ia64/linux-2.6.18-xen.hg

view Documentation/ibm-acpi.txt @ 897:329ea0ccb344

balloon: try harder to balloon up under memory pressure.

Currently if the balloon driver is unable to increase the guest's
reservation it assumes the failure was due to reaching its full
allocation, gives up on the ballooning operation and records the limit
it reached as the "hard limit". The driver will not try again until
the target is set again (even to the same value).

However it is possible that ballooning has in fact failed due to
memory pressure in the host and therefore it is desirable to keep
attempting to reach the target in case memory becomes available. The
most likely scenario is that some guests are ballooning down while
others are ballooning up and therefore there is temporary memory
pressure while things stabilise. You would not expect a well behaved
toolstack to ask a domain to balloon to more than its allocation nor
would you expect it to deliberately over-commit memory by setting
balloon targets which exceed the total host memory.

This patch drops the concept of a hard limit and causes the balloon
driver to retry increasing the reservation on a timer in the same
manner as when decreasing the reservation.

Also if we partially succeed in increasing the reservation
(i.e. receive less pages than we asked for) then we may as well keep
those pages rather than returning them to Xen.

Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
author Keir Fraser <keir.fraser@citrix.com>
date Fri Jun 05 14:01:20 2009 +0100 (2009-06-05)
parents 831230e53067
children
line source
1 IBM ThinkPad ACPI Extras Driver
3 Version 0.12
4 17 August 2005
6 Borislav Deianov <borislav@users.sf.net>
7 http://ibm-acpi.sf.net/
10 This is a Linux ACPI driver for the IBM ThinkPad laptops. It supports
11 various features of these laptops which are accessible through the
12 ACPI framework but not otherwise supported by the generic Linux ACPI
13 drivers.
16 Status
17 ------
19 The features currently supported are the following (see below for
20 detailed description):
22 - Fn key combinations
23 - Bluetooth enable and disable
24 - video output switching, expansion control
25 - ThinkLight on and off
26 - limited docking and undocking
27 - UltraBay eject
28 - CMOS control
29 - LED control
30 - ACPI sounds
31 - temperature sensors
32 - Experimental: embedded controller register dump
33 - Experimental: LCD brightness control
34 - Experimental: volume control
35 - Experimental: fan speed, fan enable/disable
37 A compatibility table by model and feature is maintained on the web
38 site, http://ibm-acpi.sf.net/. I appreciate any success or failure
39 reports, especially if they add to or correct the compatibility table.
40 Please include the following information in your report:
42 - ThinkPad model name
43 - a copy of your DSDT, from /proc/acpi/dsdt
44 - which driver features work and which don't
45 - the observed behavior of non-working features
47 Any other comments or patches are also more than welcome.
50 Installation
51 ------------
53 If you are compiling this driver as included in the Linux kernel
54 sources, simply enable the CONFIG_ACPI_IBM option (Power Management /
55 ACPI / IBM ThinkPad Laptop Extras). The rest of this section describes
56 how to install this driver when downloaded from the web site.
58 First, you need to get a kernel with ACPI support up and running.
59 Please refer to http://acpi.sourceforge.net/ for help with this
60 step. How successful you will be depends a lot on you ThinkPad model,
61 the kernel you are using and any additional patches applied. The
62 kernel provided with your distribution may not be good enough. I
63 needed to compile a 2.6.7 kernel with the 20040715 ACPI patch to get
64 ACPI working reliably on my ThinkPad X40. Old ThinkPad models may not
65 be supported at all.
67 Assuming you have the basic ACPI support working (e.g. you can see the
68 /proc/acpi directory), follow the following steps to install this
69 driver:
71 - unpack the archive:
73 tar xzvf ibm-acpi-x.y.tar.gz; cd ibm-acpi-x.y
75 - compile the driver:
77 make
79 - install the module in your kernel modules directory:
81 make install
83 - load the module:
85 modprobe ibm_acpi
87 After loading the module, check the "dmesg" output for any error messages.
90 Features
91 --------
93 The driver creates the /proc/acpi/ibm directory. There is a file under
94 that directory for each feature described below. Note that while the
95 driver is still in the alpha stage, the exact proc file format and
96 commands supported by the various features is guaranteed to change
97 frequently.
