/[gentoo]/xml/htdocs/doc/en/power-management-guide.xml
Gentoo

Diff of /xml/htdocs/doc/en/power-management-guide.xml

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

Revision 1.15 Revision 1.35
1<?xml version='1.0' encoding="UTF-8"?> 1<?xml version='1.0' encoding="UTF-8"?>
2<!DOCTYPE guide SYSTEM "/dtd/guide.dtd"> 2<!DOCTYPE guide SYSTEM "/dtd/guide.dtd">
3<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/power-management-guide.xml,v 1.15 2005/10/28 14:48:06 so Exp $ --> 3<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/power-management-guide.xml,v 1.35 2008/01/31 23:38:05 nightmorph Exp $ -->
4<guide link="/doc/en/power-management-guide.xml"> 4<guide link="/doc/en/power-management-guide.xml">
5<title>Power Management Guide</title> 5<title>Power Management Guide</title>
6 6
7<author title="Author"> 7<author title="Author">
8 <mail link="fragfred@gmx.de">Dennis Nienhüser</mail> 8 <mail link="earthwings@gentoo.org">Dennis Nienhüser</mail>
9</author>
10<author title="Editor">
11 <mail link="chriswhite@gentoo.org">Chris White</mail>
12</author>
13<author title="Editor">
14 <mail link="nightmorph@gentoo.org">Joshua Saddler</mail>
9</author> 15</author>
10 16
11<abstract> 17<abstract>
12Power Management is the key to extend battery run time on mobile systems like 18Power Management is the key to extend battery run time on mobile systems like
13laptops. This guide assists you setting it up on your laptop. 19laptops. This guide assists you setting it up on your laptop.
15 21
16<!-- The content of this document is licensed under the CC-BY-SA license --> 22<!-- The content of this document is licensed under the CC-BY-SA license -->
17<!-- See http://creativecommons.org/licenses/by-sa/2.5 --> 23<!-- See http://creativecommons.org/licenses/by-sa/2.5 -->
18<license/> 24<license/>
19 25
20<version>1.25</version> 26<version>1.39</version>
21<date>2005-10-02</date> 27<date>2008-01-31</date>
22 28
23<chapter> 29<chapter>
24<title>Introduction</title> 30<title>Introduction</title>
25<section> 31<section>
26<title>Why Power Management?</title>
27<body> 32<body>
28 33
29<p> 34<p>
30Capacity and lifetime of laptop batteries have improved much in the last years. 35Capacity and lifetime of laptop batteries have improved much in the last years.
31Nevertheless modern processors consume much more energy than older ones and 36Nevertheless modern processors consume much more energy than older ones and
35intelligent Power Management policies. 40intelligent Power Management policies.
36</p> 41</p>
37 42
38</body> 43</body>
39</section> 44</section>
40
41<section> 45<section>
42<title>A quick overview</title> 46<title>A Quick Overview</title>
43<body> 47<body>
44 48
45<p> 49<p>
46Please notice that this guide describes Power Management for <e>laptops</e>. 50Please notice that this guide describes Power Management for <e>laptops</e>.
47While some sections might also suite for <e>servers</e>, others do not and may 51While some sections might also suite for <e>servers</e>, others do not and may
53As this guide has become rather long, here's a short overview helping you to 57As this guide has become rather long, here's a short overview helping you to
54find your way through it. 58find your way through it.
55</p> 59</p>
56 60
57<p> 61<p>
58The <e>Prerequisites</e> chapter talks about some requirements that should be 62The <uri link="#doc_chap2">Prerequisites</uri> chapter talks about some
59met before any of the following device individual sections will work. This 63requirements that should be met before any of the following device individual
60includes BIOS settings, kernel configuration and some simplifications in user 64sections will work. This includes BIOS settings, kernel configuration and some
61land. The following three chapters focus on devices that typically consume 65simplifications in user land. The following three chapters focus on devices
62most energy - processor, display and hard drive. Each can be configured 66that typically consume most energy - processor, display and hard drive. Each
63seperately. <e>CPU Power Management</e> shows how to adjust the processor's 67can be configured seperately. <uri link="#doc_chap3">CPU Power Management</uri>
64frequency to save a maximum of energy whithout losing too much performance. A 68shows how to adjust the processor's frequency to save a maximum of energy
65few different tricks prevent your hard drive from working unnecessarily often 69without losing too much performance. A few different tricks prevent your hard
66in <e>Disk Power Management</e> (decreasing noise level as a nice side 70drive from working unnecessarily often in <uri link="#doc_chap5">Disk Power
67effect). Some notes on graphics cards, Wireless LAN and USB finish the device 71Management</uri> (decreasing noise level as a nice side effect). Some notes on
68section in <e>Power Management for other devices</e> while another chapter is 72graphics cards, Wireless LAN and USB finish the device section in <uri
69dedicated to the (rather experimental) <e>sleep states</e>. Last not least 73link="#doc_chap6">Power Management For Other Devices</uri> while another
70<e>Troubleshooting</e> lists common pitfalls. 74chapter is dedicated to the (rather experimental) <uri link="#doc_chap7">sleep
75states</uri>. Last not least <uri link="#doc_chap8">Troubleshooting</uri> lists
76common pitfalls.
71</p> 77</p>
72 78
73</body> 79</body>
74</section>
75
76<section> 80</section>
81<section>
77<title>Power Budget for each component</title> 82<title>Power Budget For Each Component</title>
78<body> 83<body>
79 84
80<figure link="/images/energy-budget.png" short="Which component consumes how 85<figure link="/images/energy-budget.png" short="Which component consumes how
81much energy?" caption="Power budget for each component"/> 86much energy?" caption="Power budget for each component"/>
82 87
93</chapter> 98</chapter>
94 99
95<chapter> 100<chapter>
96<title>Prerequisites</title> 101<title>Prerequisites</title>
97<section> 102<section>
98<title>What has to be done first</title>
99<body>
100
101<p>
102Before going into the details on making individual devices Power Management
103aware, make sure certain requirements are met. After controlling the BIOS
104settings, some kernel options want to be enabled - these are in short ACPI,
105sleep states and CPU frequency scaling. As power saving most of the time comes
106along with performance loss or increased latency, it should only be enabled
107when running on batteries. That's where a new runlevel <e>battery</e> comes in
108handy.
109</p>
110
111</body> 103<body>
112</section> 104
105<p>
106Before discussing the details of making individual devices Power Management
107aware, make sure certain requirements are met. After controlling BIOS settings,
108some kernel options want to be enabled - these are in short ACPI, sleep states
109and CPU frequency scaling. As power saving most of the time comes along with
110performance loss or increased latency, it should only be enabled when running
111on batteries. That's where a new runlevel <e>battery</e> comes in handy.
112</p>
113
114</body>
113<section> 115</section>
116<section>
114<title>The BIOS part</title> 117<title>The BIOS Part</title>
115<body> 118<body>
116 119
117<p> 120<p>
118First have a look into your BIOS Power Management settings. The best way is to 121First have a look into your BIOS Power Management settings. The best way is to
119combine BIOS and operating system policies, but for the moment it's better to 122combine BIOS and operating system policies, but for the moment it's better to
123</p> 126</p>
124 127
125</body> 128</body>
126</section> 129</section>
127<section> 130<section>
131<title>Setting USE Flags</title>
132<body>
133
134<p>
135Please check that the <c>acpi</c> USE flag is set in
136<path>/etc/make.conf</path>. Other USE flags that might be interesting for your
137system are <c>apm</c>, <c>lm_sensors</c>, <c>nforce2</c>, <c>nvidia</c>,
138<c>pmu</c>. See <path>/usr/portage/profiles/use*.desc</path> for details. If
139you forgot to set one of these flags, you can recompile affected packages using
140the <c>--newuse</c> flag in <c>emerge</c>, see <c>man emerge</c>.
141</p>
142
143</body>
144</section>
145<section>
128<title>Configuring the kernel</title> 146<title>Configuring The Kernel</title>
129<body> 147<body>
130 148
131<p> 149<p>
132ACPI (Advanced Configuration and Power Interface) support in the kernel is 150ACPI (Advanced Configuration and Power Interface) support in the kernel is
133still work in progress. Using a recent kernel will make sure you'll get the 151still work in progress. Using a recent kernel will make sure you'll get the
134most out of it. 152most out of it.
135</p> 153</p>
136 154
137<p> 155<p>
138There are different kernel sources in Portage. I'd recommend using 156There are different kernel sources in Portage. I'd recommend using
139<c>gentoo-sources</c> or <c>suspend2-sources</c>. The latter contains patches 157<c>gentoo-sources</c> or <c>tuxonice-sources</c>. The latter contains patches
140for Software Suspend 2, see the chapter about sleep states for details. When 158for TuxOnIce, see the chapter about <uri link="#doc_chap7">sleep states</uri>
141configuring the kernel, activate at least these options: 159for more details. When configuring the kernel, activate at least these options:
142</p> 160</p>
143 161
144<pre caption="Minimum kernel setup for Power Management (Kernel 2.6)"> 162<pre caption="Minimum kernel setup for Power Management (Kernel 2.6)">
145Power Management Options ---&gt; 163Power Management Options ---&gt;
146 [*] Power Management Support 164 [*] Power Management Support
163 &lt; &gt; Toshiba Laptop Extras 181 &lt; &gt; Toshiba Laptop Extras
164 (0) Disable ACPI for systems before Jan 1st this year 182 (0) Disable ACPI for systems before Jan 1st this year
165 [ ] Debug Statements 183 [ ] Debug Statements
166 [*] Power Management Timer Support 184 [*] Power Management Timer Support
167 &lt; &gt; ACPI0004,PNP0A05 and PNP0A06 Container Driver (EXPERIMENTAL) 185 &lt; &gt; ACPI0004,PNP0A05 and PNP0A06 Container Driver (EXPERIMENTAL)
168 186
169 CPU Frequency Scaling ---&gt; 187 CPU Frequency Scaling ---&gt;
170 [*] CPU Frequency scaling 188 [*] CPU Frequency scaling
171 [ ] Enable CPUfreq debugging 189 [ ] Enable CPUfreq debugging
172 &lt; &gt; CPU frequency translation statistics 190 &lt; &gt; CPU frequency translation statistics
173 [ ] CPU frequency translation statistics details 191 [ ] CPU frequency translation statistics details
186(see below). If you own an ASUS, Medion, IBM Thinkpad or Toshiba laptop, enable 204(see below). If you own an ASUS, Medion, IBM Thinkpad or Toshiba laptop, enable
187the appropriate section. 205the appropriate section.
188</p> 206</p>
189 207
190<p> 208<p>
191The kernel has to know how to enable CPU frequency scaling on your processor. As 209The kernel has to know how to enable CPU frequency scaling on your processor.
192each type of CPU has a different interface, you've got to choose the right 210As each type of CPU has a different interface, you've got to choose the right
193driver for your processor. Be careful here - enabling <e>Intel Pentium 4 clock 211driver for your processor. Be careful here - enabling <c>Intel Pentium 4 clock
194modulation</e> on a Pentium M system will lead to strange results for example. 212modulation</c> on a Pentium M system will lead to strange results for example.
195Consult the kernel documentation if you're unsure which one to take. 213Consult the kernel documentation if you're unsure which one to take.
196</p> 214</p>
197 215
198<p> 216<p>
199Compile your kernel, make sure the right modules get loaded at startup and boot 217Compile your kernel, make sure the right modules get loaded at startup and boot
200into your new ACPI-enabled kernel. Next run <c>emerge sys-power/acpid</c> to get 218into your new ACPI-enabled kernel. Next run <c>emerge sys-power/acpid</c> to
201the acpi daemon. This one informs you about events like switching from AC to 219get the acpi daemon. This one informs you about events like switching from AC
202battery or closing the lid. Make sure the modules are loaded if you didn't 220to battery or closing the lid. Make sure the modules are loaded if you didn't
203compile them into the kernel and start acpid by executing 221compile them into the kernel and start acpid by executing <c>/etc/init.d/acpid
204<c>/etc/init.d/acpid start</c>. Run <c>rc-update add acpid default</c> to load 222start</c>. Run <c>rc-update add acpid default</c> to load it on startup. You'll
205it on startup. You'll soon see how to use it. 223soon see how to use it.
