Attachment #43080 for bug #69734

View | Details | Raw Unified | Return to bug 69734 | Differences between
Collapse All | Expand All




<!-- See http://creativecommons.org/licenses/by-sa/2.0 -->
<license/>

<version>1.18</version>
<date>November 01, 2004</date>

<chapter>
<title>Introduction</title>

</pre>

<p>
Decide yourself whether you want to enable Software Suspend, Suspend-to-Disk and
Sleep States (see below). If you own an ASUS, Medion or Toshiba laptop, enable
the appropriate section. Recent kernel versions (2.6.9 and later) include an
<e>'ondemand' governor</e> for CPU Frequency Scaling, activate it as well when
using such a kernel.
</p>

<p>
The kernel has to know how to enable CPU frequency scaling on your processor. As
each type of CPU has a different interface, you've got to choose the right
driver for your processor. Be careful here - enabling <e>Intel Pentium 4 clock
modulation</e> on a Pentium M system will lead to strange results for example.
Consult the kernel documentation if you're unsure which one to take.
</p>

<p>


<p>
Typical ACPI events are closing the lid, changing the power source or pressing
the sleep button. Create the following files to switch between <e>default</e>
and <e>battery</e> runlevel depending on the power source:
</p>

<pre caption="/etc/acpi/switch_runlevel.conf">
#!/bin/bash

# This runlevel will be used in AC mode
#RUNLEVEL_AC="default"

# This runlevel will be used in battery mode
#RUNLEVEL_BATTERY="battery"

# This file indicates the AC state
# It's location may vary on your system. The script checks for
# /proc/acpi/ac_adapter/AC/state, /proc/acpi/ac_adapter/AC0/state,
# /proc/acpi/ac_adapter/AC1/state and /proc/acpi/ac_adapter/AC2/state
# If it is something else on your system, uncomment the next line
# and change appropriately.
#AC_STATE="/proc/acpi/ac_adapter/AC/state"
</pre>

<pre caption="/etc/acpi/switch_runlevel.sh">
#!/bin/bash

RUNLEVEL_AC="default"
RUNLEVEL_BATTERY="battery"

if [ -f '/proc/acpi/ac_adapter/AC0/state' ]
then
    AC_STATE="/proc/acpi/ac_adapter/AC0/state"
elif [ -f '/proc/acpi/ac_adapter/AC1/state' ]
then
    AC_STATE="/proc/acpi/ac_adapter/AC1/state"
elif [ -f '/proc/acpi/ac_adapter/AC2/state' ]
then
    AC_STATE="/proc/acpi/ac_adapter/AC2/state"
else
    AC_STATE="/proc/acpi/ac_adapter/AC/state"
fi


source /etc/acpi/switch_runlevel.conf

function SwitchRunlevel() {

    if [ ! -d "/etc/runlevels/${RUNLEVEL_AC}" ]
  then
        logger "${0}: Runlevel ${RUNLEVEL_AC} does not exist. Aborting."
        exit 1
    fi


    if [ ! -d "/etc/runlevels/${RUNLEVEL_BATTERY}" ]
    then
        logger "${0}: Runlevel ${RUNLEVEL_BATTERY} does not exist. Aborting."
        exit 1
    fi

    if [ ! -f "${AC_STATE}" ]
  then
        logger "${0}: File ${AC_STATE} does not exist. Aborting"
        exit 1
    fi


    if [[ "$(grep off-line ${AC_STATE})" != "" &amp;&amp; "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_BATTERY}" ]]
    then
        logger "Switching to ${RUNLEVEL_BATTERY} runlevel"
        /sbin/rc ${RUNLEVEL_BATTERY}
    elif [[ "$(grep on-line ${AC_STATE})" != "" &amp;&amp; "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_AC}" ]]
    then
        logger "Switching to ${RUNLEVEL_AC} runlevel"
        /sbin/rc ${RUNLEVEL_AC}
  fi
}
</pre>

<pre caption="/etc/acpi/events/pmg_ac_adapter">
event=ac_adapter.*
action=/etc/acpi/actions/pmg_ac_adapter.sh %e
</pre>

<pre caption="/etc/acpi/events/pmg_battery">
event=battery.*
action=/etc/acpi/actions/pmg_battery.sh %e
</pre>

