Bug 116308

Summary:	powersaved doesn't have modules for p4
Product:	[openSUSE] SUSE LINUX 10.0	Reporter:	Renan Fernandes <renanfernandes>
Component:	Mobile Devices	Assignee:	Holger Macht <hmacht>
Status:	RESOLVED INVALID	QA Contact:	E-mail List <qa-bugs>
Severity:	Normal
Priority:	P5 - None	CC:	bg, renanfernandes, trenn
Version:	Preview 4
Target Milestone:	---
Hardware:	i686
OS:	All
Whiteboard:
Found By:	Beta-Customer	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---

Description Renan Fernandes 2005-09-10 03:44:26 UTC

the file /etc/init.d/powersaved doesn't have the CPUFREQ_MODULES for 
p4_clockmod
so, adding the module p4_clockmod to CPUFREQ_MODULES variable should fix it

Comment 1 Forgotten User ZhJd0F0L3x 2005-09-10 22:40:28 UTC

no. This module is useless, it does not do anything useful wrt power saving.

Sorry to say this, but if you want power saving, you need a processor that can
do frequency scaling, not clock modulation. So we will not add this module to
the default list unless someonce comes up with a very good reason (and
measurements) that support this.

Comment 2 Thomas Renninger 2005-09-12 07:08:38 UTC

The p4_clockmod module does throttling, not cpufreq.
We use another interface for this.
Default should be: if machine is idle for 10s -> machine is throttled by 50 %.
Use:
powersave -T
and use
Yast power-management module to allow throttling in performance scheme.
Or just edit
/etc/sysconfig/powersave/scheme_performance
ALLOW_THROTTLING="no" -> "yes"

Comment 3 Holger Macht 2006-06-29 14:07:28 UTC

*** Bug 188848 has been marked as a duplicate of this bug. ***

Comment 4 Thomas Renninger 2006-06-29 15:03:54 UTC

For anyone who cares about the p4-clockmode module:
This module is deprecated!
It is also discussed to let this module vanish from mainline kernel.

BTW: This module does throttling. Throttling is not intended for runtime power saving. It is intended for thermal reduction when the machine reaches very high temperatures. The powersaving/performance ratio is very poor. If one really cares about powersaving one should look at real cpufreq techniques like (Enhanced)SpeedStep or PowerNow (on x86(_64)).

Anyone who's bug ends up here as a duplicate: Please search the database for existing bugs first.