Documentation/thermal/sysfs-api.txt

   1 Generic Thermal Sysfs driver How To
   2 ===================================
   3
   4 Written by Sujith Thomas <sujith.thomas@intel.com>, Zhang Rui <rui.zhang@intel.com>
   5
   6 Updated: 2 January 2008
   7
   8 Copyright (c)  2008 Intel Corporation
   9
  10
  11 0. Introduction
  12
  13 The generic thermal sysfs provides a set of interfaces for thermal zone
  14 devices (sensors) and thermal cooling devices (fan, processor...) to register
  15 with the thermal management solution and to be a part of it.
  16
  17 This how-to focuses on enabling new thermal zone and cooling devices to
  18 participate in thermal management.
  19 This solution is platform independent and any type of thermal zone devices
  20 and cooling devices should be able to make use of the infrastructure.
  21
  22 The main task of the thermal sysfs driver is to expose thermal zone attributes
  23 as well as cooling device attributes to the user space.
  24 An intelligent thermal management application can make decisions based on
  25 inputs from thermal zone attributes (the current temperature and trip point
  26 temperature) and throttle appropriate devices.
  27
  28 [0-*]   denotes any positive number starting from 0
  29 [1-*]   denotes any positive number starting from 1
  30
  31 1. thermal sysfs driver interface functions
  32
  33 1.1 thermal zone device interface
  34 1.1.1 struct thermal_zone_device *thermal_zone_device_register(char *type,
  35                 int trips, int mask, void *devdata,
  36                 struct thermal_zone_device_ops *ops,
  37                 const struct thermal_zone_params *tzp,
  38                 int passive_delay, int polling_delay))
  39
  40     This interface function adds a new thermal zone device (sensor) to
  41     /sys/class/thermal folder as thermal_zone[0-*]. It tries to bind all the
  42     thermal cooling devices registered at the same time.
  43
  44     type: the thermal zone type.
  45     trips: the total number of trip points this thermal zone supports.
  46     mask: Bit string: If 'n'th bit is set, then trip point 'n' is writeable.
  47     devdata: device private data
  48     ops: thermal zone device call-backs.
  49         .bind: bind the thermal zone device with a thermal cooling device.
  50         .unbind: unbind the thermal zone device with a thermal cooling device.
  51         .get_temp: get the current temperature of the thermal zone.
  52         .get_mode: get the current mode (enabled/disabled) of the thermal zone.
  53             - "enabled" means the kernel thermal management is enabled.
  54             - "disabled" will prevent kernel thermal driver action upon trip points
  55               so that user applications can take charge of thermal management.
  56         .set_mode: set the mode (enabled/disabled) of the thermal zone.
  57         .get_trip_type: get the type of certain trip point.
  58         .get_trip_temp: get the temperature above which the certain trip point
  59                         will be fired.
  60         .set_emul_temp: set the emulation temperature which helps in debugging
  61                         different threshold temperature points.
  62     tzp: thermal zone platform parameters.
  63     passive_delay: number of milliseconds to wait between polls when
  64         performing passive cooling.
  65     polling_delay: number of milliseconds to wait between polls when checking
  66         whether trip points have been crossed (0 for interrupt driven systems).
  67
  68
  69 1.1.2 void thermal_zone_device_unregister(struct thermal_zone_device *tz)
  70
  71     This interface function removes the thermal zone device.
  72     It deletes the corresponding entry from /sys/class/thermal folder and
  73     unbinds all the thermal cooling devices it uses.
  74
  75 1.1.3 struct thermal_zone_device *thermal_zone_of_sensor_register(
  76                 struct device *dev, int sensor_id, void *data,
  77                 const struct thermal_zone_of_device_ops *ops)
  78
  79         This interface adds a new sensor to a DT thermal zone.
  80         This function will search the list of thermal zones described in
  81         device tree and look for the zone that refer to the sensor device
  82         pointed by dev->of_node as temperature providers. For the zone
  83         pointing to the sensor node, the sensor will be added to the DT
  84         thermal zone device.
  85
  86         The parameters for this interface are:
  87         dev:            Device node of sensor containing valid node pointer in
  88                         dev->of_node.
