| 1 | Kernel driver w83781d |
| 2 | ===================== |
| 3 | |
| 4 | Supported chips: |
| 5 | * Winbond W83781D |
| 6 | Prefix: 'w83781d' |
| 7 | Addresses scanned: I2C 0x28 - 0x2f, ISA 0x290 (8 I/O ports) |
| 8 | Datasheet: http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83781d.pdf |
| 9 | * Winbond W83782D |
| 10 | Prefix: 'w83782d' |
| 11 | Addresses scanned: I2C 0x28 - 0x2f, ISA 0x290 (8 I/O ports) |
| 12 | Datasheet: http://www.winbond.com/PDF/sheet/w83782d.pdf |
| 13 | * Winbond W83783S |
| 14 | Prefix: 'w83783s' |
| 15 | Addresses scanned: I2C 0x2d |
| 16 | Datasheet: http://www.winbond-usa.com/products/winbond_products/pdfs/PCIC/w83783s.pdf |
| 17 | * Asus AS99127F |
| 18 | Prefix: 'as99127f' |
| 19 | Addresses scanned: I2C 0x28 - 0x2f |
| 20 | Datasheet: Unavailable from Asus |
| 21 | |
| 22 | Authors: |
| 23 | Frodo Looijaard <frodol@dds.nl>, |
| 24 | Philip Edelbrock <phil@netroedge.com>, |
| 25 | Mark Studebaker <mdsxyz123@yahoo.com> |
| 26 | |
| 27 | Module parameters |
| 28 | ----------------- |
| 29 | |
| 30 | * init int |
| 31 | (default 1) |
| 32 | Use 'init=0' to bypass initializing the chip. |
| 33 | Try this if your computer crashes when you load the module. |
| 34 | |
| 35 | * reset int |
| 36 | (default 0) |
| 37 | The driver used to reset the chip on load, but does no more. Use |
| 38 | 'reset=1' to restore the old behavior. Report if you need to do this. |
| 39 | |
| 40 | force_subclients=bus,caddr,saddr,saddr |
| 41 | This is used to force the i2c addresses for subclients of |
| 42 | a certain chip. Typical usage is `force_subclients=0,0x2d,0x4a,0x4b' |
| 43 | to force the subclients of chip 0x2d on bus 0 to i2c addresses |
| 44 | 0x4a and 0x4b. This parameter is useful for certain Tyan boards. |
| 45 | |
| 46 | Description |
| 47 | ----------- |
| 48 | |
| 49 | This driver implements support for the Winbond W83781D, W83782D, W83783S |
| 50 | chips, and the Asus AS99127F chips. We will refer to them collectively as |
| 51 | W8378* chips. |
| 52 | |
| 53 | There is quite some difference between these chips, but they are similar |
| 54 | enough that it was sensible to put them together in one driver. |
| 55 | The Asus chips are similar to an I2C-only W83782D. |
| 56 | |
| 57 | Chip #vin #fanin #pwm #temp wchipid vendid i2c ISA |
| 58 | as99127f 7 3 0 3 0x31 0x12c3 yes no |
| 59 | as99127f rev.2 (type_name = as99127f) 0x31 0x5ca3 yes no |
| 60 | w83781d 7 3 0 3 0x10-1 0x5ca3 yes yes |
| 61 | w83782d 9 3 2-4 3 0x30 0x5ca3 yes yes |
| 62 | w83783s 5-6 3 2 1-2 0x40 0x5ca3 yes no |
| 63 | |
| 64 | Detection of these chips can sometimes be foiled because they can be in |
| 65 | an internal state that allows no clean access. If you know the address |
| 66 | of the chip, use a 'force' parameter; this will put them into a more |
| 67 | well-behaved state first. |
| 68 | |
| 69 | The W8378* implements temperature sensors (three on the W83781D and W83782D, |
| 70 | two on the W83783S), three fan rotation speed sensors, voltage sensors |
| 71 | (seven on the W83781D, nine on the W83782D and six on the W83783S), VID |
| 72 | lines, alarms with beep warnings, and some miscellaneous stuff. |
| 73 | |
| 74 | Temperatures are measured in degrees Celsius. There is always one main |
| 75 | temperature sensor, and one (W83783S) or two (W83781D and W83782D) other |
| 76 | sensors. An alarm is triggered for the main sensor once when the |