<pre caption="/etc/acpi/actions/pmg_ac_adapter.sh">
#! /bin/bash

source /etc/acpi/switch_runlevel.sh
SwitchRunlevel
</pre>

<pre caption="/etc/acpi/actions/pmg_battery.sh">
#! /bin/bash

source /etc/acpi/switch_runlevel.sh
SwitchRunlevel
    case "$action" in
      battery) 
        SwitchRunlevel
        ;;
      *) 
        logger "ACPI group battery / action $action is not defined"
        ;;
    esac
    ;;

  ac_adapter)
    case "$action" in
      ac_adapter)
        SwitchRunlevel
        ;;
      *)
        logger "ACPI group ac_adapter / action $action is not defined"
        ;;
    esac
    ;;
  *)
    logger "ACPI group $group / action $action is not defined"
    ;;
esac
</pre>

<p>


<pre caption="Runlevel switch at boot time by editing local.start">
<comment># Fake acpi event to switch runlevel if running on batteries</comment>
/etc/acpi/actions/pmg_battery.sh "battery/battery"
</pre>

<p>

<chapter>
<title>CPU Power Management</title>
<section>
<title>Some technical terms</title>
<body>

<p>
CPU frequency scaling brings up some technical terms that might be unknown to
you. Here's a quick introduction.
</p>

<p>
First of all, the kernel has to be able to change the processor's frequency. The
<e>CPUfreq processor driver</e> knows the commands to do it on your CPU. Thus
it's important to choose the right one in your kernel. You should already have
done it above. Once the kernel knows how to change frequencies, it has to know
which frequency it should set. This is done according to the <e>policy</e> which
consists of <e>CPUfreq policy</e> and a <e>governor</e>. A CPUfreq policy are
just two numbers which define a range the frequency has to stay between -
minimal and maximal frequency. The governor now decides which of the available
frequencies in between minimal and maximal frequency to choose. For example, the
<e>powersave governor</e> always chooses the lowest frequency available, the
<e>performance governor</e> the highest one. The <e>userspace governor</e> makes
no decision but chooses whatever the user (or a program in userspace) wants -
which means it reads the frequency from
<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed</path>.
</p>

<p>
This doesn't sound like dynamic frequency changes yet and in fact it isn't.
Dynamics however can be accomplished with various approaches. For example,
the <e>ondemand governor</e> makes its decisions depending on the current CPU
load. The same is done by various userland tools like <c>cpudyn</c>,
<c>speedfreq</c>, <c>powernowd</c> and many more. ACPI events can be used to
enable or disable dynamic frequency changes depending on power source.
</p>

</body>
</section>
<section>
<title>Setting the frequency manually</title>
<body>



<p>
The above is quite nice, but not doable in daily life. Better let your system
set the appropriate frequency automatically. There are many different approaches
to do this. The following table gives a quick overview to help you decide on one
of them. It's roughly seperated in three categories <e>kernel</e> for approaches
that only need kernel support, <e>daemon</e> for programs that run in the
background and <e>graphical</e> for programs that provide a GUI for easy
configuration and changes.
</p>