206</p> 224</p>
207 225
208<pre caption="Installing acpid"> 226<pre caption="Installing acpid">
209# <i>emerge sys-power/acpid</i> 227# <i>emerge sys-power/acpid</i>
210# <i>/etc/init.d/acpid start</i> 228# <i>/etc/init.d/acpid start</i>
212</pre> 230</pre>
213 231
214</body> 232</body>
215</section> 233</section>
216<section> 234<section>
217<title>Creating a "battery" runlevel</title> 235<title>Creating A "battery" Runlevel</title>
218<body> 236<body>
219 237
220<p> 238<p>
221The default policy will be to enable Power Management only when needed - 239The default policy will be to enable Power Management only when needed -
222running on batteries. To make the switch between AC and battery convenient, 240running on batteries. To make the switch between AC and battery convenient,
223create a runlevel <e>battery</e> that holds all the scripts starting and 241create a runlevel <c>battery</c> that holds all the scripts starting and
224stopping Power Management. 242stopping Power Management.
225</p> 243</p>
226 244
227<note> 245<note>
228You can safely skip this section if you don't like the idea of having another 246You can safely skip this section if you don't like the idea of having another
229runlevel. However, skipping this step will make the rest a bit trickier to set 247runlevel. However, skipping this step will make the rest a bit trickier to set
230up. The next sections assume a runlevel <e>battery</e> exists. 248up. The next sections assume a runlevel <c>battery</c> exists.
231</note> 249</note>
232 250
233<pre caption="Creating a battery runlevel"> 251<pre caption="Creating a battery runlevel">
234# <i>cd /etc/runlevels</i> 252# <i>cd /etc/runlevels</i>
235# <i>cp -a default battery</i> 253# <i>cp -a default battery</i>
236</pre> 254</pre>
237 255
238<p> 256<p>
239Finished. Your new runlevel <e>battery</e> contains everything like 257Finished. Your new runlevel <c>battery</c> contains everything like
240<e>default</e>, but there is no automatic switch between both yet. Time to 258<c>default</c>, but there is no automatic switch between both yet. Time to
241change it. 259change it.
242</p> 260</p>
243 261
244</body> 262</body>
245</section> 263</section>
246<section> 264<section>
247<title>Reacting on ACPI events</title> 265<title>Reacting On ACPI Events</title>
248<body> 266<body>
249 267
250<p> 268<p>
251Typical ACPI events are closing the lid, changing the power source or pressing 269Typical ACPI events are closing the lid, changing the power source or pressing
252the sleep button. An important event is changing the power source, which should 270the sleep button. An important event is changing the power source, which should
259<c>on_ac_power</c> command from <c>sys-power/powermgmt-base</c> - make sure the 277<c>on_ac_power</c> command from <c>sys-power/powermgmt-base</c> - make sure the
260package is installed on your system. 278package is installed on your system.
261</p> 279</p>
262 280
263<pre caption="Installing powermgt-base"> 281<pre caption="Installing powermgt-base">
264<i># emerge powermgmt-base</i> 282# <i>emerge powermgmt-base</i>
265</pre> 283</pre>
266 284
267<p> 285<p>
268You are now able to determine the power source by executing 286You are now able to determine the power source by executing <c>on_ac_power
269<c>on_ac_power &amp;&amp; echo AC available || echo Running on batteries</c> in 287&amp;&amp; echo AC available || echo Running on batteries</c> in a shell. The
270a shell. The script below is responsible for changing runlevels. Save it as 288script below is responsible for changing runlevels. Save it as
271<path>/etc/acpi/actions/pmg_switch_runlevel.sh</path>. 289<path>/etc/acpi/actions/pmg_switch_runlevel.sh</path>.
272</p> 290</p>
273 291
274<pre caption="/etc/acpi/actions/pmg_switch_runlevel.sh"> 292<pre caption="/etc/acpi/actions/pmg_switch_runlevel.sh">
275#!/bin/bash 293#!/bin/bash
280<comment># END configuration</comment> 298<comment># END configuration</comment>
281 299
282 300
283if [ ! -d "/etc/runlevels/${RUNLEVEL_AC}" ] 301if [ ! -d "/etc/runlevels/${RUNLEVEL_AC}" ]
284then 302then
285 logger "${0}: Runlevel ${RUNLEVEL_AC} does not exist. Aborting." 303 logger "${0}: Runlevel ${RUNLEVEL_AC} does not exist. Aborting."
286 exit 1 304 exit 1
287fi 305fi
288 306
289if [ ! -d "/etc/runlevels/${RUNLEVEL_BATTERY}" ] 307if [ ! -d "/etc/runlevels/${RUNLEVEL_BATTERY}" ]
290then 308then
291 logger "${0}: Runlevel ${RUNLEVEL_BATTERY} does not exist. Aborting." 309 logger "${0}: Runlevel ${RUNLEVEL_BATTERY} does not exist. Aborting."
292 exit 1 310 exit 1
293fi 311fi
294 312
295if on_ac_power 313if on_ac_power
296then 314then
297 if [[ "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_AC}" ]] 315 if [[ "$(&lt;/var/lib/init.d/softlevel)" != "${RUNLEVEL_AC}" ]]
298 then 316 then
299 logger "Switching to ${RUNLEVEL_AC} runlevel" 317 logger "Switching to ${RUNLEVEL_AC} runlevel"
300 /sbin/rc ${RUNLEVEL_AC} 318 /sbin/rc ${RUNLEVEL_AC}
301 fi 319 fi
302elif [[ "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_BATTERY}" ]] 320elif [[ "$(&lt;/var/lib/init.d/softlevel)" != "${RUNLEVEL_BATTERY}" ]]
303then 321then
304 logger "Switching to ${RUNLEVEL_BATTERY} runlevel" 322 logger "Switching to ${RUNLEVEL_BATTERY} runlevel"
305 /sbin/rc ${RUNLEVEL_BATTERY} 323 /sbin/rc ${RUNLEVEL_BATTERY}
306fi 324fi
307</pre> 325</pre>
308 326
309<p> 327<p>
310Dont forget to run <c>chmod +x /etc/acpi/actions/pmg_switch_runlevel.sh</c> to 328Dont forget to run <c>chmod +x /etc/acpi/actions/pmg_switch_runlevel.sh</c> to
311make the script executable. The last thing that needs to be done is calling the 329make the script executable. The last thing that needs to be done is calling the
312script whenever the power source changes. That's done by catching ACPI events 330script whenever the power source changes. That's done by catching ACPI events
313with the help of <c>acpid</c>. First you need to know which events are 331with the help of <c>acpid</c>. First you need to know which events are
314generated when the power source changes. The events are called 332generated when the power source changes. The events are called
315<e>ac_adapter</e> and <e>battery</e> on most laptops, but it might be different 333<c>ac_adapter</c> and <c>battery</c> on most laptops, but it might be different
316on yours. 334on yours.
317</p> 335</p>
318 336
319<pre caption="Determining ACPI events for changing the power source"> 337<pre caption="Determining ACPI events for changing the power source">
320<i># tail -f /var/log/acpid | grep "received event"</i> 338# <i>tail -f /var/log/messages | grep "received event"</i>
321</pre> 339</pre>
322 340
323<p> 341<p>
324Run the command above and pull the power cable. You should see something 342Run the command above and pull the power cable. You should see something like
325like this: 343this:
326</p> 344</p>
327 345
328<pre caption="Sample output for power source changes"> 346<pre caption="Sample output for power source changes">
329[Tue Sep 20 17:39:06 2005] received event "ac_adapter AC 00000080 00000000" 347[Tue Sep 20 17:39:06 2005] received event "ac_adapter AC 00000080 00000000"
330[Tue Sep 20 17:39:06 2005] received event "battery BAT0 00000080 00000001" 348[Tue Sep 20 17:39:06 2005] received event "battery BAT0 00000080 00000001"
331</pre> 349</pre>
332 350
333<p> 351<p>
334The interesting part is the quoted string after <e>received event</e>. It will 352The interesting part is the quoted string after <c>received event</c>. It will
335be matched by the event line in the files you are going to create below. Don't 353be matched by the event line in the files you are going to create below. Don't
336worry if your system generates multiple events or always the same. As long as 354worry if your system generates multiple events or always the same. As long as
337any event is generated, runlevel changing will work. 355any event is generated, runlevel changing will work.
338</p> 356</p>
339 357
354<p> 372<p>
355Finally acpid has to be restarted to recognize the changes. 373Finally acpid has to be restarted to recognize the changes.
356</p> 374</p>
357 375
358<pre caption="Finishing runlevel switching with acpid"> 376<pre caption="Finishing runlevel switching with acpid">
359<i># /etc/init.d/acpid restart</i> 377# <i>/etc/init.d/acpid restart</i>
360</pre> 378</pre>
361 379
362<p> 380<p>
363Give it a try: Plug AC in and out and watch syslog for the "Switching to AC 381Give it a try: Plug AC in and out and watch syslog for the "Switching to AC
364mode" or "Switching to battery mode" messages. See the Troubleshooting 382mode" or "Switching to battery mode" messages. See the <uri
365section if the script is not able to detect the power source correctly. 383link="#doc_chap8">Troubleshooting section</uri> if the script is not able to
384detect the power source correctly.
366</p> 385</p>
367 386
368<p> 387<p>
369Due to the nature of the event mechanism, your laptop will boot into runlevel 388Due to the nature of the event mechanism, your laptop will boot into runlevel
370<e>default</e> regardless of the AC/battery state. This is fine when running 389<c>default</c> regardless of the AC/battery state. This is fine when running
371from AC, but we'd like to boot into the battery runlevel otherwise. One 390from AC, but we'd like to boot into the battery runlevel otherwise. One
372solution would be to add another entry to the boot loader with the parameter 391solution would be to add another entry to the boot loader with the parameter
373<c>softlevel=battery</c>, but it's likely to forget choosing it. A better way 392<c>softlevel=battery</c>, but it's likely to forget choosing it. A better way
374is faking an ACPI event in the end of the boot process and letting 393is faking an ACPI event in the end of the boot process and letting
375<path>pmg_switch_runlevel.sh</path> script decide whether a 394<path>pmg_switch_runlevel.sh</path> script decide whether a runlevel change is
376runlevel change is necessary. Open <path>/etc/conf.d/local.start</path> in your 395necessary. Open <path>/etc/conf.d/local.start</path> in your favourite editor
377favourite editor and add these lines: 396and add these lines:
378</p> 397</p>
379 398
380<pre caption="Runlevel adjustment at boot time by editing local.start"> 399<pre caption="Runlevel adjustment at boot time by editing local.start">
381<comment># Fake acpi event to switch runlevel if running on batteries</comment> 400<comment># Fake acpi event to switch runlevel if running on batteries</comment>
382/etc/acpi/actions/pmg_switch_runlevel.sh "battery/battery" 401/etc/acpi/actions/pmg_switch_runlevel.sh "battery/battery"
383</pre> 402</pre>
384 403
385<p> 404<p>
386Prepared like this you can activate Power Management policies for individual 405Prepared like this you can activate Power Management policies for individual
387devices. 406devices.
388</p> 407</p>
389 408
390</body> 409</body>
391</section> 410</section>
392</chapter> 411</chapter>
393 412
394<chapter> 413<chapter>
395<title>CPU Power Management</title> 414<title>CPU Power Management</title>
396<section> 415<section>
416<body>
417
418<p>
419Mobile processors can operate at different frequencies. Some allow changing
420voltage as well. Most of the time your CPU doesn't need to run at full speed
421and scaling it down will save much energy - often without any performance
422decrease.