| 77 | Overtemperature Shutdown limit is crossed; it is triggered again as soon as |
| 78 | it drops below the Hysteresis value. A more useful behavior |
| 79 | can be found by setting the Hysteresis value to +127 degrees Celsius; in |
| 80 | this case, alarms are issued during all the time when the actual temperature |
| 81 | is above the Overtemperature Shutdown value. The driver sets the |
| 82 | hysteresis value for temp1 to 127 at initialization. |
| 83 | |
| 84 | For the other temperature sensor(s), an alarm is triggered when the |
| 85 | temperature gets higher then the Overtemperature Shutdown value; it stays |
| 86 | on until the temperature falls below the Hysteresis value. But on the |
| 87 | W83781D, there is only one alarm that functions for both other sensors! |
| 88 | Temperatures are guaranteed within a range of -55 to +125 degrees. The |
| 89 | main temperature sensors has a resolution of 1 degree; the other sensor(s) |
| 90 | of 0.5 degree. |
| 91 | |
| 92 | Fan rotation speeds are reported in RPM (rotations per minute). An alarm is |
| 93 | triggered if the rotation speed has dropped below a programmable limit. Fan |
| 94 | readings can be divided by a programmable divider (1, 2, 4 or 8 for the |
| 95 | W83781D; 1, 2, 4, 8, 16, 32, 64 or 128 for the others) to give |
| 96 | the readings more range or accuracy. Not all RPM values can accurately |
| 97 | be represented, so some rounding is done. With a divider of 2, the lowest |
| 98 | representable value is around 2600 RPM. |
| 99 | |
| 100 | Voltage sensors (also known as IN sensors) report their values in volts. |
| 101 | An alarm is triggered if the voltage has crossed a programmable minimum |
| 102 | or maximum limit. Note that minimum in this case always means 'closest to |
| 103 | zero'; this is important for negative voltage measurements. All voltage |
| 104 | inputs can measure voltages between 0 and 4.08 volts, with a resolution |
| 105 | of 0.016 volt. |
| 106 | |
| 107 | The VID lines encode the core voltage value: the voltage level your processor |
| 108 | should work with. This is hardcoded by the mainboard and/or processor itself. |
| 109 | It is a value in volts. When it is unconnected, you will often find the |
| 110 | value 3.50 V here. |
| 111 | |
| 112 | The W83782D and W83783S temperature conversion machine understands about |
| 113 | several kinds of temperature probes. You can program the so-called |
| 114 | beta value in the sensor files. '1' is the PII/Celeron diode, '2' is the |
| 115 | TN3904 transistor, and 3435 the default thermistor value. Other values |
| 116 | are (not yet) supported. |
| 117 | |
| 118 | In addition to the alarms described above, there is a CHAS alarm on the |
| 119 | chips which triggers if your computer case is open. |
| 120 | |
| 121 | When an alarm goes off, you can be warned by a beeping signal through |
| 122 | your computer speaker. It is possible to enable all beeping globally, |
| 123 | or only the beeping for some alarms. |
| 124 | |
| 125 | Individual alarm and beep bits: |
| 126 | |
| 127 | 0x000001: in0 |
| 128 | 0x000002: in1 |
| 129 | 0x000004: in2 |
| 130 | 0x000008: in3 |
| 131 | 0x000010: temp1 |
| 132 | 0x000020: temp2 (+temp3 on W83781D) |
| 133 | 0x000040: fan1 |
| 134 | 0x000080: fan2 |
| 135 | 0x000100: in4 |
| 136 | 0x000200: in5 |
| 137 | 0x000400: in6 |
| 138 | 0x000800: fan3 |
| 139 | 0x001000: chassis |
| 140 | 0x002000: temp3 (W83782D only) |
