blob: e2ddcdeb61b6f5caffa1f80f901ebf5619f93dd7 [file] [log] [blame]
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -03001 ThinkPad ACPI Extras Driver
Linus Torvalds1da177e2005-04-16 15:20:36 -07002
Henrique de Moraes Holschuhb57f7e72009-04-14 02:44:14 +00003 Version 0.23
4 April 10th, 2009
Linus Torvalds1da177e2005-04-16 15:20:36 -07005
6 Borislav Deianov <borislav@users.sf.net>
Henrique de Moraes Holschuhc78d5c92007-07-18 23:45:47 -03007 Henrique de Moraes Holschuh <hmh@hmh.eng.br>
8 http://ibm-acpi.sf.net/
Linus Torvalds1da177e2005-04-16 15:20:36 -07009
10
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -030011This is a Linux driver for the IBM and Lenovo ThinkPad laptops. It
12supports various features of these laptops which are accessible
13through the ACPI and ACPI EC framework, but not otherwise fully
14supported by the generic Linux ACPI drivers.
15
16This driver used to be named ibm-acpi until kernel 2.6.21 and release
170.13-20070314. It used to be in the drivers/acpi tree, but it was
18moved to the drivers/misc tree and renamed to thinkpad-acpi for kernel
Henrique de Moraes Holschuh9abf0ee2009-01-11 03:00:58 -0200192.6.22, and release 0.14. It was moved to drivers/platform/x86 for
Henrique de Moraes Holschuhaa2fbce2009-01-11 03:01:10 -020020kernel 2.6.29 and release 0.22.
Linus Torvalds1da177e2005-04-16 15:20:36 -070021
Henrique de Moraes Holschuh95e57ab2008-04-26 01:02:22 -030022The driver is named "thinkpad-acpi". In some places, like module
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000023names and log messages, "thinkpad_acpi" is used because of userspace
24issues.
Henrique de Moraes Holschuh95e57ab2008-04-26 01:02:22 -030025
26"tpacpi" is used as a shorthand where "thinkpad-acpi" would be too
27long due to length limitations on some Linux kernel versions.
Linus Torvalds1da177e2005-04-16 15:20:36 -070028
29Status
30------
31
32The features currently supported are the following (see below for
33detailed description):
34
35 - Fn key combinations
36 - Bluetooth enable and disable
Henrique de Moraes Holschuh837ca6d2007-03-23 17:33:54 -030037 - video output switching, expansion control
Linus Torvalds1da177e2005-04-16 15:20:36 -070038 - ThinkLight on and off
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000039 - CMOS/UCMS control
Borislav Deianov78f81cc2005-08-17 00:00:00 -040040 - LED control
41 - ACPI sounds
42 - temperature sensors
43 - Experimental: embedded controller register dump
Stefan Schmidt24f7ff02006-09-22 12:19:15 +020044 - LCD brightness control
45 - Volume control
Henrique de Moraes Holschuhecf2a802007-04-27 22:00:09 -030046 - Fan control and monitoring: fan speed, fan enable/disable
Jeremy Fitzhardinge9662e082008-08-27 21:04:49 -070047 - WAN enable and disable
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000048 - UWB enable and disable
Linus Torvalds1da177e2005-04-16 15:20:36 -070049
50A compatibility table by model and feature is maintained on the web
51site, http://ibm-acpi.sf.net/. I appreciate any success or failure
52reports, especially if they add to or correct the compatibility table.
53Please include the following information in your report:
54
55 - ThinkPad model name
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000056 - a copy of your ACPI tables, using the "acpidump" utility
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -030057 - a copy of the output of dmidecode, with serial numbers
58 and UUIDs masked off
Linus Torvalds1da177e2005-04-16 15:20:36 -070059 - which driver features work and which don't
60 - the observed behavior of non-working features
61
62Any other comments or patches are also more than welcome.
63
64
65Installation
66------------
67
68If you are compiling this driver as included in the Linux kernel
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000069sources, look for the CONFIG_THINKPAD_ACPI Kconfig option.
70It is located on the menu path: "Device Drivers" -> "X86 Platform
71Specific Device Drivers" -> "ThinkPad ACPI Laptop Extras".
72
Linus Torvalds1da177e2005-04-16 15:20:36 -070073
74Features
75--------
76
Henrique de Moraes Holschuh54ae1502007-04-24 11:48:12 -030077The driver exports two different interfaces to userspace, which can be
78used to access the features it provides. One is a legacy procfs-based
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +000079interface, which will be removed at some time in the future. The other
80is a new sysfs-based interface which is not complete yet.
Henrique de Moraes Holschuh54ae1502007-04-24 11:48:12 -030081
82The procfs interface creates the /proc/acpi/ibm directory. There is a
83file under that directory for each feature it supports. The procfs
84interface is mostly frozen, and will change very little if at all: it
85will not be extended to add any new functionality in the driver, instead
86all new functionality will be implemented on the sysfs interface.
87
88The sysfs interface tries to blend in the generic Linux sysfs subsystems
89and classes as much as possible. Since some of these subsystems are not
90yet ready or stabilized, it is expected that this interface will change,
91and any and all userspace programs must deal with it.
92
93
94Notes about the sysfs interface:
95
96Unlike what was done with the procfs interface, correctness when talking
97to the sysfs interfaces will be enforced, as will correctness in the
98thinkpad-acpi's implementation of sysfs interfaces.
99
100Also, any bugs in the thinkpad-acpi sysfs driver code or in the
101thinkpad-acpi's implementation of the sysfs interfaces will be fixed for
102maximum correctness, even if that means changing an interface in
103non-compatible ways. As these interfaces mature both in the kernel and
104in thinkpad-acpi, such changes should become quite rare.
105
106Applications interfacing to the thinkpad-acpi sysfs interfaces must
107follow all sysfs guidelines and correctly process all errors (the sysfs
108interface makes extensive use of errors). File descriptors and open /
109close operations to the sysfs inodes must also be properly implemented.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700110
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -0300111The version of thinkpad-acpi's sysfs interface is exported by the driver
112as a driver attribute (see below).
113
114Sysfs driver attributes are on the driver's sysfs attribute space,
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000115for 2.6.23+ this is /sys/bus/platform/drivers/thinkpad_acpi/ and
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -0300116/sys/bus/platform/drivers/thinkpad_hwmon/
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -0300117
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -0300118Sysfs device attributes are on the thinkpad_acpi device sysfs attribute
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000119space, for 2.6.23+ this is /sys/devices/platform/thinkpad_acpi/.
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -0300120
121Sysfs device attributes for the sensors and fan are on the
122thinkpad_hwmon device's sysfs attribute space, but you should locate it
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000123looking for a hwmon device with the name attribute of "thinkpad", or
124better yet, through libsensors.
125
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -0300126
127Driver version
128--------------
129
130procfs: /proc/acpi/ibm/driver
131sysfs driver attribute: version
Linus Torvalds1da177e2005-04-16 15:20:36 -0700132
133The driver name and version. No commands can be written to this file.
134
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000135
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -0300136Sysfs interface version
137-----------------------
138
139sysfs driver attribute: interface_version
140
141Version of the thinkpad-acpi sysfs interface, as an unsigned long
142(output in hex format: 0xAAAABBCC), where:
143 AAAA - major revision
144 BB - minor revision
145 CC - bugfix revision
146
147The sysfs interface version changelog for the driver can be found at the
148end of this document. Changes to the sysfs interface done by the kernel
149subsystems are not documented here, nor are they tracked by this
150attribute.
151
Henrique de Moraes Holschuh94b08712007-07-18 23:45:32 -0300152Changes to the thinkpad-acpi sysfs interface are only considered
153non-experimental when they are submitted to Linux mainline, at which
154point the changes in this interface are documented and interface_version
155may be updated. If you are using any thinkpad-acpi features not yet
156sent to mainline for merging, you do so on your own risk: these features
157may disappear, or be implemented in a different and incompatible way by
158the time they are merged in Linux mainline.
159
160Changes that are backwards-compatible by nature (e.g. the addition of
161attributes that do not change the way the other attributes work) do not
162always warrant an update of interface_version. Therefore, one must
163expect that an attribute might not be there, and deal with it properly
164(an attribute not being there *is* a valid way to make it clear that a
165feature is not available in sysfs).
166
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000167
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300168Hot keys
169--------
170
171procfs: /proc/acpi/ibm/hotkey
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300172sysfs device attribute: hotkey_*
Linus Torvalds1da177e2005-04-16 15:20:36 -0700173
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200174In a ThinkPad, the ACPI HKEY handler is responsible for communicating
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300175some important events and also keyboard hot key presses to the operating
176system. Enabling the hotkey functionality of thinkpad-acpi signals the
177firmware that such a driver is present, and modifies how the ThinkPad
178firmware will behave in many situations.
179
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000180The driver enables the HKEY ("hot key") event reporting automatically
181when loaded, and disables it when it is removed.
