1 Naming and data format standards for sysfs files
2 ------------------------------------------------
4 The libsensors library offers an interface to the raw sensors data
5 through the sysfs interface. Since lm-sensors 3.0.0, libsensors is
6 completely chip-independent. It assumes that all the kernel drivers
7 implement the standard sysfs interface described in this document.
8 This makes adding or updating support for any given chip very easy, as
9 libsensors, and applications using it, do not need to be modified.
10 This is a major improvement compared to lm-sensors 2.
12 Note that motherboards vary widely in the connections to sensor chips.
13 There is no standard that ensures, for example, that the second
14 temperature sensor is connected to the CPU, or that the second fan is on
15 the CPU. Also, some values reported by the chips need some computation
16 before they make full sense. For example, most chips can only measure
17 voltages between 0 and +4V. Other voltages are scaled back into that
18 range using external resistors. Since the values of these resistors
19 can change from motherboard to motherboard, the conversions cannot be
20 hard coded into the driver and have to be done in user space.
22 For this reason, even if we aim at a chip-independent libsensors, it will
23 still require a configuration file (e.g. /etc/sensors.conf) for proper
24 values conversion, labeling of inputs and hiding of unused inputs.
26 An alternative method that some programs use is to access the sysfs
27 files directly. This document briefly describes the standards that the
28 drivers follow, so that an application program can scan for entries and
29 access this data in a simple and consistent way. That said, such programs
30 will have to implement conversion, labeling and hiding of inputs. For
31 this reason, it is still not recommended to bypass the library.
33 Each chip gets its own directory in the sysfs /sys/devices tree. To
34 find all sensor chips, it is easier to follow the device symlinks from
35 /sys/class/hwmon/hwmon*.
37 Up to lm-sensors 3.0.0, libsensors looks for hardware monitoring attributes
38 in the "physical" device directory. Since lm-sensors 3.0.1, attributes found
39 in the hwmon "class" device directory are also supported. Complex drivers
40 (e.g. drivers for multifunction chips) may want to use this possibility to
41 avoid namespace pollution. The only drawback will be that older versions of
42 libsensors won't support the driver in question.
44 All sysfs values are fixed point numbers.
46 There is only one value per file, unlike the older /proc specification.
47 The common scheme for files naming is: <type><number>_<item>. Usual
48 types for sensor chips are "in" (voltage), "temp" (temperature) and
49 "fan" (fan). Usual items are "input" (measured value), "max" (high
50 threshold, "min" (low threshold). Numbering usually starts from 1,
51 except for voltages which start from 0 (because most data sheets use
52 this). A number is always used for elements that can be present more
53 than once, even if there is a single element of the given type on the
54 specific chip. Other files do not refer to a specific element, so
55 they have a simple name, and no number.
57 Alarms are direct indications read from the chips. The drivers do NOT
58 make comparisons of readings to thresholds. This allows violations
59 between readings to be caught and alarmed. The exact definition of an
60 alarm (for example, whether a threshold must be met or must be exceeded
61 to cause an alarm) is chip-dependent.
63 When setting values of hwmon sysfs attributes, the string representation of
64 the desired value must be written, note that strings which are not a number
65 are interpreted as 0! For more on how written strings are interpreted see the
66 "sysfs attribute writes interpretation" section at the end of this file.
68 -------------------------------------------------------------------------
70 [0-*] denotes any positive number starting from 0
71 [1-*] denotes any positive number starting from 1
76 Read/write values may be read-only for some chips, depending on the
77 hardware implementation.
79 All entries (except name) are optional, and should only be created in a
80 given driver if the chip has the feature.
88 This should be a short, lowercase string, not containing
89 whitespace, dashes, or the wildcard character '*'.
90 This attribute represents the chip name. It is the only
92 I2C devices get this attribute created automatically.
95 update_interval The interval at which the chip will update readings.
98 Some devices have a variable update rate or interval.
99 This attribute can be used to change it to the desired value.
106 in[0-*]_min Voltage min value.
110 in[0-*]_lcrit Voltage critical min value.
113 If voltage drops to or below this limit, the system may
114 take drastic action such as power down or reset. At the very
115 least, it should report a fault.
117 in[0-*]_max Voltage max value.
121 in[0-*]_crit Voltage critical max value.
124 If voltage reaches or exceeds this limit, the system may
125 take drastic action such as power down or reset. At the very
126 least, it should report a fault.
128 in[0-*]_input Voltage input value.
131 Voltage measured on the chip pin.
132 Actual voltage depends on the scaling resistors on the
133 motherboard, as recommended in the chip datasheet.
134 This varies by chip and by motherboard.
135 Because of this variation, values are generally NOT scaled
136 by the chip driver, and must be done by the application.