  89         sensor_id:      a sensor identifier, in case the sensor IP has more
  90                         than one sensors
  91         data:           a private pointer (owned by the caller) that will be
  92                         passed back, when a temperature reading is needed.
  93         ops:            struct thermal_zone_of_device_ops *.
  94
  95                         get_temp:       a pointer to a function that reads the
  96                                         sensor temperature. This is mandatory
  97                                         callback provided by sensor driver.
  98                         get_trend:      a pointer to a function that reads the
  99                                         sensor temperature trend.
 100                         set_emul_temp:  a pointer to a function that sets
 101                                         sensor emulated temperature.
 102         The thermal zone temperature is provided by the get_temp() function
 103         pointer of thermal_zone_of_device_ops. When called, it will
 104         have the private pointer @data back.
 105
 106         It returns error pointer if fails otherwise valid thermal zone device
 107         handle. Caller should check the return handle with IS_ERR() for finding
 108         whether success or not.
 109
 110 1.1.4 void thermal_zone_of_sensor_unregister(struct device *dev,
 111                 struct thermal_zone_device *tzd)
 112
 113         This interface unregisters a sensor from a DT thermal zone which was
 114         successfully added by interface thermal_zone_of_sensor_register().
 115         This function removes the sensor callbacks and private data from the
 116         thermal zone device registered with thermal_zone_of_sensor_register()
 117         interface. It will also silent the zone by remove the .get_temp() and
 118         get_trend() thermal zone device callbacks.
 119
 120 1.1.5 struct thermal_zone_device *devm_thermal_zone_of_sensor_register(
 121                 struct device *dev, int sensor_id,
 122                 void *data, const struct thermal_zone_of_device_ops *ops)
 123
 124         This interface is resource managed version of
 125         thermal_zone_of_sensor_register().
 126         All details of thermal_zone_of_sensor_register() described in
 127         section 1.1.3 is applicable here.
 128         The benefit of using this interface to register sensor is that it
 129         is not require to explicitly call thermal_zone_of_sensor_unregister()
 130         in error path or during driver unbinding as this is done by driver
 131         resource manager.
 132
 133 1.1.6 void devm_thermal_zone_of_sensor_unregister(struct device *dev,
 134                 struct thermal_zone_device *tzd)
 135
 136         This interface is resource managed version of
 137         thermal_zone_of_sensor_unregister().
 138         All details of thermal_zone_of_sensor_unregister() described in
 139         section 1.1.4 is applicable here.
 140         Normally this function will not need to be called and the resource
 141         management code will ensure that the resource is freed.
 142
 143 1.2 thermal cooling device interface
 144 1.2.1 struct thermal_cooling_device *thermal_cooling_device_register(char *name,
 145                 void *devdata, struct thermal_cooling_device_ops *)
 146
 147     This interface function adds a new thermal cooling device (fan/processor/...)
 148     to /sys/class/thermal/ folder as cooling_device[0-*]. It tries to bind itself
 149     to all the thermal zone devices registered at the same time.
 150     name: the cooling device name.
 151     devdata: device private data.
 152     ops: thermal cooling devices call-backs.
 153         .get_max_state: get the Maximum throttle state of the cooling device.
 154         .get_cur_state: get the Currently requested throttle state of the cooling device.
 155         .set_cur_state: set the Current throttle state of the cooling device.
 156
 157 1.2.2 void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
 158
 159     This interface function removes the thermal cooling device.
 160     It deletes the corresponding entry from /sys/class/thermal folder and
 161     unbinds itself from all the thermal zone devices using it.
 162
 163 1.3 interface for binding a thermal zone device with a thermal cooling device
 164 1.3.1 int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
 165         int trip, struct thermal_cooling_device *cdev,
 166         unsigned long upper, unsigned long lower, unsigned int weight);
 167
 168     This interface function binds a thermal cooling device to a particular trip
 169     point of a thermal zone device.
 170     This function is usually called in the thermal zone device .bind callback.
 171     tz: the thermal zone device
 172     cdev: thermal cooling device
 173     trip: indicates which trip point in this thermal zone the cooling device
 174           is associated with.
 175     upper:the Maximum cooling state for this trip point.
 176           THERMAL_NO_LIMIT means no upper limit,
 177           and the cooling device can be in max_state.