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300182
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000183The driver will report HKEY events in the following format:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700184
185 ibm/hotkey HKEY 00000080 0000xxxx
186
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000187Some of these events refer to hot key presses, but not all of them.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300188
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300189The driver will generate events over the input layer for hot keys and
190radio switches, and over the ACPI netlink layer for other events. The
191input layer support accepts the standard IOCTLs to remap the keycodes
192assigned to each hot key.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300193
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300194The hot key bit mask allows some control over which hot keys generate
195events. If a key is "masked" (bit set to 0 in the mask), the firmware
196will handle it. If it is "unmasked", it signals the firmware that
197thinkpad-acpi would prefer to handle it, if the firmware would be so
198kind to allow it (and it often doesn't!).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700199
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300200Not all bits in the mask can be modified. Not all bits that can be
201modified do anything. Not all hot keys can be individually controlled
202by the mask. Some models do not support the mask at all, and in those
203models, hot keys cannot be controlled individually. The behaviour of
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200204the mask is, therefore, highly dependent on the ThinkPad model.
Henrique de Moraes Holschuhae92bd12007-07-18 23:45:29 -0300205
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300206Note that unmasking some keys prevents their default behavior. For
207example, if Fn+F5 is unmasked, that key will no longer enable/disable
208Bluetooth by itself.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700209
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300210Note also that not all Fn key combinations are supported through ACPI.
211For example, on the X40, the brightness, volume and "Access IBM" buttons
212do not generate ACPI events even with this driver. They *can* be used
213through the "ThinkPad Buttons" utility, see http://www.nongnu.org/tpb/
Linus Torvalds1da177e2005-04-16 15:20:36 -0700214
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300215procfs notes:
216
217The following commands can be written to the /proc/acpi/ibm/hotkey file:
218
Henrique de Moraes Holschuhae92bd12007-07-18 23:45:29 -0300219 echo 0xffffffff > /proc/acpi/ibm/hotkey -- enable all hot keys
220 echo 0 > /proc/acpi/ibm/hotkey -- disable all possible hot keys
221 ... any other 8-hex-digit mask ...
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300222 echo reset > /proc/acpi/ibm/hotkey -- restore the original mask
223
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000224The following commands have been deprecated and will cause the kernel
225to log a warning:
226
227 echo enable > /proc/acpi/ibm/hotkey -- does nothing
228 echo disable > /proc/acpi/ibm/hotkey -- returns an error
229
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -0200230The procfs interface does not support NVRAM polling control. So as to
231maintain maximum bug-to-bug compatibility, it does not report any masks,
232nor does it allow one to manipulate the hot key mask when the firmware
233does not support masks at all, even if NVRAM polling is in use.
234
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300235sysfs notes:
236
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300237 hotkey_bios_enabled:
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000238 DEPRECATED, WILL BE REMOVED SOON.
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300239
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000240 Returns 0.
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300241
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300242 hotkey_bios_mask:
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300243 Returns the hot keys mask when thinkpad-acpi was loaded.
244 Upon module unload, the hot keys mask will be restored
245 to this value.
246
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300247 hotkey_enable:
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000248 DEPRECATED, WILL BE REMOVED SOON.
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300249
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +0000250 0: returns -EPERM
251 1: does nothing
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300252
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300253 hotkey_mask:
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -0200254 bit mask to enable driver-handling (and depending on
255 the firmware, ACPI event generation) for each hot key
256 (see above). Returns the current status of the hot keys
257 mask, and allows one to modify it.
258
259 Note: when NVRAM polling is active, the firmware mask
260 will be different from the value returned by
261 hotkey_mask. The driver will retain enabled bits for
262 hotkeys that are under NVRAM polling even if the
263 firmware refuses them, and will not set these bits on
264 the firmware hot key mask.
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300265
Henrique de Moraes Holschuh9b010de2007-07-18 23:45:30 -0300266 hotkey_all_mask:
267 bit mask that should enable event reporting for all
268 supported hot keys, when echoed to hotkey_mask above.
269 Unless you know which events need to be handled
270 passively (because the firmware *will* handle them
271 anyway), do *not* use hotkey_all_mask. Use
272 hotkey_recommended_mask, instead. You have been warned.
273
274 hotkey_recommended_mask:
275 bit mask that should enable event reporting for all
Henrique de Moraes Holschuh1a343762007-07-18 23:45:36 -0300276 supported hot keys, except those which are always
277 handled by the firmware anyway. Echo it to
278 hotkey_mask above, to use.
Henrique de Moraes Holschuh9b010de2007-07-18 23:45:30 -0300279
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -0200280 hotkey_source_mask:
281 bit mask that selects which hot keys will the driver
282 poll the NVRAM for. This is auto-detected by the driver
283 based on the capabilities reported by the ACPI firmware,
284 but it can be overridden at runtime.
285
286 Hot keys whose bits are set in both hotkey_source_mask
287 and also on hotkey_mask are polled for in NVRAM. Only a
288 few hot keys are available through CMOS NVRAM polling.
289
290 Warning: when in NVRAM mode, the volume up/down/mute
291 keys are synthesized according to changes in the mixer,
292 so you have to use volume up or volume down to unmute,
293 as per the ThinkPad volume mixer user interface. When
294 in ACPI event mode, volume up/down/mute are reported as
295 separate events, but this behaviour may be corrected in
296 future releases of this driver, in which case the
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200297 ThinkPad volume mixer user interface semantics will be
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -0200298 enforced.
299
300 hotkey_poll_freq:
301 frequency in Hz for hot key polling. It must be between
302 0 and 25 Hz. Polling is only carried out when strictly
303 needed.
304
305 Setting hotkey_poll_freq to zero disables polling, and
306 will cause hot key presses that require NVRAM polling
307 to never be reported.
308
309 Setting hotkey_poll_freq too low will cause repeated
310 pressings of the same hot key to be misreported as a
311 single key press, or to not even be detected at all.
312 The recommended polling frequency is 10Hz.
313
Henrique de Moraes Holschuh74941a62007-07-18 23:45:31 -0300314 hotkey_radio_sw:
Henrique de Moraes Holschuhd147da72008-02-16 02:17:57 -0200315 If the ThinkPad has a hardware radio switch, this
Henrique de Moraes Holschuh74941a62007-07-18 23:45:31 -0300316 attribute will read 0 if the switch is in the "radios
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200317 disabled" position, and 1 if the switch is in the
Henrique de Moraes Holschuh74941a62007-07-18 23:45:31 -0300318 "radios enabled" position.
319
Henrique de Moraes Holschuh50ebec02008-01-08 13:02:55 -0200320 This attribute has poll()/select() support.
321
Henrique de Moraes Holschuh6c231bd2008-02-16 02:17:58 -0200322 hotkey_tablet_mode:
323 If the ThinkPad has tablet capabilities, this attribute
324 will read 0 if the ThinkPad is in normal mode, and
325 1 if the ThinkPad is in tablet mode.
326
327 This attribute has poll()/select() support.
328
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300329 hotkey_report_mode:
330 Returns the state of the procfs ACPI event report mode
331 filter for hot keys. If it is set to 1 (the default),
332 all hot key presses are reported both through the input
333 layer and also as ACPI events through procfs (but not
334 through netlink). If it is set to 2, hot key presses
335 are reported only through the input layer.
336
337 This attribute is read-only in kernels 2.6.23 or later,
338 and read-write on earlier kernels.
339
340 May return -EPERM (write access locked out by module
341 parameter) or -EACCES (read-only).
342
Henrique de Moraes Holschuha713b4d2008-01-08 13:02:52 -0200343 wakeup_reason:
344 Set to 1 if the system is waking up because the user
345 requested a bay ejection. Set to 2 if the system is
346 waking up because the user requested the system to
347 undock. Set to zero for normal wake-ups or wake-ups
348 due to unknown reasons.
349
Henrique de Moraes Holschuh50ebec02008-01-08 13:02:55 -0200350 This attribute has poll()/select() support.
351
Henrique de Moraes Holschuha713b4d2008-01-08 13:02:52 -0200352 wakeup_hotunplug_complete:
353 Set to 1 if the system was waken up because of an
354 undock or bay ejection request, and that request
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200355 was successfully completed. At this point, it might
Henrique de Moraes Holschuha713b4d2008-01-08 13:02:52 -0200356 be useful to send the system back to sleep, at the
357 user's choice. Refer to HKEY events 0x4003 and
358 0x3003, below.
359
Henrique de Moraes Holschuh50ebec02008-01-08 13:02:55 -0200360 This attribute has poll()/select() support.
361
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300362input layer notes:
363
364A Hot key is mapped to a single input layer EV_KEY event, possibly
365followed by an EV_MSC MSC_SCAN event that shall contain that key's scan
366code. An EV_SYN event will always be generated to mark the end of the
367event block.
368
369Do not use the EV_MSC MSC_SCAN events to process keys. They are to be
370used as a helper to remap keys, only. They are particularly useful when
371remapping KEY_UNKNOWN keys.
372
373The events are available in an input device, with the following id:
374
375 Bus: BUS_HOST
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300376 vendor: 0x1014 (PCI_VENDOR_ID_IBM) or
377 0x17aa (PCI_VENDOR_ID_LENOVO)
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300378 product: 0x5054 ("TP")
379 version: 0x4101
380
381The version will have its LSB incremented if the keymap changes in a
382backwards-compatible way. The MSB shall always be 0x41 for this input
383device. If the MSB is not 0x41, do not use the device as described in
384this section, as it is either something else (e.g. another input device
385exported by a thinkpad driver, such as HDAPS) or its functionality has
386been changed in a non-backwards compatible way.