423</p>
424
425</body>
426</section>
427<section>
397<title>Some technical terms</title> 428<title>Some Technical Terms</title>
398<body> 429<body>
399 430
400<p> 431<p>
401CPU frequency scaling brings up some technical terms that might be unknown to 432CPU frequency scaling brings up some technical terms that might be unknown to
402you. Here's a quick introduction. 433you. Here's a quick introduction.
403</p> 434</p>
404 435
405<p> 436<p>
406First of all, the kernel has to be able to change the processor's frequency. 437First of all, the kernel has to be able to change the processor's frequency.
407The <e>CPUfreq processor driver</e> knows the commands to do it on your CPU. 438The <b>CPUfreq processor driver</b> knows the commands to do it on your CPU.
408Thus it's important to choose the right one in your kernel. You should 439Thus it's important to choose the right one in your kernel. You should already
409already have done it above. Once the kernel knows how to change frequencies, 440have done it above. Once the kernel knows how to change frequencies, it has to
410it has to know which frequency it should set. This is done according to the 441know which frequency it should set. This is done according to the <b>policy</b>
411<e>policy</e> which consists of a <e>CPUfreq policy</e> and a 442which consists of a <b>CPUfreq policy</b> and a <b>governor</b>. A CPUfreq
412<e>governor</e>. A CPUfreq policy are just two numbers which define a range 443policy are just two numbers which define a range the frequency has to stay
413the frequency has to stay between - minimal and maximal frequency. The 444between - minimal and maximal frequency. The governor now decides which of the
414governor now decides which of the available frequencies in between minimal 445available frequencies in between minimal and maximal frequency to choose. For
415and maximal frequency to choose. For example, the <e>powersave governor</e> 446example, the <b>powersave governor</b> always chooses the lowest frequency
416always chooses the lowest frequency available, the <e>performance 447available, the <b>performance governor</b> the highest one. The <b>userspace
417governor</e> the highest one. The <e>userspace governor</e> makes no decision 448governor</b> makes no decision but chooses whatever the user (or a program in
418but chooses whatever the user (or a program in userspace) wants - which means 449userspace) wants - which means it reads the frequency from
419it reads the frequency from
420<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed</path>. 450<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed</path>.
421</p> 451</p>
422 452
423<p> 453<p>
424This doesn't sound like dynamic frequency changes yet and in fact it isn't. 454This doesn't sound like dynamic frequency changes yet and in fact it isn't.
425Dynamics however can be accomplished with various approaches. For example, 455Dynamics however can be accomplished with various approaches. For example, the
426the <e>ondemand governor</e> makes its decisions depending on the current CPU 456<b>ondemand governor</b> makes its decisions depending on the current CPU load.
427load. The same is done by various userland tools like <c>cpudyn</c>, 457The same is done by various userland tools like <c>cpudyn</c>, <c>cpufreqd</c>,
428<c>cpufreqd</c>, <c>powernowd</c> and many more. ACPI events can be used to 458<c>powernowd</c> and many more. ACPI events can be used to enable or disable
429enable or disable dynamic frequency changes depending on power source. 459dynamic frequency changes depending on power source.
430</p> 460</p>
431 461
432</body> 462</body>
433</section>
434<section> 463</section>
464<section>
435<title>Setting the frequency manually</title> 465<title>Setting The Frequency Manually</title>
436<body> 466<body>
437 467
438<p> 468<p>
439Decreasing CPU speed and voltage has two advantages: On the one hand less 469Decreasing CPU speed and voltage has two advantages: On the one hand less
440energy is consumed, on the other hand there is thermal improvement as your 470energy is consumed, on the other hand there is thermal improvement as your
443between performance loss and energy saving. 473between performance loss and energy saving.
444</p> 474</p>
445 475
446<note> 476<note>
447Not every laptop supports frequency scaling. If unsure, have a look at the list 477Not every laptop supports frequency scaling. If unsure, have a look at the list
448of supported processors in the <e>Troubleshooting</e> section to verify your's 478of supported processors in the <uri link="#doc_chap8">Troubleshooting</uri>
449is supported. 479section to verify yours is supported.
450</note> 480</note>
451 481
452<p> 482<p>
453It's time to test whether CPU frequency changing works. Let's install another 483It's time to test whether CPU frequency changing works. Let's install another
454tool which is very handy for debugging purposes: <c>sys-power/cpufrequtils</c> 484tool which is very handy for debugging purposes: <c>sys-power/cpufrequtils</c>
480 510
481<p> 511<p>
482Now play around with <c>cpufreq-set</c> to make sure frequency switching works. 512Now play around with <c>cpufreq-set</c> to make sure frequency switching works.
483Run <c>cpufreq-set -g ondemand</c> for example to activate the ondemand 513Run <c>cpufreq-set -g ondemand</c> for example to activate the ondemand
484governor and verify the change with <c>cpufreq-info</c>. If it doesn't work as 514governor and verify the change with <c>cpufreq-info</c>. If it doesn't work as
485expected, you might find help in the Troubleshooting section in the end of this 515expected, you might find help in the <uri link="#doc_chap8">Troubleshooting
486guide. 516section</uri> in the end of this guide.
487</p> 517</p>
488 518
489</body> 519</body>
490</section> 520</section>
491<section> 521<section>
494 524
495<p> 525<p>
496The above is quite nice, but not doable in daily life. Better let your system 526The above is quite nice, but not doable in daily life. Better let your system
497set the appropriate frequency automatically. There are many different 527set the appropriate frequency automatically. There are many different
498approaches to do this. The following table gives a quick overview to help you 528approaches to do this. The following table gives a quick overview to help you
499decide on one of them. It's roughly seperated in three categories 529decide on one of them. It's roughly separated in three categories <b>kernel</b>
500<e>kernel</e> for approaches that only need kernel support, <e>daemon</e> for 530for approaches that only need kernel support, <b>daemon</b> for programs that
501programs that run in the background and <e>graphical</e> for programs that 531run in the background and <b>graphical</b> for programs that provide a GUI for
502provide a GUI for easy configuration and changes. 532easy configuration and changes.
503</p> 533</p>
504 534
505<table> 535<table>
506<tr> 536<tr>
507 <th>Name</th> 537 <th>Name</th>
530 <ti>Kernel</ti> 560 <ti>Kernel</ti>
531 <ti>CPU load</ti> 561 <ti>CPU load</ti>
532 <ti>N.A.</ti> 562 <ti>N.A.</ti>
533 <ti>N.A.</ti> 563 <ti>N.A.</ti>
534 <ti> 564 <ti>
535 Unlike the ondemand governor, conversative doesn't jump to maximum 565 Unlike the ondemand governor, conversative doesn't jump to maximum
536 frequency when CPU load is high, but increases the frequency step by 566 frequency when CPU load is high, but increases the frequency step by step.
537 step. Further tuning through files in 567 Further tuning through files in
538 <path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still 568 <path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still requires
539 requires userland tools (programs, scripts) if governor switching or 569 userland tools (programs, scripts) if governor switching or similar is
540 similar is desired. 570 desired.
541 </ti> 571 </ti>
542</tr> 572</tr>
543<tr> 573<tr>
544 <ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti> 574 <ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti>
545 <ti>Daemon</ti> 575 <ti>Daemon</ti>
552 </ti> 582 </ti>
553</tr> 583</tr>
554<tr> 584<tr>
555 <ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreqd</uri></ti> 585 <ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreqd</uri></ti>
556 <ti>Daemon</ti> 586 <ti>Daemon</ti>
557 <ti>Battery state, CPU load, temperature, running programs</ti> 587 <ti>Battery state, CPU load, temperature, running programs and more</ti>
558 <ti>All available</ti> 588 <ti>All available</ti>
559 <ti>None</ti> 589 <ti>None</ti>
560 <ti> 590 <ti>
561 Sophisticated (but also complicated) setup. 591 Sophisticated (but somewhat complicated) setup. Extendible through plugins
592 like sensor monitoring (lm_sensors) or coordinating some NVidia based
593 graphics card memory and core. Cpufreqd is SMP aware and can optionally be
594 controlled manually at runtime.
562 </ti> 595 </ti>
563</tr> 596</tr>
564<tr> 597<tr>
565 <ti> 598 <ti>
566 <uri link="http://www.deater.net/john/powernowd.html">powernowd</uri> 599 <uri link="http://www.deater.net/john/powernowd.html">powernowd</uri>
573 Supports SMP. 606 Supports SMP.
574 </ti> 607 </ti>
575</tr> 608</tr>
576<tr> 609<tr>
577 <ti> 610 <ti>
578 <uri link="http://fatcat.ftj.agh.edu.pl/~nelchael/index.php?cat=projs&amp;subcat=ncpufreqd&amp;language=en">ncpufreqd</uri> 611 <uri
612 link="http://projects.simpledesigns.com.pl/project/ncpufreqd/">ncpufreqd</uri>
579 </ti> 613 </ti>
580 <ti>Daemon</ti> 614 <ti>Daemon</ti>
581 <ti>Temperature</ti> 615 <ti>Temperature</ti>
582 <ti>None</ti> 616 <ti>None</ti>
583 <ti>Powersave, performance</ti> 617 <ti>Powersave, performance</ti>
623 657
624<p> 658<p>
625While adjusting the frequency to the current load looks simple at a first 659While adjusting the frequency to the current load looks simple at a first
626glance, it's not such a trivial task. A bad algorithm can cause switching 660glance, it's not such a trivial task. A bad algorithm can cause switching
627between two frequencies all the time or wasting energy when setting frequency 661between two frequencies all the time or wasting energy when setting frequency
628to an unnecessary high level. 662to an unnecessary high level.
629</p> 663</p>
630 664
631<p> 665<p>
632Which one to choose? If you have no idea about it, try <c>cpufreqd</c>: 666Which one to choose? If you have no idea about it, try <c>cpufreqd</c>:
633</p> 667</p>
637</pre> 671</pre>
638 672
639<p> 673<p>
640<c>cpufreqd</c> can be configured by editing <path>/etc/cpufreqd.conf</path>. 674<c>cpufreqd</c> can be configured by editing <path>/etc/cpufreqd.conf</path>.
641The default one that ships with cpufreqd may look a bit confusing. I recommend 675The default one that ships with cpufreqd may look a bit confusing. I recommend
642replacing it with the one from Gentoo developer Henrik Brix Andersen (see 676replacing it with the one from former Gentoo developer Henrik Brix Andersen
643below). 677(see below). Please notice that you need cpufreqd-2.0.0 or later. Earlier
678versions have a different syntax for the config file.
644</p> 679</p>
645 680
646<pre caption="/etc/cpufreqd.conf"> 681<pre caption="/etc/cpufreqd.conf (cpufreqd-2.0.0 and later)">
647[General] 682[General]
648pidfile=/var/run/cpufreqd.pid 683pidfile=/var/run/cpufreqd.pid
649poll_interval=2 684poll_interval=3
650pm_type=acpi 685enable_plugins=acpi_ac, acpi_battery
686enable_remote=1
687remote_group=wheel
651verbosity=5 688verbosity=5
689[/General]
652 690
653[Profile] 691[Profile]
654name=ondemand 692name=ondemand
655minfreq=0% 693minfreq=0%
656maxfreq=100% 694maxfreq=100%
657policy=ondemand 695policy=ondemand
696[/Profile]
658 697
659[Profile] 698[Profile]
660name=conservative 699name=conservative
661minfreq=0% 700minfreq=0%
662maxfreq=100% 701maxfreq=100%
663policy=conservative 702policy=conservative
703[/Profile]
664 704
665[Profile] 705[Profile]
666name=powersave 706name=powersave
667minfreq=0% 707minfreq=0%
668maxfreq=100% 708maxfreq=100%
669policy=powersave 709policy=powersave
710[/Profile]
670 711
671[Profile] 712[Profile]
672name=performance 713name=performance
673minfreq=0% 714minfreq=0%
674maxfreq=100% 715maxfreq=100%
675policy=performance 716policy=performance
717[/Profile]
676 718
677[Rule] 719[Rule]
678name=battery 720name=battery
679ac=off 721ac=off
680profile=conservative 722profile=conservative
723[/Rule]
681 724
682[Rule] 725[Rule]
683name=battery_low 726name=battery_low
684ac=off 727ac=off
685battery_interval=0-10 728battery_interval=0-10
686profile=powersave 729profile=powersave
730[/Rule]
687 731
688[Rule] 732[Rule]
689name=ac 733name=ac
690ac=on 734ac=on
691profile=ondemand 735profile=ondemand
736[/Rule]
692</pre> 737</pre>
693 738
694<p> 739<p>
695Now you can start the cpufreqd daemon. Add it to the <e>default</e> and 740Now you can start the cpufreqd daemon. Add it to the <c>default</c> and
696<e>battery</e> runlevel as well. 741<c>battery</c> runlevel as well.