<table>
<tr>
  <th>Name</th>
  <th>Category</th>
  <th>Switch decision</th>
  <th>Kernel governors</th>
  <th>Further governors</th>
  <th>Comments</th>
</tr>
<tr>
  <ti>'ondemand' governor</ti>
  <ti>Kernel</ti>
  <ti>CPU load</ti>
  <ti>N.A.</ti>
  <ti>N.A.</ti>
  <ti>
    Further tuning through files in
    <path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still requires
    userland tools (programs, scripts) if governor switching or similar is
    desired.
  </ti>
</tr>
<tr>
  <ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti>
  <ti>Daemon</ti>
  <ti>CPU load</ti>
  <ti>None</ti>
  <ti>Dynamic</ti>
  <ti>
    Also supports disk standby - notice however that <e>laptop mode</e> in most
    cases will do a better job.
  </ti>
</tr>
<tr>
  <ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreqd</uri></ti>
  <ti>Daemon</ti>
  <ti>Battery state, CPU load, running programs</ti>
  <ti>All available</ti>
  <ti>None</ti>
  <ti>
    Sophisticated (but also complicated) setup. An optimal configuration
    requires detailed knowledge of your system.
  </ti>
</tr>
<tr>
  <ti>
    <!-- FIXME: bug #68953 -->
    <uri
    link="http://fatcat.ftj.agh.edu.pl/~nelchael/index.php?id=ncpufreqd&amp;language=en">
    ncpufreqd</uri>
  </ti>
  <ti>Daemon</ti>
  <ti>CPU temperature</ti>
  <ti>Powersave, performance</ti>
  <ti>temperature governor</ti>
  <ti>
    It's main purpose is thermal management - whenever CPU temperature exceeds
    a user defined trip point, processor frequency will be lowered through
    frequency scaling or ACPI throttling. Very useful for laptops with mobile
    versions of desktops CPUs and other ones that have heat problems. See the
    "Thermal Management" chapter for details.
  </ti>
</tr>
<tr>
  <ti>
    <uri link="http://www.deater.net/john/powernowd.html">powernowd</uri>
  </ti>
  <ti>Daemon</ti>
  <ti>CPU load</ti>
  <ti>None</ti>
  <ti>Passive, sine, aggressive</ti>
  <ti>
    Supports SMP.
  </ti>
</tr>
<tr>
  <ti><uri link="http://www.goop.org/~jeremy/speedfreq/">speedfreq</uri></ti>
  <ti>Daemon</ti>
  <ti>CPU load</ti>
  <ti>None</ti>
  <ti>Dynamic, powersave, performance, fixed speed</ti>
  <ti>
    Small yet powerful with an useful client/server interface. Requires a 2.6
    kernel.
  </ti>
  <ti>Kernel 2.6 series only</ti>
</tr>
<tr>
  <ti><uri link="http://cpuspeedy.sourceforge.net/">gtk-cpuspeedy</uri></ti>
  <ti>Graphical</ti>
  <ti>None</ti>
  <ti>None</ti>
  <ti>None</ti>
  <ti>
    Gnome application, a graphical tool to set CPU frequency manually. It does
    not offer any automation and is mainly listed for the sake of completeness.
  </ti>
</tr>
<tr>
  <ti>klaptopdaemon</ti>
  <ti>Graphical</ti>
  <ti>Battery state</ti>
  <ti>All available</ti>
  <ti>None</ti>
  <ti>
    KDE only, 'ondemand' governor required for dynamic frequency scaling.
  </ti>
</tr>
</table>


<c>speedfreq</c>.
</p>

</body>
</section>
<section>
<title>Thermal Management</title>

<body>

<p>
There's another area where throttling CPU frequency is useful. If your laptop
(or Desktop system) has heat problems, frequency throttling can be the only
alternative to a shutdown. <c>ncpufreqd</c> is a tool that keeps an eye on
temperature and throttles CPU frequency whenever it gets too high. Make sure
that no other daemon like speedfreqd, cpufreqd, powernowd or cpudyn is running
and install <c>ncpufreqd</c>:
</p>

<pre caption="Installing ncpufreqd">
# <i> emerge ncpufreqd</i>
# <i> rc-update add ncpufreqd default battery </i>
</pre>

<p>
<c>ncpufreqd</c> is configured through <path>/etc/ncpufreqd.conf</path>. Modify
it and start ncpufreqd afterwards: <c>/etc/init.d/ncpufreqd start</c>.
</p>

</body>
</section>

<section>
<title>Verifying the result</title>

<body>

<p>
The last thing to check is that your new policies do a good job. An easy way to
do so is monitoring CPU speed while working with your laptop:
</p>

<pre caption="Monitoring CPU speed">


<p>
<!-- FIXME: bug #45593 -->
Besides kernel support you also need a script that controls starting and
stopping of laptop-mode. The package <c>laptop-mode-tools</c> contains it.
<path>/usr/src/linux/Documentation/laptop-mode.txt</path>. Onces it's ready,
make sure it gets called.
</p>

<pre caption="Automatic start of laptop-mode">
# <i>emerge laptop-mode-tools</i>
</pre>

<p>
<c>laptop-mode-tools</c> has it's configuration file in
<path>/etc/laptop-mode/laptop-mode.conf</path>. Adjust it the way you like it,
it's well commented. If you have <e>apm</e> or <e>acpi</e> in your USE flags,
laptop-mode will be started automatically in battery mode. Otherwise you can
automate it by running <c>rc-update add laptop-mode battery</c>.
</p>

