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 spaces nor dashes, representing the chip name. This is
90 the only mandatory attribute.
91 I2C devices get this attribute created automatically.
94 update_interval The interval at which the chip will update readings.
97 Some devices have a variable update rate or interval.
98 This attribute can be used to change it to the desired value.
105 in[0-*]_min Voltage min value.
109 in[0-*]_lcrit Voltage critical min value.
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.
120 in[0-*]_crit Voltage critical max value.
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.
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
142 in[0-*]_label Suggested voltage channel label.
144 Should only be created if the driver has hints about what
145 this voltage channel is being used for, and user-space
146 doesn't. In all other cases, the label is provided by
150 cpu[0-*]_vid CPU core reference voltage.
155 vrm Voltage Regulator Module version number.
156 RW (but changing it should no more be necessary)
157 Originally the VRM standard version multiplied by 10, but now
158 an arbitrary number, as not all standards have a version
160 Affects the way the driver calculates the CPU core reference
161 voltage from the vid pins.
163 Also see the Alarms section for status flags associated with voltages.
170 fan[1-*]_min Fan minimum value
171 Unit: revolution/min (RPM)
174 fan[1-*]_max Fan maximum value
175 Unit: revolution/min (RPM)
176 Only rarely supported by the hardware.
179 fan[1-*]_input Fan input value.
180 Unit: revolution/min (RPM)
183 fan[1-*]_div Fan divisor.
184 Integer value in powers of two (1, 2, 4, 8, 16, 32, 64, 128).
186 Some chips only support values 1, 2, 4 and 8.
187 Note that this is actually an internal clock divisor, which
188 affects the measurable speed range, not the read value.
192 Unit: revolution/min (RPM)
194 Only makes sense if the chip supports closed-loop fan speed
195 control based on the measured fan speed.
197 fan[1-*]_label Suggested fan channel label.
199 Should only be created if the driver has hints about what
200 this fan channel is being used for, and user-space doesn't.
201 In all other cases, the label is provided by user-space.
204 Also see the Alarms section for status flags associated with fans.
211 pwm[1-*] Pulse width modulation fan control.
212 Integer value in the range 0 to 255
217 Fan speed control method:
218 0: no fan speed control (i.e. fan at full speed)
219 1: manual fan speed control enabled (using pwm[1-*])
220 2+: automatic fan speed control enabled
221 Check individual chip documentation files for automatic mode
225 pwm[1-*]_mode 0: DC mode (direct current)
226 1: PWM mode (pulse-width modulation)
229 pwm[1-*]_freq Base PWM frequency in Hz.
230 Only possibly available when pwmN_mode is PWM, but not always
234 pwm[1-*]_auto_channels_temp
235 Select which temperature channels affect this PWM output in
236 auto mode. Bitfield, 1 is temp1, 2 is temp2, 4 is temp3 etc...
237 Which values are possible depend on the chip used.
240 pwm[1-*]_auto_point[1-*]_pwm
241 pwm[1-*]_auto_point[1-*]_temp
242 pwm[1-*]_auto_point[1-*]_temp_hyst
243 Define the PWM vs temperature curve. Number of trip points is
244 chip-dependent. Use this for chips which associate trip points
245 to PWM output channels.
248 temp[1-*]_auto_point[1-*]_pwm
249 temp[1-*]_auto_point[1-*]_temp
250 temp[1-*]_auto_point[1-*]_temp_hyst
251 Define the PWM vs temperature curve. Number of trip points is
252 chip-dependent. Use this for chips which associate trip points
253 to temperature channels.
256 There is a third case where trip points are associated to both PWM output
257 channels and temperature channels: the PWM values are associated to PWM
258 output channels while the temperature values are associated to temperature
259 channels. In that case, the result is determined by the mapping between
260 temperature inputs and PWM outputs. When several temperature inputs are
261 mapped to a given PWM output, this leads to several candidate PWM values.