99 Driver version -- /proc/acpi/ibm/driver
100 ---------------------------------------
102 The driver name and version. No commands can be written to this file.
104 Hot keys -- /proc/acpi/ibm/hotkey
105 ---------------------------------
107 Without this driver, only the Fn-F4 key (sleep button) generates an
108 ACPI event. With the driver loaded, the hotkey feature enabled and the
109 mask set (see below), the various hot keys generate ACPI events in the
110 following format:
112 ibm/hotkey HKEY 00000080 0000xxxx
114 The last four digits vary depending on the key combination pressed.
115 All labeled Fn-Fx key combinations generate distinct events. In
116 addition, the lid microswitch and some docking station buttons may
117 also generate such events.
119 The following commands can be written to this file:
121 echo enable > /proc/acpi/ibm/hotkey -- enable the hot keys feature
122 echo disable > /proc/acpi/ibm/hotkey -- disable the hot keys feature
123 echo 0xffff > /proc/acpi/ibm/hotkey -- enable all possible hot keys
124 echo 0x0000 > /proc/acpi/ibm/hotkey -- disable all possible hot keys
125 ... any other 4-hex-digit mask ...
126 echo reset > /proc/acpi/ibm/hotkey -- restore the original mask
128 The bit mask allows some control over which hot keys generate ACPI
129 events. Not all bits in the mask can be modified. Not all bits that
130 can be modified do anything. Not all hot keys can be individually
131 controlled by the mask. Most recent ThinkPad models honor the
132 following bits (assuming the hot keys feature has been enabled):
134 key bit behavior when set behavior when unset
136 Fn-F3 always generates ACPI event
137 Fn-F4 always generates ACPI event
138 Fn-F5 0010 generate ACPI event enable/disable Bluetooth
139 Fn-F7 0040 generate ACPI event switch LCD and external display
140 Fn-F8 0080 generate ACPI event expand screen or none
141 Fn-F9 0100 generate ACPI event none
142 Fn-F12 always generates ACPI event
144 Some models do not support all of the above. For example, the T30 does
145 not support Fn-F5 and Fn-F9. Other models do not support the mask at
146 all. On those models, hot keys cannot be controlled individually.
148 Note that enabling ACPI events for some keys prevents their default
149 behavior. For example, if events for Fn-F5 are enabled, that key will
150 no longer enable/disable Bluetooth by itself. This can still be done
151 from an acpid handler for the ibm/hotkey event.
153 Note also that not all Fn key combinations are supported through
154 ACPI. For example, on the X40, the brightness, volume and "Access IBM"
155 buttons do not generate ACPI events even with this driver. They *can*
156 be used through the "ThinkPad Buttons" utility, see
157 http://www.nongnu.org/tpb/
159 Bluetooth -- /proc/acpi/ibm/bluetooth
160 -------------------------------------
162 This feature shows the presence and current state of a Bluetooth
163 device. If Bluetooth is installed, the following commands can be used:
165 echo enable > /proc/acpi/ibm/bluetooth
166 echo disable > /proc/acpi/ibm/bluetooth
168 Video output control -- /proc/acpi/ibm/video
169 --------------------------------------------
171 This feature allows control over the devices used for video output -
172 LCD, CRT or DVI (if available). The following commands are available:
174 echo lcd_enable > /proc/acpi/ibm/video
175 echo lcd_disable > /proc/acpi/ibm/video
176 echo crt_enable > /proc/acpi/ibm/video
177 echo crt_disable > /proc/acpi/ibm/video
178 echo dvi_enable > /proc/acpi/ibm/video
179 echo dvi_disable > /proc/acpi/ibm/video
180 echo auto_enable > /proc/acpi/ibm/video
181 echo auto_disable > /proc/acpi/ibm/video
182 echo expand_toggle > /proc/acpi/ibm/video
183 echo video_switch > /proc/acpi/ibm/video
185 Each video output device can be enabled or disabled individually.
186 Reading /proc/acpi/ibm/video shows the status of each device.
188 Automatic video switching can be enabled or disabled. When automatic
189 video switching is enabled, certain events (e.g. opening the lid,
190 docking or undocking) cause the video output device to change
191 automatically. While this can be useful, it also causes flickering
192 and, on the X40, video corruption. By disabling automatic switching,
193 the flickering or video corruption can be avoided.
