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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1.18 - (show annotations) (download) (as text)
Thu Feb 16 18:45:09 2006 UTC (8 years, 7 months ago) by nightmorph
Branch: MAIN
Changes since 1.17: +4 -2 lines
File MIME type: application/xml
Fixed <no cpufreqd found> error for bug 123062

1 <?xml version='1.0' encoding="UTF-8"?>
2 <!DOCTYPE guide SYSTEM "/dtd/guide.dtd">
3 <!-- $Header: $ -->
4 <guide link="/doc/en/power-management-guide.xml">
5 <title>Power Management Guide</title>
6
7 <author title="Author">
8 <mail link="earthwings@gentoo.org">Dennis Nienhüser</mail>
9 </author>
10
11 <abstract>
12 Power Management is the key to extend battery run time on mobile systems like
13 laptops. This guide assists you setting it up on your laptop.
14 </abstract>
15
16 <!-- The content of this document is licensed under the CC-BY-SA license -->
17 <!-- See http://creativecommons.org/licenses/by-sa/2.5 -->
18 <license/>
19
20 <version>1.27</version>
21 <date>2006-02-16</date>
22
23 <chapter>
24 <title>Introduction</title>
25 <section>
26 <body>
27
28 <p>
29 Capacity and lifetime of laptop batteries have improved much in the last years.
30 Nevertheless modern processors consume much more energy than older ones and
31 each laptop generation introduces more devices hungry for energy. That's why
32 Power Management is more important than ever. Increasing battery run time
33 doesn't necessarily mean buying another battery. Much can be achieved applying
34 intelligent Power Management policies.
35 </p>
36
37 </body>
38 </section>
39
40 <section>
41 <title>A quick overview</title>
42 <body>
43
44 <p>
45 Please notice that this guide describes Power Management for <e>laptops</e>.
46 While some sections might also suite for <e>servers</e>, others do not and may
47 even cause harm. Please do not apply anything from this guide to a server
48 unless you really know what you are doing.
49 </p>
50
51 <p>
52 As this guide has become rather long, here's a short overview helping you to
53 find your way through it.
54 </p>
55
56 <p>
57 The <e>Prerequisites</e> chapter talks about some requirements that should be
58 met before any of the following device individual sections will work. This
59 includes BIOS settings, kernel configuration and some simplifications in user
60 land. The following three chapters focus on devices that typically consume
61 most energy - processor, display and hard drive. Each can be configured
62 seperately. <e>CPU Power Management</e> shows how to adjust the processor's
63 frequency to save a maximum of energy whithout losing too much performance. A
64 few different tricks prevent your hard drive from working unnecessarily often
65 in <e>Disk Power Management</e> (decreasing noise level as a nice side
66 effect). Some notes on graphics cards, Wireless LAN and USB finish the device
67 section in <e>Power Management for other devices</e> while another chapter is
68 dedicated to the (rather experimental) <e>sleep states</e>. Last not least
69 <e>Troubleshooting</e> lists common pitfalls.
70 </p>
71
72 </body>
73 </section>
74
75 <section>
76 <title>Power Budget for each component</title>
77 <body>
78
79 <figure link="/images/energy-budget.png" short="Which component consumes how
80 much energy?" caption="Power budget for each component"/>
81
82 <p>
83 Nearly every component can operate in different states - off, sleep, idle,
84 active to name a few - consuming a different amount of energy. Major parts are
85 consumed by the LCD display, CPU, chipset and hard drives. Often one is able to
86 activate OS-independent Power Management in the BIOS, but an intelligent setup
87 in the operating system adapting to different situations can achieve much more.
88 </p>
89
90 </body>
91 </section>
92 </chapter>
93
94 <chapter>
95 <title>Prerequisites</title>
96 <section>
97 <body>
98
99 <p>
100 Before discussing the details of making individual devices Power Management
101 aware, make sure certain requirements are met. After controlling BIOS
102 settings, some kernel options want to be enabled - these are in short ACPI,
103 sleep states and CPU frequency scaling. As power saving most of the time comes
104 along with performance loss or increased latency, it should only be enabled
105 when running on batteries. That's where a new runlevel <e>battery</e> comes in
106 handy.
107 </p>
108
109 </body>
110 </section>
111 <section>
112 <title>The BIOS part</title>
113 <body>
114
115 <p>
116 First have a look into your BIOS Power Management settings. The best way is to
117 combine BIOS and operating system policies, but for the moment it's better to
118 disable most of the BIOS part. This makes sure it doesn't interfere with your
119 policies. Don't forget to re-check BIOS settings after you configured
120 everything else.
121 </p>
122
123 </body>
124 </section>
125 <section>
126 <title>Setting USE flags</title>
127 <body>
128
129 <p>
130 Please check that the <c>acpi</c> USE flag is set in
131 <path>/etc/make.conf</path>. Other USE flags that might be interesting for your
132 system are <c>apm</c>, <c>lm_sensors</c>, <c>nforce2</c>, <c>nvidia</c>,
133 <c>pmu</c>. See <path>/usr/portage/profiles/use*.desc</path> for details. If
134 you forgot to set one of these flags, you can recompile affected packages using
135 the <c>--newuse</c> flag in <c>emerge</c>, see <c>man 1 emerge</c>.
136 </p>
137
138 </body>
139 </section>
140 <section>
141 <title>Configuring the kernel</title>
142 <body>
143
144 <p>
145 ACPI (Advanced Configuration and Power Interface) support in the kernel is
146 still work in progress. Using a recent kernel will make sure you'll get the
147 most out of it.
148 </p>
149
150 <p>
151 There are different kernel sources in Portage. I'd recommend using
152 <c>gentoo-sources</c> or <c>suspend2-sources</c>. The latter contains patches
153 for Software Suspend 2, see the chapter about sleep states for details. When
154 configuring the kernel, activate at least these options:
155 </p>
156
157 <pre caption="Minimum kernel setup for Power Management (Kernel 2.6)">
158 Power Management Options ---&gt;
159 [*] Power Management Support
160 [ ] Software Suspend
161
162 ACPI( Advanced Configuration and Power Interface ) Support ---&gt;
163 [*] ACPI Support
164 [ ] Sleep States
165 [ ] /proc/acpi/sleep (deprecated)
166 [*] AC Adapter
167 [*] Battery
168 &lt;M&gt; Button
169 &lt;M&gt; Video
170 [ ] Generic Hotkey
171 &lt;M&gt; Fan
172 &lt;M&gt; Processor
173 &lt;M&gt; Thermal Zone
174 &lt; &gt; ASUS/Medion Laptop Extras
175 &lt; &gt; IBM ThinkPad Laptop Extras
176 &lt; &gt; Toshiba Laptop Extras
177 (0) Disable ACPI for systems before Jan 1st this year
178 [ ] Debug Statements
179 [*] Power Management Timer Support
180 &lt; &gt; ACPI0004,PNP0A05 and PNP0A06 Container Driver (EXPERIMENTAL)
181
182 CPU Frequency Scaling ---&gt;
183 [*] CPU Frequency scaling
184 [ ] Enable CPUfreq debugging
185 &lt; &gt; CPU frequency translation statistics
186 [ ] CPU frequency translation statistics details
187 Default CPUFreq governor (userspace)
188 &lt;*&gt; 'performance' governor
189 &lt;*&gt; 'powersave' governor
190 &lt;*&gt; 'ondemand' cpufreq policy governor
191 &lt;*&gt; 'conservative' cpufreq governor
192 &lt;*&gt; CPU frequency table helpers
193 &lt;M&gt; ACPI Processor P-States driver
194 &lt;*&gt; <i>CPUFreq driver for your processor</i>
195 </pre>
196
197 <p>
198 Decide yourself whether you want to enable Software Suspend, and Sleep States
199 (see below). If you own an ASUS, Medion, IBM Thinkpad or Toshiba laptop, enable
200 the appropriate section.