137 However, some drivers (notably lm87 and via686a)
138 do scale, because of internal resistors built into a chip.
139 These drivers will output the actual voltage. Rule of
140 thumb: drivers should report the voltage values at the
149 Historical minimum voltage
154 Historical maximum voltage
158 in[0-*]_reset_history
159 Reset inX_lowest and inX_highest
163 Reset inX_lowest and inX_highest for all sensors
166 in[0-*]_label Suggested voltage channel label.
168 Should only be created if the driver has hints about what
169 this voltage channel is being used for, and user-space
170 doesn't. In all other cases, the label is provided by
175 Enable or disable the sensors.
176 When disabled the sensor read will return -ENODATA.
181 cpu[0-*]_vid CPU core reference voltage.
186 vrm Voltage Regulator Module version number.
187 RW (but changing it should no more be necessary)
188 Originally the VRM standard version multiplied by 10, but now
189 an arbitrary number, as not all standards have a version
191 Affects the way the driver calculates the CPU core reference
192 voltage from the vid pins.
194 Also see the Alarms section for status flags associated with voltages.
201 fan[1-*]_min Fan minimum value
202 Unit: revolution/min (RPM)
205 fan[1-*]_max Fan maximum value
206 Unit: revolution/min (RPM)
207 Only rarely supported by the hardware.
210 fan[1-*]_input Fan input value.
211 Unit: revolution/min (RPM)
214 fan[1-*]_div Fan divisor.
215 Integer value in powers of two (1, 2, 4, 8, 16, 32, 64, 128).
217 Some chips only support values 1, 2, 4 and 8.
218 Note that this is actually an internal clock divisor, which
219 affects the measurable speed range, not the read value.
221 fan[1-*]_pulses Number of tachometer pulses per fan revolution.
222 Integer value, typically between 1 and 4.
224 This value is a characteristic of the fan connected to the
225 device's input, so it has to be set in accordance with the fan
227 Should only be created if the chip has a register to configure
228 the number of pulses. In the absence of such a register (and
229 thus attribute) the value assumed by all devices is 2 pulses
234 Unit: revolution/min (RPM)
236 Only makes sense if the chip supports closed-loop fan speed
237 control based on the measured fan speed.
239 fan[1-*]_label Suggested fan channel label.
241 Should only be created if the driver has hints about what
242 this fan channel is being used for, and user-space doesn't.
243 In all other cases, the label is provided by user-space.
247 Enable or disable the sensors.
248 When disabled the sensor read will return -ENODATA.
253 Also see the Alarms section for status flags associated with fans.
260 pwm[1-*] Pulse width modulation fan control.
261 Integer value in the range 0 to 255
266 Fan speed control method:
267 0: no fan speed control (i.e. fan at full speed)
268 1: manual fan speed control enabled (using pwm[1-*])
269 2+: automatic fan speed control enabled
270 Check individual chip documentation files for automatic mode
274 pwm[1-*]_mode 0: DC mode (direct current)
275 1: PWM mode (pulse-width modulation)
278 pwm[1-*]_freq Base PWM frequency in Hz.
279 Only possibly available when pwmN_mode is PWM, but not always
283 pwm[1-*]_auto_channels_temp
284 Select which temperature channels affect this PWM output in
285 auto mode. Bitfield, 1 is temp1, 2 is temp2, 4 is temp3 etc...
286 Which values are possible depend on the chip used.
289 pwm[1-*]_auto_point[1-*]_pwm
290 pwm[1-*]_auto_point[1-*]_temp
291 pwm[1-*]_auto_point[1-*]_temp_hyst
292 Define the PWM vs temperature curve. Number of trip points is
293 chip-dependent. Use this for chips which associate trip points
294 to PWM output channels.
297 temp[1-*]_auto_point[1-*]_pwm
298 temp[1-*]_auto_point[1-*]_temp
299 temp[1-*]_auto_point[1-*]_temp_hyst
300 Define the PWM vs temperature curve. Number of trip points is
301 chip-dependent. Use this for chips which associate trip points
302 to temperature channels.
305 There is a third case where trip points are associated to both PWM output
306 channels and temperature channels: the PWM values are associated to PWM
307 output channels while the temperature values are associated to temperature
308 channels. In that case, the result is determined by the mapping between
309 temperature inputs and PWM outputs. When several temperature inputs are
310 mapped to a given PWM output, this leads to several candidate PWM values.
311 The actual result is up to the chip, but in general the highest candidate
312 value (fastest fan speed) wins.
319 temp[1-*]_type Sensor type selection.
322 1: CPU embedded diode
328 Not all types are supported by all chips
330 temp[1-*]_max Temperature max value.