697</p> 742</p>
698 743
699<pre caption="Starting cpufreqd"> 744<pre caption="Starting cpufreqd">
700# <i>rc-update add cpufreqd default battery</i> 745# <i>rc-update add cpufreqd default battery</i>
701# <i>rc</i> 746# <i>rc</i>
702</pre> 747</pre>
748
749<p>
750Sometimes it can be desirable to select another policy than the daemon chooses,
751for example when battery power is low, but you know that AC will be available
752soon. In that case you can turn on cpufreqd's manual mode with <c>cpufreqd-set
753manual</c> and select one of your configured policies (as listed by
754<c>cpufreqd-get</c>). You can leave manual mode by executing <c>cpufreqd-set
755dynamic</c>.
756</p>
703 757
704<warn> 758<warn>
705Do not run more than one of the above programs at the same time. It may cause 759Do not run more than one of the above programs at the same time. It may cause
706confusion like switching between two frequencies all the time. 760confusion like switching between two frequencies all the time.
707</warn> 761</warn>
708 762
709</body> 763</body>
710</section> 764</section>
711
712<section> 765<section>
713<title>Verifying the result</title> 766<title>Verifying the result</title>
714
715<body> 767<body>
716 768
717<p> 769<p>
718The last thing to check is that your new policies do a good job. An easy way to 770The last thing to check is that your new policies do a good job. An easy way to
719do so is monitoring CPU speed while working with your laptop: 771do so is monitoring CPU speed while working with your laptop:
722<pre caption="Monitoring CPU speed"> 774<pre caption="Monitoring CPU speed">
723# <i>watch grep \"cpu MHz\" /proc/cpuinfo</i> 775# <i>watch grep \"cpu MHz\" /proc/cpuinfo</i>
724</pre> 776</pre>
725 777
726<p> 778<p>
727If <path>/proc/cpuinfo</path> doesn't get updated (see Troubleshooting), 779If <path>/proc/cpuinfo</path> doesn't get updated (see <uri
728monitor the CPU frequency with: 780link="#doc_chap8">Troubleshooting</uri>), monitor the CPU frequency with
781<c>sys-apps/x86info</c>:
729</p> 782</p>
730 783
731<pre caption="Alternative CPU speed monitoring"> 784<pre caption="Alternative CPU speed monitoring">
732# <i>watch x86info -mhz</i> 785# <i>watch x86info -mhz</i>
733</pre> 786</pre>
734 787
735<p> 788<p>
736Depending on your setup, CPU speed should increase on heavy load, decrease on 789Depending on your setup, CPU speed should increase on heavy load, decrease on
737no activity or just stay at the same level. When using cpufreqd and verbosity 790no activity or just stay at the same level. When using <c>cpufreqd</c> and
738set to 5 or higher in <path>cpufreqd.conf</path> you'll get additional 791verbosity set to 5 or higher in <path>cpufreqd.conf</path> you'll get
739information about what's happening reported to syslog. 792additional information about what's happening reported to <c>syslog</c>.
740</p> 793</p>
741 794
742</body> 795</body>
743</section> 796</section>
744</chapter> 797</chapter>
745 798
746<chapter> 799<chapter>
747<title>LCD Power Management</title> 800<title>LCD Power Management</title>
748<section> 801<section>
749<title>Energy consumer no. 1</title>
750<body> 802<body>
751 803
752<p> 804<p>
753As you can see in <uri link="#doc_chap1_fig1">figure 1.1</uri>, the LCD display 805As you can see in <uri link="#doc_chap1_fig1">figure 1.1</uri>, the LCD
754consumes the biggest part of energy (might not be the case for non-mobile 806display consumes the biggest part of energy (might not be the case for
755CPU's). Thus it's quite important not only to shut the display off when not 807non-mobile CPU's). Thus it's quite important not only to shut the display off
756needed, but also to reduce it's backlight if possible. Most laptops offer the 808when not needed, but also to reduce it's backlight if possible. Most laptops
757possibility to control the backlight dimming. 809offer the possibility to control the backlight dimming.
758</p>
759
760<p> 810</p>
811
812</body>
813</section>
814<section>
815<title>Standby settings</title>
816<body>
817
818<p>
761First thing to check is the standby/suspend/off timings of the display. As this 819The first thing to check is the standby/suspend/off timings of the display. As
762depends heavily on your windowmanager, I'll let you figure it out yourself. 820this depends heavily on your windowmanager, I'll let you figure it out
763Just two common places: Blanking the terminal can be done with <c>setterm 821yourself. Just two common places: Blanking the terminal can be done with
764-blank &lt;number-of-minutesM&gt;</c>, <c>setterm -powersave on</c> and 822<c>setterm -blank &lt;number-of-minutesM&gt;</c>, <c>setterm -powersave on</c>
765<c>setterm -powerdown &lt;number-of-minutesM&gt;</c>. 823and <c>setterm -powerdown &lt;number-of-minutesM&gt;</c>. For X.org, modify
766For X.org, modify <path>/etc/X11/xorg.conf</path> similar to this: 824<path>/etc/X11/xorg.conf</path> similar to this:
767</p> 825</p>
768 826
769<pre caption="LCD suspend settings in X.org and XFree86"> 827<pre caption="LCD suspend settings in X.org">
770Section "ServerLayout" 828Section "ServerFlags"
771 Identifier [...]
772 [...]
773 Option "BlankTime" "5" <comment># Blank the screen after 5 minutes (Fake)</comment> 829 Option "blank time" "5" <comment># Blank the screen after 5 minutes (Fake)</comment>
774 Option "StandbyTime" "10" <comment># Turn off screen after 10 minutes (DPMS)</comment> 830 Option "standby time" "10" <comment># Turn off screen after 10 minutes (DPMS)</comment>
775 Option "SuspendTime" "20" <comment># Full suspend after 20 minutes</comment> 831 Option "suspend time" "20" <comment># Full suspend after 20 minutes</comment>
776 Option "OffTime" "30" <comment># Turn off after half an hour</comment> 832 Option "off time" "30" <comment># Turn off after half an hour</comment>
777 [...] 833 [...]
778EndSection 834EndSection
779 835
780[...] 836[...]
781 837
782Section "Monitor" 838Section "Monitor"
783 Identifier [...] 839 Identifier [...]
784 Option "DPMS" "true" 840 Option "DPMS"
785 [...] 841 [...]
786EndSection 842EndSection
787</pre> 843</pre>
788 844
789<p> 845</body>
790This is the same for XFree86 and <path>/etc/X11/XF86Config</path>. 846</section>
791</p> 847<section>
848<title>Backlight dimming</title>
849<body>
792 850
793<p> 851<p>
794Probably more important is the backlight dimming. If you have access to the 852Probably more important is the backlight dimming. If you have access to the
795dimming settings via a tool, write a small script that dims the backlight in 853dimming settings via a tool, write a small script that dims the backlight in
796battery mode and place it in your <e>battery</e> runlevel. The following script 854battery mode and place it in your <c>battery</c> runlevel. The following script
797should work on most IBM Thinkpads. It needs the <c>app-laptop/ibm-acpi</c> 855should work on most IBM Thinkpads and Toshiba laptops. You've got to enable the
798package or the appropriate option in your kernel has to be enabled. 856appropriate option in your kernel (IBM Thinkpads only). For Toshiba laptops,
857install <c>sys-power/acpitool</c> and skip configuration of <c>thinkpad_acpi</c>
858(formerly called <c>ibm_acpi</c>) as described below.
799</p> 859</p>
800 860
801<warn> 861<warn>
802Support for setting brightness is marked experimental in ibm-acpi. It accesses 862Support for setting brightness is marked experimental in thinkpad_acpi. It accesses
803hardware directly and may cause severe harm to your system. Please read the 863hardware directly and may cause severe harm to your system. Please read the
804<uri link="http://ibm-acpi.sourceforge.net/">ibm-acpi website</uri> 864<uri link="http://ibm-acpi.sourceforge.net/">thinkpad_acpi website</uri>
805</warn> 865</warn>
806 866
807<p> 867<p>
808To be able to set the brightness level, the ibm_acpi module has to be loaded 868To be able to set the brightness level, the thinkpad_acpi module has to be loaded
809with the experimental parameter. 869with the experimental parameter.
810</p> 870</p>
811 871
812<pre caption="automatically loading the ibm_acpi module"> 872<pre caption="automatically loading the thinkpad_acpi module">
813<comment>(Please read the warnings above before doing this!)</comment> 873<comment>(Please read the warnings above before doing this!)</comment>
814<i># emerge ibm-acpi</i>
815<i># echo "options ibm_acpi experimental=1" >> /etc/modules.d/ibm_acpi</i> 874# <i>echo "options thinkpad_acpi experimental=1" >> /etc/modules.d/thinkpad_acpi</i>
816<i># /sbin/modules-update</i> 875# <i>/sbin/update-modules</i>
817<i># echo ibm_acpi >> /etc/modules.autoload.d/kernel-2.6</i> 876# <i>echo thinkpad_acpi >> /etc/modules.autoload.d/kernel-2.6</i>
818<i># modprobe ibm_acpi</i> 877# <i>modprobe thinkpad_acpi</i>
819</pre> 878</pre>
820 879
821<p> 880<p>
822This should work without error messages and a file 881This should work without error messages and a file
823<path>/proc/acpi/ibm/brightness</path> should be created after loading the 882<path>/proc/acpi/ibm/brightness</path> should be created after loading the
824module. An init script will take care of choosing the brightness according 883module. An init script will take care of choosing the brightness according to
825to the power source. 884the power source.
826</p> 885</p>
827 886
828<pre caption="/etc/conf.d/lcd-brightness"> 887<pre caption="/etc/conf.d/lcd-brightness">
829<comment># See /proc/acpi/ibm/brightness for available values</comment> 888<comment># See /proc/acpi/ibm/brightness for available values</comment>
830<comment># Please read /usr/share/doc/ibm-acpi-*/README.gz</comment> 889<comment># Please read /usr/src/linux/Documentation/thinkpad-acpi.txt</comment>
831 890
832<comment># brigthness level in ac mode. Default is 7.</comment> 891<comment># brightness level in ac mode. Default is 7.</comment>
833BRIGHTNESS_AC=7 892BRIGHTNESS_AC=7
834 893
835<comment># brightness level in battery mode. Default is 4.</comment> 894<comment># brightness level in battery mode. Default is 4.</comment>
836BRIGHTNESS_BATTERY=4 895BRIGHTNESS_BATTERY=4
837</pre> 896</pre>
850 if [ -f /proc/acpi/ibm/brightness ] 909 if [ -f /proc/acpi/ibm/brightness ]
851 then 910 then
852 ebegin "Setting LCD brightness" 911 ebegin "Setting LCD brightness"
853 echo "level ${LEVEL}" > /proc/acpi/ibm/brightness 912 echo "level ${LEVEL}" > /proc/acpi/ibm/brightness
854 eend $? 913 eend $?