201 </p>
202
203 <p>
204 The kernel has to know how to enable CPU frequency scaling on your processor. As
205 each type of CPU has a different interface, you've got to choose the right
206 driver for your processor. Be careful here - enabling <e>Intel Pentium 4 clock
207 modulation</e> on a Pentium M system will lead to strange results for example.
208 Consult the kernel documentation if you're unsure which one to take.
209 </p>
210
211 <p>
212 Compile your kernel, make sure the right modules get loaded at startup and boot
213 into your new ACPI-enabled kernel. Next run <c>emerge sys-power/acpid</c> to get
214 the acpi daemon. This one informs you about events like switching from AC to
215 battery or closing the lid. Make sure the modules are loaded if you didn't
216 compile them into the kernel and start acpid by executing
217 <c>/etc/init.d/acpid start</c>. Run <c>rc-update add acpid default</c> to load
218 it on startup. You'll soon see how to use it.
219 </p>
220
221 <pre caption="Installing acpid">
222 # <i>emerge sys-power/acpid</i>
223 # <i>/etc/init.d/acpid start</i>
224 # <i>rc-update add acpid default</i>
225 </pre>
226
227 </body>
228 </section>
229 <section>
230 <title>Creating a "battery" runlevel</title>
231 <body>
232
233 <p>
234 The default policy will be to enable Power Management only when needed -
235 running on batteries. To make the switch between AC and battery convenient,
236 create a runlevel <e>battery</e> that holds all the scripts starting and
237 stopping Power Management.
238 </p>
239
240 <note>
241 You can safely skip this section if you don't like the idea of having another
242 runlevel. However, skipping this step will make the rest a bit trickier to set
243 up. The next sections assume a runlevel <e>battery</e> exists.
244 </note>
245
246 <pre caption="Creating a battery runlevel">
247 # <i>cd /etc/runlevels</i>
248 # <i>cp -a default battery</i>
249 </pre>
250
251 <p>
252 Finished. Your new runlevel <e>battery</e> contains everything like
253 <e>default</e>, but there is no automatic switch between both yet. Time to
254 change it.
255 </p>
256
257 </body>
258 </section>
259 <section>
260 <title>Reacting on ACPI events</title>
261 <body>
262
263 <p>
264 Typical ACPI events are closing the lid, changing the power source or pressing
265 the sleep button. An important event is changing the power source, which should
266 cause a runlevel switch. A small script will take care of it.
267 </p>
268
269 <p>
270 First you need a script which changes the runlevel to <c>default</c>
271 respectively <c>battery</c> depending on the power source. The script uses the
272 <c>on_ac_power</c> command from <c>sys-power/powermgmt-base</c> - make sure the
273 package is installed on your system.
274 </p>
275
276 <pre caption="Installing powermgt-base">
277 <i># emerge powermgmt-base</i>
278 </pre>
279
280 <p>
281 You are now able to determine the power source by executing
282 <c>on_ac_power &amp;&amp; echo AC available || echo Running on batteries</c> in
283 a shell. The script below is responsible for changing runlevels. Save it as
284 <path>/etc/acpi/actions/pmg_switch_runlevel.sh</path>.
285 </p>
286
287 <pre caption="/etc/acpi/actions/pmg_switch_runlevel.sh">
288 #!/bin/bash
289
290 <comment># BEGIN configuration</comment>
291 RUNLEVEL_AC="default"
292 RUNLEVEL_BATTERY="battery"
293 <comment># END configuration</comment>
294
295
296 if [ ! -d "/etc/runlevels/${RUNLEVEL_AC}" ]
297 then
298 logger "${0}: Runlevel ${RUNLEVEL_AC} does not exist. Aborting."
299 exit 1
300 fi
301
302 if [ ! -d "/etc/runlevels/${RUNLEVEL_BATTERY}" ]
303 then
304 logger "${0}: Runlevel ${RUNLEVEL_BATTERY} does not exist. Aborting."
305 exit 1
306 fi
307
308 if on_ac_power
309 then
310 if [[ "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_AC}" ]]
311 then
312 logger "Switching to ${RUNLEVEL_AC} runlevel"
313 /sbin/rc ${RUNLEVEL_AC}
314 fi
315 elif [[ "$(cat /var/lib/init.d/softlevel)" != "${RUNLEVEL_BATTERY}" ]]
316 then
317 logger "Switching to ${RUNLEVEL_BATTERY} runlevel"
318 /sbin/rc ${RUNLEVEL_BATTERY}
319 fi
320 </pre>
321
322 <p>
323 Dont forget to run <c>chmod +x /etc/acpi/actions/pmg_switch_runlevel.sh</c> to
324 make the script executable. The last thing that needs to be done is calling the
325 script whenever the power source changes. That's done by catching ACPI events
326 with the help of <c>acpid</c>. First you need to know which events are
327 generated when the power source changes. The events are called
328 <e>ac_adapter</e> and <e>battery</e> on most laptops, but it might be different
329 on yours.
330 </p>
331
332 <pre caption="Determining ACPI events for changing the power source">
333 <i># tail -f /var/log/acpid | grep "received event"</i>
334 </pre>
335
336 <p>
337 Run the command above and pull the power cable. You should see something
338 like this:
339 </p>
340
341 <pre caption="Sample output for power source changes">
342 [Tue Sep 20 17:39:06 2005] received event "ac_adapter AC 00000080 00000000"
343 [Tue Sep 20 17:39:06 2005] received event "battery BAT0 00000080 00000001"
344 </pre>
345
346 <p>
347 The interesting part is the quoted string after <e>received event</e>. It will
348 be matched by the event line in the files you are going to create below. Don't
349 worry if your system generates multiple events or always the same. As long as
350 any event is generated, runlevel changing will work.
351 </p>
352
353 <pre caption="/etc/acpi/events/pmg_ac_adapter">
354 <comment># replace "ac_adapter" below with the event generated on your laptop</comment>
355 <comment># For example, ac_adapter.* will match ac_adapter AC 00000080 00000000</comment>
356 event=ac_adapter.*
357 action=/etc/acpi/actions/pmg_switch_runlevel.sh %e
358 </pre>
359
360 <pre caption="/etc/acpi/events/pmg_battery">
361 <comment># replace "battery" below with the event generated on your laptop</comment>
362 <comment># For example, battery.* will match battery BAT0 00000080 00000001</comment>
363 event=battery.*
364 action=/etc/acpi/actions/pmg_switch_runlevel.sh %e
365 </pre>
366
367 <p>
368 Finally acpid has to be restarted to recognize the changes.
369 </p>
370
371 <pre caption="Finishing runlevel switching with acpid">
372 <i># /etc/init.d/acpid restart</i>
373 </pre>
374
375 <p>
376 Give it a try: Plug AC in and out and watch syslog for the "Switching to AC
377 mode" or "Switching to battery mode" messages. See the Troubleshooting
378 section if the script is not able to detect the power source correctly.
379 </p>
380
381 <p>
382 Due to the nature of the event mechanism, your laptop will boot into runlevel
383 <e>default</e> regardless of the AC/battery state. This is fine when running
384 from AC, but we'd like to boot into the battery runlevel otherwise. One
385 solution would be to add another entry to the boot loader with the parameter
386 <c>softlevel=battery</c>, but it's likely to forget choosing it. A better way
387 is faking an ACPI event in the end of the boot process and letting
388 <path>pmg_switch_runlevel.sh</path> script decide whether a
389 runlevel change is necessary. Open <path>/etc/conf.d/local.start</path> in your
390 favourite editor and add these lines:
391 </p>
392
393 <pre caption="Runlevel adjustment at boot time by editing local.start">
394 <comment># Fake acpi event to switch runlevel if running on batteries</comment>
395 /etc/acpi/actions/pmg_switch_runlevel.sh "battery/battery"
396 </pre>
397
398 <p>
399 Prepared like this you can activate Power Management policies for individual
400 devices.
401 </p>
402
403 </body>
404 </section>
405 </chapter>
406
407 <chapter>
408 <title>CPU Power Management</title>
409 <section>
410 <body>
411
412 <p>
413 Mobile processors can operate at different frequencies. Some allow changing
414 voltage as well. Most of the time your CPU doesn't need to run at full speed
415 and scaling it down will save much energy - often without any performance
416 decrease.