195 The video_switch command cycles through the available video outputs
196 (it simulates the behavior of Fn-F7).
198 Video expansion can be toggled through this feature. This controls
199 whether the display is expanded to fill the entire LCD screen when a
200 mode with less than full resolution is used. Note that the current
201 video expansion status cannot be determined through this feature.
203 Note that on many models (particularly those using Radeon graphics
204 chips) the X driver configures the video card in a way which prevents
205 Fn-F7 from working. This also disables the video output switching
206 features of this driver, as it uses the same ACPI methods as
207 Fn-F7. Video switching on the console should still work.
209 UPDATE: There's now a patch for the X.org Radeon driver which
210 addresses this issue. Some people are reporting success with the patch
211 while others are still having problems. For more information:
213 https://bugs.freedesktop.org/show_bug.cgi?id=2000
215 ThinkLight control -- /proc/acpi/ibm/light
216 ------------------------------------------
218 The current status of the ThinkLight can be found in this file. A few
219 models which do not make the status available will show it as
220 "unknown". The available commands are:
222 echo on > /proc/acpi/ibm/light
223 echo off > /proc/acpi/ibm/light
225 Docking / undocking -- /proc/acpi/ibm/dock
226 ------------------------------------------
228 Docking and undocking (e.g. with the X4 UltraBase) requires some
229 actions to be taken by the operating system to safely make or break
230 the electrical connections with the dock.
232 The docking feature of this driver generates the following ACPI events:
234 ibm/dock GDCK 00000003 00000001 -- eject request
235 ibm/dock GDCK 00000003 00000002 -- undocked
236 ibm/dock GDCK 00000000 00000003 -- docked
238 NOTE: These events will only be generated if the laptop was docked
239 when originally booted. This is due to the current lack of support for
240 hot plugging of devices in the Linux ACPI framework. If the laptop was
241 booted while not in the dock, the following message is shown in the
242 logs:
244 Mar 17 01:42:34 aero kernel: ibm_acpi: dock device not present
246 In this case, no dock-related events are generated but the dock and
247 undock commands described below still work. They can be executed
248 manually or triggered by Fn key combinations (see the example acpid
249 configuration files included in the driver tarball package available
250 on the web site).
252 When the eject request button on the dock is pressed, the first event
253 above is generated. The handler for this event should issue the
254 following command:
256 echo undock > /proc/acpi/ibm/dock
258 After the LED on the dock goes off, it is safe to eject the laptop.
259 Note: if you pressed this key by mistake, go ahead and eject the
260 laptop, then dock it back in. Otherwise, the dock may not function as
261 expected.
263 When the laptop is docked, the third event above is generated. The
264 handler for this event should issue the following command to fully
265 enable the dock:
267 echo dock > /proc/acpi/ibm/dock
269 The contents of the /proc/acpi/ibm/dock file shows the current status
270 of the dock, as provided by the ACPI framework.
272 The docking support in this driver does not take care of enabling or
273 disabling any other devices you may have attached to the dock. For
274 example, a CD drive plugged into the UltraBase needs to be disabled or
275 enabled separately. See the provided example acpid configuration files
276 for how this can be accomplished.
278 There is no support yet for PCI devices that may be attached to a
279 docking station, e.g. in the ThinkPad Dock II. The driver currently
280 does not recognize, enable or disable such devices. This means that
281 the only docking stations currently supported are the X-series
282 UltraBase docks and "dumb" port replicators like the Mini Dock (the
283 latter don't need any ACPI support, actually).
285 UltraBay eject -- /proc/acpi/ibm/bay
286 ------------------------------------
288 Inserting or ejecting an UltraBay device requires some actions to be
289 taken by the operating system to safely make or break the electrical
290 connections with the device.