262 The actual result is up to the chip, but in general the highest candidate
263 value (fastest fan speed) wins.
270 temp[1-*]_type Sensor type selection.
279 Not all types are supported by all chips
281 temp[1-*]_max Temperature max value.
282 Unit: millidegree Celsius (or millivolt, see below)
285 temp[1-*]_min Temperature min value.
286 Unit: millidegree Celsius
290 Temperature hysteresis value for max limit.
291 Unit: millidegree Celsius
292 Must be reported as an absolute temperature, NOT a delta
296 temp[1-*]_input Temperature input value.
297 Unit: millidegree Celsius
300 temp[1-*]_crit Temperature critical max value, typically greater than
301 corresponding temp_max values.
302 Unit: millidegree Celsius
306 Temperature hysteresis value for critical limit.
307 Unit: millidegree Celsius
308 Must be reported as an absolute temperature, NOT a delta
309 from the critical value.
312 temp[1-*]_lcrit Temperature critical min value, typically lower than
313 corresponding temp_min values.
314 Unit: millidegree Celsius
318 Temperature offset which is added to the temperature reading
320 Unit: millidegree Celsius
323 temp[1-*]_label Suggested temperature channel label.
325 Should only be created if the driver has hints about what
326 this temperature channel is being used for, and user-space
327 doesn't. In all other cases, the label is provided by
332 Historical minimum temperature
333 Unit: millidegree Celsius
337 Historical maximum temperature
338 Unit: millidegree Celsius
341 temp[1-*]_reset_history
342 Reset temp_lowest and temp_highest
346 Reset temp_lowest and temp_highest for all sensors
349 Some chips measure temperature using external thermistors and an ADC, and
350 report the temperature measurement as a voltage. Converting this voltage
351 back to a temperature (or the other way around for limits) requires
352 mathematical functions not available in the kernel, so the conversion
353 must occur in user space. For these chips, all temp* files described
354 above should contain values expressed in millivolt instead of millidegree
355 Celsius. In other words, such temperature channels are handled as voltage
356 channels by the driver.
358 Also see the Alarms section for status flags associated with temperatures.
365 curr[1-*]_max Current max value
369 curr[1-*]_min Current min value.
373 curr[1-*]_input Current input value
381 power[1-*]_average Average power use
385 power[1-*]_average_interval Power use averaging interval. A poll
386 notification is sent to this file if the
387 hardware changes the averaging interval.
391 power[1-*]_average_interval_max Maximum power use averaging interval
395 power[1-*]_average_interval_min Minimum power use averaging interval
399 power[1-*]_average_highest Historical average maximum power use
403 power[1-*]_average_lowest Historical average minimum power use
407 power[1-*]_average_max A poll notification is sent to
408 power[1-*]_average when power use
409 rises above this value.
413 power[1-*]_average_min A poll notification is sent to
414 power[1-*]_average when power use
415 sinks below this value.
419 power[1-*]_input Instantaneous power use
423 power[1-*]_input_highest Historical maximum power use
427 power[1-*]_input_lowest Historical minimum power use
431 power[1-*]_reset_history Reset input_highest, input_lowest,
432 average_highest and average_lowest.
435 power[1-*]_accuracy Accuracy of the power meter.
439 power[1-*]_alarm 1 if the system is drawing more power than the
440 cap allows; 0 otherwise. A poll notification is
441 sent to this file when the power use exceeds the
442 cap. This file only appears if the cap is known
443 to be enforced by hardware.
446 power[1-*]_cap If power use rises above this limit, the
447 system should take action to reduce power use.
448 A poll notification is sent to this file if the
449 cap is changed by the hardware. The *_cap
450 files only appear if the cap is known to be
451 enforced by hardware.
455 power[1-*]_cap_hyst Margin of hysteresis built around capping and
460 power[1-*]_cap_max Maximum cap that can be set.