331 Unit: millidegree Celsius (or millivolt, see below)
334 temp[1-*]_min Temperature min value.
335 Unit: millidegree Celsius
339 Temperature hysteresis value for max limit.
340 Unit: millidegree Celsius
341 Must be reported as an absolute temperature, NOT a delta
346 Temperature hysteresis value for min limit.
347 Unit: millidegree Celsius
348 Must be reported as an absolute temperature, NOT a delta
352 temp[1-*]_input Temperature input value.
353 Unit: millidegree Celsius
356 temp[1-*]_crit Temperature critical max value, typically greater than
357 corresponding temp_max values.
358 Unit: millidegree Celsius
362 Temperature hysteresis value for critical limit.
363 Unit: millidegree Celsius
364 Must be reported as an absolute temperature, NOT a delta
365 from the critical value.
369 Temperature emergency max value, for chips supporting more than
370 two upper temperature limits. Must be equal or greater than
371 corresponding temp_crit values.
372 Unit: millidegree Celsius
375 temp[1-*]_emergency_hyst
376 Temperature hysteresis value for emergency limit.
377 Unit: millidegree Celsius
378 Must be reported as an absolute temperature, NOT a delta
379 from the emergency value.
382 temp[1-*]_lcrit Temperature critical min value, typically lower than
383 corresponding temp_min values.
384 Unit: millidegree Celsius
388 Temperature hysteresis value for critical min limit.
389 Unit: millidegree Celsius
390 Must be reported as an absolute temperature, NOT a delta
391 from the critical min value.
395 Temperature offset which is added to the temperature reading
397 Unit: millidegree Celsius
400 temp[1-*]_label Suggested temperature channel label.
402 Should only be created if the driver has hints about what
403 this temperature channel is being used for, and user-space
404 doesn't. In all other cases, the label is provided by
409 Historical minimum temperature
410 Unit: millidegree Celsius
414 Historical maximum temperature
415 Unit: millidegree Celsius
418 temp[1-*]_reset_history
419 Reset temp_lowest and temp_highest
423 Reset temp_lowest and temp_highest for all sensors
427 Enable or disable the sensors.
428 When disabled the sensor read will return -ENODATA.
433 Some chips measure temperature using external thermistors and an ADC, and
434 report the temperature measurement as a voltage. Converting this voltage
435 back to a temperature (or the other way around for limits) requires
436 mathematical functions not available in the kernel, so the conversion
437 must occur in user space. For these chips, all temp* files described
438 above should contain values expressed in millivolt instead of millidegree
439 Celsius. In other words, such temperature channels are handled as voltage
440 channels by the driver.
442 Also see the Alarms section for status flags associated with temperatures.
449 curr[1-*]_max Current max value
453 curr[1-*]_min Current min value.
457 curr[1-*]_lcrit Current critical low value
461 curr[1-*]_crit Current critical high value.
465 curr[1-*]_input Current input value
475 Historical minimum current
480 Historical maximum current
484 curr[1-*]_reset_history
485 Reset currX_lowest and currX_highest
489 Reset currX_lowest and currX_highest for all sensors
493 Enable or disable the sensors.
494 When disabled the sensor read will return -ENODATA.
499 Also see the Alarms section for status flags associated with currents.
505 power[1-*]_average Average power use
509 power[1-*]_average_interval Power use averaging interval. A poll
510 notification is sent to this file if the
511 hardware changes the averaging interval.
515 power[1-*]_average_interval_max Maximum power use averaging interval
519 power[1-*]_average_interval_min Minimum power use averaging interval
523 power[1-*]_average_highest Historical average maximum power use
527 power[1-*]_average_lowest Historical average minimum power use
531 power[1-*]_average_max A poll notification is sent to
532 power[1-*]_average when power use
533 rises above this value.
537 power[1-*]_average_min A poll notification is sent to
538 power[1-*]_average when power use
539 sinks below this value.
543 power[1-*]_input Instantaneous power use
547 power[1-*]_input_highest Historical maximum power use
551 power[1-*]_input_lowest Historical minimum power use
555 power[1-*]_reset_history Reset input_highest, input_lowest,
556 average_highest and average_lowest.
559 power[1-*]_accuracy Accuracy of the power meter.
563 power[1-*]_cap If power use rises above this limit, the
564 system should take action to reduce power use.
565 A poll notification is sent to this file if the
566 cap is changed by the hardware. The *_cap
567 files only appear if the cap is known to be
568 enforced by hardware.
572 power[1-*]_cap_hyst Margin of hysteresis built around capping and
577 power[1-*]_cap_max Maximum cap that can be set.
581 power[1-*]_cap_min Minimum cap that can be set.
