lkml.org 
[lkml]   [2006]   [Feb]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRE: acpi_cpufreq broken after _PDC patch
Date
From

>-----Original Message-----
>From: Gerhard Schrenk [mailto:gerhard.schrenk@uni-konstanz.de]
>Sent: Saturday, February 04, 2006 9:43 AM
>To: Pallipadi, Venkatesh
>Cc: Gerhard Schrenk; linux-kernel@vger.kernel.org
>Subject: Re: acpi_cpufreq broken after _PDC patch
>
>* Pallipadi, Venkatesh <venkatesh.pallipadi@intel.com>
>[2006-02-04 16:32]:
>>
>> Please use Enhanced speedstep config option as well under Ppower
>> Management->cpufreq (with ACPI option under it enabled too).
>That will
>> configure speedstep-centrino driver, which should work here.
>> Speedstep-centrino driver is a better option than acpi-cpufreq, as it
>> will do much faster P-state transition.
>
>Thank you for your instantaneous reply! Do you mean?
>
>CONFIG_X86_SPEEDSTEP_CENTRINO=y
>CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE=y
>
>I *always* had those drivers statically compiled in my kernel
>along with
>
>CONFIG_X86_ACPI_CPUFREQ=m
>
>These settings always "worked" in the sense that I saw frequencies down
>to 800Mhz in /proc/cpuinfo and I could set different governors in
>/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor.
>
>Probably the speedstep-centrino driver actually never worked for me and
>only the acpi_cpufreq driver did it's job until your patch!?
>
>To give you a chance to reproduce this or to avoid missunderstandings I
>attach you my .config. I didn't touch my cpufreq settings since
>2.6.14.x.
>
>The acpi-tables of my MSI S260 (branded as Medion sim 2100) seem to be
>problematic. Here are my boot options
>


You need to configure X86_SPEEDSTEP_CENTRINO_ACPI as well. That is
required to make speedstep-centrino work with ACPI.

I understand that my change caused the change of behaviour here. But,
the code originally was wrong and causing crashes on few other systems.
Besides, as I mentioned earlier, using speedstep-centrino driver rather
than acpi-cpufreq is a better option.

If it doesn't work with this new config either, we will need to root
cause the failure of these two drivers with CPU_FREQ_DEBUG and also
possible acpidump.

Thanks,
Venki
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2006-02-04 19:09    [from the cache]
©2003-2011 Jasper Spaans