 178     lower:the Minimum cooling state can be used for this trip point.
 179           THERMAL_NO_LIMIT means no lower limit,
 180           and the cooling device can be in cooling state 0.
 181     weight: the influence of this cooling device in this thermal
 182             zone.  See 1.4.1 below for more information.
 183
 184 1.3.2 int thermal_zone_unbind_cooling_device(struct thermal_zone_device *tz,
 185                 int trip, struct thermal_cooling_device *cdev);
 186
 187     This interface function unbinds a thermal cooling device from a particular
 188     trip point of a thermal zone device. This function is usually called in
 189     the thermal zone device .unbind callback.
 190     tz: the thermal zone device
 191     cdev: thermal cooling device
 192     trip: indicates which trip point in this thermal zone the cooling device
 193           is associated with.
 194
 195 1.4 Thermal Zone Parameters
 196 1.4.1 struct thermal_bind_params
 197     This structure defines the following parameters that are used to bind
 198     a zone with a cooling device for a particular trip point.
 199     .cdev: The cooling device pointer
 200     .weight: The 'influence' of a particular cooling device on this
 201              zone. This is relative to the rest of the cooling
 202              devices. For example, if all cooling devices have a
 203              weight of 1, then they all contribute the same. You can
 204              use percentages if you want, but it's not mandatory. A
 205              weight of 0 means that this cooling device doesn't
 206              contribute to the cooling of this zone unless all cooling
 207              devices have a weight of 0. If all weights are 0, then
 208              they all contribute the same.
 209     .trip_mask:This is a bit mask that gives the binding relation between
 210                this thermal zone and cdev, for a particular trip point.
 211                If nth bit is set, then the cdev and thermal zone are bound
 212                for trip point n.
 213     .binding_limits: This is an array of cooling state limits. Must have
 214                      exactly 2 * thermal_zone.number_of_trip_points. It is an
 215                      array consisting of tuples <lower-state upper-state> of
 216                      state limits. Each trip will be associated with one state
 217                      limit tuple when binding. A NULL pointer means
 218                      <THERMAL_NO_LIMITS THERMAL_NO_LIMITS> on all trips.
 219                      These limits are used when binding a cdev to a trip point.
 220     .match: This call back returns success(0) if the 'tz and cdev' need to
 221             be bound, as per platform data.
 222 1.4.2 struct thermal_zone_params
 223     This structure defines the platform level parameters for a thermal zone.
 224     This data, for each thermal zone should come from the platform layer.
 225     This is an optional feature where some platforms can choose not to
 226     provide this data.
 227     .governor_name: Name of the thermal governor used for this zone
 228     .no_hwmon: a boolean to indicate if the thermal to hwmon sysfs interface
 229                is required. when no_hwmon == false, a hwmon sysfs interface
 230                will be created. when no_hwmon == true, nothing will be done.
 231                In case the thermal_zone_params is NULL, the hwmon interface
 232                will be created (for backward compatibility).
 233     .num_tbps: Number of thermal_bind_params entries for this zone
 234     .tbp: thermal_bind_params entries
 235
 236 2. sysfs attributes structure
 237
 238 RO      read only value
 239 RW      read/write value
 240
 241 Thermal sysfs attributes will be represented under /sys/class/thermal.
 242 Hwmon sysfs I/F extension is also available under /sys/class/hwmon
 243 if hwmon is compiled in or built as a module.
 244
 245 Thermal zone device sys I/F, created once it's registered:
 246 /sys/class/thermal/thermal_zone[0-*]:
 247     |---type:                   Type of the thermal zone
 248     |---temp:                   Current temperature
 249     |---mode:                   Working mode of the thermal zone
 250     |---policy:                 Thermal governor used for this zone
 251     |---available_policies:     Available thermal governors for this zone
 252     |---trip_point_[0-*]_temp:  Trip point temperature
 253     |---trip_point_[0-*]_type:  Trip point type
 254     |---trip_point_[0-*]_hyst:  Hysteresis value for this trip point
 255     |---emul_temp:              Emulated temperature set node
 256     |---sustainable_power:      Sustainable dissipatable power
 257     |---k_po:                   Proportional term during temperature overshoot
 258     |---k_pu:                   Proportional term during temperature undershoot
 259     |---k_i:                    PID's integral term in the power allocator gov
 260     |---k_d:                    PID's derivative term in the power allocator
 261     |---integral_cutoff:        Offset above which errors are accumulated
 262     |---slope:                  Slope constant applied as linear extrapolation
 263     |---offset:                 Offset constant applied as linear extrapolation
 264
 265 Thermal cooling device sys I/F, created once it's registered:
 266 /sys/class/thermal/cooling_device[0-*]:
 267     |---type:                   Type of the cooling device(processor/fan/...)