914 elif [[ -e /usr/bin/acpitool &amp;&amp; -n $(acpitool -T | grep "LCD brightness") ]]
915 then
916 ebegin "Setting LCD brightness"
917 acpitool -l $LEVEL >/dev/null || ewarn "Unable to set lcd brightness"
918 eend $?
855 else 919 else
856 ewarn "Setting LCD brightness is not supported." 920 ewarn "Setting LCD brightness is not supported."
857 ewarn "Check that ibm_acpi is loaded into the kernel" 921 ewarn "For IBM Thinkpads, check that thinkpad_acpi is loaded into the kernel"
922 ewarn "For Toshiba laptops, you've got to install sys-power/acpitool"
858 fi 923 fi
859} 924}
860 925
861start() { 926start() {
862 set_brightness 927 set_brightness
871When done, make sure brightness is adjusted automatically by adding it to the 936When done, make sure brightness is adjusted automatically by adding it to the
872battery runlevel. 937battery runlevel.
873</p> 938</p>
874 939
875<pre caption="Enabling automatic brightness adjustment"> 940<pre caption="Enabling automatic brightness adjustment">
876<i># chmod +x /etc/init.d/lcd-brightness</i> 941# <i>chmod +x /etc/init.d/lcd-brightness</i>
877<i># rc-update add lcd-brightness battery</i> 942# <i>rc-update add lcd-brightness battery</i>
878<i># rc</i> 943# <i>rc</i>
879</pre> 944</pre>
880 945
881</body> 946</body>
882</section> 947</section>
883</chapter> 948</chapter>
884 949
885<chapter> 950<chapter>
886<title>Disk Power Management</title> 951<title>Disk Power Management</title>
887<section> 952<section>
888<title>Sleep when idle</title> 953<body>
954
955<p>
956Hard disks consume less energy in sleep mode. Therefore it makes sense to
957activate power saving features whenever the hard disk is not used for a certain
958amount of time. I'll show you two alternative possibilities to do it. First,
959laptop-mode will save most energy due to several measures which prevent or at
960least delay write accesses. The drawback is that due to the delayed write
961accesses a power outage or kernel crash will be more dangerous for data loss.
962If you don't like this, you have to make sure that there are no processes which
963write to your hard disk frequently. Afterwards you can enable power saving
964features of your hard disk with <c>hdparm</c> as the second alternative.
965</p>
966
889<body> 967</body>
968</section>
969<section>
970<title>Increasing idle time - laptop-mode</title>
971<body>
890 972
891<p>
892Let's bring the hard disk to sleep as early as possible whenever it is not
893needed. I'll show you two possibilities to do it. First <c>cpudyn</c> supports
894Disk Power Management. Uncomment the lines in the "Disk Options" section in
895<path>/etc/conf.d/cpudyn</path>. To put your first disk to sleep after 60
896seconds of no activity, you would modify it like this:
897</p> 973<p>
898 974Recent 2.6 kernels include the so-called <c>laptop-mode</c>. When activated,
899<pre caption="Using cpudyn for disk standby"> 975dirty buffers are written to disk on read calls or after 10 minutes (instead of
900<comment>################################################ 97630 seconds). This minimizes the time the hard disk needs to be spun up.
901# DISK OPTIONS
902# (disabled by default)
903################################################
904
905#
906# Timeout to put the disk in standby mode if there was no
907# io during that period (in seconds)
908#
909</comment>
910TIMEOUT=60
911<comment>
912#
913# Specified disks to spindown (comma separated devices)
914#
915</comment>
916DISKS=/dev/hda
917</pre>
918
919<p> 977</p>
920The second possibility is using a small script and hdparm. Create 978
921<path>/etc/init.d/pm.hda</path> like this: 979<pre caption="Automated start of laptop-mode">
980# <i>emerge laptop-mode-tools</i>
981</pre>
982
922</p> 983<p>
984<c>laptop-mode-tools</c> has its configuration file in
985<path>/etc/laptop-mode/laptop-mode.conf</path>. Adjust it the way you like it,
986it's well commented. Run <c>rc-update add laptop_mode battery</c> to start it
987automatically.
988</p>
923 989
990<p>
991Recent versions (1.11 and later) of laptop-mode-tools include a new tool
992<c>lm-profiler</c>. It will monitor your system's disk usage and running
993network services and suggests to disable unneeded ones. You can either disable
994them through laptop-mode-tools builtin runlevel support (which will be reverted
995by Gentoo's <c>/sbin/rc</c>) or use your <c>default</c>/<c>battery</c>
996runlevels (recommended).
997</p>
998
999<pre caption="Sample output from running lm-profiler">
1000# <i>lm-profiler</i>
1001Profiling session started.
1002Time remaining: 600 seconds
1003[4296896.602000] amarokapp
1004Time remaining: 599 seconds
1005[4296897.714000] sort
1006[4296897.970000] mv
1007Time remaining: 598 seconds
1008Time remaining: 597 seconds
1009[4296900.482000] reiserfs/0
1010</pre>
1011
1012<p>
1013After profiling your system for ten minutes, lm-profiler will present a list of
1014services which might have caused disk accesses during that time.
1015</p>
1016
1017<pre caption="lm-profiler suggests to disable some services">
1018Program: "atd"
1019Reason: standard recommendation (program may not be running)
1020Init script: /etc/init.d/atd (GUESSED)
1021
1022Do you want to disable this service in battery mode? [y/N]: <i>n</i>
1023</pre>
1024
1025<p>
1026To disable atd as suggested in the example above, you would run <c>rc-update
1027del atd battery</c>. Be careful not to disable services that are needed for
1028your system to run properly - <c>lm-profiler</c> is likely to generate some
1029false positives. Do not disable a service if you are unsure whether it's
1030needed.
1031</p>
1032
1033</body>
1034</section>
1035<section>
1036<title>Limiting Write Accesses</title>
1037<body>
1038
1039<p>
1040If you don't want to use laptop-mode, you must take special care to disable
1041services that write to your disk frequently - <c>syslogd</c> is a good
1042candidate, for example. You probably don't want to shut it down completely, but
1043it's possible to modify the config file so that "unnecessary" things don't get
1044logged and thus don't create disk traffic. <c>Cups</c> writes to disk
1045periodically, so consider shutting it down and only enable it manually when
1046needed.
1047</p>
1048
1049<pre caption="Disabling cups in battery mode">
1050# <i>rc-update del cupsd battery</i>
1051</pre>
1052
1053<p>
1054You can also use <c>lm-profiler</c> from laptop-mode-tools (see above) to find
1055services to disable. Once you eliminated all of them, go on with configuring
1056hdparm.
1057</p>
1058
1059</body>
1060</section>
1061<section>
1062<title>hdparm</title>
1063<body>
1064
1065<p>
1066The second possibility is using <c>hdparm</c>. Skip this if
1067you are using laptop-mode. Otherwise, edit <path>/etc/conf.d/hdparm</path> and
1068add the following values to your drive entries. This example assumes your hard
1069drive is called <b>hda</b>:
1070</p>
1071
924<pre caption="Using hdparm for disk standby"> 1072<pre caption="Using /etc/conf.d/hdparm for disk standby">
925#!/sbin/runscript 1073hda_args="-q -S12"
926
927depend() {
928 after hdparm
929}
930
931start() {
932 ebegin "Activating Power Management for Hard Drives"
933 hdparm -q -S12 /dev/hda
934 eend $?
935}
936
937stop () {
938 ebegin "Deactivating Power Management for Hard Drives"
939 hdparm -q -S253 /dev/hda
940 eend $?
941}
942</pre> 1074</pre>
943 1075
1076<p>
1077This will activate power management for your hard drive. If you ever want to
1078deactivate power management, you can edit <path>/etc/conf.d/hdparm</path> and
1079change the values to <c>-q -S0</c>, or just run <c>hdparm -q -S0 /dev/hda</c>.
944<p> 1080</p>
945See <c>man hdparm</c> for the options. If your script is ready, add it to the 1081
946battery runlevel. 1082<p>
1083See <c>man hdparm</c> for the options. Though you can always start <c>hdparm</c>
1084manually when you are on battery power by running <c>/etc/init.d/hdparm
1085start</c>, it's much easier to automate its startup and shutdown. To do so, add
1086<c>hdparm</c> to the battery runlevel so that it will automatically enable power
1087management.
947</p> 1088</p>
948 1089
949<pre caption="Automate disk standby settings"> 1090<pre caption="Automate disk standby settings">
950# <i>chmod +x /etc/init.d/pm.hda</i>
951# <i>/sbin/depscan.sh</i>
952# <i>rc-update add pm.hda battery</i> 1091# <i>rc-update add hdparm battery</i>
953</pre> 1092</pre>
954 1093
955<impo> 1094<impo>
956Be careful with sleep/spin down settings of your hard drive. Setting it to 1095Be careful with sleep/spin down settings of your hard drive. Setting it to
957small values might wear out your drive and lose warranty. 1096small values might wear out your drive and lose warranty.
958</impo> 1097</impo>
959 1098
960</body> 1099</body>
961</section> 1100</section>
962<section> 1101<section>
963<title>Increasing idle time - laptop-mode</title>
964<body>
965
966<p>
967Recent kernels (2.6.6 and greater, recent 2.4 ones and others with patches)
968include the so-called <e>laptop-mode</e>. When activated, dirty buffers are
969written to disk on read calls or after 10 minutes (instead of 30 seconds). This
970minimizes the time the hard disk needs to be spun up.
971</p>
972
973<pre caption="Automated start of laptop-mode">
974# <i>emerge laptop-mode-tools</i>
975</pre>
976
977<p>
978<c>laptop-mode-tools</c> has it's configuration file in
979<path>/etc/laptop-mode/laptop-mode.conf</path>. Adjust it the way you like it,
980it's well commented. Run <c>rc-update add laptop_mode battery</c> to start it
981automatically.
982</p>
983
984</body>
985</section>
986<section>
987<title>Other tricks</title> 1102<title>Other tricks</title>
988<body> 1103<body>
989
990<p>
991Besides putting your disk to sleep state as early as possible, it is a good
992idea to minimize disk accesses. Have a look at processes that write to your
993disk frequently - the syslogd is a good candidate. You probably don't want to
994shut it down completely, but it's possible to modify the config file so that
995"unnecessary" things don't get logged and thus don't create disk traffic. Cups
996writes to disk periodically, so consider shutting it down and only enable it
997manually when needed.
998</p>
999
1000<pre caption="Disabling cups in battery mode">
1001# <i>rc-update del cupsd battery</i>
1002</pre>
1003 1104
1004<p> 1105<p>
1005Another possibility is to deactivate swap in battery mode. Before writing a 1106Another possibility is to deactivate swap in battery mode. Before writing a
1006swapon/swapoff switcher, make sure there is enough RAM and swap isn't used 1107swapon/swapoff switcher, make sure there is enough RAM and swap isn't used
1007heavily, otherwise you'll be in big problems. 1108heavily, otherwise you'll be in big problems.
