE 10.1 and the powersave
tool.  The system goes into S3 and comes out fine.  Both cores are
available and seem to be switching.

I'll set up suspend2 this afternoon and try again.

-Mark Langsdorf
AMD, Inc.

Re: cpufreq and suspend

[lucke]

On Wednesday 20 September 2006 20:53, you wrote:
> > >>http://lists.suspend2.net/lurker/message/20060827.234345.486
> >
> > 617dc.en.html
> >
> > >> Forgive me if this issue has been already mentioned, I can't
> > >> seem to find a way to search through archives.
> > >
> > >I'll go take a look at this.
> >
> > Is there any hope for a solution to this problem, Mark et al?
>
> It's been a busy week, and I just secured the system necessary to
> do the testing.
>
> Currently, I cannot reproduce with OpenSUSE 10.1 and the powersave
> tool.  The system goes into S3 and comes out fine.  Both cores are
> available and seem to be switching.
>
> I'll set up suspend2 this afternoon and try again.

Well, S3 doesn't work for me (doesn't want to wake up), thus I can't check how 
cpufreq behaves with it. The problem occurs with suspend to disk (just 
checked, not only with suspend2, but also with swsusp/powersave) - quite 
logically it's where the problem could occur, as it has to go through BIOS 
and kernel initialization phases again (I suppose suspend to RAM doesn't go 
through that). I should've mentioned it earlier.

For completeness, MB is Asus A8N-E with latest BIOS.

I wish you a fruitful hunting,

Luke

RE: cpufreq and suspend

[Langsdorf, Mark]

> > It's been a busy week, and I just secured the system necessary to
> > do the testing.
> >
> > Currently, I cannot reproduce with OpenSUSE 10.1 and the powersave
> > tool.  The system goes into S3 and comes out fine.  Both cores are
> > available and seem to be switching.
> >
> > I'll set up suspend2 this afternoon and try again.
> 
> Well, S3 doesn't work for me (doesn't want to wake up), thus 
> I can't check how cpufreq behaves with it.

> The problem occurs with suspend to disk (just 
> checked, not only with suspend2, but also with 
> swsusp/powersave)

I've tested "powersave -U" on a 4P/4C AMD internal reference
design and on an HP Turion X2 laptop, both using the 1.60.2
driver in the SUSE 2.6.16.13 kernel.  Both systems go into
S4 properly and come out properly, and PowerNow! has entries
for each core before and after suspend/resume.

This may be a MB specific problem.
 
> For completeness, MB is Asus A8N-E with latest BIOS.

Okay, I'll go try to dig one of those up.

-Mark Langsdorf
AMD, Inc.

Re: cpufreq and suspend

[lucke]

>> For completeness, MB is Asus A8N-E with latest BIOS.

>Okay, I'll go try to dig one of those up.

Any success in obtaining this MB, Mark?

If you fail to get one, is there perhaps anything I could do to help you debug 
this issue?

Regards,

Luke