</body>
</section>

</p>

<p>
<e>Q:</e> When configuring the kernel, powersave, performance and userspace
governors show up, but that ondemand thing is missing. Where do I get it?
</p>

<p>
<e>A:</e> The ondemand governor is only included in recent kernel sources. Try
updating them.
</p>

<p>
<e>Q:</e> Battery life time seems to be worse than before.
</p>


Return to bug 69734

Lines 17-24 Link Here

(-)power-management-guide.xml.org (-91 / +297 lines)
17	<!-- See http://creativecommons.org/licenses/by-sa/2.0 -->	17	<!-- See http://creativecommons.org/licenses/by-sa/2.0 -->
18	<license/>	18	<license/>
19		19
20	<version>1.17</version>	20	<version>1.18</version>
21	<date>October 10, 2004</date>	21	<date>November 01, 2004</date>
22		22
23	<chapter>	23	<chapter>
24	<title>Introduction</title>	24	<title>Introduction</title>
Lines 170-178 Link Here
170	</pre>	170	</pre>
171		171
172	<p>	172	<p>
173	Decide yourself whether you want to enable Software Suspend, Suspend-to-Disk	173	Decide yourself whether you want to enable Software Suspend, Suspend-to-Disk and
174	and Sleep States (see below). If you own an ASUS, Medion or Toshiba laptop,	174	Sleep States (see below). If you own an ASUS, Medion or Toshiba laptop, enable
175	enable the appropriate section.	175	the appropriate section. Recent kernel versions (2.6.9 and later) include an
		176	<e>'ondemand' governor</e> for CPU Frequency Scaling, activate it as well when
		177	using such a kernel.
		178	</p>
		179
		180	<p>
		181	The kernel has to know how to enable CPU frequency scaling on your processor. As
		182	each type of CPU has a different interface, you've got to choose the right
		183	driver for your processor. Be careful here - enabling <e>Intel Pentium 4 clock
		184	modulation</e> on a Pentium M system will lead to strange results for example.
		185	Consult the kernel documentation if you're unsure which one to take.
176	</p>	186	</p>
177		187
178	<p>	188	<p>
Lines 230-299 Link Here
230		240
231	<p>	241	<p>
232	Typical ACPI events are closing the lid, changing the power source or pressing	242	Typical ACPI events are closing the lid, changing the power source or pressing
233	the sleep button. Every acpi event recognized by the kernel is catched by acpid	243	the sleep button. Create the following files to switch between <e>default</e>
234	which calls <path>/etc/acpi/default.sh</path>. Here is a basic modification	244	and <e>battery</e> runlevel depending on the power source:
235	supporting runlevel switching:	245	</p>
236	</p>	246
237		247	<pre caption="/etc/acpi/switch_runlevel.conf">
238	<pre caption="Event driven runlevel switching with acpid">	248	#!/bin/bash
239	#!/bin/sh	249
240		250	# This runlevel will be used in AC mode
241	set $*	251	#RUNLEVEL_AC="default"
242		252
243	group=${1/\/*/}	253	# This runlevel will be used in battery mode
244	action=${1/*\//}	254	#RUNLEVEL_BATTERY="battery"
245		255
246	<comment># runlevel to use in AC mode</comment>	256	# This file indicates the AC state
247	RLVL_AC="default"	257	# It's location may vary on your system. The script checks for
248	<comment># runlevel to use in battery mode</comment>	258	# /proc/acpi/ac_adapter/AC/state, /proc/acpi/ac_adapter/AC0/state,
249	RLVL_BATTERY="battery"	259	# /proc/acpi/ac_adapter/AC1/state and /proc/acpi/ac_adapter/AC2/state
250		260	# If it is something else on your system, uncomment the next line
251	<comment># file indicating the AC state. Verify the filename before using</comment>	261	# and change appropriately.
252	AC_STATE="/proc/acpi/ac_adapter/AC/state"	262	#AC_STATE="/proc/acpi/ac_adapter/AC/state"