292 This feature generates the following ACPI events:
294 ibm/bay MSTR 00000003 00000000 -- eject request
295 ibm/bay MSTR 00000001 00000000 -- eject lever inserted
297 NOTE: These events will only be generated if the UltraBay was present
298 when the laptop was originally booted (on the X series, the UltraBay
299 is in the dock, so it may not be present if the laptop was undocked).
300 This is due to the current lack of support for hot plugging of devices
301 in the Linux ACPI framework. If the laptop was booted without the
302 UltraBay, the following message is shown in the logs:
304 Mar 17 01:42:34 aero kernel: ibm_acpi: bay device not present
306 In this case, no bay-related events are generated but the eject
307 command described below still works. It can be executed manually or
308 triggered by a hot key combination.
310 Sliding the eject lever generates the first event shown above. The
311 handler for this event should take whatever actions are necessary to
312 shut down the device in the UltraBay (e.g. call idectl), then issue
313 the following command:
315 echo eject > /proc/acpi/ibm/bay
317 After the LED on the UltraBay goes off, it is safe to pull out the
318 device.
320 When the eject lever is inserted, the second event above is
321 generated. The handler for this event should take whatever actions are
322 necessary to enable the UltraBay device (e.g. call idectl).
324 The contents of the /proc/acpi/ibm/bay file shows the current status
325 of the UltraBay, as provided by the ACPI framework.
327 EXPERIMENTAL warm eject support on the 600e/x, A22p and A3x (To use
328 this feature, you need to supply the experimental=1 parameter when
329 loading the module):
331 These models do not have a button near the UltraBay device to request
332 a hot eject but rather require the laptop to be put to sleep
333 (suspend-to-ram) before the bay device is ejected or inserted).
334 The sequence of steps to eject the device is as follows:
336 echo eject > /proc/acpi/ibm/bay
337 put the ThinkPad to sleep
338 remove the drive
339 resume from sleep
340 cat /proc/acpi/ibm/bay should show that the drive was removed
342 On the A3x, both the UltraBay 2000 and UltraBay Plus devices are
343 supported. Use "eject2" instead of "eject" for the second bay.
345 Note: the UltraBay eject support on the 600e/x, A22p and A3x is
346 EXPERIMENTAL and may not work as expected. USE WITH CAUTION!
348 CMOS control -- /proc/acpi/ibm/cmos
349 -----------------------------------
351 This feature is used internally by the ACPI firmware to control the
352 ThinkLight on most newer ThinkPad models. It may also control LCD
353 brightness, sounds volume and more, but only on some models.
355 The commands are non-negative integer numbers:
357 echo 0 >/proc/acpi/ibm/cmos
358 echo 1 >/proc/acpi/ibm/cmos
359 echo 2 >/proc/acpi/ibm/cmos
360 ...
362 The range of valid numbers is 0 to 21, but not all have an effect and
363 the behavior varies from model to model. Here is the behavior on the
364 X40 (tpb is the ThinkPad Buttons utility):
366 0 - no effect but tpb reports "Volume down"
367 1 - no effect but tpb reports "Volume up"
368 2 - no effect but tpb reports "Mute on"
369 3 - simulate pressing the "Access IBM" button
370 4 - LCD brightness up
371 5 - LCD brightness down
372 11 - toggle screen expansion
373 12 - ThinkLight on
374 13 - ThinkLight off
375 14 - no effect but tpb reports ThinkLight status change
377 LED control -- /proc/acpi/ibm/led
378 ---------------------------------
380 Some of the LED indicators can be controlled through this feature. The
381 available commands are:
383 echo '<led number> on' >/proc/acpi/ibm/led
384 echo '<led number> off' >/proc/acpi/ibm/led
385 echo '<led number> blink' >/proc/acpi/ibm/led
387 The <led number> range is 0 to 7. The set of LEDs that can be
388 controlled varies from model to model. Here is the mapping on the X40:
390 0 - power
391 1 - battery (orange)
392 2 - battery (green)
393 3 - UltraBase
394 4 - UltraBay
395 7 - standby
397 All of the above can be turned on and off and can be made to blink.
399 ACPI sounds -- /proc/acpi/ibm/beep
400 ----------------------------------
402 The BEEP method is used internally by the ACPI firmware to provide
403 audible alerts in various situations. This feature allows the same
404 sounds to be triggered manually.