1008</p> 1109</p>
1009 1110
1010<p> 1111<p>
1011If you don't want to use laptop-mode, it's still possible to minimize disk 1112If you don't want to use laptop-mode, it's still possible to minimize disk
1012access by mounting certain directories as <e>tmpfs</e> - write accesses are not 1113access by mounting certain directories as <c>tmpfs</c> - write accesses are not
1013stored on a disk, but in main memory and get lost with unmounting. Often it's 1114stored on a disk, but in main memory and get lost with unmounting. Often it's
1014useful to mount <path>/tmp</path> like this - you don't have to pay special 1115useful to mount <path>/tmp</path> like this - you don't have to pay special
1015attention as it gets cleared on every reboot regardless whether it was mounted 1116attention as it gets cleared on every reboot regardless whether it was mounted
1016on disk or in RAM. Just make sure you have enough RAM and no program (like a 1117on disk or in RAM. Just make sure you have enough RAM and no program (like a
1017download client or compress utility) needs extraordinary much space in 1118download client or compress utility) needs extraordinary much space in
1023none /tmp tmpfs size=32m 0 0 1124none /tmp tmpfs size=32m 0 0
1024</pre> 1125</pre>
1025 1126
1026<warn> 1127<warn>
1027Pay attention to the size parameter and modify it for your system. If you're 1128Pay attention to the size parameter and modify it for your system. If you're
1028unsure, don't try this at all, it can become a perfomance bottleneck easily. In 1129unsure, don't try this at all, it can become a performance bottleneck easily. In
1029case you want to mount <path>/var/log</path> like this, make sure to merge the 1130case you want to mount <path>/var/log</path> like this, make sure to merge the
1030log files to disk before unmounting. They are essential. Don't attempt to mount 1131log files to disk before unmounting. They are essential. Don't attempt to mount
1031/var/tmp like this. Portage uses it for compiling... 1132<path>/var/tmp</path> like this. Portage uses it for compiling...
1032</warn> 1133</warn>
1033 1134
1034</body> 1135</body>
1035</section> 1136</section>
1036</chapter> 1137</chapter>
1037 1138
1038<chapter> 1139<chapter>
1039<title>Power Management for other devices</title> 1140<title>Power Management For Other Devices</title>
1040<section> 1141<section>
1041<title>Graphics cards</title> 1142<title>Graphics Cards</title>
1042<body> 1143<body>
1043 1144
1044<p> 1145<p>
1045In case you own an ATI graphics card supporting PowerPlay (dynamic clock 1146In case you own an ATI graphics card supporting PowerPlay (dynamic clock
1046scaling for the the graphics processing unit GPU), you can activate this 1147scaling for the graphics processing unit GPU), you can activate this
1047feature in X.org. Open <path>/etc/X11/xorg.conf</path> and add (or enable) 1148feature in X.org. Open <path>/etc/X11/xorg.conf</path> and add (or enable) the
1048the <c>DynamicClocks</c> option in the Device section. Please notice that 1149<c>DynamicClocks</c> option in the Device section. Please notice that this
1049this feature will lead to crashes on some systems. 1150feature will lead to crashes on some systems.
1050</p> 1151</p>
1051 1152
1052<pre caption="Enabling ATI PowerPlay support in X.org"> 1153<pre caption="Enabling ATI PowerPlay support in X.org">
1053Section "Device" 1154Section "Device"
1054[...] 1155[...]
1061<section> 1162<section>
1062<title>Wireless Power Management</title> 1163<title>Wireless Power Management</title>
1063<body> 1164<body>
1064 1165
1065<p> 1166<p>
1066Wireless LAN cards consume quite a few energy. Put them in Power Management 1167Wireless LAN cards consume quite a bit of energy. Put them in Power Management
1067mode in analogy to the pm.hda script. 1168mode just like your hard drives.
1068</p>
1069
1070<pre caption="WLAN Power Management automated">
1071#!/sbin/runscript
1072start() {
1073 ebegin "Activating Power Management for Wireless LAN"
1074 iwconfig wlan0 power on power max period 3
1075 eend $?
1076}
1077
1078stop () {
1079 ebegin "Deactivating Power Management for Wireless LAN"
1080 iwconfig wlan0 power off
1081 eend $?
1082}
1083</pre>
1084
1085<p> 1169</p>
1086Starting this script will put wlan0 in Power Management mode, going to sleep at 1170
1087the latest three seconds after no traffic. 1171<note>
1088Save it as <path>/etc/init.d/pm.wlan0</path> and add it to the battery runlevel 1172This script assumes your wireless interface is called <c>wlan0</c>; replace
1089like the disk script above. See <c>man iwconfig</c> for details and more 1173this with the actual name of your interface.
1090options. If your driver and access point support changing the beacon time, this 1174</note>
1091is a good starting point to save even more energy. 1175
1092</p> 1176<p>
1093 1177Add the following option to <path>/etc/conf.d/net</path> to automatically enable
1094<pre caption="Power Management for WLAN"> 1178power management for your wireless card:
1095# <i>chmod +x /etc/init.d/pm.wlan0</i>
1096# <i>/sbin/depscan.sh</i>
1097# <i>rc-update add pm.wlan0 battery</i>
1098</pre> 1179</p>
1180
1181<pre caption="Automated WLAN Power Management">
1182iwconfig_wlan0="power on"
1183</pre>
1184
1185<p>
1186See <c>man iwconfig</c> for details and more options like the period between
1187wakeups or timeout settings. If your driver and access point support changing
1188the beacon time, this is a good starting point to save even more energy.
1189</p>
1099 1190
1100</body> 1191</body>
1101</section> 1192</section>
1102<section> 1193<section>
1103<title>USB Power Management</title> 1194<title>USB Power Management</title>
1107There are two problems with USB devices regarding energy consumption: First, 1198There are two problems with USB devices regarding energy consumption: First,
1108devices like USB mice, digital cameras or USB sticks consume energy while 1199devices like USB mice, digital cameras or USB sticks consume energy while
1109plugged in. You cannot avoid this (nevertheless remove them in case they're not 1200plugged in. You cannot avoid this (nevertheless remove them in case they're not
1110needed). Second, when there are USB devices plugged in, the USB host controller 1201needed). Second, when there are USB devices plugged in, the USB host controller
1111periodically accesses the bus which in turn prevents the CPU from going into 1202periodically accesses the bus which in turn prevents the CPU from going into
1112sleep mode. The kernel offers an experimental option to enable suspension of 1203sleep mode. The kernel offers an experimental option to enable suspension of
1113USB devices through driver calls or one of the <path>power/state</path> files 1204USB devices through driver calls or one of the <path>power/state</path> files
1114in <path>/sys</path>. 1205in <path>/sys</path>.
1115</p> 1206</p>
1116 1207
1117<pre caption="Enabling USB suspend support in the kernel"> 1208<pre caption="Enabling USB suspend support in the kernel">
1124</body> 1215</body>
1125</section> 1216</section>
1126</chapter> 1217</chapter>
1127 1218
1128<chapter> 1219<chapter>
1129<title>Sleep states: sleep, standby, suspend to disk</title> 1220<title>Sleep States: sleep, standby, and suspend to disk</title>
1130<section> 1221<section>
1131<title>Overview</title>
1132<body> 1222<body>
1133 1223
1134<p> 1224<p>
1135ACPI defines different sleep states. The more important ones are 1225ACPI defines different sleep states. The more important ones are
1136</p> 1226</p>
1137 1227
1138<ul> 1228<ul>
1139 <li>S1 aka Standby</li> 1229 <li>S1 aka Standby</li>
1140 <li>S3 aka Suspend to RAM aka Sleep</li> 1230 <li>S3 aka Suspend to RAM aka Sleep</li>
1141 <li>S4 aka Suspend to Disk aka Hibernate</li> 1231 <li>S4 aka Suspend to Disk aka Hibernate</li>
1142</ul> 1232</ul>
1143 1233
1165 [*] ACPI Support 1255 [*] ACPI Support
1166 [*] Sleep States 1256 [*] Sleep States
1167</pre> 1257</pre>
1168 1258
1169<p> 1259<p>
1170Once your kernel is prepared like above, you can use the 1260Once your kernel is properly configured, you can use the
1171<c>hibernate-script</c> to activate suspend or sleep mode. Let's install that 1261<c>hibernate-script</c> to activate suspend or sleep mode. Let's install that
1172first. 1262first.
1173</p> 1263</p>
1174 1264
1175<pre caption="Installing the hibernate-script"> 1265<pre caption="Installing the hibernate-script">
1176<i># emerge hibernate-script</i> 1266# <i>emerge hibernate-script</i>
1177</pre> 1267</pre>
1178 1268
1179<p> 1269<p>
1180Some configuration has to be done in <path>/etc/hibernate</path> The default 1270Some configuration has to be done in <path>/etc/hibernate</path>. The default
1181package introduces two configuration files <path>hibernate.conf</path> and 1271package introduces a few configuration files for each sleep state. Options that
1182<path>ram.conf</path>. 1272are common to all suspend methods are placed in <path>common.conf</path>; make
1183</p> 1273sure this file is properly set up for your system.
1184
1185<p> 1274</p>
1186To configure sleep, edit <path>ram.conf</path> in <path>/etc/hibernate</path>. 1275
1187<c>UseSysfsPowerState mem</c> is already setup correctly, but you have to go
1188through the rest of the configuration file and set it up for your system. The
1189comments and option names will guide you. If you use nfs or samba shares over
1190the network, make sure to shutdown the appropriate init scripts to avoid
1191timeouts.
1192</p> 1276<p>
1277To configure sleep, edit <path>sysfs-ram.conf</path> in
1278<path>/etc/hibernate</path>. <c>UseSysfsPowerState mem</c> is already setup
1279correctly, but if you need to make further changes to this particular sleep
1280state (or any other sleep state) you should add them to
1281<path>/etc/hibernate/hibernate.conf</path>. The comments and option names will
1282guide you. If you use nfs or samba shares over the network, make sure to
1283shutdown the appropriate init scripts to avoid timeouts.
1284</p>
1285
1286<note>
1287For more information on setting up sleep states, read <c>man
1288hibernate.conf</c>.
1289</note>
1193 1290
1194<p> 1291<p>
1195Ready? Now is the last chance to backup any data you want to keep after 1292Ready? Now is the last chance to backup any data you want to keep after
1196executing the next command. Notice that you probably have to hit a special key 1293executing the next command. Notice that you probably have to hit a special key
1197like <e>Fn</e> to resume from sleep. 1294like <c>Fn</c> to resume from sleep.
1198</p> 1295</p>
1199 1296
1200<pre caption="Calling sleep"> 1297<pre caption="Calling sleep">
1201<i># hibernate-ram</i> 1298# <i>hibernate-ram</i>
1202</pre> 1299</pre>
1203 1300
1204<p> 1301<p>
1205If you're still reading, it seems to work. You can also setup standby (S1) in 1302If you're still reading, it seems to work. You can also setup standby (S1) in a
1206a similar way by copying <path>ram.conf</path> to <path>standby.conf</path> 1303similar way by editing <path>sysfs-ram.conf</path> and changing
1207and creating a symlink <path>/usr/sbin/hibernate-standby</path> pointing to 1304"UseSysfsPowerState mem" to "UseSysfsPowerState standby". S3 and S4 are the more
1208<path>/usr/sbin/hibernate</path>. S3 and S4 are the more interesting sleep
1209states due to greater energy savings however. 1305interesting sleep states due to greater energy savings however.
1210</p> 1306</p>
1211 1307
1212</body> 1308</body>
1213</section> 1309</section>
1214<section> 1310<section>
1227Shutdown any NFS or samba server/client before hibernating. 1323Shutdown any NFS or samba server/client before hibernating.
1228</warn> 1324</warn>
1229 1325
1230<p> 1326<p>
1231There are two different implementations for S4. The original one is swsusp, 1327There are two different implementations for S4. The original one is swsusp,
1232then there is the newer suspend2 with a nicer interface (including 1328then there is the newer tuxonice (former suspend2) with a nicer interface (including fbsplash
1233fbsplash support). A <uri link="http://suspend2.net/features.html#compare"> 1329support). A <uri link="http://tuxonice.net/features.html#compare"> feature
1234feature comparison</uri> is available at the <uri link="http://suspend2.net"> 1330comparison</uri> is available at the <uri link="http://www.tuxonice.net">tuxonice
1235suspend2 Homepage</uri>. There used to be Suspend-to-Disk (pmdisk), a fork of 1331Homepage</uri>. There used to be Suspend-to-Disk (pmdisk), a fork of swsusp,
1236swsusp, but it has been merged back. 1332but it has been merged back.
