Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 1 | Naming and data format standards for sysfs files |
| 2 | ------------------------------------------------ |
| 3 | |
| 4 | The libsensors library offers an interface to the raw sensors data |
| 5 | through the sysfs interface. See libsensors documentation and source for |
| 6 | more further information. As of writing this document, libsensors |
| 7 | (from lm_sensors 2.8.3) is heavily chip-dependant. Adding or updating |
| 8 | support for any given chip requires modifying the library's code. |
| 9 | This is because libsensors was written for the procfs interface |
| 10 | older kernel modules were using, which wasn't standardized enough. |
| 11 | Recent versions of libsensors (from lm_sensors 2.8.2 and later) have |
| 12 | support for the sysfs interface, though. |
| 13 | |
| 14 | The new sysfs interface was designed to be as chip-independant as |
| 15 | possible. |
| 16 | |
| 17 | Note that motherboards vary widely in the connections to sensor chips. |
| 18 | There is no standard that ensures, for example, that the second |
| 19 | temperature sensor is connected to the CPU, or that the second fan is on |
| 20 | the CPU. Also, some values reported by the chips need some computation |
| 21 | before they make full sense. For example, most chips can only measure |
| 22 | voltages between 0 and +4V. Other voltages are scaled back into that |
| 23 | range using external resistors. Since the values of these resistors |
| 24 | can change from motherboard to motherboard, the conversions cannot be |
| 25 | hard coded into the driver and have to be done in user space. |
| 26 | |
| 27 | For this reason, even if we aim at a chip-independant libsensors, it will |
| 28 | still require a configuration file (e.g. /etc/sensors.conf) for proper |
| 29 | values conversion, labeling of inputs and hiding of unused inputs. |
| 30 | |
| 31 | An alternative method that some programs use is to access the sysfs |
| 32 | files directly. This document briefly describes the standards that the |
| 33 | drivers follow, so that an application program can scan for entries and |
| 34 | access this data in a simple and consistent way. That said, such programs |
| 35 | will have to implement conversion, labeling and hiding of inputs. For |
| 36 | this reason, it is still not recommended to bypass the library. |
| 37 | |
| 38 | If you are developing a userspace application please send us feedback on |
| 39 | this standard. |
| 40 | |
| 41 | Note that this standard isn't completely established yet, so it is subject |
| 42 | to changes, even important ones. One more reason to use the library instead |
| 43 | of accessing sysfs files directly. |
| 44 | |
| 45 | Each chip gets its own directory in the sysfs /sys/devices tree. To |
| 46 | find all sensor chips, it is easier to follow the symlinks from |
| 47 | /sys/i2c/devices/ |
| 48 | |
| 49 | All sysfs values are fixed point numbers. To get the true value of some |
| 50 | of the values, you should divide by the specified value. |
| 51 | |
| 52 | There is only one value per file, unlike the older /proc specification. |
| 53 | The common scheme for files naming is: <type><number>_<item>. Usual |
| 54 | types for sensor chips are "in" (voltage), "temp" (temperature) and |
| 55 | "fan" (fan). Usual items are "input" (measured value), "max" (high |
| 56 | threshold, "min" (low threshold). Numbering usually starts from 1, |
| 57 | except for voltages which start from 0 (because most data sheets use |
| 58 | this). A number is always used for elements that can be present more |
| 59 | than once, even if there is a single element of the given type on the |