406 The commands are non-negative integer numbers:
408 echo <number> >/proc/acpi/ibm/beep
410 The valid <number> range is 0 to 17. Not all numbers trigger sounds
411 and the sounds vary from model to model. Here is the behavior on the
412 X40:
414 0 - stop a sound in progress (but use 17 to stop 16)
415 2 - two beeps, pause, third beep ("low battery")
416 3 - single beep
417 4 - high, followed by low-pitched beep ("unable")
418 5 - single beep
419 6 - very high, followed by high-pitched beep ("AC/DC")
420 7 - high-pitched beep
421 9 - three short beeps
422 10 - very long beep
423 12 - low-pitched beep
424 15 - three high-pitched beeps repeating constantly, stop with 0
425 16 - one medium-pitched beep repeating constantly, stop with 17
426 17 - stop 16
428 Temperature sensors -- /proc/acpi/ibm/thermal
429 ---------------------------------------------
431 Most ThinkPads include six or more separate temperature sensors but
432 only expose the CPU temperature through the standard ACPI methods.
433 This feature shows readings from up to eight different sensors. Some
434 readings may not be valid, e.g. may show large negative values. For
435 example, on the X40, a typical output may be:
437 temperatures: 42 42 45 41 36 -128 33 -128
439 Thomas Gruber took his R51 apart and traced all six active sensors in
440 his laptop (the location of sensors may vary on other models):
442 1: CPU
443 2: Mini PCI Module
444 3: HDD
445 4: GPU
446 5: Battery
447 6: N/A
448 7: Battery
449 8: N/A
451 No commands can be written to this file.
453 EXPERIMENTAL: Embedded controller reigster dump -- /proc/acpi/ibm/ecdump
454 ------------------------------------------------------------------------
456 This feature is marked EXPERIMENTAL because the implementation
457 directly accesses hardware registers and may not work as expected. USE
458 WITH CAUTION! To use this feature, you need to supply the
459 experimental=1 parameter when loading the module.
461 This feature dumps the values of 256 embedded controller
462 registers. Values which have changed since the last time the registers
463 were dumped are marked with a star:
465 [root@x40 ibm-acpi]# cat /proc/acpi/ibm/ecdump
466 EC +00 +01 +02 +03 +04 +05 +06 +07 +08 +09 +0a +0b +0c +0d +0e +0f
467 EC 0x00: a7 47 87 01 fe 96 00 08 01 00 cb 00 00 00 40 00
468 EC 0x10: 00 00 ff ff f4 3c 87 09 01 ff 42 01 ff ff 0d 00
469 EC 0x20: 00 00 00 00 00 00 00 00 00 00 00 03 43 00 00 80
470 EC 0x30: 01 07 1a 00 30 04 00 00 *85 00 00 10 00 50 00 00
471 EC 0x40: 00 00 00 00 00 00 14 01 00 04 00 00 00 00 00 00
472 EC 0x50: 00 c0 02 0d 00 01 01 02 02 03 03 03 03 *bc *02 *bc
473 EC 0x60: *02 *bc *02 00 00 00 00 00 00 00 00 00 00 00 00 00
474 EC 0x70: 00 00 00 00 00 12 30 40 *24 *26 *2c *27 *20 80 *1f 80
475 EC 0x80: 00 00 00 06 *37 *0e 03 00 00 00 0e 07 00 00 00 00
476 EC 0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
477 EC 0xa0: *ff 09 ff 09 ff ff *64 00 *00 *00 *a2 41 *ff *ff *e0 00
478 EC 0xb0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
479 EC 0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
480 EC 0xd0: 03 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
481 EC 0xe0: 00 00 00 00 00 00 00 00 11 20 49 04 24 06 55 03
482 EC 0xf0: 31 55 48 54 35 38 57 57 08 2f 45 73 07 65 6c 1a
484 This feature can be used to determine the register holding the fan
485 speed on some models. To do that, do the following:
487 - make sure the battery is fully charged
488 - make sure the fan is running
489 - run 'cat /proc/acpi/ibm/ecdump' several times, once per second or so
491 The first step makes sure various charging-related values don't