1237</p>
1238
1239<p> 1333</p>
1334
1335<p>
1240Suspend2 is not included in the mainline kernel yet, therefore you either have 1336TuxOnIce is not included in the mainline kernel yet, therefore you either have
1241to patch your kernel sources with the patches provided by 1337to patch your kernel sources with the patches provided by <uri
1242<uri link="http://suspend2.net">suspend2.net</uri> or use 1338link="http://tuxonice.net">tuxonice.net</uri> or use
1243<c>sys-kernel/suspend2-sources</c>. 1339<c>sys-kernel/tuxonice-sources</c>.
1244</p>
1245
1246<p> 1340</p>
1341
1342<p>
1247The kernel part for both swusp and suspend2 is as follows: 1343The kernel part for both swusp and TuxOnIce is as follows:
1248</p> 1344</p>
1249 1345
1250<pre caption="Kernel configuration for the various suspend types"> 1346<pre caption="Kernel configuration for the various suspend types">
1251Power Management Options ---&gt; 1347Power Management Options ---&gt;
1252 <comment>(hibernate with swsusp)</comment> 1348 <comment>(hibernate with swsusp)</comment>
1253 [*] Software Suspend 1349 [*] Software Suspend
1254 <comment>(replace /dev/SWAP with your swap partition)</comment> 1350 <comment>(replace /dev/SWAP with your swap partition)</comment>
1255 (/dev/SWAP) Default resume partition 1351 (/dev/SWAP) Default resume partition
1256 1352
1257 <comment>(hibernate with suspend2)</comment> 1353 <comment>(hibernate with TuxOnIce)</comment>
1258 Software Suspend 2 1354 Enhanced Hibernation (TuxOnIce)
1259 --- Image Storage (you need at least one writer) 1355 --- Image Storage (you need at least one writer)
1260 [*] File Writer 1356 [*] File Writer
1261 [*] Swap Writer 1357 [*] Swap Writer
1262 --- General Options 1358 --- General Options
1263 [*] LZF image compression 1359 [*] LZF image compression
1269<p> 1365<p>
1270The configuration for swsusp is rather easy. If you didn't store the location 1366The configuration for swsusp is rather easy. If you didn't store the location
1271of your swap partition in the kernel config, you can also pass it as a 1367of your swap partition in the kernel config, you can also pass it as a
1272parameter with the <c>resume=/dev/SWAP</c> directive. If booting is not 1368parameter with the <c>resume=/dev/SWAP</c> directive. If booting is not
1273possible due to a broken image, use the <c>noresume</c> kernel parameter. The 1369possible due to a broken image, use the <c>noresume</c> kernel parameter. The
1274<c>hibernate-cleanup</c> init script invalidates swsusp images during the 1370<c>hibernate-cleanup</c> init script invalidates swsusp images during the boot
1275boot process. 1371process.
1276</p> 1372</p>
1277 1373
1278<pre caption="Invalidating swsusp images during the boot process"> 1374<pre caption="Invalidating swsusp images during the boot process">
1279<i># rc-update add hibernate-cleanup boot</i> 1375# <i>rc-update add hibernate-cleanup boot</i>
1280</pre> 1376</pre>
1281 1377
1282<p> 1378<p>
1283To activate hibernate with swsusp, use the hibernate script and set 1379To activate hibernate with swsusp, use the hibernate script and set
1284<c>UseSysfsPowerState disk</c> in <path>/etc/hibernate/hibernate.conf</path>. 1380<c>UseSysfsPowerState disk</c> in <path>/etc/hibernate/sysfs-disk</path>.
1285</p> 1381</p>
1286 1382
1287<warn> 1383<warn>
1288Backup your data before doing this. Run <c>sync</c> before executing one of the 1384Backup your data before doing this. Run <c>sync</c> before executing one of the
1289commands to have cached data written to disk. First try it outside of X, then 1385commands to have cached data written to disk. First try it outside of X, then
1290with X running, but not logged in. 1386with X running, but not logged in.
1291</warn> 1387</warn>
1292 1388
1293<p> 1389<p>
1294If you experience kernel panics due to uhci or similar, try to compile USB 1390If you experience kernel panics due to uhci or similar, try to compile USB
1295support as module and unload the modules before sending your laptop to sleep 1391support as module and unload the modules before sending your laptop to sleep
1296mode. There are configuration options for this in <path>hibernate.conf</path> 1392mode. There are configuration options for this in <path>common.conf</path>
1297</p> 1393</p>
1298 1394
1299<pre caption="Hibernating with swsusp"> 1395<pre caption="Hibernating with swsusp">
1300<i># nano -w /etc/hibernate.conf</i> 1396# <i>nano -w /etc/hibernate/common.conf</i>
1301<comment>(Make sure you have a backup of your data)</comment> 1397<comment>(Make sure you have a backup of your data)</comment>
1302<i># hibernate</i> 1398# <i>hibernate</i>
1303</pre> 1399</pre>
1304 1400
1305<p> 1401<p>
1306The following section discusses the setup of suspend2 including fbsplash 1402The following section discusses the setup of TuxOnIce including fbsplash support
1307support for a nice graphical progress bar during suspend and resume. 1403for a nice graphical progress bar during suspend and resume.
1308</p>
1309
1310<p> 1404</p>
1405
1406<p>
1311The first part of the configuration is similar to the configuration of 1407The first part of the configuration is similar to the configuration of swsusp.
1312swsusp. In case you didn't store the location of your swap partition in the 1408In case you didn't store the location of your swap partition in the kernel
1313kernel config, you have to pass it as a kernel parameter with the 1409config, you have to pass it as a kernel parameter with the
1314<c>resume2=swap:/dev/SWAP</c> directive. If booting is not possible due to a 1410<c>resume=swap:/dev/SWAP</c> directive. If booting is not possible due to a
1315broken image, append the <c>noresume2</c> parameter. Additionally, the 1411broken image, append the <c>noresume</c> parameter. Additionally, the
1316<c>hibernate-cleanup</c> init script invalidates suspend2 images during the 1412<c>hibernate-cleanup</c> init script invalidates TuxOnIce images during the boot
1317boot process. 1413process.
1318</p> 1414</p>
1319 1415
1320<pre caption="Invalidating suspend2 images during the boot process"> 1416<pre caption="Invalidating TuxOnIce images during the boot process">
1321<i># rc-update add hibernate-cleanup boot</i> 1417# <i>rc-update add hibernate-cleanup boot</i>
1322</pre> 1418</pre>
1323 1419
1324<p>Now edit <path>/etc/hibernate/hibernate.conf</path>, enable the
1325<e>suspend2</e> section and comment everything in the <e>sysfs_power_state</e>
1326and <e>acpi_sleep</e> sections. Do not enable the fbsplash part in global
1327options yet.
1328</p> 1420<p>
1421Now edit <path>/etc/hibernate/suspend2.conf</path>, enable the <c>TuxOnIce</c>
1422options you need. Do not enable the <c>fbsplash</c> options in
1423<c>common.conf</c> just yet.
1424</p>
1329 1425
1330<pre caption="Hibernating with suspend2"> 1426<pre caption="Hibernating with TuxOnIce">
1331<i># nano -w /etc/hibernate.conf</i> 1427# <i>nano -w /etc/hibernate/suspend2.conf</i>
1332<comment>(Make sure you have a backup of your data)</comment> 1428<comment>(Make sure you have a backup of your data)</comment>
1333<i># hibernate</i> 1429# <i>hibernate</i>
1334</pre> 1430</pre>
1335 1431
1336<p> 1432<p>
1337Please configure fbsplash now if you didn't do already. To enable fbsplash 1433Please configure <c>fbsplash</c> now if you didn't do already. To enable
1338support during hibernation, the <c>sys-apps/suspend2-userui</c> package is 1434fbsplash support during hibernation, the <c>sys-apps/tuxonice-userui</c> package
1339needed. Additionally, you've got to enable the <e>fbsplash</e> USE flag. 1435is needed. Additionally, you've got to enable the <c>fbsplash</c> USE flag.
1340</p> 1436</p>
1341 1437
1342<pre caption="Installing suspend2-userui"> 1438<pre caption="Installing tuxonice-userui">
1343<i># mkdir -p /etc/portage</i>
1344<i># echo sys-apps/suspend2-userui fbsplash >> /etc/portage/package.use</i> 1439# <i>echo "sys-apps/tuxonice-userui fbsplash" >> /etc/portage/package.use</i>
1345<i># emerge suspend2-userui</i> 1440<comment>(It may be marked ~arch, so first it must be keyworded)</comment>
1441# <i>echo "sys-apps/tuxonice-userui" >> /etc/portage/package.keywords</i>
1442# <i>emerge tuxonice-userui</i>
1346</pre> 1443</pre>
1347 1444
1348<p> 1445<p>
1349The ebuild tells you to make a symlink to the theme you want to use. For 1446The ebuild tells you to make a symlink to the theme you want to use. For
1350example, to use the <c>livecd-2005.1</c> theme, run the following command: 1447example, to use the <c>livecd-2005.1</c> theme, run the following command:
1351</p> 1448</p>
1352 1449
1353<pre caption="Using the livecd-2005.1 theme during hibernation"> 1450<pre caption="Using the livecd-2005.1 theme during hibernation">
1354<i># ln -sfn /etc/splash/livecd-2005.1 /etc/splash/suspend2</i> 1451# <i>ln -sfn /etc/splash/livecd-2005.1 /etc/splash/suspend2</i>
1355</pre> 1452</pre>
1356 1453
1357<p> 1454<p>
1358If you don't want a black screen in the first part of the resume process, you 1455If you don't want a black screen in the first part of the resume process, you
1359have to add the <c>suspend2ui_fbsplash</c> tool to your initrd image. Assuming 1456have to add the <c>tuxoniceui_fbsplash</c> tool to your initrd image. Assuming
1360you created the initrd image with <c>splash_geninitramfs</c> and saved it as 1457you created the initrd image with <c>splash_geninitramfs</c> and saved it as
1361<path>/boot/fbsplash-emergence-1024x768</path>, here's how to do 1458<path>/boot/fbsplash-emergence-1024x768</path>, here's how to do that.
1362that.
1363</p> 1459</p>
1364 1460
1365<pre caption="Adding suspend2ui_fbsplash to an initrd image"> 1461<pre caption="Adding tuxoniceui_fbsplash to an initrd image">
1366<i># mount /boot</i> 1462# <i>mount /boot</i>
1367<i># mkdir ~/initrd.d</i> 1463# <i>mkdir ~/initrd.d</i>
1368<i># cp /boot/fbsplash-emergence-1024x768 ~/initrd.d/</i> 1464# <i>cp /boot/fbsplash-emergence-1024x768 ~/initrd.d/</i>
1369<i># cd ~/initrd.d</i> 1465# <i>cd ~/initrd.d</i>
1370<i># gunzip -c fbsplash-emergence-1024x768 | cpio -idm --quiet -H newc</i> 1466# <i>gunzip -c fbsplash-emergence-1024x768 | cpio -idm --quiet -H newc</i>
1371<i># rm fbsplash-emergence-1024x768</i> 1467# <i>rm fbsplash-emergence-1024x768</i>
1372<i># cp /usr/sbin/suspend2ui_fbsplash sbin/</i> 1468# <i>cp /usr/sbin/tuxoniceui_fbsplash sbin/</i>
1373<i># find . | cpio --quiet --dereference -o -H newc | gzip -9 > /boot/fbsplash-suspend2-emergence-1024x768</i> 1469# <i>find . | cpio --quiet --dereference -o -H newc | gzip -9 > /boot/fbsplash-tuxonice-emergence-1024x768</i>
1374</pre> 1470</pre>
1375 1471
1376<p> 1472<p>
1377Afterwards adjust <path>grub.conf</path> respectively <path>lilo.conf</path> 1473Afterwards adjust <path>grub.conf</path> (or <path>lilo.conf</path>) so that
1378so that your suspend2 kernel uses 1474your TuxOnIce kernel uses
1379<path>/boot/fbsplash-suspend2-emergence-1024x768</path> as initrd image. You 1475<path>/boot/fbsplash-tuxonice-emergence-1024x768</path> as initrd image. You can
1380can now test a dry run to see if everything is setup correctly. 1476now test a dry run to see if everything is setup correctly.