| 141 | 0x010000: in7 (W83782D only) |
| 142 | 0x020000: in8 (W83782D only) |
| 143 | |
| 144 | If an alarm triggers, it will remain triggered until the hardware register |
| 145 | is read at least once. This means that the cause for the alarm may |
| 146 | already have disappeared! Note that in the current implementation, all |
| 147 | hardware registers are read whenever any data is read (unless it is less |
| 148 | than 1.5 seconds since the last update). This means that you can easily |
| 149 | miss once-only alarms. |
| 150 | |
| 151 | The chips only update values each 1.5 seconds; reading them more often |
| 152 | will do no harm, but will return 'old' values. |
| 153 | |
| 154 | AS99127F PROBLEMS |
| 155 | ----------------- |
| 156 | The as99127f support was developed without the benefit of a datasheet. |
| 157 | In most cases it is treated as a w83781d (although revision 2 of the |
| 158 | AS99127F looks more like a w83782d). |
| 159 | This support will be BETA until a datasheet is released. |
| 160 | One user has reported problems with fans stopping |
| 161 | occasionally. |
| 162 | |
| 163 | Note that the individual beep bits are inverted from the other chips. |
| 164 | The driver now takes care of this so that user-space applications |
| 165 | don't have to know about it. |
| 166 | |
| 167 | Known problems: |
| 168 | - Problems with diode/thermistor settings (supported?) |
| 169 | - One user reports fans stopping under high server load. |
| 170 | - Revision 2 seems to have 2 PWM registers but we don't know |
| 171 | how to handle them. More details below. |
| 172 | |
| 173 | These will not be fixed unless we get a datasheet. |
| 174 | If you have problems, please lobby Asus to release a datasheet. |
| 175 | Unfortunately several others have without success. |
| 176 | Please do not send mail to us asking for better as99127f support. |
| 177 | We have done the best we can without a datasheet. |
| 178 | Please do not send mail to the author or the sensors group asking for |
| 179 | a datasheet or ideas on how to convince Asus. We can't help. |
| 180 | |
| 181 | |
| 182 | NOTES: |
| 183 | ----- |
| 184 | 783s has no in1 so that in[2-6] are compatible with the 781d/782d. |
| 185 | |
| 186 | 783s pin is programmable for -5V or temp1; defaults to -5V, |
| 187 | no control in driver so temp1 doesn't work. |
| 188 | |
| 189 | 782d and 783s datasheets differ on which is pwm1 and which is pwm2. |
| 190 | We chose to follow 782d. |
| 191 | |
| 192 | 782d and 783s pin is programmable for fan3 input or pwm2 output; |
| 193 | defaults to fan3 input. |
| 194 | If pwm2 is enabled (with echo 255 1 > pwm2), then |
| 195 | fan3 will report 0. |
| 196 | |
| 197 | 782d has pwm1-2 for ISA, pwm1-4 for i2c. (pwm3-4 share pins with |
| 198 | the ISA pins) |
| 199 | |
| 200 | Data sheet updates: |
| 201 | ------------------ |
| 202 | - PWM clock registers: |
| 203 | |
| 204 | 000: master / 512 |
| 205 | 001: master / 1024 |
| 206 | 010: master / 2048 |
| 207 | 011: master / 4096 |
| 208 | 100: master / 8192 |
| 209 | |
| 210 | |
| 211 | Answers from Winbond tech support |
| 212 | --------------------------------- |
| 213 | > |
| 214 | > 1) In the W83781D data sheet section 7.2 last paragraph, it talks about |
| 215 | > reprogramming the R-T table if the Beta of the thermistor is not |
| 216 | > 3435K. The R-T table is described briefly in section 8.20. |