492 vary. The second ensures that the fan-related values do vary, since
493 the fan speed fluctuates a bit. The third will (hopefully) mark the
494 fan register with a star:
496 [root@x40 ibm-acpi]# cat /proc/acpi/ibm/ecdump
497 EC +00 +01 +02 +03 +04 +05 +06 +07 +08 +09 +0a +0b +0c +0d +0e +0f
498 EC 0x00: a7 47 87 01 fe 96 00 08 01 00 cb 00 00 00 40 00
499 EC 0x10: 00 00 ff ff f4 3c 87 09 01 ff 42 01 ff ff 0d 00
500 EC 0x20: 00 00 00 00 00 00 00 00 00 00 00 03 43 00 00 80
501 EC 0x30: 01 07 1a 00 30 04 00 00 85 00 00 10 00 50 00 00
502 EC 0x40: 00 00 00 00 00 00 14 01 00 04 00 00 00 00 00 00
503 EC 0x50: 00 c0 02 0d 00 01 01 02 02 03 03 03 03 bc 02 bc
504 EC 0x60: 02 bc 02 00 00 00 00 00 00 00 00 00 00 00 00 00
505 EC 0x70: 00 00 00 00 00 12 30 40 24 27 2c 27 21 80 1f 80
506 EC 0x80: 00 00 00 06 *be 0d 03 00 00 00 0e 07 00 00 00 00
507 EC 0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
508 EC 0xa0: ff 09 ff 09 ff ff 64 00 00 00 a2 41 ff ff e0 00
509 EC 0xb0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
510 EC 0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
511 EC 0xd0: 03 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
512 EC 0xe0: 00 00 00 00 00 00 00 00 11 20 49 04 24 06 55 03
513 EC 0xf0: 31 55 48 54 35 38 57 57 08 2f 45 73 07 65 6c 1a
515 Another set of values that varies often is the temperature
516 readings. Since temperatures don't change vary fast, you can take
517 several quick dumps to eliminate them.
519 You can use a similar method to figure out the meaning of other
520 embedded controller registers - e.g. make sure nothing else changes
521 except the charging or discharging battery to determine which
522 registers contain the current battery capacity, etc. If you experiment
523 with this, do send me your results (including some complete dumps with
524 a description of the conditions when they were taken.)
526 EXPERIMENTAL: LCD brightness control -- /proc/acpi/ibm/brightness
527 -----------------------------------------------------------------
529 This feature is marked EXPERIMENTAL because the implementation
530 directly accesses hardware registers and may not work as expected. USE
531 WITH CAUTION! To use this feature, you need to supply the
532 experimental=1 parameter when loading the module.
534 This feature allows software control of the LCD brightness on ThinkPad
535 models which don't have a hardware brightness slider. The available
536 commands are:
538 echo up >/proc/acpi/ibm/brightness
539 echo down >/proc/acpi/ibm/brightness
540 echo 'level <level>' >/proc/acpi/ibm/brightness
542 The <level> number range is 0 to 7, although not all of them may be
543 distinct. The current brightness level is shown in the file.
545 EXPERIMENTAL: Volume control -- /proc/acpi/ibm/volume
546 -----------------------------------------------------
548 This feature is marked EXPERIMENTAL because the implementation
549 directly accesses hardware registers and may not work as expected. USE
550 WITH CAUTION! To use this feature, you need to supply the
551 experimental=1 parameter when loading the module.
553 This feature allows volume control on ThinkPad models which don't have
554 a hardware volume knob. The available commands are:
556 echo up >/proc/acpi/ibm/volume
557 echo down >/proc/acpi/ibm/volume
558 echo mute >/proc/acpi/ibm/volume
559 echo 'level <level>' >/proc/acpi/ibm/volume
561 The <level> number range is 0 to 15 although not all of them may be
562 distinct. The unmute the volume after the mute command, use either the
563 up or down command (the level command will not unmute the volume).