417 </p>
418
419 </body>
420 </section>
421
422 <section>
423 <title>Some technical terms</title>
424 <body>
425
426 <p>
427 CPU frequency scaling brings up some technical terms that might be unknown to
428 you. Here's a quick introduction.
429 </p>
430
431 <p>
432 First of all, the kernel has to be able to change the processor's frequency.
433 The <e>CPUfreq processor driver</e> knows the commands to do it on your CPU.
434 Thus it's important to choose the right one in your kernel. You should
435 already have done it above. Once the kernel knows how to change frequencies,
436 it has to know which frequency it should set. This is done according to the
437 <e>policy</e> which consists of a <e>CPUfreq policy</e> and a
438 <e>governor</e>. A CPUfreq policy are just two numbers which define a range
439 the frequency has to stay between - minimal and maximal frequency. The
440 governor now decides which of the available frequencies in between minimal
441 and maximal frequency to choose. For example, the <e>powersave governor</e>
442 always chooses the lowest frequency available, the <e>performance
443 governor</e> the highest one. The <e>userspace governor</e> makes no decision
444 but chooses whatever the user (or a program in userspace) wants - which means
445 it reads the frequency from
446 <path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed</path>.
447 </p>
448
449 <p>
450 This doesn't sound like dynamic frequency changes yet and in fact it isn't.
451 Dynamics however can be accomplished with various approaches. For example,
452 the <e>ondemand governor</e> makes its decisions depending on the current CPU
453 load. The same is done by various userland tools like <c>cpudyn</c>,
454 <c>cpufreqd</c>, <c>powernowd</c> and many more. ACPI events can be used to
455 enable or disable dynamic frequency changes depending on power source.
456 </p>
457
458 </body>
459 </section>
460 <section>
461 <title>Setting the frequency manually</title>
462 <body>
463
464 <p>
465 Decreasing CPU speed and voltage has two advantages: On the one hand less
466 energy is consumed, on the other hand there is thermal improvement as your
467 system doesn't get as hot as running on full speed. The main disadvantage is
468 obviously the loss of performance. Decreasing processor speed is a trade off
469 between performance loss and energy saving.
470 </p>
471
472 <note>
473 Not every laptop supports frequency scaling. If unsure, have a look at the list
474 of supported processors in the <e>Troubleshooting</e> section to verify your's
475 is supported.
476 </note>
477
478 <p>
479 It's time to test whether CPU frequency changing works. Let's install another
480 tool which is very handy for debugging purposes: <c>sys-power/cpufrequtils</c>
481 </p>
482
483 <pre caption="Checking CPU frequency">
484 # <i>emerge cpufrequtils</i>
485 # <i>cpufreq-info</i>
486 </pre>
487
488 <p>
489 Here is an example output:
490 </p>
491
492 <pre caption="Sample output from cpufreq-info">
493 cpufrequtils 0.3: cpufreq-info (C) Dominik Brodowski 2004
494 Report errors and bugs to linux@brodo.de, please.
495 analyzing CPU 0:
496 driver: centrino
497 CPUs which need to switch frequency at the same time: 0
498 hardware limits: 600 MHz - 1.40 GHz
499 available frequency steps: 600 MHz, 800 MHz, 1000 MHz, 1.20 GHz, 1.40 GHz
500 available cpufreq governors: conservative, ondemand, powersave, userspace, performance
501 current policy: frequency should be within 924 MHz and 1.40 GHz.
502 The governor "performance" may decide which speed to use
503 within this range.
504 current CPU frequency is 1.40 GHz.
505 </pre>
506
507 <p>
508 Now play around with <c>cpufreq-set</c> to make sure frequency switching works.
509 Run <c>cpufreq-set -g ondemand</c> for example to activate the ondemand
510 governor and verify the change with <c>cpufreq-info</c>. If it doesn't work as
511 expected, you might find help in the Troubleshooting section in the end of this
512 guide.
513 </p>
514
515 </body>
516 </section>
517 <section>
518 <title>Automated frequency adaption</title>
519 <body>
520
521 <p>
522 The above is quite nice, but not doable in daily life. Better let your system
523 set the appropriate frequency automatically. There are many different
524 approaches to do this. The following table gives a quick overview to help you
525 decide on one of them. It's roughly seperated in three categories
526 <e>kernel</e> for approaches that only need kernel support, <e>daemon</e> for
527 programs that run in the background and <e>graphical</e> for programs that
528 provide a GUI for easy configuration and changes.
529 </p>
530
531 <table>
532 <tr>
533 <th>Name</th>
534 <th>Category</th>
535 <th>Switch decision</th>
536 <th>Kernel governors</th>
537 <th>Further governors</th>
538 <th>Comments</th>
539 </tr>
540 <tr>
541 <ti>'ondemand' governor</ti>
542 <ti>Kernel</ti>
543 <ti>CPU load</ti>
544 <ti>N.A.</ti>
545 <ti>N.A.</ti>
546 <ti>
547 Chooses maximal frequency on CPU load and slowly steps down when the CPU is
548 idle. Further tuning through files in
549 <path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still requires
550 userland tools (programs, scripts) if governor switching or similar is
551 desired.
552 </ti>
553 </tr>
554 <tr>
555 <ti>'conservative' governor</ti>
556 <ti>Kernel</ti>
557 <ti>CPU load</ti>
558 <ti>N.A.</ti>
559 <ti>N.A.</ti>
560 <ti>
561 Unlike the ondemand governor, conversative doesn't jump to maximum
562 frequency when CPU load is high, but increases the frequency step by step.
563 Further tuning through files in
564 <path>/sys/devices/system/cpu/cpu0/cpufreq/ondemand/</path>. Still requires
565 userland tools (programs, scripts) if governor switching or similar is
566 desired.
567 </ti>
568 </tr>
569 <tr>
570 <ti><uri link="http://mnm.uib.es/~gallir/cpudyn/">cpudyn</uri></ti>
571 <ti>Daemon</ti>
572 <ti>CPU load</ti>
573 <ti>Performance, powersave</ti>
574 <ti>Dynamic</ti>
575 <ti>
576 Also supports disk standby - notice however that <e>laptop mode</e> in most
577 cases will do a better job.
578 </ti>
579 </tr>
580 <tr>
581 <ti><uri link="http://sourceforge.net/projects/cpufreqd/">cpufreqd</uri></ti>
582 <ti>Daemon</ti>
583 <ti>Battery state, CPU load, temperature, running programs and more</ti>
584 <ti>All available</ti>
585 <ti>None</ti>
586 <ti>
587 Sophisticated (but somewhat complicated) setup. Extendible through plugins
588 like sensor monitoring (lm_sensors) or coordinating some NVidia based
589 graphics card memory and core. Cpufreqd is SMP aware and can optionally be
590 controlled manually at runtime.
591 </ti>
592 </tr>
593 <tr>
594 <ti>
595 <uri link="http://www.deater.net/john/powernowd.html">powernowd</uri>
596 </ti>
597 <ti>Daemon</ti>
598 <ti>CPU load</ti>
599 <ti>None</ti>
600 <ti>Passive, sine, aggressive</ti>
601 <ti>
602 Supports SMP.
603 </ti>
604 </tr>
605 <tr>
606 <ti>
607 <uri link="http://fatcat.ftj.agh.edu.pl/~nelchael/index.php?cat=projs&amp;subcat=ncpufreqd&amp;language=en">ncpufreqd</uri>
608 </ti>
609 <ti>Daemon</ti>
610 <ti>Temperature</ti>
611 <ti>None</ti>
612 <ti>Powersave, performance</ti>
613 <ti>
614 Toggles the used governor between performance and powersave depending on
615 system temperature. Very useful on laptops with notorious heat problems.
