Some questions about ACPI and laptops

Some questions about ACPI and laptops

[Hanno Böck]

I'm currently running a 2.4.18-kernel with newest acpi and the nocorruption-2-patch.

I have some questions:
- Most time I use my laptop simply to work, i don't need the full power of my system. Is it possible to slowdown my pc to make the battery last longer?
- What else can I do to let my laptop work longer?
- Currently I don't have any suspend mode. Only thing is that my monitor is black after some time, but it still gives some light. Can I configure my system that it will switch the monitor completely off and also stop the harddisk?

Another thing is that I still get errors like
* Error: Could not install Pci_config handler for PCI0,AE_ALREADY_EXISTS
on boot and on halt.
Has anyone found out why they are there? (They have been posted some times in the past, mostly on Sony Vaio Gr-Series)


-------------------------------------------------------
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

Re: Some questions about ACPI and laptops

[Mathias Kolehmainen]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: TEXT/PLAIN; charset=X-UNKNOWN, Size: 645 bytes --]

You may be able to slow down your machine if you have support for chaning
the either the performance or the throttling.

See:

  http://www.brodo.de/english/pub/acpi/proc/processor.html


-mathias

On Thu, 25 Jul 2002, Hanno Böck wrote:

> - Most time I use my laptop simply to work, i don't need the full
> power of my system. Is it possible to slowdown my pc to make the
> battery last longer?




-------------------------------------------------------
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

Re: Some questions about ACPI and laptops

[Dominik Brodowski]

On Thu, Jul 25, 2002 at 07:44:52PM +0200, Hanno Bvck wrote:
> I have some questions:
> - Most time I use my laptop simply to work, i don't need the full power of my system. Is it possible to slowdown my pc to make the battery last longer?
> - What else can I do to let my laptop work longer?
You might want to check
http://www.brodo.de/english/pub/acpi/proc/processor.html
(watch out for "performance" and "throttling"), and if you can't find no
P-States, you might get lucky here:
http://www.brodo.de/cpufreq/

Dominik


-------------------------------------------------------
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

Re: Some questions about ACPI and laptops

[Hanno Böck]

> http://www.brodo.de/cpufreq/

I've tried to apply the cpufreq-patch to my kernel (vanilla+acpi+nocor-2), but I get rejects according arch/i386/config.in

Is this patch available for the acpi-kernel?


-------------------------------------------------------
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

Re: Some questions about ACPI and laptops

[P. Christeas]

> You may be able to slow down your machine if you have support for chaning
> the either the performance or the throttling.

> On Thu, 25 Jul 2002, Hanno Bφck wrote:
> > - Most time I use my laptop simply to work, i don't need the full
> > power of my system. Is it possible to slowdown my pc to make the
> > battery last longer?
>

We have had some conversation in this list about that a few months ago. It 
seems the processor will 'fall' by itself in the 'C' states (see 
/proc/acpi/processor/CPU0/power ) when the load is low.  That way, it will 
run slower.
The question is: will a throttling (= not allow the processor to run at max/A 
speed) result in a lower consumption?
The answer is : the 'C' states theoretically have the ideal consumption. Some 
manufacturers may have gotten it wrong. We should add a daemon which will 
throttle. Then plot a graph of the power and performance and see if any 
method is better than the other.
I have seen such a daemon, but I don't think its results are reliable, since 
it polls the /proc/load?? . I think we could have results only if the 
'throttling daemon' resides in the very heart of the kernel (in academic: 
never trust an experiment done in a coarse way). 

That's the current state of power efficiency (IMHO). The ACPI team may not 
have better results soon, but rest asure that there is already some power 
conservation (remember that Linux itself makes very good use of the 
processing power).

P. Christeas


-------------------------------------------------------
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