| 60 | specific chip. Other files do not refer to a specific element, so |
| 61 | they have a simple name, and no number. |
| 62 | |
| 63 | Alarms are direct indications read from the chips. The drivers do NOT |
| 64 | make comparisons of readings to thresholds. This allows violations |
| 65 | between readings to be caught and alarmed. The exact definition of an |
| 66 | alarm (for example, whether a threshold must be met or must be exceeded |
| 67 | to cause an alarm) is chip-dependent. |
| 68 | |
| 69 | |
| 70 | ------------------------------------------------------------------------- |
| 71 | |
| 72 | ************ |
| 73 | * Voltages * |
| 74 | ************ |
| 75 | |
| 76 | in[0-8]_min Voltage min value. |
| 77 | Unit: millivolt |
| 78 | Read/Write |
| 79 | |
| 80 | in[0-8]_max Voltage max value. |
| 81 | Unit: millivolt |
| 82 | Read/Write |
| 83 | |
| 84 | in[0-8]_input Voltage input value. |
| 85 | Unit: millivolt |
| 86 | Read only |
| 87 | Actual voltage depends on the scaling resistors on the |
| 88 | motherboard, as recommended in the chip datasheet. |
| 89 | This varies by chip and by motherboard. |
| 90 | Because of this variation, values are generally NOT scaled |
| 91 | by the chip driver, and must be done by the application. |
| 92 | However, some drivers (notably lm87 and via686a) |
| 93 | do scale, with various degrees of success. |
| 94 | These drivers will output the actual voltage. |
| 95 | |
| 96 | Typical usage: |
| 97 | in0_* CPU #1 voltage (not scaled) |
| 98 | in1_* CPU #2 voltage (not scaled) |
| 99 | in2_* 3.3V nominal (not scaled) |
| 100 | in3_* 5.0V nominal (scaled) |
| 101 | in4_* 12.0V nominal (scaled) |
| 102 | in5_* -12.0V nominal (scaled) |
| 103 | in6_* -5.0V nominal (scaled) |
| 104 | in7_* varies |
| 105 | in8_* varies |
| 106 | |
| 107 | cpu[0-1]_vid CPU core reference voltage. |
| 108 | Unit: millivolt |
| 109 | Read only. |
| 110 | Not always correct. |
| 111 | |
| 112 | vrm Voltage Regulator Module version number. |
| 113 | Read only. |
| 114 | Two digit number, first is major version, second is |
| 115 | minor version. |
| 116 | Affects the way the driver calculates the CPU core reference |
| 117 | voltage from the vid pins. |
| 118 | |
| 119 | |
| 120 | ******** |
| 121 | * Fans * |
| 122 | ******** |
| 123 | |
| 124 | fan[1-3]_min Fan minimum value |
| 125 | Unit: revolution/min (RPM) |
| 126 | Read/Write. |
| 127 | |
| 128 | fan[1-3]_input Fan input value. |
| 129 | Unit: revolution/min (RPM) |
| 130 | Read only. |
| 131 | |
| 132 | fan[1-3]_div Fan divisor. |
| 133 | Integer value in powers of two (1, 2, 4, 8, 16, 32, 64, 128). |
| 134 | Some chips only support values 1, 2, 4 and 8. |
| 135 | Note that this is actually an internal clock divisor, which |
| 136 | affects the measurable speed range, not the read value. |
| 137 | |
| 138 | ******* |
| 139 | * PWM * |
| 140 | ******* |
| 141 | |
| 142 | pwm[1-3] Pulse width modulation fan control. |
| 143 | Integer value in the range 0 to 255 |
| 144 | Read/Write |
| 145 | 255 is max or 100%. |
| 146 | |
| 147 | pwm[1-3]_enable |
| 148 | Switch PWM on and off. |
| 149 | Not always present even if fan*_pwm is. |
| 150 | 0 to turn off |
| 151 | 1 to turn on in manual mode |
| 152 | 2 to turn on in automatic mode |
| 153 | Read/Write |
| 154 | |
| 155 | pwm[1-*]_auto_channels_temp |
| 156 | Select which temperature channels affect this PWM output in |
| 157 | auto mode. Bitfield, 1 is temp1, 2 is temp2, 4 is temp3 etc... |
| 158 | Which values are possible depend on the chip used. |
| 159 | |
| 160 | pwm[1-*]_auto_point[1-*]_pwm |
| 161 | pwm[1-*]_auto_point[1-*]_temp |
