Re: found unsupported CPU with Enhanced SpeedStep

On Thu, Dec 30, 2004 at 02:38:14PM -0800, Jeremy Fitzhardinge wrote:
> On Wed, 2004-12-29 at 20:40 -0500, Matthew Harrell wrote:
> > Hi.  I get the following message so I figured I would send you an email
> > 
> >   speedstep-centrino: invalid ACPI data
> >   speedstep-centrino: found unsupported CPU with Enhanced SpeedStep: send /proc/cpuinfo to Jeremy
Fitzhardinge <jeremy <at> goop.org>
> > 
> > The cpu info file is attached.  Is there a simple change I can do to get this
> > to work on this laptop?
> 
> The "invalid ACPI data" suggests that a BIOS update might help.
> Unfortunately the driver can't do anything for an EST-enabled mobile P4
> without the ACPI data.
> 
> I don't know too much about ACPI, so I've cc:'d this to the cpufreq
> mailing list.

Can you make the disassembled DSDT (see http://acpi.sourceforge.net for
details and/or http://forums.gentoo.org/viewtopic.php?t=122145 for a howto;
proceed until step "7. Diagnosing a Buggy DSDT - Disassemble the DSDT"
please) as well as a full dmesg output available on the 'net somewhere, please?
If that's not possible, please send these two files off-list to me. 

Thanks,
	Dominik

[ANNOUNCE] cpufrequtils-0.2-pre1 released

cpufrequtils-0.2-pre1 is available at

http://kernel.org/pub/linux/utils/kernel/cpufreq/

Changes since cpufrequtils-0.1 include:
- content for the README file
- direct centrino-decode by accessing MSRs (Bruno Ducrot)
- building outside of srcdir (Mike Castle)
- kHz/MHz/GHz conversion cleanup (Harald Mild)

For more information about cpufrequtils, please visit
http://kernel.org/pub/linux/utils/kernel/cpufreq/cpufrequtils.html

	Dominik Brodowski

_______________________________________________
Cpufreq mailing list
Cpufreq <at> www.linux.org.uk
http://www.linux.org.uk/mailman/listinfo/cpufreq

"CPU frequency out of sync" error

I'm trying to get cpu frequency scaling to work with a desktop system using 
a Mobile Athlon (XP2200+) but so far I'm not having much luck.  Incase it 
helps, the motherboard being used is a Soltek 75frv which has a KT400 
chipset.

The bios didn't have the relevant PST table for the cpu so I have patched 
powernow-k7 with a valid table (which I found on the internet.)  After 
patching this, if I try to execute powernowd I get numerous errors in the 
form of "Warning: CPU frequency is 1795, cpufreq assumed 1662 kHz" where the 
first number is always 1795 and the second number decreases as cpufreq tries 
to clock down through the frequency steps defined in my custom PST.  After 
going down 4 "steps" the error message then changes to "Warning: CPU 
frequency out of sync: cpufreq and timing core thinks of 798, is 1795 kHz."
The above happens when using ACPI or APM.

If I try to use cpufreqd instead of powernowd, no errors are thrown up but 
I'm not convinced it is working.  cat /proc/cpuinfo reports a frequency of 
532MHz (which is the lowest step I've defined.)  However, running x86info -a 
reports:
Voltage ID codes: Maximum=1.450V Startup=1.450V Currently=1.450V
Frequency ID codes: Maximum=13.5x Startup=6.0x Currently=13.5x
Therefore, there seems to be conflicting information on whether the 
processor has been downclocked or not.  (Note that although the startup 
multiplier is defined as 6.0x, it seems that as my bios is made by AMI, it 
actually writes the maximum multiplier (rather than the startup multiplier) 
to the FID MSR on boot up and thus the startup multiplier is really 13.5x.)
I should also mention that similar behaviour was shown with /proc/cpuinfo 
and x86info when I tried powernowd (ie. x86info showing constant multiplier 
of 13.5 which cpuinfo showed the expected, downclocked frequency.)

cpufreq for older Toshibas with P3-M and ALi chipset?

Greetings.

