| blob | 2cc95ad466047b055d45f748619cf2d7231d5e93 |
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
72 RO read only value
73 WO write only value
74 RW read/write value
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.
83 *********************
84 * Global attributes *
85 *********************
87 name The chip name.
88 This should be a short, lowercase string, not containing
89 spaces nor dashes, representing the chip name. This is
90 the only mandatory attribute.
91 I2C devices get this attribute created automatically.
92 RO
94 update_interval The interval at which the chip will update readings.
95 Unit: millisecond
96 RW
97 Some devices have a variable update rate or interval.
98 This attribute can be used to change it to the desired value.
101 ************
102 * Voltages *
103 ************
105 in[0-*]_min Voltage min value.
106 Unit: millivolt
107 RW
109 in[0-*]_lcrit Voltage critical min value.
110 Unit: millivolt
111 RW
112 If voltage drops to or below this limit, the system may
113 take drastic action such as power down or reset. At the very
114 least, it should report a fault.
116 in[0-*]_max Voltage max value.
117 Unit: millivolt
118 RW
120 in[0-*]_crit Voltage critical max value.
121 Unit: millivolt
122 RW
123 If voltage reaches or exceeds this limit, the system may
124 take drastic action such as power down or reset. At the very
125 least, it should report a fault.
127 in[0-*]_input Voltage input value.
128 Unit: millivolt
129 RO
130 Voltage measured on the chip pin.
131 Actual voltage depends on the scaling resistors on the
132 motherboard, as recommended in the chip datasheet.
133 This varies by chip and by motherboard.
134 Because of this variation, values are generally NOT scaled
135 by the chip driver, and must be done by the application.
136 However, some drivers (notably lm87 and via686a)
137 do scale, because of internal resistors built into a chip.
138 These drivers will output the actual voltage. Rule of
139 thumb: drivers should report the voltage values at the
140 "pins" of the chip.
142 in[0-*]_average
143 Average voltage
144 Unit: millivolt
145 RO
147 in[0-*]_lowest
148 Historical minimum voltage
149 Unit: millivolt
150 RO
152 in[0-*]_highest
153 Historical maximum voltage
154 Unit: millivolt
155 RO
157 in[0-*]_reset_history
158 Reset inX_lowest and inX_highest
159 WO
161 in_reset_history
162 Reset inX_lowest and inX_highest for all sensors
163 WO
165 in[0-*]_label Suggested voltage channel label.
166 Text string
167 Should only be created if the driver has hints about what
168 this voltage channel is being used for, and user-space
169 doesn't. In all other cases, the label is provided by
170 user-space.
171 RO
173 cpu[0-*]_vid CPU core reference voltage.
174 Unit: millivolt
175 RO
176 Not always correct.
178 vrm Voltage Regulator Module version number.
179 RW (but changing it should no more be necessary)
180 Originally the VRM standard version multiplied by 10, but now
181 an arbitrary number, as not all standards have a version
182 number.
183 Affects the way the driver calculates the CPU core reference
184 voltage from the vid pins.
186 Also see the Alarms section for status flags associated with voltages.
189 ********
190 * Fans *
191 ********
193 fan[1-*]_min Fan minimum value
194 Unit: revolution/min (RPM)
195 RW
197 fan[1-*]_max Fan maximum value
198 Unit: revolution/min (RPM)
199 Only rarely supported by the hardware.
200 RW
202 fan[1-*]_input Fan input value.
203 Unit: revolution/min (RPM)
204 RO
206 fan[1-*]_div Fan divisor.
207 Integer value in powers of two (1, 2, 4, 8, 16, 32, 64, 128).
208 RW
209 Some chips only support values 1, 2, 4 and 8.
210 Note that this is actually an internal clock divisor, which
211 affects the measurable speed range, not the read value.
213 fan[1-*]_pulses Number of tachometer pulses per fan revolution.
214 Integer value, typically between 1 and 4.
215 RW
216 This value is a characteristic of the fan connected to the
217 device's input, so it has to be set in accordance with the fan
218 model.