387
388Adding other event types for other functionalities shall be considered a
389backwards-compatible change for this input device.
390
391Thinkpad-acpi Hot Key event map (version 0x4101):
392
393ACPI Scan
394event code Key Notes
395
3960x1001 0x00 FN+F1 -
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -03003970x1002 0x01 FN+F2 IBM: battery (rare)
398 Lenovo: Screen lock
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300399
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -03004000x1003 0x02 FN+F3 Many IBM models always report
401 this hot key, even with hot keys
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300402 disabled or with Fn+F3 masked
403 off
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300404 IBM: screen lock
405 Lenovo: battery
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300406
4070x1004 0x03 FN+F4 Sleep button (ACPI sleep button
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200408 semantics, i.e. sleep-to-RAM).
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300409 It is always generate some kind
410 of event, either the hot key
411 event or a ACPI sleep button
412 event. The firmware may
413 refuse to generate further FN+F4
414 key presses until a S3 or S4 ACPI
415 sleep cycle is performed or some
416 time passes.
417
4180x1005 0x04 FN+F5 Radio. Enables/disables
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200419 the internal Bluetooth hardware
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300420 and W-WAN card if left in control
421 of the firmware. Does not affect
422 the WLAN card.
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300423 Should be used to turn on/off all
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200424 radios (Bluetooth+W-WAN+WLAN),
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300425 really.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300426
4270x1006 0x05 FN+F6 -
428
4290x1007 0x06 FN+F7 Video output cycle.
430 Do you feel lucky today?
431
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -03004320x1008 0x07 FN+F8 IBM: toggle screen expand
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200433 Lenovo: configure UltraNav
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300434
4350x1009 0x08 FN+F9 -
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300436 .. .. ..
4370x100B 0x0A FN+F11 -
438
4390x100C 0x0B FN+F12 Sleep to disk. You are always
440 supposed to handle it yourself,
441 either through the ACPI event,
442 or through a hotkey event.
443 The firmware may refuse to
444 generate further FN+F4 key
445 press events until a S3 or S4
446 ACPI sleep cycle is performed,
447 or some time passes.
448
4490x100D 0x0C FN+BACKSPACE -
4500x100E 0x0D FN+INSERT -
4510x100F 0x0E FN+DELETE -
452
4530x1010 0x0F FN+HOME Brightness up. This key is
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300454 always handled by the firmware
455 in IBM ThinkPads, even when
456 unmasked. Just leave it alone.
457 For Lenovo ThinkPads with a new
458 BIOS, it has to be handled either
459 by the ACPI OSI, or by userspace.
4600x1011 0x10 FN+END Brightness down. See brightness
461 up for details.
462
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -02004630x1012 0x11 FN+PGUP ThinkLight toggle. This key is
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300464 always handled by the firmware,
465 even when unmasked.
466
4670x1013 0x12 FN+PGDOWN -
468
4690x1014 0x13 FN+SPACE Zoom key
470
4710x1015 0x14 VOLUME UP Internal mixer volume up. This
472 key is always handled by the
473 firmware, even when unmasked.
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300474 NOTE: Lenovo seems to be changing
475 this.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -03004760x1016 0x15 VOLUME DOWN Internal mixer volume up. This
477 key is always handled by the
478 firmware, even when unmasked.
Henrique de Moraes Holschuhedf0e0e2007-07-18 23:45:44 -0300479 NOTE: Lenovo seems to be changing
480 this.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -03004810x1017 0x16 MUTE Mute internal mixer. This
482 key is always handled by the
483 firmware, even when unmasked.
484
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -02004850x1018 0x17 THINKPAD ThinkPad/Access IBM/Lenovo key
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300486
4870x1019 0x18 unknown
488.. .. ..
4890x1020 0x1F unknown
490
491The ThinkPad firmware does not allow one to differentiate when most hot
492keys are pressed or released (either that, or we don't know how to, yet).
493For these keys, the driver generates a set of events for a key press and
494immediately issues the same set of events for a key release. It is
495unknown by the driver if the ThinkPad firmware triggered these events on
496hot key press or release, but the firmware will do it for either one, not
497both.
498
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300499If a key is mapped to KEY_RESERVED, it generates no input events at all.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300500If a key is mapped to KEY_UNKNOWN, it generates an input event that
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300501includes an scan code. If a key is mapped to anything else, it will
502generate input device EV_KEY events.
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300503
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200504In addition to the EV_KEY events, thinkpad-acpi may also issue EV_SW
505events for switches:
506
Matt LaPlante19f59462009-04-27 15:06:31 +0200507SW_RFKILL_ALL T60 and later hardware rfkill rocker switch
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200508SW_TABLET_MODE Tablet ThinkPads HKEY events 0x5009 and 0x500A
509
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -0300510Non hot-key ACPI HKEY event map:
5110x5001 Lid closed
5120x5002 Lid opened
Henrique de Moraes Holschuhb3ec6f92008-02-16 02:17:55 -02005130x5009 Tablet swivel: switched to tablet mode
5140x500A Tablet swivel: switched to normal mode
Henrique de Moraes Holschuh6a38abb2007-07-18 23:45:35 -03005150x7000 Radio Switch may have changed state
516
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300517The above events are not propagated by the driver, except for legacy
518compatibility purposes when hotkey_report_mode is set to 1.
519
Henrique de Moraes Holschuha713b4d2008-01-08 13:02:52 -02005200x2304 System is waking up from suspend to undock
5210x2305 System is waking up from suspend to eject bay
5220x2404 System is waking up from hibernation to undock
5230x2405 System is waking up from hibernation to eject bay
524
525The above events are never propagated by the driver.
526
5270x3003 Bay ejection (see 0x2x05) complete, can sleep again
5280x4003 Undocked (see 0x2x04), can sleep again
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -02005290x500B Tablet pen inserted into its storage bay
Henrique de Moraes Holschuhd1edb2b2008-01-08 13:02:53 -02005300x500C Tablet pen removed from its storage bay
Henrique de Moraes Holschuh3b64b512008-01-08 13:02:51 -02005310x5010 Brightness level changed (newer Lenovo BIOSes)
532
533The above events are propagated by the driver.
534
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300535Compatibility notes:
536
537ibm-acpi and thinkpad-acpi 0.15 (mainline kernels before 2.6.23) never
538supported the input layer, and sent events over the procfs ACPI event
539interface.
540
541To avoid sending duplicate events over the input layer and the ACPI
542event interface, thinkpad-acpi 0.16 implements a module parameter
543(hotkey_report_mode), and also a sysfs device attribute with the same
544name.
545
546Make no mistake here: userspace is expected to switch to using the input
547layer interface of thinkpad-acpi, together with the ACPI netlink event
548interface in kernels 2.6.23 and later, or with the ACPI procfs event
549interface in kernels 2.6.22 and earlier.
550
551If no hotkey_report_mode module parameter is specified (or it is set to
552zero), the driver defaults to mode 1 (see below), and on kernels 2.6.22
553and earlier, also allows one to change the hotkey_report_mode through
554sysfs. In kernels 2.6.23 and later, where the netlink ACPI event
555interface is available, hotkey_report_mode cannot be changed through
556sysfs (it is read-only).
557
558If the hotkey_report_mode module parameter is set to 1 or 2, it cannot
559be changed later through sysfs (any writes will return -EPERM to signal
560that hotkey_report_mode was locked. On 2.6.23 and later, where
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200561hotkey_report_mode cannot be changed at all, writes will return -EACCES).
Henrique de Moraes Holschuhff80f132007-09-04 11:13:15 -0300562
563hotkey_report_mode set to 1 makes the driver export through the procfs
564ACPI event interface all hot key presses (which are *also* sent to the
565input layer). This is a legacy compatibility behaviour, and it is also
566the default mode of operation for the driver.
567
568hotkey_report_mode set to 2 makes the driver filter out the hot key
569presses from the procfs ACPI event interface, so these events will only
570be sent through the input layer. Userspace that has been updated to use
571the thinkpad-acpi input layer interface should set hotkey_report_mode to
5722.
573
574Hot key press events are never sent to the ACPI netlink event interface.
575Really up-to-date userspace under kernel 2.6.23 and later is to use the
576netlink interface and the input layer interface, and don't bother at all
577with hotkey_report_mode.
578
Henrique de Moraes Holschuha0416422007-04-27 22:00:16 -0300579
Henrique de Moraes Holschuhb5972792008-04-26 01:02:17 -0300580Brightness hotkey notes:
581
582These are the current sane choices for brightness key mapping in
583thinkpad-acpi:
584
585For IBM and Lenovo models *without* ACPI backlight control (the ones on
586which thinkpad-acpi will autoload its backlight interface by default,
587and on which ACPI video does not export a backlight interface):
588
5891. Don't enable or map the brightness hotkeys in thinkpad-acpi, as
590 these older firmware versions unfortunately won't respect the hotkey
591 mask for brightness keys anyway, and always reacts to them. This
592 usually work fine, unless X.org drivers are doing something to block
593 the BIOS. In that case, use (3) below. This is the default mode of
594 operation.