585 power[1-*]_max Maximum power.
589 power[1-*]_crit Critical maximum power.
590 If power rises to or above this limit, the
591 system is expected take drastic action to reduce
592 power consumption, such as a system shutdown or
593 a forced powerdown of some devices.
597 power[1-*]_enable Enable or disable the sensors.
598 When disabled the sensor read will return
604 Also see the Alarms section for status flags associated with power readings.
610 energy[1-*]_input Cumulative energy use
614 energy[1-*]_enable Enable or disable the sensors.
615 When disabled the sensor read will return
625 humidity[1-*]_input Humidity
626 Unit: milli-percent (per cent mille, pcm)
630 humidity[1-*]_enable Enable or disable the sensors
631 When disabled the sensor read will return
641 Each channel or limit may have an associated alarm file, containing a
642 boolean value. 1 means than an alarm condition exists, 0 means no alarm.
644 Usually a given chip will either use channel-related alarms, or
645 limit-related alarms, not both. The driver should just reflect the hardware
666 curr[1-*]_lcrit_alarm
670 power[1-*]_crit_alarm
675 temp[1-*]_lcrit_alarm
677 temp[1-*]_emergency_alarm
683 Each input channel may have an associated fault file. This can be used
684 to notify open diodes, unconnected fans etc. where the hardware
685 supports it. When this boolean has value 1, the measurement for that
686 channel should not be trusted.
690 Input fault condition
695 Some chips also offer the possibility to get beeped when an alarm occurs:
697 beep_enable Master beep enable
711 In theory, a chip could provide per-limit beep masking, but no such chip
714 Old drivers provided a different, non-standard interface to alarms and
715 beeps. These interface files are deprecated, but will be kept around
716 for compatibility reasons:
718 alarms Alarm bitmask.
720 Integer representation of one to four bytes.
721 A '1' bit means an alarm.
722 Chips should be programmed for 'comparator' mode so that
723 the alarm will 'come back' after you read the register
724 if it is still valid.
725 Generally a direct representation of a chip's internal
726 alarm registers; there is no standard for the position
727 of individual bits. For this reason, the use of this
728 interface file for new drivers is discouraged. Use
729 individual *_alarm and *_fault files instead.
730 Bits are defined in kernel/include/sensors.h.
732 beep_mask Bitmask for beep.
733 Same format as 'alarms' with the same bit locations,
734 use discouraged for the same reason. Use individual
735 *_beep files instead.
739 ***********************
740 * Intrusion detection *
741 ***********************
744 Chassis intrusion detection
746 1: intrusion detected
748 Contrary to regular alarm flags which clear themselves
749 automatically when read, this one sticks until cleared by
750 the user. This is done by writing 0 to the file. Writing
751 other values is unsupported.
754 Chassis intrusion beep
760 sysfs attribute writes interpretation
761 -------------------------------------
763 hwmon sysfs attributes always contain numbers, so the first thing to do is to
764 convert the input to a number, there are 2 ways todo this depending whether
765 the number can be negative or not:
766 unsigned long u = simple_strtoul(buf, NULL, 10);
767 long s = simple_strtol(buf, NULL, 10);
769 With buf being the buffer with the user input being passed by the kernel.
770 Notice that we do not use the second argument of strto[u]l, and thus cannot
771 tell when 0 is returned, if this was really 0 or is caused by invalid input.
772 This is done deliberately as checking this everywhere would add a lot of
775 Notice that it is important to always store the converted value in an
776 unsigned long or long, so that no wrap around can happen before any further
779 After the input string is converted to an (unsigned) long, the value should be
780 checked if its acceptable. Be careful with further conversions on the value
781 before checking it for validity, as these conversions could still cause a wrap
782 around before the check. For example do not multiply the result, and only
783 add/subtract if it has been divided before the add/subtract.
785 What to do if a value is found to be invalid, depends on the type of the
786 sysfs attribute that is being set. If it is a continuous setting like a
787 tempX_max or inX_max attribute, then the value should be clamped to its
788 limits using clamp_val(value, min_limit, max_limit). If it is not continuous
789 like for example a tempX_type, then when an invalid value is written,
790 -EINVAL should be returned.
792 Example1, temp1_max, register is a signed 8 bit value (-128 - 127 degrees):
794 long v = simple_strtol(buf, NULL, 10) / 1000;
795 v = clamp_val(v, -128, 127);
796 /* write v to register */
798 Example2, fan divider setting, valid values 2, 4 and 8:
800 unsigned long v = simple_strtoul(buf, NULL, 10);
803 case 2: v = 1; break;
804 case 4: v = 2; break;
805 case 8: v = 3; break;
809 /* write v to register */