616 </ti>
617 </tr>
618 <tr>
619 <ti><uri link="http://www.goop.org/~jeremy/speedfreq/">speedfreq</uri></ti>
620 <ti>Daemon</ti>
621 <ti>CPU load</ti>
622 <ti>None</ti>
623 <ti>Dynamic, powersave, performance, fixed speed</ti>
624 <ti>
625 Easy to configure with a nice client/server interface. Requires a 2.6
626 kernel. Unmaintained, broken and thus removed from Portage. Please switch
627 to cpufreqd if you're still using it.
628 </ti>
629 </tr>
630 <tr>
631 <ti><uri link="http://cpuspeedy.sourceforge.net/">gtk-cpuspeedy</uri></ti>
632 <ti>Graphical</ti>
633 <ti>None</ti>
634 <ti>None</ti>
635 <ti>None</ti>
636 <ti>
637 Gnome application, a graphical tool to set CPU frequency manually. It does
638 not offer any automation.
639 </ti>
640 </tr>
641 <tr>
642 <ti>klaptopdaemon</ti>
643 <ti>Graphical</ti>
644 <ti>Battery state</ti>
645 <ti>All available</ti>
646 <ti>None</ti>
647 <ti>
648 KDE only, 'ondemand' governor required for dynamic frequency scaling.
649 </ti>
650 </tr>
651 </table>
652
653 <p>
654 While adjusting the frequency to the current load looks simple at a first
655 glance, it's not such a trivial task. A bad algorithm can cause switching
656 between two frequencies all the time or wasting energy when setting frequency
657 to an unnecessary high level.
658 </p>
659
660 <p>
661 Which one to choose? If you have no idea about it, try <c>cpufreqd</c>:
662 </p>
663
664 <pre caption="Installing cpufreqd">
665 # <i>emerge cpufreqd</i>
666 </pre>
667
668 <p>
669 <c>cpufreqd</c> can be configured by editing <path>/etc/cpufreqd.conf</path>.
670 The default one that ships with cpufreqd may look a bit confusing. I recommend
671 replacing it with the one from Gentoo developer Henrik Brix Andersen (see
672 below). Please notice that you need cpufreqd-2.0.0 or later. Earlier versions
673 have a different syntax for the config file.
674 </p>
675
676 <pre caption="/etc/cpufreqd.conf (cpufreqd-2.0.0 and later)">
677 [General]
678 pidfile=/var/run/cpufreqd.pid
679 poll_interval=3
680 enable_plugins=acpi_ac, acpi_battery
681 enable_remote=1
682 remote_group=wheel
683 verbosity=5
684 [/General]
685
686 [Profile]
687 name=ondemand
688 minfreq=0%
689 maxfreq=100%
690 policy=ondemand
691 [/Profile]
692
693 [Profile]
694 name=conservative
695 minfreq=0%
696 maxfreq=100%
697 policy=conservative
698 [/Profile]
699
700 [Profile]
701 name=powersave
702 minfreq=0%
703 maxfreq=100%
704 policy=powersave
705 [/Profile]
706
707 [Profile]
708 name=performance
709 minfreq=0%
710 maxfreq=100%
711 policy=performance
712 [/Profile]
713
714 [Rule]
715 name=battery
716 ac=off
717 profile=conservative
718 [/Rule]
719
720 [Rule]
721 name=battery_low
722 ac=off
723 battery_interval=0-10
724 profile=powersave
725 [/Rule]
726
727 [Rule]
728 name=ac
729 ac=on
730 profile=ondemand
731 [/Rule]
732 </pre>
733
734 <p>
735 Now you can start the cpufreqd daemon. Add it to the <e>default</e> and
736 <e>battery</e> runlevel as well.
737 </p>
738
739 <pre caption="Starting cpufreqd">
740 # <i>rc-update add cpufreqd default battery</i>
741 # <i>rc</i>
742 </pre>
743
744 <p>
745 Sometimes it can be desirable to select another policy than the daemon chooses,
746 for example when battery power is low, but you know that AC will be available
747 soon. In that case you can turn on cpufreqd's manual mode with
748 <c>cpufreqd-set manual</c> and select one of your configured policies (as
749 listed by <c>cpufreqd-get</c>). You can leave manual mode by executing
750 <c>cpufreqd-set dynamic</c>.
751 </p>
752
753 <warn>
754 Do not run more than one of the above programs at the same time. It may cause
755 confusion like switching between two frequencies all the time.
756 </warn>
757
758 </body>
759 </section>
760
761 <section>
762 <title>Verifying the result</title>
763
764 <body>
765
766 <p>
767 The last thing to check is that your new policies do a good job. An easy way to
768 do so is monitoring CPU speed while working with your laptop:
769 </p>
770
771 <pre caption="Monitoring CPU speed">
772 # <i>watch grep \"cpu MHz\" /proc/cpuinfo</i>
773 </pre>
774
775 <p>
776 If <path>/proc/cpuinfo</path> doesn't get updated (see Troubleshooting),
777 monitor the CPU frequency with:
778 </p>
779
780 <pre caption="Alternative CPU speed monitoring">
781 # <i>watch x86info -mhz</i>
782 </pre>
783
784 <p>
785 Depending on your setup, CPU speed should increase on heavy load, decrease on
786 no activity or just stay at the same level. When using cpufreqd and verbosity
787 set to 5 or higher in <path>cpufreqd.conf</path> you'll get additional
788 information about what's happening reported to syslog.
789 </p>
790
791 </body>
792 </section>
793 </chapter>
794
795 <chapter>
796 <title>LCD Power Management</title>
797 <section>
798 <body>
799
800 <p>
801 As you can see in <uri link="#doc_chap1_fig1">figure 1.1</uri>, the LCD display
802 consumes the biggest part of energy (might not be the case for non-mobile
803 CPU's). Thus it's quite important not only to shut the display off when not
804 needed, but also to reduce it's backlight if possible. Most laptops offer the
805 possibility to control the backlight dimming.
806 </p>
807
808 </body>
809 </section>
810 <section>
811 <title>Standby settings</title>
812 <body>
813
814 <p>
815 The first thing to check is the standby/suspend/off timings of the display. As
816 this depends heavily on your windowmanager, I'll let you figure it out
817 yourself. Just two common places: Blanking the terminal can be done with
818 <c>setterm -blank &lt;number-of-minutesM&gt;</c>, <c>setterm -powersave on</c>
819 and <c>setterm -powerdown &lt;number-of-minutesM&gt;</c>. For X.org, modify
820 <path>/etc/X11/xorg.conf</path> similar to this:
821 </p>
822
823 <pre caption="LCD suspend settings in X.org and XFree86">
824 Section "ServerLayout"
825 Identifier [...]
826 [...]
827 Option "BlankTime" "5" <comment># Blank the screen after 5 minutes (Fake)</comment>
828 Option "StandbyTime" "10" <comment># Turn off screen after 10 minutes (DPMS)</comment>
829 Option "SuspendTime" "20" <comment># Full suspend after 20 minutes</comment>
830 Option "OffTime" "30" <comment># Turn off after half an hour</comment>
831 [...]
832 EndSection
833
834 [...]
835
836 Section "Monitor"
837 Identifier [...]
838 Option "DPMS" "true"
839 [...]
840 EndSection
841 </pre>
842
843 <p>
844 This is the same for XFree86 and <path>/etc/X11/XF86Config</path>.
845 </p>
846
847 </body>
848 </section>
849 <section>
850 <title>Backlight dimming</title>
851 <body>
852
853 <p>
854 Probably more important is the backlight dimming. If you have access to the
855 dimming settings via a tool, write a small script that dims the backlight in
856 battery mode and place it in your <e>battery</e> runlevel. The following script
857 should work on most IBM Thinkpads and Toshiba laptops. You've got to enable the
858 appropriate option in your kernel (IBM Thinkpads only). For Toshiba laptops, install
859 <c>app-laptop/acpitool</c> and skip configuration of ibm_acpi as described below.
860 </p>
861
862 <warn>
863 Support for setting brightness is marked experimental in ibm-acpi. It accesses
864 hardware directly and may cause severe harm to your system. Please read the
865 <uri link="http://ibm-acpi.sourceforge.net/">ibm-acpi website</uri>
866 </warn>
867
868 <p>
869 To be able to set the brightness level, the ibm_acpi module has to be loaded
870 with the experimental parameter.