253	<comment># this string means running on AC</comment>	263	</pre>
254	AC_ON="on-line"	264
255	<comment># this string means running on batteries</comment>	265	<pre caption="/etc/acpi/switch_runlevel.sh">
256	AC_OFF="off-line"	266	#!/bin/bash
		267
		268	RUNLEVEL_AC="default"
		269	RUNLEVEL_BATTERY="battery"
		270
		271	if [ -f '/proc/acpi/ac_adapter/AC0/state' ]
		272	then
		273	AC_STATE="/proc/acpi/ac_adapter/AC0/state"
		274	elif [ -f '/proc/acpi/ac_adapter/AC1/state' ]
		275	then
		276	AC_STATE="/proc/acpi/ac_adapter/AC1/state"
		277	elif [ -f '/proc/acpi/ac_adapter/AC2/state' ]
		278	then
		279	AC_STATE="/proc/acpi/ac_adapter/AC2/state"
		280	else
		281	AC_STATE="/proc/acpi/ac_adapter/AC/state"
		282	fi
		283
		284
		285	source /etc/acpi/switch_runlevel.conf
257		286
258	function SwitchRunlevel() {	287	function SwitchRunlevel() {
259	if [[ "$(grep ${AC_OFF} ${AC_STATE})" != "" && "$(cat /var/lib/init.d/softlevel)" != "${RLVL_BATTERY}" ]]	288
		289	if [ ! -d "/etc/runlevels/${RUNLEVEL_AC}" ]
260	then	290	then
261	logger "Switching to ${RLVL_BATTERY} runlevel"	291	logger "${0}: Runlevel ${RUNLEVEL_AC} does not exist. Aborting."
262	/sbin/rc ${RLVL_BATTERY}	292	exit 1
263	elif [[ "$(grep ${AC_ON} ${AC_STATE})" != "" && "$(cat /var/lib/init.d/softlevel)" != "${RLVL_AC}" ]]	293	fi
		294
		295
		296	if [ ! -d "/etc/runlevels/${RUNLEVEL_BATTERY}" ]
		297	then
		298	logger "${0}: Runlevel ${RUNLEVEL_BATTERY} does not exist. Aborting."
		299	exit 1
		300	fi
		301
		302	if [ ! -f "${AC_STATE}" ]
264	then	303	then
265	logger "Switching to ${RLVL_AC} runlevel"	304	logger "${0}: File ${AC_STATE} does not exist. Aborting"
266	/sbin/rc ${RLVL_AC}	305	exit 1
		306	fi
		307
		308
		309	if [[ "$(grep off-line ${AC_STATE})" != "" && "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_BATTERY}" ]]
		310	then
		311	logger "Switching to ${RUNLEVEL_BATTERY} runlevel"
		312	/sbin/rc ${RUNLEVEL_BATTERY}
		313	elif [[ "$(grep on-line ${AC_STATE})" != "" && "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_AC}" ]]
		314	then
		315	logger "Switching to ${RUNLEVEL_AC} runlevel"
		316	/sbin/rc ${RUNLEVEL_AC}
267	fi	317	fi
268	}	318	}
		319	</pre>
269		320
		321	<pre caption="/etc/acpi/events/pmg_ac_adapter">
		322	event=ac_adapter.*
		323	action=/etc/acpi/actions/pmg_ac_adapter.sh %e
		324	</pre>
		325
		326	<pre caption="/etc/acpi/events/pmg_battery">
		327	event=battery.*
		328	action=/etc/acpi/actions/pmg_battery.sh %e
		329	</pre>
		330
		331	<pre caption="/etc/acpi/actions/pmg_ac_adapter.sh">
		332	#! /bin/bash
		333
		334	source /etc/acpi/switch_runlevel.sh
		335	SwitchRunlevel
		336	</pre>
		337
		338	<pre caption="/etc/acpi/actions/pmg_battery.sh">
		339	#! /bin/bash
270		340
271	case "$group" in	341	source /etc/acpi/switch_runlevel.sh
272	battery)	342	SwitchRunlevel
273	case "$action" in
274	battery)
275	SwitchRunlevel
276	;;
277	*)
278	logger "ACPI group battery / action $action is not defined"
279	;;
280	esac
281	;;
282
283	ac_adapter)
284	case "$action" in
285	ac_adapter)
286	SwitchRunlevel
287	;;
288	*)
289	logger "ACPI group ac_adapter / action $action is not defined"
290	;;
291	esac
292	;;
293	*)
294	logger "ACPI group $group / action $action is not defined"
295	;;
296	esac
297	</pre>	343	</pre>
298		344
299	<p>	345	<p>
Lines 313-319 Link Here
313		359
314	<pre caption="Runlevel switch at boot time by editing local.start">	360	<pre caption="Runlevel switch at boot time by editing local.start">
315	<comment># Fake acpi event to switch runlevel if running on batteries</comment>	361	<comment># Fake acpi event to switch runlevel if running on batteries</comment>
