Re: [PATCH] amd76x_pm: C3 powersaving for AMD K7

[Erik Slagter <erik@slagter.name>]

[-- Attachment #1: Type: text/plain, Size: 933 bytes --]

On Fri, 2006-02-03 at 10:33 +0100, Pavel Troller wrote:

> > Isn't there any other way to tell the operating system that the C2/C3
> > states are unusable (e.g. by simply telling they're not there?). This
> > looks so obfuscated...
> 
>   The problem is that the table (RSDT) has a fixed format and the values have
> reserved positions. Putting 0 there would mean zero (unmeasureble) latency,
> which would be the best case, of course. So to disable the feature, "out of
> range" method was selected. Yes, it would be nice to have some "special" value
> there, like FFFF, but it is not a common practice.

Still I don't get it. Current linux acpi implementation suggests there
are 8 C states. My laptop says the processor supports 4 C states, my
other laptop says it has 3 C states. Do you mean the table always has 8
C entries, and all of the unusable states are marked "unusable"? In that
case I would understand.

[-- Attachment #2: smime.p7s --]
[-- Type: application/x-pkcs7-signature, Size: 2771 bytes --]