| 162 | pwm[1-*]_auto_point[1-*]_temp_hyst |
| 163 | Define the PWM vs temperature curve. Number of trip points is |
| 164 | chip-dependent. Use this for chips which associate trip points |
| 165 | to PWM output channels. |
| 166 | |
| 167 | OR |
| 168 | |
| 169 | temp[1-*]_auto_point[1-*]_pwm |
| 170 | temp[1-*]_auto_point[1-*]_temp |
| 171 | temp[1-*]_auto_point[1-*]_temp_hyst |
| 172 | Define the PWM vs temperature curve. Number of trip points is |
| 173 | chip-dependent. Use this for chips which associate trip points |
| 174 | to temperature channels. |
| 175 | |
| 176 | |
| 177 | **************** |
| 178 | * Temperatures * |
| 179 | **************** |
| 180 | |
| 181 | temp[1-3]_type Sensor type selection. |
Jean Delvare | e53004e | 2006-01-09 23:26:14 +0100 | [diff] [blame] | 182 | Integers 1 to 4 or thermistor Beta value (typically 3435) |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 183 | Read/Write. |
| 184 | 1: PII/Celeron Diode |
| 185 | 2: 3904 transistor |
| 186 | 3: thermal diode |
Jean Delvare | e53004e | 2006-01-09 23:26:14 +0100 | [diff] [blame] | 187 | 4: thermistor (default/unknown Beta) |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 188 | Not all types are supported by all chips |
| 189 | |
| 190 | temp[1-4]_max Temperature max value. |
| 191 | Unit: millidegree Celcius |
| 192 | Read/Write value. |
| 193 | |
| 194 | temp[1-3]_min Temperature min value. |
| 195 | Unit: millidegree Celcius |
| 196 | Read/Write value. |
| 197 | |
| 198 | temp[1-3]_max_hyst |
| 199 | Temperature hysteresis value for max limit. |
| 200 | Unit: millidegree Celcius |
| 201 | Must be reported as an absolute temperature, NOT a delta |
| 202 | from the max value. |
| 203 | Read/Write value. |
| 204 | |
| 205 | temp[1-4]_input Temperature input value. |
| 206 | Unit: millidegree Celcius |
| 207 | Read only value. |
| 208 | |
| 209 | temp[1-4]_crit Temperature critical value, typically greater than |
| 210 | corresponding temp_max values. |
| 211 | Unit: millidegree Celcius |
| 212 | Read/Write value. |
| 213 | |
| 214 | temp[1-2]_crit_hyst |
| 215 | Temperature hysteresis value for critical limit. |
| 216 | Unit: millidegree Celcius |
| 217 | Must be reported as an absolute temperature, NOT a delta |
| 218 | from the critical value. |
| 219 | Read/Write value. |
| 220 | |
Hartmut Rick | 59ac836 | 2006-03-23 16:37:23 +0100 | [diff] [blame] | 221 | temp[1-4]_offset |
| 222 | Temperature offset which is added to the temperature reading |
| 223 | by the chip. |
| 224 | Unit: millidegree Celsius |
| 225 | Read/Write value. |
| 226 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 227 | If there are multiple temperature sensors, temp1_* is |
| 228 | generally the sensor inside the chip itself, |
| 229 | reported as "motherboard temperature". temp2_* to |
| 230 | temp4_* are generally sensors external to the chip |
| 231 | itself, for example the thermal diode inside the CPU or |
| 232 | a thermistor nearby. |
| 233 | |
| 234 | |
| 235 | ************ |
| 236 | * Currents * |
| 237 | ************ |
| 238 | |
| 239 | Note that no known chip provides current measurements as of writing, |
| 240 | so this part is theoretical, so to say. |
| 241 | |
| 242 | curr[1-n]_max Current max value |
| 243 | Unit: milliampere |
| 244 | Read/Write. |
| 245 | |
| 246 | curr[1-n]_min Current min value. |
| 247 | Unit: milliampere |
| 248 | Read/Write. |
| 249 | |
| 250 | curr[1-n]_input Current input value |
| 251 | Unit: milliampere |
| 252 | Read only. |
| 253 | |
| 254 | |
Jean Delvare | 400b48e | 2006-03-23 16:46:47 +0100 | [diff] [blame^] | 255 | ********** |
| 256 | * Alarms * |