871 </p>
872
873 <pre caption="automatically loading the ibm_acpi module">
874 <comment>(Please read the warnings above before doing this!)</comment>
875 <i># echo "options ibm_acpi experimental=1" >> /etc/modules.d/ibm_acpi</i>
876 <i># /sbin/modules-update</i>
877 <i># echo ibm_acpi >> /etc/modules.autoload.d/kernel-2.6</i>
878 <i># modprobe ibm_acpi</i>
879 </pre>
880
881 <p>
882 This should work without error messages and a file
883 <path>/proc/acpi/ibm/brightness</path> should be created after loading the
884 module. An init script will take care of choosing the brightness according
885 to the power source.
886 </p>
887
888 <pre caption="/etc/conf.d/lcd-brightness">
889 <comment># See /proc/acpi/ibm/brightness for available values</comment>
890 <comment># Please read /usr/src/linux/Documentation/ibm-acpi.txt</comment>
891
892 <comment># brigthness level in ac mode. Default is 7.</comment>
893 BRIGHTNESS_AC=7
894
895 <comment># brightness level in battery mode. Default is 4.</comment>
896 BRIGHTNESS_BATTERY=4
897 </pre>
898
899 <pre caption="/etc/init.d/lcd-brightness">
900 #!/sbin/runscript
901
902 set_brightness() {
903 if on_ac_power
904 then
905 LEVEL=${BRIGHTNESS_AC:-7}
906 else
907 LEVEL=${BRIGHTNESS_BATTERY:-4}
908 fi
909
910 if [ -f /proc/acpi/ibm/brightness ]
911 then
912 ebegin "Setting LCD brightness"
913 echo "level ${LEVEL}" > /proc/acpi/ibm/brightness
914 eend $?
915 elif [[ -e /usr/bin/acpitool &amp;&amp; -n $(acpitool -T | grep "LCD brightness") ]]
916 then
917 ebegin "Setting LCD brightness"
918 acpitool -l $LEVEL >/dev/null || ewarn "Unable to set lcd brightness"
919 eend $?
920 else
921 ewarn "Setting LCD brightness is not supported."
922 ewarn "For IBM Thinkpads, check that ibm_acpi is loaded into the kernel"
923 ewarn "For Toshiba laptops, you've got to install app-laptop/acpitool"
924 fi
925 }
926
927 start() {
928 set_brightness
929 }
930
931 stop () {
932 set_brightness
933 }
934 </pre>
935
936 <p>
937 When done, make sure brightness is adjusted automatically by adding it to the
938 battery runlevel.
939 </p>
940
941 <pre caption="Enabling automatic brightness adjustment">
942 <i># chmod +x /etc/init.d/lcd-brightness</i>
943 <i># rc-update add lcd-brightness battery</i>
944 <i># rc</i>
945 </pre>
946
947 </body>
948 </section>
949 </chapter>
950
951 <chapter>
952 <title>Disk Power Management</title>
953 <section>
954 <body>
955 <p>
956 Hard disks consume less energy in sleep mode. Therefore it makes sense to
957 activate power saving features whenever the hard disk is not used for a certain
958 amount of time. I'll show you two alternative possibilities to do it. First,
959 laptop-mode will save most energy due to several measures which prevent or at
960 least delay write accesses. The drawback is that due to the delayed write
961 accesses a power outage or kernel crash will be more dangerous for data loss.
962 If you don't like this, you have to make sure that there are no processes which
963 write to your hard disk frequently. Afterwards you can enable power saving
964 features of your hard disk with hdparm as the second alternative.
965 </p>
966
967 </body>
968 </section>
969
970 <section>
971 <title>Increasing idle time - laptop-mode</title>
972 <body>
973
974 <p>
975 Recent kernels (2.6.6 and greater, recent 2.4 ones and others with patches)
976 include the so-called <e>laptop-mode</e>. When activated, dirty buffers are
977 written to disk on read calls or after 10 minutes (instead of 30 seconds). This
978 minimizes the time the hard disk needs to be spun up.
979 </p>
980
981 <pre caption="Automated start of laptop-mode">
982 # <i>emerge laptop-mode-tools</i>
983 </pre>
984
985 <p>
986 <c>laptop-mode-tools</c> has its configuration file in
987 <path>/etc/laptop-mode/laptop-mode.conf</path>. Adjust it the way you like it,
988 it's well commented. Run <c>rc-update add laptop_mode battery</c> to start it
989 automatically.
990 </p>
991
992 <p>
993 Recent versions (1.11 and later) of laptop-mode-tools include a new tool
994 <c>lm-profiler</c>. It will monitor your system's disk usage and running
995 network services and suggests to disable unneeded ones. You can either disable
996 them through laptop-mode-tools builtin runlevel support (which will be reverted
997 by Gentoo's <c>/sbin/rc</c>) or use your <e>default</e>/<e>battery</e>
998 runlevels (recommended).
999 </p>
1000
1001 <pre caption="Sample output from running lm-profiler">
1002 # lm-profiler
1003 Profiling session started.
1004 Time remaining: 600 seconds
1005 [4296896.602000] amarokapp
1006 Time remaining: 599 seconds
1007 [4296897.714000] sort
1008 [4296897.970000] mv
1009 Time remaining: 598 seconds
1010 Time remaining: 597 seconds
1011 [4296900.482000] reiserfs/0
1012 </pre>
1013
1014 <p>
1015 After profiling your system for ten minutes, lm-profiler will present a list of
1016 services which might have caused disk accesses during that time.
1017 </p>
1018
1019 <pre caption="lm-profiler suggests to disable some services">
1020 Program: "atd"
1021 Reason: standard recommendation (program may not be running)
1022 Init script: /etc/init.d/atd (GUESSED)
1023
1024 Do you want to disable this service in battery mode? [y/N]: n
1025 </pre>
1026
1027 <p>
1028 To disable atd as suggested in the example above, you would run <c>rc-update
1029 del atd battery</c>. Be careful not to disable services that are needed for
1030 your system to run properly - lm-profiler is likely to generate some false
1031 positives. Do not disable a service if you are unsure whether it's needed.
1032 </p>
1033
1034 </body>
1035 </section>
1036
1037 <section>
1038 <title>Limiting write accesses</title>
1039 <body>
1040
1041 <p>
1042 If you don't want to use laptop-mode, you must take special care to disable
1043 services that write to your disk frequently - <c>syslogd</c> is a good
1044 candidate, for example. You probably don't want to shut it down completely, but
1045 it's possible to modify the config file so that "unnecessary" things don't get
1046 logged and thus don't create disk traffic. Cups writes to disk periodically, so
1047 consider shutting it down and only enable it manually when needed.
1048 </p>
1049
1050 <pre caption="Disabling cups in battery mode">
1051 # <i>rc-update del cupsd battery</i>
1052 </pre>
1053
1054 <p>
1055 You can also use <c>lm-profiler</c> from laptop-mode-tools (see above) to find
1056 services to disable. Once you eliminated all of them, go on with configuring
1057 hdparm.
1058 </p>
1059
1060 </body>
1061 </section>
1062
1063 <section>
1064 <title>hdparm</title>
1065 <body>
1066
1067 <p>
1068 The second possibility is using a small script and hdparm. Skip this if you
1069 are using laptop-mode. Otherwise, create <path>/etc/init.d/pmg_hda</path>:
1070 </p>
1071
1072 <pre caption="Using hdparm for disk standby">
1073 #!/sbin/runscript
1074
1075 depend() {
1076 after hdparm
1077 }
1078
1079 start() {
1080 ebegin "Activating Power Management for Hard Drives"
1081 hdparm -q -S12 /dev/hda
1082 eend $?
1083 }
1084
1085 stop () {
1086 ebegin "Deactivating Power Management for Hard Drives"
1087 hdparm -q -S253 /dev/hda
1088 eend $?