595
5962. Enable the hotkeys, but map them to something else that is NOT
597 KEY_BRIGHTNESS_UP/DOWN or any other keycode that would cause
598 userspace to try to change the backlight level, and use that as an
599 on-screen-display hint.
600
6013. IF AND ONLY IF X.org drivers find a way to block the firmware from
602 automatically changing the brightness, enable the hotkeys and map
603 them to KEY_BRIGHTNESS_UP and KEY_BRIGHTNESS_DOWN, and feed that to
604 something that calls xbacklight. thinkpad-acpi will not be able to
605 change brightness in that case either, so you should disable its
606 backlight interface.
607
608For Lenovo models *with* ACPI backlight control:
609
6101. Load up ACPI video and use that. ACPI video will report ACPI
611 events for brightness change keys. Do not mess with thinkpad-acpi
612 defaults in this case. thinkpad-acpi should not have anything to do
613 with backlight events in a scenario where ACPI video is loaded:
614 brightness hotkeys must be disabled, and the backlight interface is
615 to be kept disabled as well. This is the default mode of operation.
616
6172. Do *NOT* load up ACPI video, enable the hotkeys in thinkpad-acpi,
618 and map them to KEY_BRIGHTNESS_UP and KEY_BRIGHTNESS_DOWN. Process
619 these keys on userspace somehow (e.g. by calling xbacklight).
620
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000621
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300622Bluetooth
623---------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700624
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300625procfs: /proc/acpi/ibm/bluetooth
Henrique de Moraes Holschuh0e74dc22008-07-21 09:15:51 -0300626sysfs device attribute: bluetooth_enable (deprecated)
627sysfs rfkill class: switch "tpacpi_bluetooth_sw"
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300628
629This feature shows the presence and current state of a ThinkPad
630Bluetooth device in the internal ThinkPad CDC slot.
631
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000632If the ThinkPad supports it, the Bluetooth state is stored in NVRAM,
633so it is kept across reboots and power-off.
634
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300635Procfs notes:
636
637If Bluetooth is installed, the following commands can be used:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700638
639 echo enable > /proc/acpi/ibm/bluetooth
640 echo disable > /proc/acpi/ibm/bluetooth
641
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300642Sysfs notes:
643
644 If the Bluetooth CDC card is installed, it can be enabled /
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -0300645 disabled through the "bluetooth_enable" thinkpad-acpi device
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300646 attribute, and its current status can also be queried.
647
648 enable:
649 0: disables Bluetooth / Bluetooth is disabled
650 1: enables Bluetooth / Bluetooth is enabled.
651
Henrique de Moraes Holschuh0e74dc22008-07-21 09:15:51 -0300652 Note: this interface has been superseded by the generic rfkill
653 class. It has been deprecated, and it will be removed in year
654 2010.
655
656 rfkill controller switch "tpacpi_bluetooth_sw": refer to
657 Documentation/rfkill.txt for details.
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -0300658
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000659
Linus Torvalds1da177e2005-04-16 15:20:36 -0700660Video output control -- /proc/acpi/ibm/video
661--------------------------------------------
662
663This feature allows control over the devices used for video output -
664LCD, CRT or DVI (if available). The following commands are available:
665
666 echo lcd_enable > /proc/acpi/ibm/video
667 echo lcd_disable > /proc/acpi/ibm/video
668 echo crt_enable > /proc/acpi/ibm/video
669 echo crt_disable > /proc/acpi/ibm/video
670 echo dvi_enable > /proc/acpi/ibm/video
671 echo dvi_disable > /proc/acpi/ibm/video
672 echo auto_enable > /proc/acpi/ibm/video
673 echo auto_disable > /proc/acpi/ibm/video
674 echo expand_toggle > /proc/acpi/ibm/video
675 echo video_switch > /proc/acpi/ibm/video
676
677Each video output device can be enabled or disabled individually.
678Reading /proc/acpi/ibm/video shows the status of each device.
679
680Automatic video switching can be enabled or disabled. When automatic
681video switching is enabled, certain events (e.g. opening the lid,
682docking or undocking) cause the video output device to change
683automatically. While this can be useful, it also causes flickering
684and, on the X40, video corruption. By disabling automatic switching,
685the flickering or video corruption can be avoided.
686
687The video_switch command cycles through the available video outputs
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400688(it simulates the behavior of Fn-F7).
Linus Torvalds1da177e2005-04-16 15:20:36 -0700689
690Video expansion can be toggled through this feature. This controls
691whether the display is expanded to fill the entire LCD screen when a
692mode with less than full resolution is used. Note that the current
693video expansion status cannot be determined through this feature.
694
695Note that on many models (particularly those using Radeon graphics
696chips) the X driver configures the video card in a way which prevents
697Fn-F7 from working. This also disables the video output switching
698features of this driver, as it uses the same ACPI methods as
699Fn-F7. Video switching on the console should still work.
700
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000701UPDATE: refer to https://bugs.freedesktop.org/show_bug.cgi?id=2000
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400702
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400703
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300704ThinkLight control
705------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700706
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300707procfs: /proc/acpi/ibm/light
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300708sysfs attributes: as per LED class, for the "tpacpi::thinklight" LED
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300709
710procfs notes:
711
712The ThinkLight status can be read and set through the procfs interface. A
713few models which do not make the status available will show the ThinkLight
714status as "unknown". The available commands are:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700715
716 echo on > /proc/acpi/ibm/light
717 echo off > /proc/acpi/ibm/light
718
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300719sysfs notes:
720
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300721The ThinkLight sysfs interface is documented by the LED class
722documentation, in Documentation/leds-class.txt. The ThinkLight LED name
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300723is "tpacpi::thinklight".
724
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000725Due to limitations in the sysfs LED class, if the status of the ThinkLight
Henrique de Moraes Holschuhe3065012008-04-26 01:02:24 -0300726cannot be read or if it is unknown, thinkpad-acpi will report it as "off".
727It is impossible to know if the status returned through sysfs is valid.
728
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000729
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000730CMOS/UCMS control
731-----------------
Henrique de Moraes Holschuhb6160042007-04-24 11:48:19 -0300732
733procfs: /proc/acpi/ibm/cmos
734sysfs device attribute: cmos_command
Linus Torvalds1da177e2005-04-16 15:20:36 -0700735
Henrique de Moraes Holschuhd54b7d72007-07-18 23:45:33 -0300736This feature is mostly used internally by the ACPI firmware to keep the legacy
737CMOS NVRAM bits in sync with the current machine state, and to record this
738state so that the ThinkPad will retain such settings across reboots.
739
740Some of these commands actually perform actions in some ThinkPad models, but
741this is expected to disappear more and more in newer models. As an example, in
742a T43 and in a X40, commands 12 and 13 still control the ThinkLight state for
743real, but commands 0 to 2 don't control the mixer anymore (they have been
744phased out) and just update the NVRAM.
Linus Torvalds1da177e2005-04-16 15:20:36 -0700745
Henrique de Moraes Holschuhb6160042007-04-24 11:48:19 -0300746The range of valid cmos command numbers is 0 to 21, but not all have an
747effect and the behavior varies from model to model. Here is the behavior
748on the X40 (tpb is the ThinkPad Buttons utility):
Linus Torvalds1da177e2005-04-16 15:20:36 -0700749
Henrique de Moraes Holschuhd54b7d72007-07-18 23:45:33 -0300750 0 - Related to "Volume down" key press
751 1 - Related to "Volume up" key press
752 2 - Related to "Mute on" key press
753 3 - Related to "Access IBM" key press
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200754 4 - Related to "LCD brightness up" key press
Henrique de Moraes Holschuhd54b7d72007-07-18 23:45:33 -0300755 5 - Related to "LCD brightness down" key press
756 11 - Related to "toggle screen expansion" key press/function
757 12 - Related to "ThinkLight on"
758 13 - Related to "ThinkLight off"
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -0200759 14 - Related to "ThinkLight" key press (toggle ThinkLight)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700760
Henrique de Moraes Holschuhb6160042007-04-24 11:48:19 -0300761The cmos command interface is prone to firmware split-brain problems, as
Henrique de Moraes Holschuhd54b7d72007-07-18 23:45:33 -0300762in newer ThinkPads it is just a compatibility layer. Do not use it, it is
763exported just as a debug tool.
Henrique de Moraes Holschuhb6160042007-04-24 11:48:19 -0300764
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000765
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300766LED control
767-----------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700768
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300769procfs: /proc/acpi/ibm/led
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300770sysfs attributes: as per LED class, see below for names
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300771
772Some of the LED indicators can be controlled through this feature. On
773some older ThinkPad models, it is possible to query the status of the
774LED indicators as well. Newer ThinkPads cannot query the real status
775of the LED indicators.
776
Henrique de Moraes Holschuha4d5eff2009-04-04 04:25:49 +0000777Because misuse of the LEDs could induce an unaware user to perform
778dangerous actions (like undocking or ejecting a bay device while the
779buses are still active), or mask an important alarm (such as a nearly
780empty battery, or a broken battery), access to most LEDs is
781restricted.