| 257 | ********** |
| 258 | |
| 259 | Each channel or limit may have an associated alarm file, containing a |
| 260 | boolean value. 1 means than an alarm condition exists, 0 means no alarm. |
| 261 | |
| 262 | Usually a given chip will either use channel-related alarms, or |
| 263 | limit-related alarms, not both. The driver should just reflect the hardware |
| 264 | implementation. |
| 265 | |
| 266 | in[0-n]_alarm |
| 267 | fan[1-n]_alarm |
| 268 | temp[1-n]_alarm |
| 269 | Channel alarm |
| 270 | Boolean |
| 271 | Read-only |
| 272 | |
| 273 | OR |
| 274 | |
| 275 | in[0-n]_min_alarm |
| 276 | in[0-n]_max_alarm |
| 277 | fan[1-n]_min_alarm |
| 278 | temp[1-n]_min_alarm |
| 279 | temp[1-n]_max_alarm |
| 280 | temp[1-n]_crit_alarm |
| 281 | Limit alarm |
| 282 | Boolean |
| 283 | Read-only |
| 284 | |
| 285 | Each input channel may have an associated fault file. This can be used |
| 286 | to notify open diodes, unconnected fans etc. where the hardware |
| 287 | supports it. When this boolean has value 1, the measurement for that |
| 288 | channel should not be trusted. |
| 289 | |
| 290 | in[0-n]_input_fault |
| 291 | fan[1-n]_input_fault |
| 292 | temp[1-n]_input_fault |
| 293 | Input fault condition |
| 294 | Boolean |
| 295 | Read-only |
| 296 | |
| 297 | Some chips also offer the possibility to get beeped when an alarm occurs: |
| 298 | |
| 299 | beep_enable Master beep enable |
| 300 | Boolean |
| 301 | Read/Write |
| 302 | |
| 303 | in[0-n]_beep |
| 304 | fan[1-n]_beep |
| 305 | temp[1-n]_beep |
| 306 | Channel beep |
| 307 | 0 to disable. |
| 308 | 1 to enable. |
| 309 | Read/write |
| 310 | |
| 311 | In theory, a chip could provide per-limit beep masking, but no such chip |
| 312 | was seen so far. |
| 313 | |
| 314 | Old drivers provided a different, non-standard interface to alarms and |
| 315 | beeps. These interface files are deprecated, but will be kept around |
| 316 | for compatibility reasons: |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 317 | |
| 318 | alarms Alarm bitmask. |
| 319 | Read only. |
| 320 | Integer representation of one to four bytes. |
| 321 | A '1' bit means an alarm. |
| 322 | Chips should be programmed for 'comparator' mode so that |
| 323 | the alarm will 'come back' after you read the register |
| 324 | if it is still valid. |
| 325 | Generally a direct representation of a chip's internal |
| 326 | alarm registers; there is no standard for the position |
Jean Delvare | 400b48e | 2006-03-23 16:46:47 +0100 | [diff] [blame^] | 327 | of individual bits. For this reason, the use of this |
| 328 | interface file for new drivers is discouraged. Use |
| 329 | individual *_alarm and *_fault files instead. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 330 | Bits are defined in kernel/include/sensors.h. |
| 331 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 332 | beep_mask Bitmask for beep. |
Jean Delvare | 400b48e | 2006-03-23 16:46:47 +0100 | [diff] [blame^] | 333 | Same format as 'alarms' with the same bit locations, |
| 334 | use discouraged for the same reason. Use individual |
| 335 | *_beep files instead. |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 336 | Read/Write |
| 337 | |
Jean Delvare | 400b48e | 2006-03-23 16:46:47 +0100 | [diff] [blame^] | 338 | |
| 339 | ********* |
| 340 | * Other * |
| 341 | ********* |
| 342 | |
Linus Torvalds | 1da177e | 2005-04-16 15:20:36 -0700 | [diff] [blame] | 343 | eeprom Raw EEPROM data in binary form. |
| 344 | Read only. |
Jean Delvare | c3df580 | 2005-10-26 21:39:40 +0200 | [diff] [blame] | 345 | |
| 346 | pec Enable or disable PEC (SMBus only) |
| 347 | Read/Write |