316	/etc/acpi/default.sh "battery/battery"	362	/etc/acpi/actions/pmg_battery.sh "battery/battery"
317	</pre>	363	</pre>
318		364
319	<p>	365	<p>
Lines 328-333 Link Here
328	<chapter>	374	<chapter>
329	<title>CPU Power Management</title>	375	<title>CPU Power Management</title>
330	<section>	376	<section>
		377	<title>Some technical terms</title>
		378	<body>
		379
		380	<p>
		381	CPU frequency scaling brings up some technical terms that might be unknown to
		382	you. Here's a quick introduction.
		383	</p>
		384
		385	<p>
		386	First of all, the kernel has to be able to change the processor's frequency. The
		387	<e>CPUfreq processor driver</e> knows the commands to do it on your CPU. Thus
		388	it's important to choose the right one in your kernel. You should already have
		389	done it above. Once the kernel knows how to change frequencies, it has to know
		390	which frequency it should set. This is done according to the <e>policy</e> which
		391	consists of <e>CPUfreq policy</e> and a <e>governor</e>. A CPUfreq policy are
		392	just two numbers which define a range the frequency has to stay between -
		393	minimal and maximal frequency. The governor now decides which of the available
		394	frequencies in between minimal and maximal frequency to choose. For example, the
		395	<e>powersave governor</e> always chooses the lowest frequency available, the
		396	<e>performance governor</e> the highest one. The <e>userspace governor</e> makes
		397	no decision but chooses whatever the user (or a program in userspace) wants -
		398	which means it reads the frequency from
		399	<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed</path>.
		400	</p>
		401
		402	<p>
		403	This doesn't sound like dynamic frequency changes yet and in fact it isn't.
		404	Dynamics however can be accomplished with various approaches. For example,
		405	the <e>ondemand governor</e> makes its decisions depending on the current CPU
		406	load. The same is done by various userland tools like <c>cpudyn</c>,
		407	<c>speedfreq</c>, <c>powernowd</c> and many more. ACPI events can be used to
		408	enable or disable dynamic frequency changes depending on power source.
		409	</p>
		410
		411	</body>
		412	</section>
		413	<section>
331	<title>Setting the frequency manually</title>	414	<title>Setting the frequency manually</title>
332	<body>	415	<body>
333		416
Lines 394-432 Link Here
394		477
395	<p>	478	<p>
396	The above is quite nice, but not doable in daily life. Better let your system	479	The above is quite nice, but not doable in daily life. Better let your system
397	set the appropriate frequency automatically. A couple of user space programs	480	set the appropriate frequency automatically. There are many different approaches
398	like to do it for you. The following table gives a quick overview to help you	481	to do this. The following table gives a quick overview to help you decide on one
399	decide on one of them.	482	of them. It's roughly seperated in three categories <e>kernel</e> for approaches
		483	that only need kernel support, <e>daemon</e> for programs that run in the
		484	background and <e>graphical</e> for programs that provide a GUI for easy
		485	configuration and changes.
400	</p>	486	</p>
401		487
402	<table>	488	<table>
403	<tr>	489	<tr>
404	<th>Name</th>	490	<th>Name</th>
405	<th>Pro</th>	491	<th>Category</th>
406	<th>Con</th>	492	<th>Switch decision</th>
		493	<th>Kernel governors</th>
		494	<th>Further governors</th>
		495	<th>Comments</th>
		496	</tr>
		497	<tr>
		498	<ti>'ondemand' governor</ti>
		499	<ti>Kernel</ti>
		500	<ti>CPU load</ti>
		501	<ti>N.A.</ti>
		502	<ti>N.A.</ti>
		503	<ti>
		504	Further tuning through files in
