Re: CX > 1 support for Athlon SMP systems

Re: CX > 1 support for Athlon SMP systems

[Carl Thompson]

Wow.  This actually seems to work on my Sony laptops.  But I wonder if 
it is really puting the CPU in C2 because there is no additional power 
savings or temperature reduction than using ACPI.  Is there any way to 
test if CPU is going in C2 state? 
 
At any rate, since the power savings is comparable to ACPI and the 
only reason I was using ACPI on these laptops was for the power 
savings, I have switched to using lvcool + APM instead of ACPI.  
Battery life seems about the same, and it eliminates the issues I was 
having with ACPI (CPU usage reading battery status, dropped 
characters, USB weirdness, undocumented inteface, etc). 
 
BTW, APM uses about 1/50th to 1/100th the processor time reading 
battery status on my laptop compared to using ACPI.  No joke. 
 
Carl Thompson 
 
 
Quoting Johnathan Hicks <thetech-Pn2JTpibVOrk1uMJSBkQmQ@public.gmane.org>: 
 
> Carl Thompson wrote: 
> > Any idea how to get C2 on a UP Via Athlon board?  
> >   
> > Thanks,  
> > Carl Thompson  
>  
> Tony mentions VCool and LVCool in his comments so you could try 
> investigate  
> those. Looks like LVCool is the linux version of VCool. If that 
> doesn't work  
> for you, you might be able to grab the data sheets for your chipset 
> and modify  
> some code to make things work. ;-) 
>  
> --John 
>  
>  
>  
>  
> ------------------------------------------------------- 
> This sf.net email is sponsored by:ThinkGeek 
> Welcome to geek heaven. 
> http://thinkgeek.com/sf 
> _______________________________________________ 
> Acpi-devel mailing list 
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org 
> https://lists.sourceforge.net/lists/listinfo/acpi-devel 
>  
>  
 
 


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: CX > 1 support for Athlon SMP systems

[Johnathan Hicks]

Carl Thompson wrote:
> Wow.  This actually seems to work on my Sony laptops.  But I wonder if 
> it is really puting the CPU in C2 because there is no additional power 
> savings or temperature reduction than using ACPI.  Is there any way to 
> test if CPU is going in C2 state? 

I can't think of anything at the hardware level, but you do a printk or 
something inside the while loop that contains the inb() in the idle loop.

> At any rate, since the power savings is comparable to ACPI and the 
> only reason I was using ACPI on these laptops was for the power 
> savings, I have switched to using lvcool + APM instead of ACPI.  
> Battery life seems about the same, and it eliminates the issues I was 
> having with ACPI (CPU usage reading battery status, dropped 
> characters, USB weirdness, undocumented inteface, etc). 

ACPI will likely be fixed eventually... ;-)

> BTW, APM uses about 1/50th to 1/100th the processor time reading 
> battery status on my laptop compared to using ACPI.  No joke. 

Yikes! That's no good, when under CPU load anyway...

--John




-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: CX > 1 support for Athlon SMP systems

[Carl Thompson]

Quoting Johnathan Hicks <thetech-Pn2JTpibVOrk1uMJSBkQmQ@public.gmane.org>: 
 
> Carl Thompson wrote: 
> > Wow.  This actually seems to work on my Sony laptops.  But I 
> > wonder if it is really puting the CPU in C2 because there is no 
> > additional power savings or temperature reduction than using ACPI.  
> > Is there any way to test if CPU is going in C2 state?  
>  
> I can't think of anything at the hardware level, but you do a printk 
> or something inside the while loop that contains the inb() in the 
> idle loop. 
 
What I mean is I know the loop is executing, but how do I know the 
processor is _really_ in C2?  I was expecting a significant change in 
batterly life and CPU temp vs. the C1 state that ACPI put my CPU in.  
I haven't run definitive tests, but at best I'm only at most 15-20 
minutes more battery life (if any at all). 
 
> ... 
 
> --John 
 
Carl Thompson 
 


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: CX > 1 support for Athlon SMP systems

[Johnathan Hicks]

Carl Thompson wrote:
<SNIP>
> What I mean is I know the loop is executing, but how do I know the 
> processor is _really_ in C2?  I was expecting a significant change in 
> batterly life and CPU temp vs. the C1 state that ACPI put my CPU in.  
> I haven't run definitive tests, but at best I'm only at most 15-20 
> minutes more battery life (if any at all). 

What I would do is grab the data sheet(s) for the chipset in the laptop and 
compare what the code is doing with what you see in the docs. That's about as 
far as I can help you here, I'm rather busy unfortunately. :-(

--John




-------------------------------------------------------
This sf.net email is sponsored by: Jabber - The world's fastest growing 
real-time communications platform! Don't just IM. Build it in! 
http://www.jabber.com/osdn/xim