RE: [2.5. PATCH] cpufreq: correct initialization on Intel Copperm ines

RE: [2.5. PATCH] cpufreq: correct initialization on Intel Copperm ines

[Perez-Gonzalez, Inaky]

> [2.5. PATCH] cpufreq: Intel Coppermines -- the saga continues.
> 
> The detection process for speedstep-enabled Pentium III Coppermines is
> considered proprietary by Intel. The attempt to detect this
> capability using MSRs failed. So, users need to pass the option
> "speedstep_coppermine=1" to the kernel (boot option or parameter) if
> they own a SpeedStep capable PIII Coppermine processor. Tualatins work
> as before.

Cannot you use ACPI to detect that? AFAIK, if the machine supports it, it is
doable.

Cheers, 

Inaky Perez-Gonzalez -- Not speaking for Intel - opinions are my own [or my
fault]

Re: [2.5. PATCH] cpufreq: correct initialization on Intel Copperm ines

[Dominik Brodowski]

Hi,

On Sat, Nov 09, 2002 at 04:54:27PM -0800, Perez-Gonzalez, Inaky wrote:
> > [2.5. PATCH] cpufreq: Intel Coppermines -- the saga continues.
> > 
> > The detection process for speedstep-enabled Pentium III Coppermines is
> > considered proprietary by Intel. The attempt to detect this
> > capability using MSRs failed. So, users need to pass the option
> > "speedstep_coppermine=1" to the kernel (boot option or parameter) if
> > they own a SpeedStep capable PIII Coppermine processor. Tualatins work
> > as before.
> 
> Cannot you use ACPI to detect that? AFAIK, if the machine supports it, it is
> doable.

Most PIII Coppermine notebooks only have ACPI 1.x which does not include
"Performance States" - an interface to Intel SpeedStep and other, similar 
technologies. And when there is a working "Performance States" support in
ACPI, this itself registered as cpufreq driver and then there's no need for 
the legacy speedstep driver. So, no, unfortunately your suggestion doesn't 
solve the problem.

	Dominik

RE: [2.5. PATCH] cpufreq: correct initialization on Intel Copperm ines

[Perez-Gonzalez, Inaky]

> > Cannot you use ACPI to detect that? AFAIK, if the machine 
> > supports it, it is doable.
> 
> Most PIII Coppermine notebooks only have ACPI 1.x which does 
> not include
> "Performance States" - an interface to Intel SpeedStep and 
> other, similar 
> technologies. 

Bummer - that was something I did not know; thanks for your clarification -
this was the only sollution I found (without IP concerns) when I was
enquiring how to get the specifications to support performance states.

Thanks,

Inaky Perez-Gonzalez -- Not speaking for Intel - opinions are my own [or my
fault]