| 217 | > What formulas do I use to program a new R-T table for a given Beta? |
| 218 | > |
| 219 | We are sorry that the calculation for R-T table value is |
| 220 | confidential. If you have another Beta value of thermistor, we can help |
| 221 | to calculate the R-T table for you. But you should give us real R-T |
| 222 | Table which can be gotten by thermistor vendor. Therefore we will calculate |
| 223 | them and obtain 32-byte data, and you can fill the 32-byte data to the |
| 224 | register in Bank0.CR51 of W83781D. |
| 225 | |
| 226 | |
| 227 | > 2) In the W83782D data sheet, it mentions that pins 38, 39, and 40 are |
| 228 | > programmable to be either thermistor or Pentium II diode inputs. |
| 229 | > How do I program them for diode inputs? I can't find any register |
| 230 | > to program these to be diode inputs. |
| 231 | --> You may program Bank0 CR[5Dh] and CR[59h] registers. |
| 232 | |
| 233 | CR[5Dh] bit 1(VTIN1) bit 2(VTIN2) bit 3(VTIN3) |
| 234 | |
| 235 | thermistor 0 0 0 |
| 236 | diode 1 1 1 |
| 237 | |
| 238 | |
| 239 | (error) CR[59h] bit 4(VTIN1) bit 2(VTIN2) bit 3(VTIN3) |
| 240 | (right) CR[59h] bit 4(VTIN1) bit 5(VTIN2) bit 6(VTIN3) |
| 241 | |
| 242 | PII thermal diode 1 1 1 |
| 243 | 2N3904 diode 0 0 0 |
| 244 | |
| 245 | |
| 246 | Asus Clones |
| 247 | ----------- |
| 248 | |
| 249 | We have no datasheets for the Asus clones (AS99127F and ASB100 Bach). |
| 250 | Here are some very useful information that were given to us by Alex Van |
| 251 | Kaam about how to detect these chips, and how to read their values. He |
| 252 | also gives advice for another Asus chipset, the Mozart-2 (which we |
| 253 | don't support yet). Thanks Alex! |
| 254 | I reworded some parts and added personal comments. |
| 255 | |
| 256 | # Detection: |
| 257 | |
| 258 | AS99127F rev.1, AS99127F rev.2 and ASB100: |
| 259 | - I2C address range: 0x29 - 0x2F |
| 260 | - If register 0x58 holds 0x31 then we have an Asus (either ASB100 or |
| 261 | AS99127F) |
| 262 | - Which one depends on register 0x4F (manufacturer ID): |
| 263 | 0x06 or 0x94: ASB100 |
| 264 | 0x12 or 0xC3: AS99127F rev.1 |
| 265 | 0x5C or 0xA3: AS99127F rev.2 |
| 266 | Note that 0x5CA3 is Winbond's ID (WEC), which let us think Asus get their |
| 267 | AS99127F rev.2 direct from Winbond. The other codes mean ATT and DVC, |
| 268 | respectively. ATT could stand for Asustek something (although it would be |
| 269 | very badly chosen IMHO), I don't know what DVC could stand for. Maybe |
| 270 | these codes simply aren't meant to be decoded that way. |
| 271 | |
| 272 | Mozart-2: |
| 273 | - I2C address: 0x77 |
| 274 | - If register 0x58 holds 0x56 or 0x10 then we have a Mozart-2 |
| 275 | - Of the Mozart there are 3 types: |
| 276 | 0x58=0x56, 0x4E=0x94, 0x4F=0x36: Asus ASM58 Mozart-2 |
| 277 | 0x58=0x56, 0x4E=0x94, 0x4F=0x06: Asus AS2K129R Mozart-2 |
| 278 | 0x58=0x10, 0x4E=0x5C, 0x4F=0xA3: Asus ??? Mozart-2 |
| 279 | You can handle all 3 the exact same way :) |
| 280 | |
| 281 | # Temperature sensors: |
| 282 | |
| 283 | ASB100: |
| 284 | - sensor 1: register 0x27 |
| 285 | - sensor 2 & 3 are the 2 LM75's on the SMBus |
| 286 | - sensor 4: register 0x17 |
| 287 | Remark: I noticed that on Intel boards sensor 2 is used for the CPU |
| 288 | and 4 is ignored/stuck, on AMD boards sensor 4 is the CPU and sensor 2 is |
| 289 | either ignored or a socket temperature. |
| 290 | |
| 291 | AS99127F (rev.1 and 2 alike): |