782
783Unrestricted access to all LEDs requires that thinkpad-acpi be
784compiled with the CONFIG_THINKPAD_ACPI_UNSAFE_LEDS option enabled.
785Distributions must never enable this option. Individual users that
786are aware of the consequences are welcome to enabling it.
787
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300788procfs notes:
789
790The available commands are:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700791
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300792 echo '<LED number> on' >/proc/acpi/ibm/led
793 echo '<LED number> off' >/proc/acpi/ibm/led
794 echo '<LED number> blink' >/proc/acpi/ibm/led
Linus Torvalds1da177e2005-04-16 15:20:36 -0700795
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -0300796The <LED number> range is 0 to 15. The set of LEDs that can be
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300797controlled varies from model to model. Here is the common ThinkPad
798mapping:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700799
800 0 - power
801 1 - battery (orange)
802 2 - battery (green)
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300803 3 - UltraBase/dock
Linus Torvalds1da177e2005-04-16 15:20:36 -0700804 4 - UltraBay
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300805 5 - UltraBase battery slot
806 6 - (unknown)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700807 7 - standby
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -0300808 8 - dock status 1
809 9 - dock status 2
810 10, 11 - (unknown)
811 12 - thinkvantage
812 13, 14, 15 - (unknown)
Linus Torvalds1da177e2005-04-16 15:20:36 -0700813
814All of the above can be turned on and off and can be made to blink.
815
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300816sysfs notes:
817
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300818The ThinkPad LED sysfs interface is described in detail by the LED class
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300819documentation, in Documentation/leds-class.txt.
820
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -0300821The LEDs are named (in LED ID order, from 0 to 12):
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300822"tpacpi::power", "tpacpi:orange:batt", "tpacpi:green:batt",
823"tpacpi::dock_active", "tpacpi::bay_active", "tpacpi::dock_batt",
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -0300824"tpacpi::unknown_led", "tpacpi::standby", "tpacpi::dock_status1",
825"tpacpi::dock_status2", "tpacpi::unknown_led2", "tpacpi::unknown_led3",
826"tpacpi::thinkvantage".
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300827
Henrique de Moraes Holschuh65807cc2008-04-26 01:02:26 -0300828Due to limitations in the sysfs LED class, if the status of the LED
Henrique de Moraes Holschuhaf116102008-04-26 01:02:25 -0300829indicators cannot be read due to an error, thinkpad-acpi will report it as
830a brightness of zero (same as LED off).
831
832If the thinkpad firmware doesn't support reading the current status,
833trying to read the current LED brightness will just return whatever
834brightness was last written to that attribute.
835
836These LEDs can blink using hardware acceleration. To request that a
837ThinkPad indicator LED should blink in hardware accelerated mode, use the
838"timer" trigger, and leave the delay_on and delay_off parameters set to
839zero (to request hardware acceleration autodetection).
840
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -0300841LEDs that are known not to exist in a given ThinkPad model are not
842made available through the sysfs interface. If you have a dock and you
843notice there are LEDs listed for your ThinkPad that do not exist (and
844are not in the dock), or if you notice that there are missing LEDs,
845a report to ibm-acpi-devel@lists.sourceforge.net is appreciated.
846
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000847
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400848ACPI sounds -- /proc/acpi/ibm/beep
849----------------------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -0700850
851The BEEP method is used internally by the ACPI firmware to provide
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400852audible alerts in various situations. This feature allows the same
Linus Torvalds1da177e2005-04-16 15:20:36 -0700853sounds to be triggered manually.
854
855The commands are non-negative integer numbers:
856
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400857 echo <number> >/proc/acpi/ibm/beep
Linus Torvalds1da177e2005-04-16 15:20:36 -0700858
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400859The valid <number> range is 0 to 17. Not all numbers trigger sounds
860and the sounds vary from model to model. Here is the behavior on the
861X40:
Linus Torvalds1da177e2005-04-16 15:20:36 -0700862
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400863 0 - stop a sound in progress (but use 17 to stop 16)
864 2 - two beeps, pause, third beep ("low battery")
Linus Torvalds1da177e2005-04-16 15:20:36 -0700865 3 - single beep
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400866 4 - high, followed by low-pitched beep ("unable")
Linus Torvalds1da177e2005-04-16 15:20:36 -0700867 5 - single beep
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400868 6 - very high, followed by high-pitched beep ("AC/DC")
Linus Torvalds1da177e2005-04-16 15:20:36 -0700869 7 - high-pitched beep
870 9 - three short beeps
871 10 - very long beep
872 12 - low-pitched beep
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400873 15 - three high-pitched beeps repeating constantly, stop with 0
874 16 - one medium-pitched beep repeating constantly, stop with 17
875 17 - stop 16
Linus Torvalds1da177e2005-04-16 15:20:36 -0700876
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +0000877
Henrique de Moraes Holschuh2c37aa42007-04-24 11:48:16 -0300878Temperature sensors
879-------------------
880
881procfs: /proc/acpi/ibm/thermal
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -0300882sysfs device attributes: (hwmon "thinkpad") temp*_input
Linus Torvalds1da177e2005-04-16 15:20:36 -0700883
Henrique de Moraes Holschuh3d6f99c2007-07-18 23:45:46 -0300884Most ThinkPads include six or more separate temperature sensors but only
885expose the CPU temperature through the standard ACPI methods. This
886feature shows readings from up to eight different sensors on older
887ThinkPads, and up to sixteen different sensors on newer ThinkPads.
Henrique de Moraes Holschuh60eb0b32006-11-24 11:47:08 -0200888
889For example, on the X40, a typical output may be:
890temperatures: 42 42 45 41 36 -128 33 -128
891
Henrique de Moraes Holschuh3d6f99c2007-07-18 23:45:46 -0300892On the T43/p, a typical output may be:
Henrique de Moraes Holschuh60eb0b32006-11-24 11:47:08 -0200893temperatures: 48 48 36 52 38 -128 31 -128 48 52 48 -128 -128 -128 -128 -128
894
895The mapping of thermal sensors to physical locations varies depending on
896system-board model (and thus, on ThinkPad model).
897
898http://thinkwiki.org/wiki/Thermal_Sensors is a public wiki page that
899tries to track down these locations for various models.
900
901Most (newer?) models seem to follow this pattern:
902
9031: CPU
9042: (depends on model)
9053: (depends on model)
9064: GPU
9075: Main battery: main sensor
9086: Bay battery: main sensor
9097: Main battery: secondary sensor
9108: Bay battery: secondary sensor
9119-15: (depends on model)
912
913For the R51 (source: Thomas Gruber):
9142: Mini-PCI
9153: Internal HDD
916
917For the T43, T43/p (source: Shmidoax/Thinkwiki.org)
918http://thinkwiki.org/wiki/Thermal_Sensors#ThinkPad_T43.2C_T43p
9192: System board, left side (near PCMCIA slot), reported as HDAPS temp
9203: PCMCIA slot
9219: MCH (northbridge) to DRAM Bus
Henrique de Moraes Holschuhb8b26402007-07-18 23:45:28 -030092210: Clock-generator, mini-pci card and ICH (southbridge), under Mini-PCI
923 card, under touchpad
Henrique de Moraes Holschuh60eb0b32006-11-24 11:47:08 -020092411: Power regulator, underside of system board, below F2 key
925
Henrique de Moraes Holschuh88679a12006-11-24 11:47:09 -0200926The A31 has a very atypical layout for the thermal sensors
927(source: Milos Popovic, http://thinkwiki.org/wiki/Thermal_Sensors#ThinkPad_A31)
9281: CPU
9292: Main Battery: main sensor
9303: Power Converter
9314: Bay Battery: main sensor
9325: MCH (northbridge)
9336: PCMCIA/ambient
9347: Main Battery: secondary sensor
9358: Bay Battery: secondary sensor
936
Henrique de Moraes Holschuh60eb0b32006-11-24 11:47:08 -0200937
Henrique de Moraes Holschuh2c37aa42007-04-24 11:48:16 -0300938Procfs notes:
939 Readings from sensors that are not available return -128.
940 No commands can be written to this file.
941
942Sysfs notes:
943 Sensors that are not available return the ENXIO error. This
944 status may change at runtime, as there are hotplug thermal
945 sensors, like those inside the batteries and docks.
946
947 thinkpad-acpi thermal sensors are reported through the hwmon
948 subsystem, and follow all of the hwmon guidelines at
949 Documentation/hwmon.
950
951
Matt LaPlanted6bc8ac2006-10-03 22:54:15 +0200952EXPERIMENTAL: Embedded controller register dump -- /proc/acpi/ibm/ecdump
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400953------------------------------------------------------------------------
954
955This feature is marked EXPERIMENTAL because the implementation
956directly accesses hardware registers and may not work as expected. USE
957WITH CAUTION! To use this feature, you need to supply the
958experimental=1 parameter when loading the module.