 268     |---max_state:              Maximum cooling state of the cooling device
 269     |---cur_state:              Current cooling state of the cooling device
 270
 271
 272 Then next two dynamic attributes are created/removed in pairs. They represent
 273 the relationship between a thermal zone and its associated cooling device.
 274 They are created/removed for each successful execution of
 275 thermal_zone_bind_cooling_device/thermal_zone_unbind_cooling_device.
 276
 277 /sys/class/thermal/thermal_zone[0-*]:
 278     |---cdev[0-*]:              [0-*]th cooling device in current thermal zone
 279     |---cdev[0-*]_trip_point:   Trip point that cdev[0-*] is associated with
 280     |---cdev[0-*]_weight:       Influence of the cooling device in
 281                                 this thermal zone
 282
 283 Besides the thermal zone device sysfs I/F and cooling device sysfs I/F,
 284 the generic thermal driver also creates a hwmon sysfs I/F for each _type_
 285 of thermal zone device. E.g. the generic thermal driver registers one hwmon
 286 class device and build the associated hwmon sysfs I/F for all the registered
 287 ACPI thermal zones.
 288
 289 /sys/class/hwmon/hwmon[0-*]:
 290     |---name:                   The type of the thermal zone devices
 291     |---temp[1-*]_input:        The current temperature of thermal zone [1-*]
 292     |---temp[1-*]_critical:     The critical trip point of thermal zone [1-*]
 293
 294 Please read Documentation/hwmon/sysfs-interface for additional information.
 295
 296 ***************************
 297 * Thermal zone attributes *
 298 ***************************
 299
 300 type
 301         Strings which represent the thermal zone type.
 302         This is given by thermal zone driver as part of registration.
 303         E.g: "acpitz" indicates it's an ACPI thermal device.
 304         In order to keep it consistent with hwmon sys attribute; this should
 305         be a short, lowercase string, not containing spaces nor dashes.
 306         RO, Required
 307
 308 temp
 309         Current temperature as reported by thermal zone (sensor).
 310         Unit: millidegree Celsius
 311         RO, Required
 312
 313 mode
 314         One of the predefined values in [enabled, disabled].
 315         This file gives information about the algorithm that is currently
 316         managing the thermal zone. It can be either default kernel based
 317         algorithm or user space application.
 318         enabled         = enable Kernel Thermal management.
 319         disabled        = Preventing kernel thermal zone driver actions upon
 320                           trip points so that user application can take full
 321                           charge of the thermal management.
 322         RW, Optional
 323
 324 policy
 325         One of the various thermal governors used for a particular zone.
 326         RW, Required
 327
 328 available_policies
 329         Available thermal governors which can be used for a particular zone.
 330         RO, Required
 331
 332 trip_point_[0-*]_temp
 333         The temperature above which trip point will be fired.
 334         Unit: millidegree Celsius
 335         RO, Optional
 336
 337 trip_point_[0-*]_type
 338         Strings which indicate the type of the trip point.
 339         E.g. it can be one of critical, hot, passive, active[0-*] for ACPI
 340         thermal zone.
 341         RO, Optional
 342
 343 trip_point_[0-*]_hyst
 344         The hysteresis value for a trip point, represented as an integer
 345         Unit: Celsius
 346         RW, Optional
 347
 348 cdev[0-*]
 349         Sysfs link to the thermal cooling device node where the sys I/F
 350         for cooling device throttling control represents.
 351         RO, Optional
 352
 353 cdev[0-*]_trip_point
 354         The trip point in this thermal zone which cdev[0-*] is associated
 355         with; -1 means the cooling device is not associated with any trip
 356         point.