I have for quite some time used Peter T Breuer's 'cpuswitch' [1] utility
to change the processor frequency on my Toshiba Satellite Pro 6000 [3]. 
This utility makes use of HCI/SCI to do its work, and must(?) be built 
together with Jonathan Buzzard's toshiba-utils. [2]
It gives me 8 different CPU frequencies to choose from. (Though the 
frequency in /proc/cpuinfo is never updated.)

I'd like to make use of cpufreq with the 'ondemand' governor instead. 
The laptop appears to work with the 'ACPI Processor P-States driver', 
but this only gives me two frequencies to choose between. Is this to be 
expected?

None of the cpufreq processor drivers appears to fit the bill for a 
PIII-M on an ALi chipset, so I wonder if I am stuck with the cpuswitch 
utility if I want the better granularity it offers over the ACPI driver?
Is it at all possible to fit the functions 'cpuswitch' makes use of into 
a cpufreq driver?

BTW:
I do notice that if I fiddle with the BIOS settings, I can make the 
laptop boot at "half speed", coming up at 500 MHz instead of 1000. (As 
displayed in /proc/cpuinfo.) And the P-states driver still works, so 
presumably it is possible to combine 'native support' and P-states to 
obtain 16 steps? 

Thank you to Jonathan, Peter, Dominik, Len and all the others who 
contribute to making laptops useable under Linux.

Re: cpufreq for older Toshibas with P3-M and ALi chipset?

Hi,

On Sun, Jan 02, 2005 at 05:24:47PM +0100, Dag Bakke wrote:
> I have for quite some time used Peter T Breuer's 'cpuswitch' [1] utility
> to change the processor frequency on my Toshiba Satellite Pro 6000 [3]. 
> This utility makes use of HCI/SCI to do its work, and must(?) be built 
> together with Jonathan Buzzard's toshiba-utils. [2]
> It gives me 8 different CPU frequencies to choose from. (Though the 
> frequency in /proc/cpuinfo is never updated.)
> 
> I'd like to make use of cpufreq with the 'ondemand' governor instead. 
> The laptop appears to work with the 'ACPI Processor P-States driver', 
> but this only gives me two frequencies to choose between. Is this to be 
> expected?

Yes, it is. Mobile Pentium 3 CPU only support two frequency scaling states.
The 8 states you're seeing using cpuswitch are throttling states -- i.e. the
CPU is stopped for short periods of time. However, during these periods the
voltag isn't scaled, so it only leads to a linear reduction of power
compared to a ~ ^3 reduction if voltage is reduced as well. Furthermore, the
state the CPU is in when stopped equals the state the CPU is put into when
there is no work to do -- all this leads to the fact that throttling is
useless in most cases.

> None of the cpufreq processor drivers appears to fit the bill for a 
> PIII-M on an ALi chipset, so I wonder if I am stuck with the cpuswitch 
> utility if I want the better granularity it offers over the ACPI driver?
> Is it at all possible to fit the functions 'cpuswitch' makes use of into 
> a cpufreq driver?

Re: cpufreq for older Toshibas with P3-M and ALi chipset?

Dominik Brodowski wrote:

>Hi,
>
>On Sun, Jan 02, 2005 at 05:24:47PM +0100, Dag Bakke wrote:
>  
>
[snip]

>>The laptop appears to work with the 'ACPI Processor P-States driver', 
>>but this only gives me two frequencies to choose between. Is this to be 
>>expected?
>>    
>>
>
>Yes, it is. Mobile Pentium 3 CPU only support two frequency scaling states.
>The 8 states you're seeing using cpuswitch are throttling states -- i.e. the
>CPU is stopped for short periods of time. However, during these periods the
>voltag isn't scaled, so it only leads to a linear reduction of power
>compared to a ~ ^3 reduction if voltage is reduced as well. Furthermore, the
>state the CPU is in when stopped equals the state the CPU is put into when
>there is no work to do -- all this leads to the fact that throttling is
>useless in most cases.
>  
>
I'll test and see how much the drain from the battery changes when using 
cpuswitch. And compare it with the two P-states. Should have done that 
anyway.

>>None of the cpufreq processor drivers appears to fit the bill for a 

Re: cpufreq for older Toshibas with P3-M and ALi chipset?