959
960This feature dumps the values of 256 embedded controller
961registers. Values which have changed since the last time the registers
962were dumped are marked with a star:
963
Henrique de Moraes Holschuh837ca6d2007-03-23 17:33:54 -0300964[root@x40 ibm-acpi]# cat /proc/acpi/ibm/ecdump
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400965EC +00 +01 +02 +03 +04 +05 +06 +07 +08 +09 +0a +0b +0c +0d +0e +0f
966EC 0x00: a7 47 87 01 fe 96 00 08 01 00 cb 00 00 00 40 00
967EC 0x10: 00 00 ff ff f4 3c 87 09 01 ff 42 01 ff ff 0d 00
968EC 0x20: 00 00 00 00 00 00 00 00 00 00 00 03 43 00 00 80
969EC 0x30: 01 07 1a 00 30 04 00 00 *85 00 00 10 00 50 00 00
970EC 0x40: 00 00 00 00 00 00 14 01 00 04 00 00 00 00 00 00
971EC 0x50: 00 c0 02 0d 00 01 01 02 02 03 03 03 03 *bc *02 *bc
972EC 0x60: *02 *bc *02 00 00 00 00 00 00 00 00 00 00 00 00 00
973EC 0x70: 00 00 00 00 00 12 30 40 *24 *26 *2c *27 *20 80 *1f 80
974EC 0x80: 00 00 00 06 *37 *0e 03 00 00 00 0e 07 00 00 00 00
975EC 0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
976EC 0xa0: *ff 09 ff 09 ff ff *64 00 *00 *00 *a2 41 *ff *ff *e0 00
977EC 0xb0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
978EC 0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
979EC 0xd0: 03 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
980EC 0xe0: 00 00 00 00 00 00 00 00 11 20 49 04 24 06 55 03
981EC 0xf0: 31 55 48 54 35 38 57 57 08 2f 45 73 07 65 6c 1a
982
983This feature can be used to determine the register holding the fan
984speed on some models. To do that, do the following:
985
986 - make sure the battery is fully charged
987 - make sure the fan is running
988 - run 'cat /proc/acpi/ibm/ecdump' several times, once per second or so
989
990The first step makes sure various charging-related values don't
991vary. The second ensures that the fan-related values do vary, since
992the fan speed fluctuates a bit. The third will (hopefully) mark the
993fan register with a star:
994
Henrique de Moraes Holschuh837ca6d2007-03-23 17:33:54 -0300995[root@x40 ibm-acpi]# cat /proc/acpi/ibm/ecdump
Borislav Deianov78f81cc2005-08-17 00:00:00 -0400996EC +00 +01 +02 +03 +04 +05 +06 +07 +08 +09 +0a +0b +0c +0d +0e +0f
997EC 0x00: a7 47 87 01 fe 96 00 08 01 00 cb 00 00 00 40 00
998EC 0x10: 00 00 ff ff f4 3c 87 09 01 ff 42 01 ff ff 0d 00
999EC 0x20: 00 00 00 00 00 00 00 00 00 00 00 03 43 00 00 80
1000EC 0x30: 01 07 1a 00 30 04 00 00 85 00 00 10 00 50 00 00
1001EC 0x40: 00 00 00 00 00 00 14 01 00 04 00 00 00 00 00 00
1002EC 0x50: 00 c0 02 0d 00 01 01 02 02 03 03 03 03 bc 02 bc
1003EC 0x60: 02 bc 02 00 00 00 00 00 00 00 00 00 00 00 00 00
1004EC 0x70: 00 00 00 00 00 12 30 40 24 27 2c 27 21 80 1f 80
1005EC 0x80: 00 00 00 06 *be 0d 03 00 00 00 0e 07 00 00 00 00
1006EC 0x90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
1007EC 0xa0: ff 09 ff 09 ff ff 64 00 00 00 a2 41 ff ff e0 00
1008EC 0xb0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
1009EC 0xc0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
1010EC 0xd0: 03 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
1011EC 0xe0: 00 00 00 00 00 00 00 00 11 20 49 04 24 06 55 03
1012EC 0xf0: 31 55 48 54 35 38 57 57 08 2f 45 73 07 65 6c 1a
1013
1014Another set of values that varies often is the temperature
1015readings. Since temperatures don't change vary fast, you can take
1016several quick dumps to eliminate them.
1017
1018You can use a similar method to figure out the meaning of other
1019embedded controller registers - e.g. make sure nothing else changes
1020except the charging or discharging battery to determine which
1021registers contain the current battery capacity, etc. If you experiment
1022with this, do send me your results (including some complete dumps with
1023a description of the conditions when they were taken.)
1024
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001025
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001026LCD brightness control
1027----------------------
1028
1029procfs: /proc/acpi/ibm/brightness
1030sysfs backlight device "thinkpad_screen"
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001031
1032This feature allows software control of the LCD brightness on ThinkPad
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001033models which don't have a hardware brightness slider.
1034
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001035It has some limitations: the LCD backlight cannot be actually turned
1036on or off by this interface, it just controls the backlight brightness
1037level.
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001038
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001039On IBM (and some of the earlier Lenovo) ThinkPads, the backlight control
1040has eight brightness levels, ranging from 0 to 7. Some of the levels
1041may not be distinct. Later Lenovo models that implement the ACPI
1042display backlight brightness control methods have 16 levels, ranging
1043from 0 to 15.
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001044
Henrique de Moraes Holschuhd7880f12009-06-18 00:40:16 -03001045For IBM ThinkPads, there are two interfaces to the firmware for direct
1046brightness control, EC and UCMS (or CMOS). To select which one should be
1047used, use the brightness_mode module parameter: brightness_mode=1 selects
1048EC mode, brightness_mode=2 selects UCMS mode, brightness_mode=3 selects EC
1049mode with NVRAM backing (so that brightness changes are remembered across
1050shutdown/reboot).
Henrique de Moraes Holschuh0e501832009-04-04 04:25:53 +00001051
1052The driver tries to select which interface to use from a table of
1053defaults for each ThinkPad model. If it makes a wrong choice, please
1054report this as a bug, so that we can fix it.
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001055
Henrique de Moraes Holschuhd7880f12009-06-18 00:40:16 -03001056Lenovo ThinkPads only support brightness_mode=2 (UCMS).
1057
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001058When display backlight brightness controls are available through the
1059standard ACPI interface, it is best to use it instead of this direct
Henrique de Moraes Holschuhe11e2112007-10-30 17:46:22 -02001060ThinkPad-specific interface. The driver will disable its native
1061backlight brightness control interface if it detects that the standard
1062ACPI interface is available in the ThinkPad.
Henrique de Moraes Holschuh24d3b772007-07-18 23:45:43 -03001063
Henrique de Moraes Holschuh87cc5372007-10-30 18:02:07 -02001064The brightness_enable module parameter can be used to control whether
1065the LCD brightness control feature will be enabled when available.
Henrique de Moraes Holschuhe11e2112007-10-30 17:46:22 -02001066brightness_enable=0 forces it to be disabled. brightness_enable=1
1067forces it to be enabled when available, even if the standard ACPI
1068interface is also available.
Henrique de Moraes Holschuh87cc5372007-10-30 18:02:07 -02001069
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001070Procfs notes:
1071
1072 The available commands are:
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001073
1074 echo up >/proc/acpi/ibm/brightness
1075 echo down >/proc/acpi/ibm/brightness
1076 echo 'level <level>' >/proc/acpi/ibm/brightness
1077
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001078Sysfs notes:
1079
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001080The interface is implemented through the backlight sysfs class, which is
1081poorly documented at this time.
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001082
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001083Locate the thinkpad_screen device under /sys/class/backlight, and inside
1084it there will be the following attributes:
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001085
1086 max_brightness:
1087 Reads the maximum brightness the hardware can be set to.
1088 The minimum is always zero.
1089
1090 actual_brightness:
1091 Reads what brightness the screen is set to at this instant.
1092
1093 brightness:
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001094 Writes request the driver to change brightness to the
1095 given value. Reads will tell you what brightness the
1096 driver is trying to set the display to when "power" is set
1097 to zero and the display has not been dimmed by a kernel
1098 power management event.
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001099
1100 power:
Henrique de Moraes Holschuha3f104c2007-10-30 17:46:20 -02001101 power management mode, where 0 is "display on", and 1 to 3
1102 will dim the display backlight to brightness level 0
1103 because thinkpad-acpi cannot really turn the backlight
1104 off. Kernel power management events can temporarily
1105 increase the current power management level, i.e. they can
1106 dim the display.
Henrique de Moraes Holschuh7d5a0152007-04-24 11:48:20 -03001107
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001108
Henrique de Moraes Holschuhb5972792008-04-26 01:02:17 -03001109WARNING:
1110
1111 Whatever you do, do NOT ever call thinkpad-acpi backlight-level change
1112 interface and the ACPI-based backlight level change interface
1113 (available on newer BIOSes, and driven by the Linux ACPI video driver)
1114 at the same time. The two will interact in bad ways, do funny things,
1115 and maybe reduce the life of the backlight lamps by needlessly kicking
1116 its level up and down at every change.
1117
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001118
Stefan Schmidt24f7ff02006-09-22 12:19:15 +02001119Volume control -- /proc/acpi/ibm/volume
1120---------------------------------------
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001121
1122This feature allows volume control on ThinkPad models which don't have
1123a hardware volume knob. The available commands are:
1124
1125 echo up >/proc/acpi/ibm/volume
1126 echo down >/proc/acpi/ibm/volume
1127 echo mute >/proc/acpi/ibm/volume
1128 echo 'level <level>' >/proc/acpi/ibm/volume
1129
1130The <level> number range is 0 to 15 although not all of them may be
1131distinct. The unmute the volume after the mute command, use either the
1132up or down command (the level command will not unmute the volume).