219 Should only be created if the chip has a register to configure
220 the number of pulses. In the absence of such a register (and
221 thus attribute) the value assumed by all devices is 2 pulses
222 per fan revolution.
224 fan[1-*]_target
225 Desired fan speed
226 Unit: revolution/min (RPM)
227 RW
228 Only makes sense if the chip supports closed-loop fan speed
229 control based on the measured fan speed.
231 fan[1-*]_label Suggested fan channel label.
232 Text string
233 Should only be created if the driver has hints about what
234 this fan channel is being used for, and user-space doesn't.
235 In all other cases, the label is provided by user-space.
236 RO
238 Also see the Alarms section for status flags associated with fans.
241 *******
242 * PWM *
243 *******
245 pwm[1-*] Pulse width modulation fan control.
246 Integer value in the range 0 to 255
247 RW
248 255 is max or 100%.
250 pwm[1-*]_enable
251 Fan speed control method:
252 0: no fan speed control (i.e. fan at full speed)
253 1: manual fan speed control enabled (using pwm[1-*])
254 2+: automatic fan speed control enabled
255 Check individual chip documentation files for automatic mode
256 details.
257 RW
259 pwm[1-*]_mode 0: DC mode (direct current)
260 1: PWM mode (pulse-width modulation)
261 RW
263 pwm[1-*]_freq Base PWM frequency in Hz.
264 Only possibly available when pwmN_mode is PWM, but not always
265 present even then.
266 RW
268 pwm[1-*]_auto_channels_temp
269 Select which temperature channels affect this PWM output in
270 auto mode. Bitfield, 1 is temp1, 2 is temp2, 4 is temp3 etc...
271 Which values are possible depend on the chip used.
272 RW
274 pwm[1-*]_auto_point[1-*]_pwm
275 pwm[1-*]_auto_point[1-*]_temp
276 pwm[1-*]_auto_point[1-*]_temp_hyst
277 Define the PWM vs temperature curve. Number of trip points is
278 chip-dependent. Use this for chips which associate trip points
279 to PWM output channels.
280 RW
282 temp[1-*]_auto_point[1-*]_pwm
283 temp[1-*]_auto_point[1-*]_temp
284 temp[1-*]_auto_point[1-*]_temp_hyst
285 Define the PWM vs temperature curve. Number of trip points is
286 chip-dependent. Use this for chips which associate trip points
287 to temperature channels.
288 RW
290 There is a third case where trip points are associated to both PWM output
291 channels and temperature channels: the PWM values are associated to PWM
292 output channels while the temperature values are associated to temperature
293 channels. In that case, the result is determined by the mapping between
294 temperature inputs and PWM outputs. When several temperature inputs are
295 mapped to a given PWM output, this leads to several candidate PWM values.
296 The actual result is up to the chip, but in general the highest candidate
297 value (fastest fan speed) wins.
300 ****************
301 * Temperatures *
302 ****************
304 temp[1-*]_type Sensor type selection.
305 Integers 1 to 6
306 RW
307 1: CPU embedded diode
308 2: 3904 transistor
309 3: thermal diode
310 4: thermistor
311 5: AMD AMDSI
312 6: Intel PECI
313 Not all types are supported by all chips
315 temp[1-*]_max Temperature max value.
316 Unit: millidegree Celsius (or millivolt, see below)
317 RW
319 temp[1-*]_min Temperature min value.
320 Unit: millidegree Celsius
321 RW
323 temp[1-*]_max_hyst
324 Temperature hysteresis value for max limit.
325 Unit: millidegree Celsius
326 Must be reported as an absolute temperature, NOT a delta
327 from the max value.
328 RW
330 temp[1-*]_min_hyst
331 Temperature hysteresis value for min limit.
332 Unit: millidegree Celsius
333 Must be reported as an absolute temperature, NOT a delta
334 from the min value.
335 RW
337 temp[1-*]_input Temperature input value.