1089 }
1090 </pre>
1091
1092 <p>
1093 See <c>man hdparm</c> for the options. If your script is ready, add it to the
1094 battery runlevel.
1095 </p>
1096
1097 <pre caption="Automate disk standby settings">
1098 # <i>chmod +x /etc/init.d/pmg_hda</i>
1099 # <i>/sbin/depscan.sh</i>
1100 # <i>rc-update add pmg_hda battery</i>
1101 </pre>
1102
1103 <impo>
1104 Be careful with sleep/spin down settings of your hard drive. Setting it to
1105 small values might wear out your drive and lose warranty.
1106 </impo>
1107
1108 </body>
1109 </section>
1110
1111 <section>
1112 <title>Other tricks</title>
1113 <body>
1114
1115 <p>
1116 Another possibility is to deactivate swap in battery mode. Before writing a
1117 swapon/swapoff switcher, make sure there is enough RAM and swap isn't used
1118 heavily, otherwise you'll be in big problems.
1119 </p>
1120
1121 <p>
1122 If you don't want to use laptop-mode, it's still possible to minimize disk
1123 access by mounting certain directories as <e>tmpfs</e> - write accesses are not
1124 stored on a disk, but in main memory and get lost with unmounting. Often it's
1125 useful to mount <path>/tmp</path> like this - you don't have to pay special
1126 attention as it gets cleared on every reboot regardless whether it was mounted
1127 on disk or in RAM. Just make sure you have enough RAM and no program (like a
1128 download client or compress utility) needs extraordinary much space in
1129 <path>/tmp</path>. To activate this, enable tmpfs support in your kernel and
1130 add a line to <path>/etc/fstab</path> like this:
1131 </p>
1132
1133 <pre caption="Editing /etc/fstab to make /tmp even more volatile">
1134 none /tmp tmpfs size=32m 0 0
1135 </pre>
1136
1137 <warn>
1138 Pay attention to the size parameter and modify it for your system. If you're
1139 unsure, don't try this at all, it can become a perfomance bottleneck easily. In
1140 case you want to mount <path>/var/log</path> like this, make sure to merge the
1141 log files to disk before unmounting. They are essential. Don't attempt to mount
1142 /var/tmp like this. Portage uses it for compiling...
1143 </warn>
1144
1145 </body>
1146 </section>
1147 </chapter>
1148
1149 <chapter>
1150 <title>Power Management for other devices</title>
1151 <section>
1152 <title>Graphics cards</title>
1153 <body>
1154
1155 <p>
1156 In case you own an ATI graphics card supporting PowerPlay (dynamic clock
1157 scaling for the the graphics processing unit GPU), you can activate this
1158 feature in X.org. Open <path>/etc/X11/xorg.conf</path> and add (or enable)
1159 the <c>DynamicClocks</c> option in the Device section. Please notice that
1160 this feature will lead to crashes on some systems.
1161 </p>
1162
1163 <pre caption="Enabling ATI PowerPlay support in X.org">
1164 Section "Device"
1165 [...]
1166 Option "DynamicClocks" "on"
1167 EndSection
1168 </pre>
1169
1170 </body>
1171 </section>
1172 <section>
1173 <title>Wireless Power Management</title>
1174 <body>
1175
1176 <p>
1177 Wireless LAN cards consume quite a bit of energy. Put them in Power Management
1178 mode in analogy to the pmg_hda script.
1179 </p>
1180
1181 <note>
1182 This script assumes your wireless interface is called <c>wlan0</c>; replace
1183 this with the actual name of your interface.
1184 </note>
1185
1186 <pre caption="WLAN Power Management automated">
1187 #!/sbin/runscript
1188 start() {
1189 ebegin "Activating Power Management for Wireless LAN"
1190 iwconfig wlan0 power on
1191 eend $?
1192 }
1193
1194 stop () {
1195 ebegin "Deactivating Power Management for Wireless LAN"
1196 iwconfig wlan0 power off
1197 eend $?
1198 }
1199 </pre>
1200
1201 <p>
1202 Starting this script will activate power saving features for wlan0. Save it as
1203 <path>/etc/init.d/pmg_wlan0</path> and add it to the battery runlevel like the
1204 disk script above. See <c>man iwconfig</c> for details and more options like
1205 the period between wakeups or timeout settings. If your driver and access point
1206 support changing the beacon time, this is a good starting point to save even
1207 more energy.
1208 </p>
1209
1210 <pre caption="Power Management for WLAN">
1211 # <i>chmod +x /etc/init.d/pmg_wlan0</i>
1212 # <i>/sbin/depscan.sh</i>
1213 # <i>rc-update add pmg_wlan0 battery</i>
1214 </pre>
1215
1216 </body>
1217 </section>
1218 <section>
1219 <title>USB Power Management</title>
1220 <body>
1221
1222 <p>
1223 There are two problems with USB devices regarding energy consumption: First,
1224 devices like USB mice, digital cameras or USB sticks consume energy while
1225 plugged in. You cannot avoid this (nevertheless remove them in case they're not
1226 needed). Second, when there are USB devices plugged in, the USB host controller
1227 periodically accesses the bus which in turn prevents the CPU from going into
1228 sleep mode. The kernel offers an experimental option to enable suspension of
1229 USB devices through driver calls or one of the <path>power/state</path> files
1230 in <path>/sys</path>.
1231 </p>
1232
1233 <pre caption="Enabling USB suspend support in the kernel">
1234 Device Drivers
1235 USB support
1236 [*] Support for Host-side USB
1237 [*] USB suspend/resume (EXPERIMENTAL)
1238 </pre>
1239
1240 </body>
1241 </section>
1242 </chapter>
1243
1244 <chapter>
1245 <title>Sleep states: sleep, standby, suspend to disk</title>
1246 <section>
1247 <body>
1248
1249 <p>
1250 ACPI defines different sleep states. The more important ones are
1251 </p>
1252
1253 <ul>
1254 <li>S1 aka Standby</li>
1255 <li>S3 aka Suspend to RAM aka Sleep</li>
1256 <li>S4 aka Suspend to Disk aka Hibernate</li>
1257 </ul>
1258
1259 <p>
1260 They can be called whenever the system is not in use, but a shutdown is not
1261 wanted due to the long boot time.
1262 </p>
1263
1264 </body>
1265 </section>
1266 <section>
1267 <title>Sleep (S3)</title>
1268 <body>
1269
1270 <p>
1271 The ACPI support for these sleep states is marked experimental for good reason.
1272 APM sleep states seem to be more stable, however you can't use APM and ACPI
1273 together.
1274 </p>
1275
1276 <pre caption="Kernel configuration for the various suspend types">
1277 Power Management Options ---&gt;
1278 [*] Power Management support
1279 ACPI (Advanced Configuration and Power Interface) Support ---&gt;
1280 [*] ACPI Support
1281 [*] Sleep States
1282 </pre>
1283
1284 <p>
1285 Once your kernel is properly configured, you can use the
1286 <c>hibernate-script</c> to activate suspend or sleep mode. Let's install that
1287 first.
1288 </p>
1289
1290 <pre caption="Installing the hibernate-script">
1291 <i># emerge hibernate-script</i>
1292 </pre>
1293
1294 <p>
1295 Some configuration has to be done in <path>/etc/hibernate</path> The default
1296 package introduces two configuration files <path>hibernate.conf</path> and
1297 <path>ram.conf</path>.
1298 </p>
1299
1300 <p>
1301 To configure sleep, edit <path>ram.conf</path> in <path>/etc/hibernate</path>.
1302 <c>UseSysfsPowerState mem</c> is already setup correctly, but you have to go
1303 through the rest of the configuration file and set it up for your system. The
1304 comments and option names will guide you. If you use nfs or samba shares over
1305 the network, make sure to shutdown the appropriate init scripts to avoid
1306 timeouts.
1307 </p>
1308
1309 <p>
1310 Ready? Now is the last chance to backup any data you want to keep after
1311 executing the next command. Notice that you probably have to hit a special key
1312 like <e>Fn</e> to resume from sleep.