| 292 | - sensor 1: register 0x27 |
| 293 | - sensor 2 & 3 are the 2 LM75's on the SMBus |
| 294 | Remark: Register 0x5b is suspected to be temperature type selector. Bit 1 |
| 295 | would control temp1, bit 3 temp2 and bit 5 temp3. |
| 296 | |
| 297 | Mozart-2: |
| 298 | - sensor 1: register 0x27 |
| 299 | - sensor 2: register 0x13 |
| 300 | |
| 301 | # Fan sensors: |
| 302 | |
| 303 | ASB100, AS99127F (rev.1 and 2 alike): |
| 304 | - 3 fans, identical to the W83781D |
| 305 | |
| 306 | Mozart-2: |
| 307 | - 2 fans only, 1350000/RPM/div |
| 308 | - fan 1: register 0x28, divisor on register 0xA1 (bits 4-5) |
| 309 | - fan 2: register 0x29, divisor on register 0xA1 (bits 6-7) |
| 310 | |
| 311 | # Voltages: |
| 312 | |
| 313 | This is where there is a difference between AS99127F rev.1 and 2. |
| 314 | Remark: The difference is similar to the difference between |
| 315 | W83781D and W83782D. |
| 316 | |
| 317 | ASB100: |
| 318 | in0=r(0x20)*0.016 |
| 319 | in1=r(0x21)*0.016 |
| 320 | in2=r(0x22)*0.016 |
| 321 | in3=r(0x23)*0.016*1.68 |
| 322 | in4=r(0x24)*0.016*3.8 |
| 323 | in5=r(0x25)*(-0.016)*3.97 |
| 324 | in6=r(0x26)*(-0.016)*1.666 |
| 325 | |
| 326 | AS99127F rev.1: |
| 327 | in0=r(0x20)*0.016 |
| 328 | in1=r(0x21)*0.016 |
| 329 | in2=r(0x22)*0.016 |
| 330 | in3=r(0x23)*0.016*1.68 |
| 331 | in4=r(0x24)*0.016*3.8 |
| 332 | in5=r(0x25)*(-0.016)*3.97 |
| 333 | in6=r(0x26)*(-0.016)*1.503 |
| 334 | |
| 335 | AS99127F rev.2: |
| 336 | in0=r(0x20)*0.016 |
| 337 | in1=r(0x21)*0.016 |
| 338 | in2=r(0x22)*0.016 |
| 339 | in3=r(0x23)*0.016*1.68 |
| 340 | in4=r(0x24)*0.016*3.8 |
| 341 | in5=(r(0x25)*0.016-3.6)*5.14+3.6 |
| 342 | in6=(r(0x26)*0.016-3.6)*3.14+3.6 |
| 343 | |
| 344 | Mozart-2: |
| 345 | in0=r(0x20)*0.016 |
| 346 | in1=255 |
| 347 | in2=r(0x22)*0.016 |
| 348 | in3=r(0x23)*0.016*1.68 |
| 349 | in4=r(0x24)*0.016*4 |
| 350 | in5=255 |
| 351 | in6=255 |
| 352 | |
| 353 | |
| 354 | # PWM |
| 355 | |
| 356 | * Additional info about PWM on the AS99127F (may apply to other Asus |
| 357 | chips as well) by Jean Delvare as of 2004-04-09: |
| 358 | |
| 359 | AS99127F revision 2 seems to have two PWM registers at 0x59 and 0x5A, |
| 360 | and a temperature sensor type selector at 0x5B (which basically means |
| 361 | that they swapped registers 0x59 and 0x5B when you compare with Winbond |
| 362 | chips). |
| 363 | Revision 1 of the chip also has the temperature sensor type selector at |
| 364 | 0x5B, but PWM registers have no effect. |
| 365 | |
| 366 | We don't know exactly how the temperature sensor type selection works. |
| 367 | Looks like bits 1-0 are for temp1, bits 3-2 for temp2 and bits 5-4 for |
| 368 | temp3, although it is possible that only the most significant bit matters |
| 369 | each time. So far, values other than 0 always broke the readings. |
| 370 | |
| 371 | PWM registers seem to be split in two parts: bit 7 is a mode selector, |
| 372 | while the other bits seem to define a value or threshold. |
| 373 | |
| 374 | When bit 7 is clear, bits 6-0 seem to hold a threshold value. If the value |
| 375 | is below a given limit, the fan runs at low speed. If the value is above |
| 376 | the limit, the fan runs at full speed. We have no clue as to what the limit |
| 377 | represents. Note that there seem to be some inertia in this mode, speed |
| 378 | changes may need some time to trigger. Also, an hysteresis mechanism is |