1133The current volume level and mute state is shown in the file.
1134
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001135The ALSA mixer interface to this feature is still missing, but patches
1136to add it exist. That problem should be addressed in the not so
1137distant future.
1138
1139
Henrique de Moraes Holschuhecf2a802007-04-27 22:00:09 -03001140Fan control and monitoring: fan speed, fan enable/disable
1141---------------------------------------------------------
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001142
1143procfs: /proc/acpi/ibm/fan
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -03001144sysfs device attributes: (hwmon "thinkpad") fan1_input, pwm1,
Henrique de Moraes Holschuhd7377242009-06-18 00:40:17 -03001145 pwm1_enable, fan2_input
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -03001146sysfs hwmon driver attributes: fan_watchdog
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001147
Henrique de Moraes Holschuhecf2a802007-04-27 22:00:09 -03001148NOTE NOTE NOTE: fan control operations are disabled by default for
1149safety reasons. To enable them, the module parameter "fan_control=1"
1150must be given to thinkpad-acpi.
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001151
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001152This feature attempts to show the current fan speed, control mode and
1153other fan data that might be available. The speed is read directly
1154from the hardware registers of the embedded controller. This is known
Henrique de Moraes Holschuhecf2a802007-04-27 22:00:09 -03001155to work on later R, T, X and Z series ThinkPads but may show a bogus
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001156value on other models.
1157
Henrique de Moraes Holschuhd7377242009-06-18 00:40:17 -03001158Some Lenovo ThinkPads support a secondary fan. This fan cannot be
1159controlled separately, it shares the main fan control.
1160
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001161Fan levels:
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001162
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001163Most ThinkPad fans work in "levels" at the firmware interface. Level 0
1164stops the fan. The higher the level, the higher the fan speed, although
1165adjacent levels often map to the same fan speed. 7 is the highest
1166level, where the fan reaches the maximum recommended speed.
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001167
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001168Level "auto" means the EC changes the fan level according to some
1169internal algorithm, usually based on readings from the thermal sensors.
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001170
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001171There is also a "full-speed" level, also known as "disengaged" level.
1172In this level, the EC disables the speed-locked closed-loop fan control,
1173and drives the fan as fast as it can go, which might exceed hardware
1174limits, so use this level with caution.
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001175
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001176The fan usually ramps up or down slowly from one speed to another, and
1177it is normal for the EC to take several seconds to react to fan
1178commands. The full-speed level may take up to two minutes to ramp up to
1179maximum speed, and in some ThinkPads, the tachometer readings go stale
1180while the EC is transitioning to the full-speed level.
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001181
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001182WARNING WARNING WARNING: do not leave the fan disabled unless you are
1183monitoring all of the temperature sensor readings and you are ready to
1184enable it if necessary to avoid overheating.
1185
1186An enabled fan in level "auto" may stop spinning if the EC decides the
1187ThinkPad is cool enough and doesn't need the extra airflow. This is
Matt LaPlante01dd2fb2007-10-20 01:34:40 +02001188normal, and the EC will spin the fan up if the various thermal readings
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001189rise too much.
1190
1191On the X40, this seems to depend on the CPU and HDD temperatures.
1192Specifically, the fan is turned on when either the CPU temperature
1193climbs to 56 degrees or the HDD temperature climbs to 46 degrees. The
1194fan is turned off when the CPU temperature drops to 49 degrees and the
1195HDD temperature drops to 41 degrees. These thresholds cannot
1196currently be controlled.
1197
Henrique de Moraes Holschuha12095c2006-11-24 11:47:13 -02001198The ThinkPad's ACPI DSDT code will reprogram the fan on its own when
1199certain conditions are met. It will override any fan programming done
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -03001200through thinkpad-acpi.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001201
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -03001202The thinkpad-acpi kernel driver can be programmed to revert the fan
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001203level to a safe setting if userspace does not issue one of the procfs
1204fan commands: "enable", "disable", "level" or "watchdog", or if there
1205are no writes to pwm1_enable (or to pwm1 *if and only if* pwm1_enable is
1206set to 1, manual mode) within a configurable amount of time of up to
1207120 seconds. This functionality is called fan safety watchdog.
Henrique de Moraes Holschuh38f996e2007-03-23 17:33:59 -03001208
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001209Note that the watchdog timer stops after it enables the fan. It will be
1210rearmed again automatically (using the same interval) when one of the
1211above mentioned fan commands is received. The fan watchdog is,
1212therefore, not suitable to protect against fan mode changes made through
1213means other than the "enable", "disable", and "level" procfs fan
1214commands, or the hwmon fan control sysfs interface.
Henrique de Moraes Holschuh38f996e2007-03-23 17:33:59 -03001215
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001216Procfs notes:
Henrique de Moraes Holschuh38f996e2007-03-23 17:33:59 -03001217
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001218The fan may be enabled or disabled with the following commands:
1219
1220 echo enable >/proc/acpi/ibm/fan
1221 echo disable >/proc/acpi/ibm/fan
1222
1223Placing a fan on level 0 is the same as disabling it. Enabling a fan
1224will try to place it in a safe level if it is too slow or disabled.
1225
1226The fan level can be controlled with the command:
1227
1228 echo 'level <level>' > /proc/acpi/ibm/fan
1229
1230Where <level> is an integer from 0 to 7, or one of the words "auto" or
1231"full-speed" (without the quotes). Not all ThinkPads support the "auto"
1232and "full-speed" levels. The driver accepts "disengaged" as an alias for
1233"full-speed", and reports it as "disengaged" for backwards
1234compatibility.
1235
1236On the X31 and X40 (and ONLY on those models), the fan speed can be
1237controlled to a certain degree. Once the fan is running, it can be
1238forced to run faster or slower with the following command:
1239
1240 echo 'speed <speed>' > /proc/acpi/ibm/fan
1241
1242The sustainable range of fan speeds on the X40 appears to be from about
12433700 to about 7350. Values outside this range either do not have any
1244effect or the fan speed eventually settles somewhere in that range. The
1245fan cannot be stopped or started with this command. This functionality
1246is incomplete, and not available through the sysfs interface.
1247
1248To program the safety watchdog, use the "watchdog" command.
1249
1250 echo 'watchdog <interval in seconds>' > /proc/acpi/ibm/fan
1251
1252If you want to disable the watchdog, use 0 as the interval.
1253
1254Sysfs notes:
1255
1256The sysfs interface follows the hwmon subsystem guidelines for the most
1257part, and the exception is the fan safety watchdog.
1258
Henrique de Moraes Holschuhb39fe582007-04-27 22:00:13 -03001259Writes to any of the sysfs attributes may return the EINVAL error if
1260that operation is not supported in a given ThinkPad or if the parameter
1261is out-of-bounds, and EPERM if it is forbidden. They may also return
1262EINTR (interrupted system call), and EIO (I/O error while trying to talk
1263to the firmware).
1264
1265Features not yet implemented by the driver return ENOSYS.
1266
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001267hwmon device attribute pwm1_enable:
1268 0: PWM offline (fan is set to full-speed mode)
1269 1: Manual PWM control (use pwm1 to set fan level)
1270 2: Hardware PWM control (EC "auto" mode)
1271 3: reserved (Software PWM control, not implemented yet)
1272
Henrique de Moraes Holschuhb39fe582007-04-27 22:00:13 -03001273 Modes 0 and 2 are not supported by all ThinkPads, and the
1274 driver is not always able to detect this. If it does know a
1275 mode is unsupported, it will return -EINVAL.
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001276
1277hwmon device attribute pwm1:
1278 Fan level, scaled from the firmware values of 0-7 to the hwmon
1279 scale of 0-255. 0 means fan stopped, 255 means highest normal
1280 speed (level 7).
1281
1282 This attribute only commands the fan if pmw1_enable is set to 1
1283 (manual PWM control).
1284
1285hwmon device attribute fan1_input:
1286 Fan tachometer reading, in RPM. May go stale on certain
1287 ThinkPads while the EC transitions the PWM to offline mode,
1288 which can take up to two minutes. May return rubbish on older
1289 ThinkPads.
1290
Henrique de Moraes Holschuhd7377242009-06-18 00:40:17 -03001291hwmon device attribute fan2_input:
1292 Fan tachometer reading, in RPM, for the secondary fan.
1293 Available only on some ThinkPads. If the secondary fan is
1294 not installed, will always read 0.
1295
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -03001296hwmon driver attribute fan_watchdog:
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001297 Fan safety watchdog timer interval, in seconds. Minimum is
1298 1 second, maximum is 120 seconds. 0 disables the watchdog.
1299
1300To stop the fan: set pwm1 to zero, and pwm1_enable to 1.
1301
1302To start the fan in a safe mode: set pwm1_enable to 2. If that fails
Henrique de Moraes Holschuhb39fe582007-04-27 22:00:13 -03001303with EINVAL, try to set pwm1_enable to 1 and pwm1 to at least 128 (255
1304would be the safest choice, though).