1313 </p>
1314
1315 <pre caption="Calling sleep">
1316 <i># hibernate-ram</i>
1317 </pre>
1318
1319 <p>
1320 If you're still reading, it seems to work. You can also setup standby (S1) in
1321 a similar way by copying <path>ram.conf</path> to <path>standby.conf</path>
1322 and creating a symlink <path>/usr/sbin/hibernate-standby</path> pointing to
1323 <path>/usr/sbin/hibernate</path>. S3 and S4 are the more interesting sleep
1324 states due to greater energy savings however.
1325 </p>
1326
1327 </body>
1328 </section>
1329 <section>
1330 <title>Hibernate (S4)</title>
1331 <body>
1332
1333 <p>
1334 This section introduces hibernation, where a snapshot of the running system is
1335 written to disk before powering off. On resume, the snapshot is loaded and you
1336 can go on working at exactly the point you called hibernate before.
1337 </p>
1338
1339 <warn>
1340 Don't exchange non hot-pluggable hardware when suspended. Don't attempt to load
1341 a snapshot with a different kernel image than the one it was created with.
1342 Shutdown any NFS or samba server/client before hibernating.
1343 </warn>
1344
1345 <p>
1346 There are two different implementations for S4. The original one is swsusp,
1347 then there is the newer suspend2 with a nicer interface (including
1348 fbsplash support). A <uri link="http://suspend2.net/features.html#compare">
1349 feature comparison</uri> is available at the <uri link="http://suspend2.net">
1350 suspend2 Homepage</uri>. There used to be Suspend-to-Disk (pmdisk), a fork of
1351 swsusp, but it has been merged back.
1352 </p>
1353
1354 <p>
1355 Suspend2 is not included in the mainline kernel yet, therefore you either have
1356 to patch your kernel sources with the patches provided by
1357 <uri link="http://suspend2.net">suspend2.net</uri> or use
1358 <c>sys-kernel/suspend2-sources</c>.
1359 </p>
1360
1361 <p>
1362 The kernel part for both swusp and suspend2 is as follows:
1363 </p>
1364
1365 <pre caption="Kernel configuration for the various suspend types">
1366 Power Management Options ---&gt;
1367 <comment>(hibernate with swsusp)</comment>
1368 [*] Software Suspend
1369 <comment>(replace /dev/SWAP with your swap partition)</comment>
1370 (/dev/SWAP) Default resume partition
1371
1372 <comment>(hibernate with suspend2)</comment>
1373 Software Suspend 2
1374 --- Image Storage (you need at least one writer)
1375 [*] File Writer
1376 [*] Swap Writer
1377 --- General Options
1378 [*] LZF image compression
1379 <comment>(replace /dev/SWAP with your swap partition)</comment>
1380 (swap:/dev/SWAP) Default resume device name
1381 [ ] Allow Keep Image Mode
1382 </pre>
1383
1384 <p>
1385 The configuration for swsusp is rather easy. If you didn't store the location
1386 of your swap partition in the kernel config, you can also pass it as a
1387 parameter with the <c>resume=/dev/SWAP</c> directive. If booting is not
1388 possible due to a broken image, use the <c>noresume</c> kernel parameter. The
1389 <c>hibernate-cleanup</c> init script invalidates swsusp images during the
1390 boot process.
1391 </p>
1392
1393 <pre caption="Invalidating swsusp images during the boot process">
1394 <i># rc-update add hibernate-cleanup boot</i>
1395 </pre>
1396
1397 <p>
1398 To activate hibernate with swsusp, use the hibernate script and set
1399 <c>UseSysfsPowerState disk</c> in <path>/etc/hibernate/hibernate.conf</path>.
1400 </p>
1401
1402 <warn>
1403 Backup your data before doing this. Run <c>sync</c> before executing one of the
1404 commands to have cached data written to disk. First try it outside of X, then
1405 with X running, but not logged in.
1406 </warn>
1407
1408 <p>
1409 If you experience kernel panics due to uhci or similar, try to compile USB
1410 support as module and unload the modules before sending your laptop to sleep
1411 mode. There are configuration options for this in <path>hibernate.conf</path>
1412 </p>
1413
1414 <pre caption="Hibernating with swsusp">
1415 <i># nano -w /etc/hibernate.conf</i>
1416 <comment>(Make sure you have a backup of your data)</comment>
1417 <i># hibernate</i>
1418 </pre>
1419
1420 <p>
1421 The following section discusses the setup of suspend2 including fbsplash
1422 support for a nice graphical progress bar during suspend and resume.
1423 </p>
1424
1425 <p>
1426 The first part of the configuration is similar to the configuration of
1427 swsusp. In case you didn't store the location of your swap partition in the
1428 kernel config, you have to pass it as a kernel parameter with the
1429 <c>resume2=swap:/dev/SWAP</c> directive. If booting is not possible due to a
1430 broken image, append the <c>noresume2</c> parameter. Additionally, the
1431 <c>hibernate-cleanup</c> init script invalidates suspend2 images during the
1432 boot process.
1433 </p>
1434
1435 <pre caption="Invalidating suspend2 images during the boot process">
1436 <i># rc-update add hibernate-cleanup boot</i>
1437 </pre>
1438
1439 <p>Now edit <path>/etc/hibernate/hibernate.conf</path>, enable the
1440 <e>suspend2</e> section and comment everything in the <e>sysfs_power_state</e>
1441 and <e>acpi_sleep</e> sections. Do not enable the fbsplash part in global
1442 options yet.
1443 </p>
1444
1445 <pre caption="Hibernating with suspend2">
1446 <i># nano -w /etc/hibernate.conf</i>
1447 <comment>(Make sure you have a backup of your data)</comment>
1448 <i># hibernate</i>
1449 </pre>
1450
1451 <p>
1452 Please configure fbsplash now if you didn't do already. To enable fbsplash
1453 support during hibernation, the <c>sys-apps/suspend2-userui</c> package is
1454 needed. Additionally, you've got to enable the <e>fbsplash</e> USE flag.
1455 </p>
1456
1457 <pre caption="Installing suspend2-userui">
1458 <i># mkdir -p /etc/portage</i>
1459 <i># echo sys-apps/suspend2-userui fbsplash >> /etc/portage/package.use</i>
1460 <i># emerge suspend2-userui</i>
1461 </pre>
1462
1463 <p>
1464 The ebuild tells you to make a symlink to the theme you want to use. For
1465 example, to use the <c>livecd-2005.1</c> theme, run the following command:
1466 </p>
1467
1468 <pre caption="Using the livecd-2005.1 theme during hibernation">
1469 <i># ln -sfn /etc/splash/livecd-2005.1 /etc/splash/suspend2</i>
1470 </pre>
1471
1472 <p>
1473 If you don't want a black screen in the first part of the resume process, you
1474 have to add the <c>suspend2ui_fbsplash</c> tool to your initrd image. Assuming
1475 you created the initrd image with <c>splash_geninitramfs</c> and saved it as
1476 <path>/boot/fbsplash-emergence-1024x768</path>, here's how to do
1477 that.
1478 </p>
1479
1480 <pre caption="Adding suspend2ui_fbsplash to an initrd image">
1481 <i># mount /boot</i>
1482 <i># mkdir ~/initrd.d</i>
1483 <i># cp /boot/fbsplash-emergence-1024x768 ~/initrd.d/</i>
1484 <i># cd ~/initrd.d</i>
1485 <i># gunzip -c fbsplash-emergence-1024x768 | cpio -idm --quiet -H newc</i>
1486 <i># rm fbsplash-emergence-1024x768</i>
1487 <i># cp /usr/sbin/suspend2ui_fbsplash sbin/</i>
1488 <i># find . | cpio --quiet --dereference -o -H newc | gzip -9 > /boot/fbsplash-suspend2-emergence-1024x768</i>
1489 </pre>
1490
1491 <p>
1492 Afterwards adjust <path>grub.conf</path> respectively <path>lilo.conf</path>
1493 so that your suspend2 kernel uses
1494 <path>/boot/fbsplash-suspend2-emergence-1024x768</path> as initrd image. You
1495 can now test a dry run to see if everything is setup correctly.