1381</p> 1477</p>
1382 1478
1383<pre caption="Test run for fbsplash hibernation"> 1479<pre caption="Test run for fbsplash hibernation">
1384<i># suspend2ui_fbsplash -t</i> 1480# <i>tuxoniceui_fbsplash -t</i>
1385</pre> 1481</pre>
1386 1482
1387<p> 1483<p>
1388Afterwards open <path>/etc/hibernate/hibernate.conf</path> again and activate 1484Afterwards open <path>/etc/hibernate/common.conf</path> and activate the
1389the fbsplash options. Execute <c>hibernate</c> and enjoy. 1485fbsplash options. Execute <c>hibernate</c> and enjoy.
1390</p> 1486</p>
1391 1487
1392</body> 1488</body>
1393</section> 1489</section>
1394</chapter> 1490</chapter>
1395 1491
1396<chapter> 1492<chapter>
1397<title>Troubleshooting</title> 1493<title>Troubleshooting</title>
1398<section> 1494<section>
1399<title>If things go wrong...</title>
1400<body> 1495<body>
1401 1496
1402<p> 1497<p>
1403<e>Q:</e> I'm trying to change the CPU frequency, but 1498<e>Q:</e> I'm trying to change the CPU frequency, but
1404<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</path> does not 1499<path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</path> does not
1406</p> 1501</p>
1407 1502
1408<p> 1503<p>
1409<e>A:</e> Make sure your processor supports CPU frequency scaling and you chose 1504<e>A:</e> Make sure your processor supports CPU frequency scaling and you chose
1410the right CPUFreq driver for your processor. Here is a list of processors that 1505the right CPUFreq driver for your processor. Here is a list of processors that
1411are supported by cpufreq (kernel 2.6.7): ARM Integrator, ARM-SA1100, 1506are supported by cpufreq (kernel 2.6.7): ARM Integrator, ARM-SA1100, ARM-SA1110,
1412ARM-SA1110, AMD Elan - SC400, SC410, AMD mobile K6-2+, AMD mobile K6-3+, AMD 1507AMD Elan - SC400, SC410, AMD mobile K6-2+, AMD mobile K6-3+, AMD mobile Duron,
1413mobile Duron, AMD mobile Athlon, AMD Opteron, AMD Athlon 64, Cyrix Media GXm, 1508AMD mobile Athlon, AMD Opteron, AMD Athlon 64, Cyrix Media GXm, Intel mobile
1414Intel mobile PIII and Intel mobile PIII-M on certain chipsets, Intel Pentium 4, 1509PIII and Intel mobile PIII-M on certain chipsets, Intel Pentium 4, Intel Xeon,
1415Intel Xeon, Intel Pentium M (Centrino), National Semiconductors Geode GX, 1510Intel Pentium M (Centrino), National Semiconductors Geode GX, Transmeta Crusoe,
1416Transmeta Crusoe, VIA Cyrix 3 / C3, UltraSPARC-III, SuperH SH-3, SH-4, several 1511VIA Cyrix 3 / C3, UltraSPARC-III, SuperH SH-3, SH-4, several "PowerBook" and
1417"PowerBook" and "iBook2" and various processors on some ACPI 2.0-compatible 1512"iBook2" and various processors on some ACPI 2.0-compatible systems (only if
1418systems (only if "ACPI Processor Performance States" are available to the 1513"ACPI Processor Performance States" are available to the ACPI/BIOS interface).
1419ACPI/BIOS interface).
1420</p> 1514</p>
1421 1515
1422<p> 1516<p>
1423<e>Q:</e> My laptop supports frequency scaling, but 1517<e>Q:</e> My laptop supports frequency scaling, but
1424<path>/sys/devices/system/cpu/cpu0/cpufreq/</path> is empty. 1518<path>/sys/devices/system/cpu/cpu0/cpufreq/</path> is empty.
1429Try to update the BIOS, especially if a broken DSDT is reported. You can also 1523Try to update the BIOS, especially if a broken DSDT is reported. You can also
1430try to fix it yourself (which is beyond the scope of this guide). 1524try to fix it yourself (which is beyond the scope of this guide).
1431</p> 1525</p>
1432 1526
1433<p> 1527<p>
1434<e>Q:</e> My laptop supports frequency scaling, but according to /proc/cpuinfo 1528<e>Q:</e> My laptop supports frequency scaling, but according to
1435the speed never changes. 1529<path>/proc/cpuinfo</path> the speed never changes.
1436</p> 1530</p>
1437 1531
1438<p> 1532<p>
1439<e>A:</e> Probably you have activated symmetric multiprocessing support 1533<e>A:</e> Probably you have activated symmetric multiprocessing support
1440(CONFIG_SMP) in your kernel. Deactivate it and it should work. Some older 1534(CONFIG_SMP) in your kernel. Deactivate it and it should work. Some older
1441kernels had a bug causing this. In that case, run <c>emerge x86info</c>, 1535kernels had a bug causing this. In that case, run <c>emerge x86info</c>, update
1442update your kernel as asked and check the current frequency with 1536your kernel as asked and check the current frequency with <c>x86info -mhz</c>.
1443<c>x86info -mhz</c>.
1444</p> 1537</p>
1445 1538
1446<p> 1539<p>
1447<e>Q:</e> I can change the CPU frequency, but the range is not as wide as in 1540<e>Q:</e> I can change the CPU frequency, but the range is not as wide as in
1448another OS. 1541another OS.
1449</p> 1542</p>
1450 1543
1451<p> 1544<p>
1452<e>A:</e> You can combine frequency scaling with ACPI throttling to get a lower 1545<e>A:</e> You can combine frequency scaling with ACPI throttling to get a lower
1453minimum frequency. Notice that throttling doesn't save much energy and is 1546minimum frequency. Notice that throttling doesn't save much energy and is mainly
1454mainly used for thermal management (keeping your laptop cool and quiet). You 1547used for thermal management (keeping your laptop cool and quiet). You can read
1455can read the current throttling state with <c>cat 1548the current throttling state with <c>cat /proc/acpi/processor/CPU/throttling</c>
1456/proc/acpi/processor/CPU/throttling</c> and change it with <c>echo -n "0:x" > 1549and change it with <c>echo -n "0:x" > /proc/acpi/processor/CPU/limit</c>, where
1457/proc/acpi/processor/CPU/limit</c>, where x is one of the Tx states listed in 1550x is one of the Tx states listed in
1458<path>/proc/acpi/processor/CPU/throttling</path>. 1551<path>/proc/acpi/processor/CPU/throttling</path>.
1459</p> 1552</p>
1460 1553
1461<p> 1554<p>
1462<e>Q:</e> When configuring the kernel, powersave, performance and userspace 1555<e>Q:</e> When configuring the kernel, powersave, performance and userspace
1486<e>A:</e> Check that battery support is compiled into your kernel. If you use 1579<e>A:</e> Check that battery support is compiled into your kernel. If you use
1487it as a module, make sure the module is loaded. 1580it as a module, make sure the module is loaded.
1488</p> 1581</p>
1489 1582
1490<p> 1583<p>
1584<e>Q:</e> My system logger reports things like "logger: ACPI group battery /
1585action battery is not defined".
1586</p>
1587
1588<p>
1589<e>A:</e> This message is generated by the <path>/etc/acpi/default.sh</path>
1590script that is shipped with acpid. You can safely ignore it. If you like to get
1591rid of it, you can comment the appropriate line in
1592<path>/etc/acpi/default.sh</path> as shown below:
1593</p>
1594
1595<pre caption="Disabling warnings about unknown acpi events">
1596 *) # logger "ACPI action $action is not defined"
1597</pre>
1598
1599<p>
1491<e>Q:</e> I have a Dell Inspiron 51XX and I don't get any ACPI events. 1600<e>Q:</e> I have a Dell Inspiron 51XX and I don't get any ACPI events.
1492</p> 1601</p>
1493 1602
1494<p> 1603<p>
1495<e>A:</e> This seems to be a kernel bug. Read on <uri 1604<e>A:</e> This seems to be a kernel bug. Read on <uri
1496link="http://bugme.osdl.org/show_bug.cgi?id=1752">here</uri>. 1605link="http://bugme.osdl.org/show_bug.cgi?id=1752">here</uri>.
1497</p> 1606</p>
1498 1607
1499<p> 1608<p>
1500<e>Q:</e> I activated the DynamicClocks option in <path>xorg.conf</path> and 1609<e>Q:</e> I activated the <c>DynamicClocks</c> option in <path>xorg.conf</path>
1501now X.org crashes / the screen stays black / my laptop doesn't shutdown 1610and now X.org crashes / the screen stays black / my laptop doesn't shutdown
1502properly. 1611properly.
1503</p> 1612</p>
1504 1613
1505<p> 1614<p>
1506<e>A:</e> This happens on some systems. You have to disable DynamicClocks. 1615<e>A:</e> This happens on some systems. You have to disable
1507</p> 1616<c>DynamicClocks</c>.
1508
1509<p> 1617</p>
1618
1619<p>
1510<e>Q:</e> I want to use suspend2, but it tells me my swap partition is too 1620<e>Q:</e> I want to use TuxOnIce, but it tells me my swap partition is too
1511small. Resizing is not an option. 1621small. Resizing is not an option.
1512</p> 1622</p>
1513 1623
1514<p> 1624<p>
1515<e>A:</e> If there is enough free space on your system, you can use the 1625<e>A:</e> If there is enough free space on your system, you can use the
1516filewriter instead of the swapwriter. The <c>hibernate-script</c> supports it 1626filewriter instead of the swapwriter. The <c>hibernate-script</c> supports it as
1517as well. More information can be found in 1627well. More information can be found in
1518<path>/usr/src/linux/Documentation/power/suspend2.txt</path>. 1628<path>/usr/src/linux/Documentation/power/tuxonice.txt</path>.
1519</p> 1629</p>
1520 1630
1521<p> 1631<p>
1522<e>Q:</e> I just bought a brand new battery, but it only lasts for some 1632<e>Q:</e> I just bought a brand new battery, but it only lasts for some
1523minutes! What am I doing wrong? 1633minutes! What am I doing wrong?
1524</p> 1634</p>
1525 1635
1526<p> 1636<p>
1527<e>A:</e> First follow your manufacturer's advice on how to charge the battery 1637<e>A:</e> First follow your manufacturer's advice on how to charge the battery
1528correctly. 1638correctly.
1529</p> 1639</p>
1530 1640
1531<p> 1641<p>
1532<e>Q:</e> The above didn't help. What should I do then? 1642<e>Q:</e> The above didn't help. What should I do then?
1533</p> 1643</p>
1550<p> 1660<p>
1551<e>Q:</e> My problem is not listed above. Where should I go next? 1661<e>Q:</e> My problem is not listed above. Where should I go next?
1552</p> 1662</p>
1553 1663
1554<p> 1664<p>
1555<e>A:</e> Don't fear to contact me, <mail link="fragfred@gmx.de">Dennis 1665<e>A:</e> Don't fear to contact me, <mail link="earthwings@gentoo.org">Dennis
1556Nienhüser</mail>, directly. 1666Nienhüser</mail>, directly. The <uri link="http://forums.gentoo.org">Gentoo
1667Forums</uri> are a good place to get help as well. If you prefer IRC, try the
1668<c>#gentoo-laptop</c> channel at <uri
1669link="irc://irc.freenode.net">irc.freenode.net</uri>.
1557</p> 1670</p>
1558 1671
1559</body> 1672</body>
1560</section> 1673</section>
1561</chapter> 1674</chapter>

Legend:
Removed from v.1.15  
changed lines
  Added in v.1.35

  ViewVC Help
Powered by ViewVC 1.1.20