 357         RO, Optional
 358
 359 cdev[0-*]_weight
 360         The influence of cdev[0-*] in this thermal zone. This value
 361         is relative to the rest of cooling devices in the thermal
 362         zone. For example, if a cooling device has a weight double
 363         than that of other, it's twice as effective in cooling the
 364         thermal zone.
 365         RW, Optional
 366
 367 passive
 368         Attribute is only present for zones in which the passive cooling
 369         policy is not supported by native thermal driver. Default is zero
 370         and can be set to a temperature (in millidegrees) to enable a
 371         passive trip point for the zone. Activation is done by polling with
 372         an interval of 1 second.
 373         Unit: millidegrees Celsius
 374         Valid values: 0 (disabled) or greater than 1000
 375         RW, Optional
 376
 377 emul_temp
 378         Interface to set the emulated temperature method in thermal zone
 379         (sensor). After setting this temperature, the thermal zone may pass
 380         this temperature to platform emulation function if registered or
 381         cache it locally. This is useful in debugging different temperature
 382         threshold and its associated cooling action. This is write only node
 383         and writing 0 on this node should disable emulation.
 384         Unit: millidegree Celsius
 385         WO, Optional
 386
 387           WARNING: Be careful while enabling this option on production systems,
 388           because userland can easily disable the thermal policy by simply
 389           flooding this sysfs node with low temperature values.
 390
 391 sustainable_power
 392         An estimate of the sustained power that can be dissipated by
 393         the thermal zone. Used by the power allocator governor. For
 394         more information see Documentation/thermal/power_allocator.txt
 395         Unit: milliwatts
 396         RW, Optional
 397
 398 k_po
 399         The proportional term of the power allocator governor's PID
 400         controller during temperature overshoot. Temperature overshoot
 401         is when the current temperature is above the "desired
 402         temperature" trip point. For more information see
 403         Documentation/thermal/power_allocator.txt
 404         RW, Optional
 405
 406 k_pu
 407         The proportional term of the power allocator governor's PID
 408         controller during temperature undershoot. Temperature undershoot
 409         is when the current temperature is below the "desired
 410         temperature" trip point. For more information see
 411         Documentation/thermal/power_allocator.txt
 412         RW, Optional
 413
 414 k_i
 415         The integral term of the power allocator governor's PID
 416         controller. This term allows the PID controller to compensate
 417         for long term drift. For more information see
 418         Documentation/thermal/power_allocator.txt
 419         RW, Optional
 420
 421 k_d
 422         The derivative term of the power allocator governor's PID
 423         controller. For more information see
 424         Documentation/thermal/power_allocator.txt
 425         RW, Optional
 426
 427 integral_cutoff
 428         Temperature offset from the desired temperature trip point
 429         above which the integral term of the power allocator
 430         governor's PID controller starts accumulating errors. For
 431         example, if integral_cutoff is 0, then the integral term only
 432         accumulates error when temperature is above the desired
 433         temperature trip point. For more information see
 434         Documentation/thermal/power_allocator.txt
 435         Unit: millidegree Celsius
 436         RW, Optional
 437
 438 slope
 439         The slope constant used in a linear extrapolation model
 440         to determine a hotspot temperature based off the sensor's
 441         raw readings. It is up to the device driver to determine
 442         the usage of these values.
 443         RW, Optional
 444
 445 offset
 446         The offset constant used in a linear extrapolation model
 447         to determine a hotspot temperature based off the sensor's
 448         raw readings. It is up to the device driver to determine
 449         the usage of these values.
 450         RW, Optional
 451
 452 *****************************
 453 * Cooling device attributes *
 454 *****************************
 455
 456 type
 457         String which represents the type of device, e.g:
 458         - for generic ACPI: should be "Fan", "Processor" or "LCD"
 459         - for memory controller device on intel_menlow platform:
 460           should be "Memory controller".
 461         RO, Required
 462
 463 max_state
 464         The maximum permissible cooling state of this cooling device.
 465         RO, Required
 466
 467 cur_state
 468         The current cooling state of this cooling device.
 469         The value can any integer numbers between 0 and max_state:
 470         - cur_state == 0 means no cooling
 471         - cur_state == max_state means the maximum cooling.
