[SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Ducrot Bruno]

Can someone please test
ftp://ftp.poupinou.org/cpufreq/speedstep-ali.trunk.diff

Note that this is agaist 2.6.0-testX, not 2.4.


-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.


-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.

Re: [SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Ducrot Bruno]

On Thu, Nov 06, 2003 at 03:09:35PM +0100, Ducrot Bruno wrote:
> 
> Can someone please test
> ftp://ftp.poupinou.org/cpufreq/speedstep-ali.trunk.diff
> 
> Note that this is agaist 2.6.0-testX, not 2.4.
> 

OK.  It do not work.  Sorry.

What it does is to turn on geyserville functionality on the PMU device,
then, (I guess) we can use the geyserville IO register (at pmbase + 0x2f),
as for the -ich case.  Unfortunately, that is wrong.  I don't
know what to do next.  For people interrested in some registers that
have to be looked, get ftp://ftp.poupinou.org/dont_use/README.ali


-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.

Re: [SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Ducrot Bruno]

On Tue, Nov 11, 2003 at 08:32:00PM +0100, Ducrot Bruno wrote:
> On Thu, Nov 06, 2003 at 03:09:35PM +0100, Ducrot Bruno wrote:
> > 
> > Can someone please test
> > ftp://ftp.poupinou.org/cpufreq/speedstep-ali.trunk.diff
> > 
> > Note that this is agaist 2.6.0-testX, not 2.4.
> > 
> 
> OK.  It do not work.  Sorry.
> 
> What it does is to turn on geyserville functionality on the PMU device,
> then, (I guess) we can use the geyserville IO register (at pmbase + 0x2f),
> as for the -ich case.  Unfortunately, that is wrong.  I don't
> know what to do next.  For people interrested in some registers that
> have to be looked, get ftp://ftp.poupinou.org/dont_use/README.ali

Grumble.  Just found a stupid bug.  Forget that email.  I still
don't know if that work, though.  Sorry for noise.

-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.

Re: [SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Rafal Dejewski]

On Thursday 06 November 2003 15:09, Ducrot Bruno wrote:
> Can someone please test
> ftp://ftp.poupinou.org/cpufreq/speedstep-ali.trunk.diff
>
Hi!
Tried this on HP VT6200 notebook with P4M at 1700MHz running 2.6.0-test9 with 
APM (without ACPI).

Here is my dmesg output after "modprobe speedstep_ali":
cpufreq: read at pmbase 0x8000 + 0x2f returned 0x1
cpufreq: writing 0x1 to pmbase 0x8000 + 0x50
cpufreq: read at pmbase 0x8000 + 0x50 returned 0x1
cpufreq: change to 1700 MHz succeeded
cpufreq: read at pmbase 0x8000 + 0x2f returned 0x1
cpufreq: writing 0x0 to pmbase 0x8000 + 0x50
cpufreq: read at pmbase 0x8000 + 0x50 returned 0x0
cpufreq: change to 1700 MHz succeeded

It didn't create cpufreq directory on /sys/devices/system/cpu/cpu0 as 
p4-clockmod does, so I don't know what to do next...

Rafal

Re: [SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Ducrot Bruno]

On Thu, Nov 13, 2003 at 02:04:31PM +0100, Rafal Dejewski wrote:
> On Thursday 06 November 2003 15:09, Ducrot Bruno wrote:
> > Can someone please test
> > ftp://ftp.poupinou.org/cpufreq/speedstep-ali.trunk.diff
> >
> Hi!
> Tried this on HP VT6200 notebook with P4M at 1700MHz running 2.6.0-test9 with 
> APM (without ACPI).
> 
> Here is my dmesg output after "modprobe speedstep_ali":
> cpufreq: read at pmbase 0x8000 + 0x2f returned 0x1
> cpufreq: writing 0x1 to pmbase 0x8000 + 0x50
> cpufreq: read at pmbase 0x8000 + 0x50 returned 0x1
> cpufreq: change to 1700 MHz succeeded
> cpufreq: read at pmbase 0x8000 + 0x2f returned 0x1
> cpufreq: writing 0x0 to pmbase 0x8000 + 0x50
> cpufreq: read at pmbase 0x8000 + 0x50 returned 0x0
> cpufreq: change to 1700 MHz succeeded
> 
> It didn't create cpufreq directory on /sys/devices/system/cpu/cpu0 as 
> p4-clockmod does, so I don't know what to do next...
> 

Can you please send me the output of lspci -xxx, thanks,

-- 
Ducrot Bruno

--  Which is worse:  ignorance or apathy?
--  Don't know.  Don't care.

Re: [SPEEDSTEP] ALi shouthbridge. alpha tester wanted.

[Rafal Dejewski]

[-- Attachment #1: Type: text/plain, Size: 373 bytes --]

On Monday 17 November 2003 20:39, Ducrot Bruno wrote:
>
> Can you please send me the output of lspci -xxx, thanks,
Lspci attached.
Don't know if this helps: when I commented out line with "return result;" from 
speedstep_cpu_init function from speedstep-ali.c I did get cpufreq directory 
in /sys. Both scalling_min_freq and scaling_max_freq where equal to 1700000.

Rafal

[-- Attachment #2: lspci_hp_vt6200.gz --]
[-- Type: application/x-gzip, Size: 1776 bytes --]

[-- Attachment #3: Type: text/plain, Size: 143 bytes --]

_______________________________________________
Cpufreq mailing list
Cpufreq@www.linux.org.uk
http://www.linux.org.uk/mailman/listinfo/cpufreq