1496 </p>
1497
1498 <pre caption="Test run for fbsplash hibernation">
1499 <i># suspend2ui_fbsplash -t</i>
1500 </pre>
1501
1502 <p>
1503 Afterwards open <path>/etc/hibernate/hibernate.conf</path> again and activate
1504 the fbsplash options. Execute <c>hibernate</c> and enjoy.
1505 </p>
1506
1507 </body>
1508 </section>
1509 </chapter>
1510
1511 <chapter>
1512 <title>Troubleshooting</title>
1513 <section>
1514 <body>
1515
1516 <p>
1517 <e>Q:</e> I'm trying to change the CPU frequency, but
1518 <path>/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor</path> does not
1519 exist.
1520 </p>
1521
1522 <p>
1523 <e>A:</e> Make sure your processor supports CPU frequency scaling and you chose
1524 the right CPUFreq driver for your processor. Here is a list of processors that
1525 are supported by cpufreq (kernel 2.6.7): ARM Integrator, ARM-SA1100,
1526 ARM-SA1110, AMD Elan - SC400, SC410, AMD mobile K6-2+, AMD mobile K6-3+, AMD
1527 mobile Duron, AMD mobile Athlon, AMD Opteron, AMD Athlon 64, Cyrix Media GXm,
1528 Intel mobile PIII and Intel mobile PIII-M on certain chipsets, Intel Pentium 4,
1529 Intel Xeon, Intel Pentium M (Centrino), National Semiconductors Geode GX,
1530 Transmeta Crusoe, VIA Cyrix 3 / C3, UltraSPARC-III, SuperH SH-3, SH-4, several
1531 "PowerBook" and "iBook2" and various processors on some ACPI 2.0-compatible
1532 systems (only if "ACPI Processor Performance States" are available to the
1533 ACPI/BIOS interface).
1534 </p>
1535
1536 <p>
1537 <e>Q:</e> My laptop supports frequency scaling, but
1538 <path>/sys/devices/system/cpu/cpu0/cpufreq/</path> is empty.
1539 </p>
1540
1541 <p>
1542 <e>A:</e> Look for ACPI related error messages with <c>dmesg | grep ACPI</c>.
1543 Try to update the BIOS, especially if a broken DSDT is reported. You can also
1544 try to fix it yourself (which is beyond the scope of this guide).
1545 </p>
1546
1547 <p>
1548 <e>Q:</e> My laptop supports frequency scaling, but according to /proc/cpuinfo
1549 the speed never changes.
1550 </p>
1551
1552 <p>
1553 <e>A:</e> Probably you have activated symmetric multiprocessing support
1554 (CONFIG_SMP) in your kernel. Deactivate it and it should work. Some older
1555 kernels had a bug causing this. In that case, run <c>emerge x86info</c>,
1556 update your kernel as asked and check the current frequency with
1557 <c>x86info -mhz</c>.
1558 </p>
1559
1560 <p>
1561 <e>Q:</e> I can change the CPU frequency, but the range is not as wide as in
1562 another OS.
1563 </p>
1564
1565 <p>
1566 <e>A:</e> You can combine frequency scaling with ACPI throttling to get a lower
1567 minimum frequency. Notice that throttling doesn't save much energy and is
1568 mainly used for thermal management (keeping your laptop cool and quiet). You
1569 can read the current throttling state with <c>cat
1570 /proc/acpi/processor/CPU/throttling</c> and change it with <c>echo -n "0:x" >
1571 /proc/acpi/processor/CPU/limit</c>, where x is one of the Tx states listed in
1572 <path>/proc/acpi/processor/CPU/throttling</path>.
1573 </p>
1574
1575 <p>
1576 <e>Q:</e> When configuring the kernel, powersave, performance and userspace
1577 governors show up, but that ondemand thing is missing. Where do I get it?
1578 </p>
1579
1580 <p>
1581 <e>A:</e> The ondemand governor is only included in recent kernel sources. Try
1582 updating them.
1583 </p>
1584
1585 <p>
1586 <e>Q:</e> Battery life time seems to be worse than before.
1587 </p>
1588
1589 <p>
1590 <e>A:</e> Check your BIOS settings. Maybe you forgot to re-enable some of the
1591 settings.
1592 </p>
1593
1594 <p>
1595 <e>Q:</e> My battery is charged, but KDE reports there would be 0% left and
1596 immediately shuts down.
1597 </p>
1598
1599 <p>
1600 <e>A:</e> Check that battery support is compiled into your kernel. If you use
1601 it as a module, make sure the module is loaded.
1602 </p>
1603
1604 <p>
1605 <e>Q:</e> My system logger reports things like "logger: ACPI group battery / action
1606 battery is not defined".
1607 </p>
1608
1609 <p>
1610 <e>A:</e> This message is generated by the /etc/acpi/default.sh script that is
1611 shipped with acpid. You can safely ignore it. If you like to get rid of it, you
1612 can comment the appropriate line in /etc/acpi/default.sh as shown below:
1613 </p>
1614
1615 <pre caption="Disabling warnings about unknown acpi events">
1616 *) # logger "ACPI action $action is not defined"
1617 </pre>
1618
1619 <p>
1620 <e>Q:</e> I have a Dell Inspiron 51XX and I don't get any ACPI events.
1621 </p>
1622
1623 <p>
1624 <e>A:</e> This seems to be a kernel bug. Read on <uri
1625 link="http://bugme.osdl.org/show_bug.cgi?id=1752">here</uri>.
1626 </p>
1627
1628 <p>
1629 <e>Q:</e> I activated the DynamicClocks option in <path>xorg.conf</path> and
1630 now X.org crashes / the screen stays black / my laptop doesn't shutdown
1631 properly.
1632 </p>
1633
1634 <p>
1635 <e>A:</e> This happens on some systems. You have to disable DynamicClocks.
1636 </p>
1637
1638 <p>
1639 <e>Q:</e> I want to use suspend2, but it tells me my swap partition is too
1640 small. Resizing is not an option.
1641 </p>
1642
1643 <p>
1644 <e>A:</e> If there is enough free space on your system, you can use the
1645 filewriter instead of the swapwriter. The <c>hibernate-script</c> supports it
1646 as well. More information can be found in
1647 <path>/usr/src/linux/Documentation/power/suspend2.txt</path>.
1648 </p>
1649
1650 <p>
1651 <e>Q:</e> I just bought a brand new battery, but it only lasts for some
1652 minutes! What am I doing wrong?
1653 </p>
1654
1655 <p>
1656 <e>A:</e> First follow your manufacturer's advice on how to charge the battery
1657 correctly.
1658 </p>
1659
1660 <p>
1661 <e>Q:</e> The above didn't help. What should I do then?
1662 </p>
1663
1664 <p>
1665 <e>A:</e> Some batteries sold as "new" are in fact old ones. Try the following:
1666 </p>
1667
1668 <pre caption="Querying battery state">
1669 $ <i>grep capacity /proc/acpi/battery/BAT0/info</i>
1670 design capacity: 47520 mWh
1671 last full capacity: 41830 mWh
1672 </pre>
1673
1674 <p>
1675 If the "last full capacity" differs significantly from the design capacity,
1676 your battery is probably broken. Try to claim your warranty.
1677 </p>
1678
1679 <p>
1680 <e>Q:</e> My problem is not listed above. Where should I go next?
1681 </p>
1682
1683 <p>
1684 <e>A:</e> Don't fear to contact me, <mail link="earthwings@gentoo.org">Dennis
1685 Nienhüser</mail>, directly. The
1686 <uri link="http://forums.gentoo.org">Gentoo Forums</uri> are a good place to
1687 get help as well. If you prefer IRC, try the <e>#gentoo-laptop</e> channel at
1688 <e>irc.freenode.net</e>.
1689 </p>
1690
1691 </body>
1692 </section>
1693 </chapter>
1694 </guide>

  ViewVC Help
Powered by ViewVC 1.1.20