338 Unit: millidegree Celsius
339 RO
341 temp[1-*]_crit Temperature critical max value, typically greater than
342 corresponding temp_max values.
343 Unit: millidegree Celsius
344 RW
346 temp[1-*]_crit_hyst
347 Temperature hysteresis value for critical limit.
348 Unit: millidegree Celsius
349 Must be reported as an absolute temperature, NOT a delta
350 from the critical value.
351 RW
353 temp[1-*]_emergency
354 Temperature emergency max value, for chips supporting more than
355 two upper temperature limits. Must be equal or greater than
356 corresponding temp_crit values.
357 Unit: millidegree Celsius
358 RW
360 temp[1-*]_emergency_hyst
361 Temperature hysteresis value for emergency limit.
362 Unit: millidegree Celsius
363 Must be reported as an absolute temperature, NOT a delta
364 from the emergency value.
365 RW
367 temp[1-*]_lcrit Temperature critical min value, typically lower than
368 corresponding temp_min values.
369 Unit: millidegree Celsius
370 RW
372 temp[1-*]_lcrit_hyst
373 Temperature hysteresis value for critical min limit.
374 Unit: millidegree Celsius
375 Must be reported as an absolute temperature, NOT a delta
376 from the critical min value.
377 RW
379 temp[1-*]_offset
380 Temperature offset which is added to the temperature reading
381 by the chip.
382 Unit: millidegree Celsius
383 Read/Write value.
385 temp[1-*]_label Suggested temperature channel label.
386 Text string
387 Should only be created if the driver has hints about what
388 this temperature channel is being used for, and user-space
389 doesn't. In all other cases, the label is provided by
390 user-space.
391 RO
393 temp[1-*]_lowest
394 Historical minimum temperature
395 Unit: millidegree Celsius
396 RO
398 temp[1-*]_highest
399 Historical maximum temperature
400 Unit: millidegree Celsius
401 RO
403 temp[1-*]_reset_history
404 Reset temp_lowest and temp_highest
405 WO
407 temp_reset_history
408 Reset temp_lowest and temp_highest for all sensors
409 WO
411 Some chips measure temperature using external thermistors and an ADC, and
412 report the temperature measurement as a voltage. Converting this voltage
413 back to a temperature (or the other way around for limits) requires
414 mathematical functions not available in the kernel, so the conversion
415 must occur in user space. For these chips, all temp* files described
416 above should contain values expressed in millivolt instead of millidegree
417 Celsius. In other words, such temperature channels are handled as voltage
418 channels by the driver.
420 Also see the Alarms section for status flags associated with temperatures.
423 ************
424 * Currents *
425 ************
427 curr[1-*]_max Current max value
428 Unit: milliampere
429 RW
431 curr[1-*]_min Current min value.
432 Unit: milliampere
433 RW
435 curr[1-*]_lcrit Current critical low value
436 Unit: milliampere
437 RW
439 curr[1-*]_crit Current critical high value.
440 Unit: milliampere
441 RW
443 curr[1-*]_input Current input value
444 Unit: milliampere
445 RO
447 curr[1-*]_average
448 Average current use
449 Unit: milliampere
450 RO
452 curr[1-*]_lowest
453 Historical minimum current
454 Unit: milliampere
455 RO
457 curr[1-*]_highest
458 Historical maximum current
459 Unit: milliampere
460 RO
462 curr[1-*]_reset_history
463 Reset currX_lowest and currX_highest
464 WO
466 curr_reset_history
467 Reset currX_lowest and currX_highest for all sensors
468 WO
470 Also see the Alarms section for status flags associated with currents.
472 *********
473 * Power *
474 *********
476 power[1-*]_average Average power use
477 Unit: microWatt
478 RO
480 power[1-*]_average_interval Power use averaging interval. A poll
481 notification is sent to this file if the
482 hardware changes the averaging interval.