Dominik Brodowski wrote:

>Hi,
>
>On Sun, Jan 02, 2005 at 05:24:47PM +0100, Dag Bakke wrote:
>  
>
[snip]

>Yes, it is. Mobile Pentium 3 CPU only support two frequency scaling states.
>The 8 states you're seeing using cpuswitch are throttling states -- i.e. the
>CPU is stopped for short periods of time. However, during these periods the
>voltag isn't scaled, so it only leads to a linear reduction of power
>compared to a ~ ^3 reduction if voltage is reduced as well. Furthermore, the
>state the CPU is in when stopped equals the state the CPU is put into when
>there is no work to do -- all this leads to the fact that throttling is
>useless in most cases.
>  
>
[snip]
OK... but this bit....

>>BTW:
>>I do notice that if I fiddle with the BIOS settings, I can make the 
>>laptop boot at "half speed", coming up at 500 MHz instead of 1000. (As 
>>displayed in /proc/cpuinfo.) And the P-states driver still works, so 
>>presumably it is possible to combine 'native support' and P-states to 
>>obtain 16 steps? 
>>    
>>

Re: cpufreq for older Toshibas with P3-M and ALi chipset?

On Mon, Jan 03, 2005 at 12:34:59AM +0100, Dag Bakke wrote:
> Well, I believe it is just the same function as cpuswitch exploits, 
> because cpuswitch reports the cpu to be running at 'level 4' when I make 
> it boot at 500Mhz.

Then the "half speed" method doesn't work using the FSB but by throttling
the CPU.

	Dominik

[PATCH]gx-suspmod: fix to return proper value when reading /sys/.../cpuinfo_cur_freq

This is a patch to fix to return proper value when reading /sys/.../cpuinfo_cur_freq
with gx-suspmod driver for Cyrix MediaGX / AMD Geode CPUs.

The patch is generated by bkexport script.

Hiroshi Miura  --- http://www.da-cha.org/  --- miura <at> da-cha.org
NTTDATA Corp. OpenSource Software Center. --- miurahr <at> nttdata.co.jp 
NTTDATA Intellilink Corp. OpenSource Engineering Dev. -- miurahr <at> intellilink.co.jp
Key fingerprint = 9117 9407 5684 FBF1 4063  15B4 401D D077 04AB 8617

_______________________________________________
Cpufreq mailing list
Cpufreq <at> www.linux.org.uk
http://www.linux.org.uk/mailman/listinfo/cpufreq

frequency switching quirks with powernow-k8

I am trying to use 2.6.10's cpufreq on my AMD64 3200+ (S939), Asus A8V deluxe 
motherboard, Nvidia 6800GL. The first couple of hours it seems to work just 
fine, but after some time (~2h), I am getting the following errors repeatedly 
in  dmesg (about 2 lines / 12 sec). 

ignoring illegal change in lo freq table-2 to 0x2
powernow-k8: transition frequency failed
ignoring illegal change in lo freq table-2 to 0x2
powernow-k8: transition frequency failed
ignoring illegal change in lo freq table-2 to 0x2
powernow-k8: transition frequency failed
ignoring illegal change in lo freq table-2 to 0x2
powernow-k8: transition frequency failed
ignoring illegal change in lo freq table-2 to 0x2
powernow-k8: transition frequency failed
powernow-k8: vid trans failed, vid 0x3, curr 0x4
powernow-k8: transition frequency failed
powernow-k8: vid trans failed, vid 0x3, curr 0x4
powernow-k8: transition frequency failed
powernow-k8: vid trans failed, vid 0x3, curr 0x4
powernow-k8: transition frequency failed
ignoring illegal change in lo freq table-2 to 0x2

It seems that the machine can't switch frequency anymore and it starts to 
behave somehow a little sluggish. In some case, it's not any more possible to 
switch frequency manually. (echo scaling_governor or alike). I have tried  
both, powernowd and the ondemand governor, both result in the same problem. 
Attached is the output of dmesg, cat /proc/acpi/processor/CPU1/*, lspci -v   
and /sys/devices/system/cpu/cpu0/cpufreq/* and of course my .config. If you'd 
like to see more information, please let me know.