464 power[1-*]_cap_min Minimum cap that can be set.
472 energy[1-*]_input Cumulative energy use
481 Each channel or limit may have an associated alarm file, containing a
482 boolean value. 1 means than an alarm condition exists, 0 means no alarm.
484 Usually a given chip will either use channel-related alarms, or
485 limit-related alarms, not both. The driver should just reflect the hardware
513 Each input channel may have an associated fault file. This can be used
514 to notify open diodes, unconnected fans etc. where the hardware
515 supports it. When this boolean has value 1, the measurement for that
516 channel should not be trusted.
520 Input fault condition
525 Some chips also offer the possibility to get beeped when an alarm occurs:
527 beep_enable Master beep enable
541 In theory, a chip could provide per-limit beep masking, but no such chip
544 Old drivers provided a different, non-standard interface to alarms and
545 beeps. These interface files are deprecated, but will be kept around
546 for compatibility reasons:
548 alarms Alarm bitmask.
550 Integer representation of one to four bytes.
551 A '1' bit means an alarm.
552 Chips should be programmed for 'comparator' mode so that
553 the alarm will 'come back' after you read the register
554 if it is still valid.
555 Generally a direct representation of a chip's internal
556 alarm registers; there is no standard for the position
557 of individual bits. For this reason, the use of this
558 interface file for new drivers is discouraged. Use
559 individual *_alarm and *_fault files instead.
560 Bits are defined in kernel/include/sensors.h.
562 beep_mask Bitmask for beep.
563 Same format as 'alarms' with the same bit locations,
564 use discouraged for the same reason. Use individual
565 *_beep files instead.
569 ***********************
570 * Intrusion detection *
571 ***********************
574 Chassis intrusion detection
576 1: intrusion detected
578 Contrary to regular alarm flags which clear themselves
579 automatically when read, this one sticks until cleared by
580 the user. This is done by writing 0 to the file. Writing
581 other values is unsupported.
584 Chassis intrusion beep
590 sysfs attribute writes interpretation
591 -------------------------------------
593 hwmon sysfs attributes always contain numbers, so the first thing to do is to
594 convert the input to a number, there are 2 ways todo this depending whether
595 the number can be negative or not:
596 unsigned long u = simple_strtoul(buf, NULL, 10);
597 long s = simple_strtol(buf, NULL, 10);
599 With buf being the buffer with the user input being passed by the kernel.
600 Notice that we do not use the second argument of strto[u]l, and thus cannot
601 tell when 0 is returned, if this was really 0 or is caused by invalid input.
602 This is done deliberately as checking this everywhere would add a lot of
605 Notice that it is important to always store the converted value in an
606 unsigned long or long, so that no wrap around can happen before any further
609 After the input string is converted to an (unsigned) long, the value should be
610 checked if its acceptable. Be careful with further conversions on the value
611 before checking it for validity, as these conversions could still cause a wrap
612 around before the check. For example do not multiply the result, and only
613 add/subtract if it has been divided before the add/subtract.
615 What to do if a value is found to be invalid, depends on the type of the
616 sysfs attribute that is being set. If it is a continuous setting like a
617 tempX_max or inX_max attribute, then the value should be clamped to its
618 limits using SENSORS_LIMIT(value, min_limit, max_limit). If it is not
619 continuous like for example a tempX_type, then when an invalid value is
620 written, -EINVAL should be returned.
622 Example1, temp1_max, register is a signed 8 bit value (-128 - 127 degrees):
624 long v = simple_strtol(buf, NULL, 10) / 1000;
625 v = SENSORS_LIMIT(v, -128, 127);
626 /* write v to register */
628 Example2, fan divider setting, valid values 2, 4 and 8:
630 unsigned long v = simple_strtoul(buf, NULL, 10);
633 case 2: v = 1; break;
634 case 4: v = 2; break;
635 case 8: v = 3; break;
639 /* write v to register */