483 Unit: milliseconds
484 RW
486 power[1-*]_average_interval_max Maximum power use averaging interval
487 Unit: milliseconds
488 RO
490 power[1-*]_average_interval_min Minimum power use averaging interval
491 Unit: milliseconds
492 RO
494 power[1-*]_average_highest Historical average maximum power use
495 Unit: microWatt
496 RO
498 power[1-*]_average_lowest Historical average minimum power use
499 Unit: microWatt
500 RO
502 power[1-*]_average_max A poll notification is sent to
503 power[1-*]_average when power use
504 rises above this value.
505 Unit: microWatt
506 RW
508 power[1-*]_average_min A poll notification is sent to
509 power[1-*]_average when power use
510 sinks below this value.
511 Unit: microWatt
512 RW
514 power[1-*]_input Instantaneous power use
515 Unit: microWatt
516 RO
518 power[1-*]_input_highest Historical maximum power use
519 Unit: microWatt
520 RO
522 power[1-*]_input_lowest Historical minimum power use
523 Unit: microWatt
524 RO
526 power[1-*]_reset_history Reset input_highest, input_lowest,
527 average_highest and average_lowest.
528 WO
530 power[1-*]_accuracy Accuracy of the power meter.
531 Unit: Percent
532 RO
534 power[1-*]_cap If power use rises above this limit, the
535 system should take action to reduce power use.
536 A poll notification is sent to this file if the
537 cap is changed by the hardware. The *_cap
538 files only appear if the cap is known to be
539 enforced by hardware.
540 Unit: microWatt
541 RW
543 power[1-*]_cap_hyst Margin of hysteresis built around capping and
544 notification.
545 Unit: microWatt
546 RW
548 power[1-*]_cap_max Maximum cap that can be set.
549 Unit: microWatt
550 RO
552 power[1-*]_cap_min Minimum cap that can be set.
553 Unit: microWatt
554 RO
556 power[1-*]_max Maximum power.
557 Unit: microWatt
558 RW
560 power[1-*]_crit Critical maximum power.
561 If power rises to or above this limit, the
562 system is expected take drastic action to reduce
563 power consumption, such as a system shutdown or
564 a forced powerdown of some devices.
565 Unit: microWatt
566 RW
568 Also see the Alarms section for status flags associated with power readings.
570 **********
571 * Energy *
572 **********
574 energy[1-*]_input Cumulative energy use
575 Unit: microJoule
576 RO
579 ************
580 * Humidity *
581 ************
583 humidity[1-*]_input Humidity
584 Unit: milli-percent (per cent mille, pcm)
585 RO
588 **********
589 * Alarms *
590 **********
592 Each channel or limit may have an associated alarm file, containing a
593 boolean value. 1 means than an alarm condition exists, 0 means no alarm.
595 Usually a given chip will either use channel-related alarms, or
596 limit-related alarms, not both. The driver should just reflect the hardware
597 implementation.
599 in[0-*]_alarm
600 curr[1-*]_alarm
601 power[1-*]_alarm
602 fan[1-*]_alarm
603 temp[1-*]_alarm
604 Channel alarm
605 0: no alarm
606 1: alarm
607 RO
609 OR
611 in[0-*]_min_alarm
612 in[0-*]_max_alarm
613 in[0-*]_lcrit_alarm
614 in[0-*]_crit_alarm
615 curr[1-*]_min_alarm
616 curr[1-*]_max_alarm
617 curr[1-*]_lcrit_alarm
618 curr[1-*]_crit_alarm
619 power[1-*]_cap_alarm
620 power[1-*]_max_alarm
621 power[1-*]_crit_alarm
622 fan[1-*]_min_alarm
623 fan[1-*]_max_alarm
624 temp[1-*]_min_alarm
625 temp[1-*]_max_alarm
626 temp[1-*]_lcrit_alarm
627 temp[1-*]_crit_alarm
628 temp[1-*]_emergency_alarm
629 Limit alarm
630 0: no alarm
631 1: alarm
632 RO
634 Each input channel may have an associated fault file. This can be used
635 to notify open diodes, unconnected fans etc. where the hardware
636 supports it. When this boolean has value 1, the measurement for that
637 channel should not be trusted.