Henrique de Moraes Holschuhfe98a522007-04-24 11:48:17 -03001305
Henrique de Moraes Holschuh38f996e2007-03-23 17:33:59 -03001306
Jeremy Fitzhardinge9662e082008-08-27 21:04:49 -07001307WAN
1308---
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -03001309
1310procfs: /proc/acpi/ibm/wan
Henrique de Moraes Holschuh0e74dc22008-07-21 09:15:51 -03001311sysfs device attribute: wwan_enable (deprecated)
1312sysfs rfkill class: switch "tpacpi_wwan_sw"
Stefan Schmidt28b779d2006-09-22 12:19:16 +02001313
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001314This feature shows the presence and current state of the built-in
1315Wireless WAN device.
1316
1317If the ThinkPad supports it, the WWAN state is stored in NVRAM,
1318so it is kept across reboots and power-off.
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -03001319
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -02001320It was tested on a Lenovo ThinkPad X60. It should probably work on other
1321ThinkPad models which come with this module installed.
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -03001322
1323Procfs notes:
1324
1325If the W-WAN card is installed, the following commands can be used:
Stefan Schmidt28b779d2006-09-22 12:19:16 +02001326
1327 echo enable > /proc/acpi/ibm/wan
1328 echo disable > /proc/acpi/ibm/wan
1329
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -03001330Sysfs notes:
1331
1332 If the W-WAN card is installed, it can be enabled /
Henrique de Moraes Holschuhcc4c24e2007-05-30 20:50:14 -03001333 disabled through the "wwan_enable" thinkpad-acpi device
Henrique de Moraes Holschuhd3a6ade2007-04-27 22:00:17 -03001334 attribute, and its current status can also be queried.
1335
1336 enable:
1337 0: disables WWAN card / WWAN card is disabled
1338 1: enables WWAN card / WWAN card is enabled.
1339
Henrique de Moraes Holschuh0e74dc22008-07-21 09:15:51 -03001340 Note: this interface has been superseded by the generic rfkill
1341 class. It has been deprecated, and it will be removed in year
1342 2010.
1343
1344 rfkill controller switch "tpacpi_wwan_sw": refer to
1345 Documentation/rfkill.txt for details.
Linus Torvalds1da177e2005-04-16 15:20:36 -07001346
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001347
Henrique de Moraes Holschuh0045c0a2009-01-11 03:01:03 -02001348EXPERIMENTAL: UWB
1349-----------------
1350
1351This feature is marked EXPERIMENTAL because it has not been extensively
1352tested and validated in various ThinkPad models yet. The feature may not
1353work as expected. USE WITH CAUTION! To use this feature, you need to supply
1354the experimental=1 parameter when loading the module.
1355
1356sysfs rfkill class: switch "tpacpi_uwb_sw"
1357
1358This feature exports an rfkill controller for the UWB device, if one is
1359present and enabled in the BIOS.
1360
1361Sysfs notes:
1362
1363 rfkill controller switch "tpacpi_uwb_sw": refer to
1364 Documentation/rfkill.txt for details.
1365
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001366
Borislav Deianov78f81cc2005-08-17 00:00:00 -04001367Multiple Commands, Module Parameters
1368------------------------------------
Linus Torvalds1da177e2005-04-16 15:20:36 -07001369
1370Multiple commands can be written to the proc files in one shot by
1371separating them with commas, for example:
1372
1373 echo enable,0xffff > /proc/acpi/ibm/hotkey
1374 echo lcd_disable,crt_enable > /proc/acpi/ibm/video
1375
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -03001376Commands can also be specified when loading the thinkpad-acpi module,
1377for example:
Linus Torvalds1da177e2005-04-16 15:20:36 -07001378
Henrique de Moraes Holschuh643f12d2007-03-29 01:58:43 -03001379 modprobe thinkpad_acpi hotkey=enable,0xffff video=auto_disable
Linus Torvalds1da177e2005-04-16 15:20:36 -07001380
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001381
Henrique de Moraes Holschuh132ce092007-04-21 11:08:30 -03001382Enabling debugging output
1383-------------------------
1384
Simon Arlott0f035b82007-10-20 01:30:25 +02001385The module takes a debug parameter which can be used to selectively
Henrique de Moraes Holschuh132ce092007-04-21 11:08:30 -03001386enable various classes of debugging output, for example:
1387
Henrique de Moraes Holschuh5f249272008-02-16 02:17:51 -02001388 modprobe thinkpad_acpi debug=0xffff
Henrique de Moraes Holschuh132ce092007-04-21 11:08:30 -03001389
1390will enable all debugging output classes. It takes a bitmask, so
1391to enable more than one output class, just add their values.
1392
Henrique de Moraes Holschuhfe08bc42007-04-21 11:08:32 -03001393 Debug bitmask Description
Henrique de Moraes Holschuh73a94d82009-04-04 04:25:47 +00001394 0x8000 Disclose PID of userspace programs
1395 accessing some functions of the driver
Henrique de Moraes Holschuhfe08bc42007-04-21 11:08:32 -03001396 0x0001 Initialization and probing
1397 0x0002 Removal
Henrique de Moraes Holschuhbee4cd92009-04-04 04:25:50 +00001398 0x0004 RF Transmitter control (RFKILL)
1399 (bluetooth, WWAN, UWB...)
Henrique de Moraes Holschuh56e2c202009-04-04 04:25:51 +00001400 0x0008 HKEY event interface, hotkeys
Henrique de Moraes Holschuh74a60c02009-04-04 04:25:52 +00001401 0x0010 Fan control
Henrique de Moraes Holschuh0e501832009-04-04 04:25:53 +00001402 0x0020 Backlight brightness
Henrique de Moraes Holschuhfe08bc42007-04-21 11:08:32 -03001403
Henrique de Moraes Holschuh132ce092007-04-21 11:08:30 -03001404There is also a kernel build option to enable more debugging
1405information, which may be necessary to debug driver problems.
Henrique de Moraes Holschuh0dcef772007-04-21 11:08:34 -03001406
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -03001407The level of debugging information output by the driver can be changed
1408at runtime through sysfs, using the driver attribute debug_level. The
1409attribute takes the same bitmask as the debug module parameter above.
1410
Henrique de Moraes Holschuh078ac192009-04-04 04:25:44 +00001411
Henrique de Moraes Holschuh0dcef772007-04-21 11:08:34 -03001412Force loading of module
1413-----------------------
1414
1415If thinkpad-acpi refuses to detect your ThinkPad, you can try to specify
1416the module parameter force_load=1. Regardless of whether this works or
1417not, please contact ibm-acpi-devel@lists.sourceforge.net with a report.
Henrique de Moraes Holschuh176750d2007-04-24 11:48:13 -03001418
1419
1420Sysfs interface changelog:
1421
14220x000100: Initial sysfs support, as a single platform driver and
1423 device.
Henrique de Moraes Holschuh94b08712007-07-18 23:45:32 -030014240x000200: Hot key support for 32 hot keys, and radio slider switch
1425 support.
Henrique de Moraes Holschuh741553c2007-07-18 23:45:39 -030014260x010000: Hot keys are now handled by default over the input
1427 layer, the radio switch generates input event EV_RADIO,
1428 and the driver enables hot key handling by default in
1429 the firmware.
Henrique de Moraes Holschuh7fd40022007-09-25 06:38:03 -03001430
14310x020000: ABI fix: added a separate hwmon platform device and
1432 driver, which must be located by name (thinkpad)
1433 and the hwmon class for libsensors4 (lm-sensors 3)
1434 compatibility. Moved all hwmon attributes to this
1435 new platform device.
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -02001436
14370x020100: Marker for thinkpad-acpi with hot key NVRAM polling
1438 support. If you must, use it to know you should not
Frederik Schwarzer0211a9c2008-12-29 22:14:56 +01001439 start a userspace NVRAM poller (allows to detect when
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -02001440 NVRAM is compiled out by the user because it is
1441 unneeded/undesired in the first place).
14420x020101: Marker for thinkpad-acpi with hot key NVRAM polling
Henrique de Moraes Holschuhd0788cf2008-02-16 02:17:56 -02001443 and proper hotkey_mask semantics (version 8 of the
Henrique de Moraes Holschuh01e88f22008-01-08 13:02:41 -02001444 NVRAM polling patch). Some development snapshots of
1445 0.18 had an earlier version that did strange things
1446 to hotkey_mask.
Henrique de Moraes Holschuh50ebec02008-01-08 13:02:55 -02001447
14480x020200: Add poll()/select() support to the following attributes:
1449 hotkey_radio_sw, wakeup_hotunplug_complete, wakeup_reason
Henrique de Moraes Holschuh2586d562009-04-04 04:25:48 +00001450
14510x020300: hotkey enable/disable support removed, attributes
1452 hotkey_bios_enabled and hotkey_enable deprecated and
1453 marked for removal.
Henrique de Moraes Holschuhf21179a2009-05-30 13:25:08 -03001454
14550x020400: Marker for 16 LEDs support. Also, LEDs that are known
1456 to not exist in a given model are not registered with
1457 the LED sysfs class anymore.