		505	<path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still requires
		506	userland tools (programs, scripts) if governor switching or similar is
		507	desired.
		508	</ti>
407	</tr>	509	</tr>
408	<tr>	510	<tr>
409	<ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti>	511	<ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti>
410	<ti>Also supports disk standby</ti>	512	<ti>Daemon</ti>
411	<ti></ti>	513	<ti>CPU load</ti>
		514	<ti>None</ti>
		515	<ti>Dynamic</ti>
		516	<ti>
		517	Also supports disk standby - notice however that <e>laptop mode</e> in most
		518	cases will do a better job.
		519	</ti>
412	</tr>	520	</tr>
413	<tr>	521	<tr>
414	<ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreq</uri></ti>	522	<ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreqd</uri></ti>
415	<ti>Sophisticated setup possible</ti>	523	<ti>Daemon</ti>
416	<ti>Complicated setup</ti>	524	<ti>Battery state, CPU load, running programs</ti>
		525	<ti>All available</ti>
		526	<ti>None</ti>
		527	<ti>
		528	Sophisticated (but also complicated) setup. An optimal configuration
		529	requires detailed knowledge of your system.
		530	</ti>
		531	</tr>
		532	<tr>
		533	<ti>
		534	<!-- FIXME: bug #68953 -->
		535	<uri
		536	link="http://fatcat.ftj.agh.edu.pl/~nelchael/index.php?id=ncpufreqd&language=en">
		537	ncpufreqd</uri>
		538	</ti>
		539	<ti>Daemon</ti>
		540	<ti>CPU temperature</ti>
		541	<ti>Powersave, performance</ti>
		542	<ti>temperature governor</ti>
		543	<ti>
		544	It's main purpose is thermal management - whenever CPU temperature exceeds
		545	a user defined trip point, processor frequency will be lowered through
		546	frequency scaling or ACPI throttling. Very useful for laptops with mobile
		547	versions of desktops CPUs and other ones that have heat problems. See the
		548	"Thermal Management" chapter for details.
		549	</ti>
		550	</tr>
		551	<tr>
		552	<ti>
		553	<uri link="http://www.deater.net/john/powernowd.html">powernowd</uri>
		554	</ti>
		555	<ti>Daemon</ti>
		556	<ti>CPU load</ti>
		557	<ti>None</ti>
		558	<ti>Passive, sine, aggressive</ti>
		559	<ti>
		560	Supports SMP.
		561	</ti>
417	</tr>	562	</tr>
418	<tr>	563	<tr>
419	<ti><uri link="http://www.goop.org/~jeremy/speedfreq/">speedfreq</uri></ti>	564	<ti><uri link="http://www.goop.org/~jeremy/speedfreq/">speedfreq</uri></ti>
		565	<ti>Daemon</ti>
		566	<ti>CPU load</ti>
		567	<ti>None</ti>
		568	<ti>Dynamic, powersave, performance, fixed speed</ti>
420	<ti>	569	<ti>
421	Small yet powerful<br />	570	Small yet powerful with an useful client/server interface. Requires a 2.6
422	Useful client/server interface	571	kernel.
423	</ti>	572	</ti>
424	<ti>Kernel 2.6 series only</ti>
425	</tr>	573	</tr>
426	<tr>	574	<tr>
427	<ti><uri link="http://www.deater.net/john/powernowd.html">powernowd</uri></ti>	575	<ti><uri link="http://cpuspeedy.sourceforge.net/">gtk-cpuspeedy</uri></ti>
428	<ti>Supports SMP</ti>	576	<ti>Graphical</ti>
429	<ti></ti>	577	<ti>None</ti>
		578	<ti>None</ti>
		579	<ti>None</ti>
		580	<ti>
		581	Gnome application, a graphical tool to set CPU frequency manually. It does
		582	not offer any automation and is mainly listed for the sake of completeness.
		583	</ti>
		584	</tr>
		585	<tr>
		586	<ti>klaptopdaemon</ti>
		587	<ti>Graphical</ti>
		588	<ti>Battery state</ti>
		589	<ti>All available</ti>
		590	<ti>None</ti>
		591	<ti>
		592	KDE only, 'ondemand' governor required for dynamic frequency scaling.
		593	</ti>
430	</tr>	594	</tr>
431	</table>	595	</table>
432		596
Lines 548-556 Link Here
548	<c>speedfreq</c>.	712	<c>speedfreq</c>.
549	</p>	713	</p>
550		714
		715	</body>