 472         RW, Required
 473
 474 3. A simple implementation
 475
 476 ACPI thermal zone may support multiple trip points like critical, hot,
 477 passive, active. If an ACPI thermal zone supports critical, passive,
 478 active[0] and active[1] at the same time, it may register itself as a
 479 thermal_zone_device (thermal_zone1) with 4 trip points in all.
 480 It has one processor and one fan, which are both registered as
 481 thermal_cooling_device. Both are considered to have the same
 482 effectiveness in cooling the thermal zone.
 483
 484 If the processor is listed in _PSL method, and the fan is listed in _AL0
 485 method, the sys I/F structure will be built like this:
 486
 487 /sys/class/thermal:
 488
 489 |thermal_zone1:
 490     |---type:                   acpitz
 491     |---temp:                   37000
 492     |---mode:                   enabled
 493     |---policy:                 step_wise
 494     |---available_policies:     step_wise fair_share
 495     |---trip_point_0_temp:      100000
 496     |---trip_point_0_type:      critical
 497     |---trip_point_1_temp:      80000
 498     |---trip_point_1_type:      passive
 499     |---trip_point_2_temp:      70000
 500     |---trip_point_2_type:      active0
 501     |---trip_point_3_temp:      60000
 502     |---trip_point_3_type:      active1
 503     |---cdev0:                  --->/sys/class/thermal/cooling_device0
 504     |---cdev0_trip_point:       1       /* cdev0 can be used for passive */
 505     |---cdev0_weight:           1024
 506     |---cdev1:                  --->/sys/class/thermal/cooling_device3
 507     |---cdev1_trip_point:       2       /* cdev1 can be used for active[0]*/
 508     |---cdev1_weight:           1024
 509
 510 |cooling_device0:
 511     |---type:                   Processor
 512     |---max_state:              8
 513     |---cur_state:              0
 514
 515 |cooling_device3:
 516     |---type:                   Fan
 517     |---max_state:              2
 518     |---cur_state:              0
 519
 520 /sys/class/hwmon:
 521
 522 |hwmon0:
 523     |---name:                   acpitz
 524     |---temp1_input:            37000
 525     |---temp1_crit:             100000
 526
 527 4. Event Notification
 528
 529 The framework includes a simple notification mechanism, in the form of a
 530 netlink event. Netlink socket initialization is done during the _init_
 531 of the framework. Drivers which intend to use the notification mechanism
 532 just need to call thermal_generate_netlink_event() with two arguments viz
 533 (originator, event). The originator is a pointer to struct thermal_zone_device
 534 from where the event has been originated. An integer which represents the
 535 thermal zone device will be used in the message to identify the zone. The
 536 event will be one of:{THERMAL_AUX0, THERMAL_AUX1, THERMAL_CRITICAL,
 537 THERMAL_DEV_FAULT}. Notification can be sent when the current temperature
 538 crosses any of the configured thresholds.
 539
 540 5. Export Symbol APIs:
 541
 542 5.1: get_tz_trend:
 543 This function returns the trend of a thermal zone, i.e the rate of change
 544 of temperature of the thermal zone. Ideally, the thermal sensor drivers
 545 are supposed to implement the callback. If they don't, the thermal
 546 framework calculated the trend by comparing the previous and the current
 547 temperature values.
 548
 549 5.2:get_thermal_instance:
 550 This function returns the thermal_instance corresponding to a given
 551 {thermal_zone, cooling_device, trip_point} combination. Returns NULL
 552 if such an instance does not exist.
 553
 554 5.3:thermal_notify_framework:
 555 This function handles the trip events from sensor drivers. It starts
 556 throttling the cooling devices according to the policy configured.
 557 For CRITICAL and HOT trip points, this notifies the respective drivers,
 558 and does actual throttling for other trip points i.e ACTIVE and PASSIVE.
 559 The throttling policy is based on the configured platform data; if no
 560 platform data is provided, this uses the step_wise throttling policy.
 561
 562 5.4:thermal_cdev_update:
 563 This function serves as an arbitrator to set the state of a cooling
 564 device. It sets the cooling device to the deepest cooling state if
 565 possible.