639 fan[1-*]_fault
640 temp[1-*]_fault
641 Input fault condition
642 0: no fault occurred
643 1: fault condition
644 RO
646 Some chips also offer the possibility to get beeped when an alarm occurs:
648 beep_enable Master beep enable
649 0: no beeps
650 1: beeps
651 RW
653 in[0-*]_beep
654 curr[1-*]_beep
655 fan[1-*]_beep
656 temp[1-*]_beep
657 Channel beep
658 0: disable
659 1: enable
660 RW
662 In theory, a chip could provide per-limit beep masking, but no such chip
663 was seen so far.
665 Old drivers provided a different, non-standard interface to alarms and
666 beeps. These interface files are deprecated, but will be kept around
667 for compatibility reasons:
669 alarms Alarm bitmask.
670 RO
671 Integer representation of one to four bytes.
672 A '1' bit means an alarm.
673 Chips should be programmed for 'comparator' mode so that
674 the alarm will 'come back' after you read the register
675 if it is still valid.
676 Generally a direct representation of a chip's internal
677 alarm registers; there is no standard for the position
678 of individual bits. For this reason, the use of this
679 interface file for new drivers is discouraged. Use
680 individual *_alarm and *_fault files instead.
681 Bits are defined in kernel/include/sensors.h.
683 beep_mask Bitmask for beep.
684 Same format as 'alarms' with the same bit locations,
685 use discouraged for the same reason. Use individual
686 *_beep files instead.
687 RW
690 ***********************
691 * Intrusion detection *
692 ***********************
694 intrusion[0-*]_alarm
695 Chassis intrusion detection
696 0: OK
697 1: intrusion detected
698 RW
699 Contrary to regular alarm flags which clear themselves
700 automatically when read, this one sticks until cleared by
701 the user. This is done by writing 0 to the file. Writing
702 other values is unsupported.
704 intrusion[0-*]_beep
705 Chassis intrusion beep
706 0: disable
707 1: enable
708 RW
711 sysfs attribute writes interpretation
712 -------------------------------------
714 hwmon sysfs attributes always contain numbers, so the first thing to do is to
715 convert the input to a number, there are 2 ways todo this depending whether
716 the number can be negative or not:
717 unsigned long u = simple_strtoul(buf, NULL, 10);
718 long s = simple_strtol(buf, NULL, 10);
720 With buf being the buffer with the user input being passed by the kernel.
721 Notice that we do not use the second argument of strto[u]l, and thus cannot
722 tell when 0 is returned, if this was really 0 or is caused by invalid input.
723 This is done deliberately as checking this everywhere would add a lot of
724 code to the kernel.
726 Notice that it is important to always store the converted value in an
727 unsigned long or long, so that no wrap around can happen before any further
728 checking.
730 After the input string is converted to an (unsigned) long, the value should be
731 checked if its acceptable. Be careful with further conversions on the value
732 before checking it for validity, as these conversions could still cause a wrap
733 around before the check. For example do not multiply the result, and only
734 add/subtract if it has been divided before the add/subtract.
736 What to do if a value is found to be invalid, depends on the type of the
737 sysfs attribute that is being set. If it is a continuous setting like a
738 tempX_max or inX_max attribute, then the value should be clamped to its
739 limits using clamp_val(value, min_limit, max_limit). If it is not continuous
740 like for example a tempX_type, then when an invalid value is written,
741 -EINVAL should be returned.
743 Example1, temp1_max, register is a signed 8 bit value (-128 - 127 degrees):
745 long v = simple_strtol(buf, NULL, 10) / 1000;
746 v = clamp_val(v, -128, 127);
747 /* write v to register */
749 Example2, fan divider setting, valid values 2, 4 and 8:
751 unsigned long v = simple_strtoul(buf, NULL, 10);
753 switch (v) {
754 case 2: v = 1; break;
755 case 4: v = 2; break;
756 case 8: v = 3; break;
757 default:
758 return -EINVAL;
759 }
760 /* write v to register */