		716	</section>
		717	<section>
		718	<title>Thermal Management</title>
		719
		720	<body>
		721
		722	<p>
		723	There's another area where throttling CPU frequency is useful. If your laptop
		724	(or Desktop system) has heat problems, frequency throttling can be the only
		725	alternative to a shutdown. <c>ncpufreqd</c> is a tool that keeps an eye on
		726	temperature and throttles CPU frequency whenever it gets too high. Make sure
		727	that no other daemon like speedfreqd, cpufreqd, powernowd or cpudyn is running
		728	and install <c>ncpufreqd</c>:
		729	</p>
		730
		731	<pre caption="Installing ncpufreqd">
		732	# <i> emerge ncpufreqd</i>
		733	# <i> rc-update add ncpufreqd default battery </i>
		734	</pre>
		735
		736	<p>
		737	<c>ncpufreqd</c> is configured through <path>/etc/ncpufreqd.conf</path>. Modify
		738	it and start ncpufreqd afterwards: <c>/etc/init.d/ncpufreqd start</c>.
		739	</p>
		740
		741	</body>
		742	</section>
		743
		744	<section>
		745	<title>Verifying the result</title>
		746
		747	<body>
		748
551	<p>	749	<p>
552	The last thing to check is that your new policies do a good job. An easy way to	750	The last thing to check is that your new policies do a good job. An easy way to
553	do so is monitoring the CPU speed while working with your laptop:	751	do so is monitoring CPU speed while working with your laptop:
554	</p>	752	</p>
555		753
556	<pre caption="Monitoring CPU speed">	754	<pre caption="Monitoring CPU speed">
Lines 716-738 Link Here
716		914
717	<p>	915	<p>
718	<!-- FIXME: bug #45593 -->	916	<!-- FIXME: bug #45593 -->
719	To start and stop laptop-mode, create a script /etc/init.d/laptop-mode. You can	917	Besides kernel support you also need a script that controls starting and
720	take the one included in	918	stopping of laptop-mode. The package <c>laptop-mode-tools</c> contains it.
721	<path>/usr/src/linux/Documentation/laptop-mode.txt</path>. Onces it's ready,
722	make sure it gets called.
723	</p>	919	</p>
724		920
725	<pre caption="Automatic start of laptop-mode">	921	<pre caption="Automatic start of laptop-mode">
726	# <i>rc-update add laptop-mode battery</i>	922	# <i>emerge laptop-mode-tools</i>
727	</pre>	923	</pre>
728		924
729	<warn>	925	<p>
730	Once again: Be careful with sleep/spin down settings of your hard drive.	926	<c>laptop-mode-tools</c> has it's configuration file in
731	Setting it to small values might wear out your drive and lose warranty. Be sure	927	<path>/etc/laptop-mode/laptop-mode.conf</path>. Adjust it the way you like it,
732	to read the documentation in laptop-mode.txt. Make sure to stop laptop-mode	928	it's well commented. If you have <e>apm</e> or <e>acpi</e> in your USE flags,
733	before your battery runs out of power and data gets written to disk - otherwise	929	laptop-mode will be started automatically in battery mode. Otherwise you can
734	you will at least lose the last 10 minutes of your work.	930	automate it by running <c>rc-update add laptop-mode battery</c>.
735	</warn>	931	</p>
736		932
737	</body>	933	</body>
738	</section>	934	</section>
Lines 1049-1054 Link Here
1049	</p>	1245	</p>
1050		1246
1051	<p>	1247	<p>
		1248	<e>Q:</e> When configuring the kernel, powersave, performance and userspace
		1249	governors show up, but that ondemand thing is missing. Where do I get it?
		1250	</p>
		1251
		1252	<p>
		1253	<e>A:</e> The ondemand governor is only included in recent kernel sources. Try
		1254	updating them.
		1255	</p>
		1256
		1257	<p>
1052	<e>Q:</e> Battery life time seems to be worse than before.	1258	<e>Q:</e> Battery life time seems to be worse than before.
1053	</p>	1259	</p>
1054		1260