| 379 | suspected since walking through all the values increasingly and then |
| 380 | decreasingly led to slightly different limits. |
| 381 | |
| 382 | When bit 7 is set, bits 3-0 seem to hold a threshold value, while bits 6-4 |
| 383 | would not be significant. If the value is below a given limit, the fan runs |
| 384 | at full speed, while if it is above the limit it runs at low speed (so this |
| 385 | is the contrary of the other mode, in a way). Here again, we don't know |
| 386 | what the limit is supposed to represent. |
| 387 | |
| 388 | One remarkable thing is that the fans would only have two or three |
| 389 | different speeds (transitional states left apart), not a whole range as |
| 390 | you usually get with PWM. |
| 391 | |
| 392 | As a conclusion, you can write 0x00 or 0x8F to the PWM registers to make |
| 393 | fans run at low speed, and 0x7F or 0x80 to make them run at full speed. |
| 394 | |
| 395 | Please contact us if you can figure out how it is supposed to work. As |
| 396 | long as we don't know more, the w83781d driver doesn't handle PWM on |
| 397 | AS99127F chips at all. |
| 398 | |
| 399 | * Additional info about PWM on the AS99127F rev.1 by Hector Martin: |
| 400 | |
| 401 | I've been fiddling around with the (in)famous 0x59 register and |
| 402 | found out the following values do work as a form of coarse pwm: |
| 403 | |
| 404 | 0x80 - seems to turn fans off after some time(1-2 minutes)... might be |
| 405 | some form of auto-fan-control based on temp? hmm (Qfan? this mobo is an |
| 406 | old ASUS, it isn't marketed as Qfan. Maybe some beta pre-attemp at Qfan |
| 407 | that was dropped at the BIOS) |
| 408 | 0x81 - off |
| 409 | 0x82 - slightly "on-ner" than off, but my fans do not get to move. I can |
| 410 | hear the high-pitched PWM sound that motors give off at too-low-pwm. |
| 411 | 0x83 - now they do move. Estimate about 70% speed or so. |
| 412 | 0x84-0x8f - full on |
| 413 | |
| 414 | Changing the high nibble doesn't seem to do much except the high bit |
| 415 | (0x80) must be set for PWM to work, else the current pwm doesn't seem to |
| 416 | change. |
| 417 | |
| 418 | My mobo is an ASUS A7V266-E. This behavior is similar to what I got |
| 419 | with speedfan under Windows, where 0-15% would be off, 15-2x% (can't |
| 420 | remember the exact value) would be 70% and higher would be full on. |
| 421 | |
| 422 | * Additional info about PWM on the AS99127F rev.1 from lm-sensors |
| 423 | ticket #2350: |
| 424 | |
| 425 | I conducted some experiment on Asus P3B-F motherboard with AS99127F |
| 426 | (Ver. 1). |
| 427 | |
| 428 | I confirm that 0x59 register control the CPU_Fan Header on this |
| 429 | motherboard, and 0x5a register control PWR_Fan. |
| 430 | |
| 431 | In order to reduce the dependency of specific fan, the measurement is |
| 432 | conducted with a digital scope without fan connected. I found out that |
| 433 | P3B-F actually output variable DC voltage on fan header center pin, |
| 434 | looks like PWM is filtered on this motherboard. |
| 435 | |
| 436 | Here are some of measurements: |
| 437 | |
| 438 | 0x80 20 mV |
| 439 | 0x81 20 mV |
| 440 | 0x82 232 mV |
| 441 | 0x83 1.2 V |
| 442 | 0x84 2.31 V |
| 443 | 0x85 3.44 V |
| 444 | 0x86 4.62 V |
| 445 | 0x87 5.81 V |
| 446 | 0x88 7.01 V |
| 447 | 9x89 8.22 V |
| 448 | 0x8a 9.42 V |
| 449 | 0x8b 10.6 V |
| 450 | 0x8c 11.9 V |
| 451 | 0x8d 12.4 V |
| 452 | 0x8e 12.4 V |
| 453 | 0x8f 12.4 V |