564 The current volume level and mute state is shown in the file.
566 EXPERIMENTAL: fan speed, fan enable/disable -- /proc/acpi/ibm/fan
567 -----------------------------------------------------------------
569 This feature is marked EXPERIMENTAL because the implementation
570 directly accesses hardware registers and may not work as expected. USE
571 WITH CAUTION! To use this feature, you need to supply the
572 experimental=1 parameter when loading the module.
574 This feature attempts to show the current fan speed. The speed is read
575 directly from the hardware registers of the embedded controller. This
576 is known to work on later R, T and X series ThinkPads but may show a
577 bogus value on other models.
579 The fan may be enabled or disabled with the following commands:
581 echo enable >/proc/acpi/ibm/fan
582 echo disable >/proc/acpi/ibm/fan
584 WARNING WARNING WARNING: do not leave the fan disabled unless you are
585 monitoring the temperature sensor readings and you are ready to enable
586 it if necessary to avoid overheating.
588 The fan only runs if it's enabled *and* the various temperature
589 sensors which control it read high enough. On the X40, this seems to
590 depend on the CPU and HDD temperatures. Specifically, the fan is
591 turned on when either the CPU temperature climbs to 56 degrees or the
592 HDD temperature climbs to 46 degrees. The fan is turned off when the
593 CPU temperature drops to 49 degrees and the HDD temperature drops to
594 41 degrees. These thresholds cannot currently be controlled.
596 On the X31 and X40 (and ONLY on those models), the fan speed can be
597 controlled to a certain degree. Once the fan is running, it can be
598 forced to run faster or slower with the following command:
600 echo 'speed <speed>' > /proc/acpi/ibm/thermal
602 The sustainable range of fan speeds on the X40 appears to be from
603 about 3700 to about 7350. Values outside this range either do not have
604 any effect or the fan speed eventually settles somewhere in that
605 range. The fan cannot be stopped or started with this command.
607 On the 570, temperature readings are not available through this
608 feature and the fan control works a little differently. The fan speed
609 is reported in levels from 0 (off) to 7 (max) and can be controlled
610 with the following command:
612 echo 'level <level>' > /proc/acpi/ibm/thermal
615 Multiple Commands, Module Parameters
616 ------------------------------------
618 Multiple commands can be written to the proc files in one shot by
619 separating them with commas, for example:
621 echo enable,0xffff > /proc/acpi/ibm/hotkey
622 echo lcd_disable,crt_enable > /proc/acpi/ibm/video
624 Commands can also be specified when loading the ibm_acpi module, for
625 example:
627 modprobe ibm_acpi hotkey=enable,0xffff video=auto_disable
630 Example Configuration
631 ---------------------
633 The ACPI support in the kernel is intended to be used in conjunction
634 with a user-space daemon, acpid. The configuration files for this
635 daemon control what actions are taken in response to various ACPI
636 events. An example set of configuration files are included in the
637 config/ directory of the tarball package available on the web
638 site. Note that these are provided for illustration purposes only and
639 may need to be adapted to your particular setup.
641 The following utility scripts are used by the example action
642 scripts (included with ibm-acpi for completeness):
644 /usr/local/sbin/idectl -- from the hdparm source distribution,
645 see http://www.ibiblio.org/pub/Linux/system/hardware
646 /usr/local/sbin/laptop_mode -- from the Linux kernel source
647 distribution, see Documentation/laptop-mode.txt
648 /sbin/service -- comes with Redhat/Fedora distributions
649 /usr/sbin/hibernate -- from the Software Suspend 2 distribution,
650 see http://softwaresuspend.berlios.de/
652 Toan T Nguyen <ntt@physics.ucla.edu> notes that Suse uses the
653 powersave program to suspend ('powersave --suspend-to-ram') or
654 hibernate ('powersave --suspend-to-disk'). This means that the
655 hibernate script is not needed on that distribution.
657 Henrik Brix Andersen <brix@gentoo.org> has written a Gentoo ACPI event
658 handler script for the X31. You can get the latest version from
659 http://dev.gentoo.org/~brix/files/x31.sh
661 David Schweikert <dws@ee.eth.ch> has written an alternative blank.sh
662 script which works on Debian systems. This scripts has now been
663 extended to also work on Fedora systems and included as the default
664 blank.sh in the distribution.