proper enabling of AMD K7 power saving

proper enabling of AMD K7 power saving

[Malte Gell]

Hi,

i have an AMD Duron 750 desktop computer with Via KT133a chipset and to 
get some noticeable temperature decrease I need to invoke a command 
from the Athlon Powersaving Howto for this chipset
(http://www.tldp.org/HOWTO/Athlon-Powersaving-HOWTO/
approaches.html#commandline )

Is it right that it is the fault of the mainboard manufacturer if one 
has to invoke such a command to get C2-level for power saving?

Will Linux-ACPI one day address this problem so a user doesn't need to 
invoke something like "setpci -v -H1 -s 0:0.0 52=$(printf %x $((0x
$(setpci -H1 -s 0:0.0 52) | 0x80)))" to enable proper power saving?

My fear is just one day I buy a new mainboard and there's no appropriate 
setpci-command on the Powersaving Howto and Linux-ACPI doesn't enable 
C2-powersaving if the mainboard doesn't do it first... 

Why the f*** do mainboard makers not offer a BIOS switch to enable this? 
Besides the lack of out-of-the-box C2-level ACPI just runs fine. I 
guess if new generation mainboards (KT400 and up) would better handle 
K7 power saving we wouldn't need the Powersaving Howto, right?

regards
Malte

(SuSE 9.0,Linux 2.4.21+ACPI 20030619 on Elitegroup K7VZA Rev3.0 + AMD 
Duron 750 (old spitfire core))



-------------------------------------------------------
This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND

Re: proper enabling of AMD K7 power saving

[Florian Hühn]

On Mon, 14 Jun 2004 23:11:53 +0200
Malte Gell <malte.gell-Mmb7MZpHnFY@public.gmane.org> wrote:

> Hi,
> 
> i have an AMD Duron 750 desktop computer with Via KT133a chipset and to 
> get some noticeable temperature decrease I need to invoke a command 
> from the Athlon Powersaving Howto for this chipset
> (http://www.tldp.org/HOWTO/Athlon-Powersaving-HOWTO/
> approaches.html#commandline )
> 
> Is it right that it is the fault of the mainboard manufacturer if one 
> has to invoke such a command to get C2-level for power saving?

Yes.

> 
> Will Linux-ACPI one day address this problem so a user doesn't need to 
> invoke something like "setpci -v -H1 -s 0:0.0 52=$(printf %x $((0x
> $(setpci -H1 -s 0:0.0 52) | 0x80)))" to enable proper power saving?

Have a look at http://vcool.occludo.net/VC_Linux.html or try the kernelpatch
on the page you mentioned. It runs great for several years now on my desktop
pc. On http://vcool.occludo.net/VC_Theory.html there is a short explanation
how it works and what it is.

> 
> My fear is just one day I buy a new mainboard and there's no appropriate 
> setpci-command on the Powersaving Howto and Linux-ACPI doesn't enable 
> C2-powersaving if the mainboard doesn't do it first... 

Don't panic. The address of the register in the north/southbridge will always
be known. At least i hope so ;)

> 
> Why the f*** do mainboard makers not offer a BIOS switch to enable this?

Some ppl say it's because some early athlos had problems with powersaving.
In fact i have never seen or heared about such a system. (I by myself tested
about 20 or 25 computers, all worked as expected.) I really can't imagine why
they don't enable it by default like all the Laptop-BIOSes does.

> Besides the lack of out-of-the-box C2-level ACPI just runs fine. I 
> guess if new generation mainboards (KT400 and up) would better handle 
> K7 power saving we wouldn't need the Powersaving Howto, right?
> 
> regards
> Malte
> 
> (SuSE 9.0,Linux 2.4.21+ACPI 20030619 on Elitegroup K7VZA Rev3.0 + AMD 
> Duron 750 (old spitfire core))

Sorry for my bad english.

PS: eh... gmx.DE? Warum zum Teufel schreib ich das auf englisch?! Damn.


-------------------------------------------------------
This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND

Re: proper enabling of AMD K7 power saving

[Malte Gell]

On Tuesday 15 June 2004 00:37, Florian Hühn wrote:
> On Mon, 14 Jun 2004 23:11:53 +0200
>
> Malte Gell <malte.gell-Mmb7MZpHnFY@public.gmane.org> wrote:

> Have a look at http://vcool.occludo.net/VC_Linux.html or try the
> kernelpatch on the page you mentioned. It runs great for several
> years now on my desktop pc. On
> http://vcool.occludo.net/VC_Theory.html there is a short explanation
> how it works and what it is.
It seems VCool just does the same what the "hack" from the Athlon 
Powersaving Howto does. Maybe a list would be fine that shows all 
mainboards which do not allow C2 or more and need such a tweak, so 
users can see what they need to get a cooler CPU. And hopefully it puts 
some pressure on the mobo makers :-)

> > Why the f*** do mainboard makers not offer a BIOS switch to enable
> > this?
>
> Some ppl say it's because some early athlos had problems with
> powersaving. In fact i have never seen or heared about such a system.
The K7 with a 1/2 multiplier are said to have this problem. But this 
problem seem to be solved long ago, my Duron is a 750 Mhz and I have 
absolutely no stability problems since I started using C2 some days 
ago.

> Sorry for my bad english.
> PS: eh... gmx.DE? Warum zum Teufel schreib ich das auf englisch?!
> Damn.
Because wir sind in a international Mailingliste hier ;-)

Malte



-------------------------------------------------------
This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference
Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer
Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA
REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND