| 1 | Sensors FAQ for lm_sensors version 2 |
|---|
| 2 | ------------------------------------ |
|---|
| 3 | |
|---|
| 4 | [Internal comments/questions/uncertainties are in square brackets.] |
|---|
| 5 | |
|---|
| 6 | Note: some questions were specific for version 1. These are put together |
|---|
| 7 | under chapter 6: version 1 questions. |
|---|
| 8 | |
|---|
| 9 | 1 What sensors are available on my PC? |
|---|
| 10 | 1.1 What can a sensor chip like the "LM78" do? |
|---|
| 11 | 1.2 Where do I find out more about any of these LMxx chips? |
|---|
| 12 | |
|---|
| 13 | 2 How are these sensors read? |
|---|
| 14 | 2.1 What is the SMBus? And the I2C bus? |
|---|
| 15 | 2.2 What sensors does the Pentium 2 (and P6) have? |
|---|
| 16 | 2.3 How often are the sensor values updated? |
|---|
| 17 | 2.4 How are alarms triggered? |
|---|
| 18 | |
|---|
| 19 | 3 Installation and management |
|---|
| 20 | 3.1 Why so many modules, and how do I cope with them? |
|---|
| 21 | 3.2 How do I know which chips I own? |
|---|
| 22 | 3.3 Which modules should I insert? |
|---|
| 23 | 3.4 Do I need the configuration file? |
|---|
| 24 | 3.5 What about the 'No such file or directory' warnings when I compile it? |
|---|
| 25 | 3.6 I get all kind of weird compilation errors? |
|---|
| 26 | 3.7 It still does not compile or patch |
|---|
| 27 | |
|---|
| 28 | 4 Problems |
|---|
| 29 | 4.1 Why do my fans report exactly half/double their values by your code |
|---|
| 30 | compared to the BIOS? |
|---|
| 31 | 4.2 Why do my two LM75's report "-48 degrees"? |
|---|
| 32 | 4.3 Why do I have two Vcore readings, I have only one processor! |
|---|
| 33 | 4.4 How do those ALARMS work? The current value is within range but there |
|---|
| 34 | is still an ALARM warning! |
|---|
| 35 | 4.5 My voltage readings seem to drift a bit. Is something wrong with my power |
|---|
| 36 | supply? |
|---|
| 37 | 4.6 Some measurements are way out of range. What happened? |
|---|
| 38 | 4.7 What are VID lines? |
|---|
| 39 | 4.8 I try to read sensor values several times a second, but it seems to be |
|---|
| 40 | updated only each second or so. Why? |
|---|
| 41 | 4.9 It sometimes seems to take a fraction of a seconds before I see the |
|---|
| 42 | sensor reading results. Why? |
|---|
| 43 | 4.10 Can I be alerted when an ALARM occurs? |
|---|
| 44 | 4.11 SMBus transactions on my PIIX4 simply don't work (timeouts happen). Why? |
|---|
| 45 | 4.12 My BIOS reports a much higher CPU temperature than your modules! |
|---|
| 46 | 4.13 I try to read the raw /proc files, but the values are strange?!? |
|---|
| 47 | 4.14 How do I set new limits? |
|---|
| 48 | 4.15 Some sensors are doubly detected? |
|---|
| 49 | 4.16 I ran sensors-detect, but now I get very strange readings?!? |
|---|
| 50 | |
|---|
| 51 | 5 How to ask for help: |
|---|
| 52 | 5.1 What to do if it won't insert? |
|---|
| 53 | 5.2 What to do if it inserts, but nothing happens? |
|---|
| 54 | 5.3 What to do if I read only bogus information? |
|---|
| 55 | 5.4 What to do if you have other problems? |
|---|
| 56 | 5.5 What if it just works like a charm? |
|---|
| 57 | |
|---|
| 58 | 6 Version 1 specific questions |
|---|
| 59 | 6.1 My manufacturer swears that my mainboard has an SMBus, but your code |
|---|
| 60 | reports that it can't find it. What's wrong? |
|---|
| 61 | 6.2 The modules won't load, saying 'SMBus not detected'. |
|---|
| 62 | 6.3 I try to read /proc/sensors, and I get a "No sensor data yet (try again in |
|---|
| 63 | a few moments)" message. Why? |
|---|
| 64 | |
|---|
| 65 | ------------------------------------------------------------- |
|---|
| 66 | |
|---|
| 67 | 1 What sensors are available on my PC? |
|---|
| 68 | |
|---|
| 69 | Most medium/high-end computers since late 1997 now come with a LM78- or |
|---|
| 70 | LM79-like hardware health monitoring chip and an SMBus. |
|---|
| 71 | |
|---|
| 72 | |
|---|
| 73 | 1.1 What can a sensor chip like the "LM78" do? |
|---|
| 74 | |
|---|
| 75 | The LM78 is a chip made by National Semiconductor which can monitor 7 |
|---|
| 76 | voltages (5 positive, 2 negative) from 0 to 4.08V. The inputs are usually in |
|---|
| 77 | series with voltage dividers which lower the 12+/-V and 5+/-V supplies to |
|---|
| 78 | measurable range. Therefore, the readings for such inputs need to be |
|---|
| 79 | re-scaled appropriately by software. |
|---|
| 80 | The LM78 also has 3 fan speed monitoring inputs, an internal |
|---|
| 81 | temperature sensor, a chassis intrusion sensor, and a couple maskable interrupt |
|---|
| 82 | inputs. The LM78 can also relay the processor's (P6 or Pent II) VID lines |
|---|
| 83 | which are hardwired and used to indicate to the power regulator (usually on |
|---|
| 84 | the mainboard close to the processor socket/slot) what voltage to supply to |
|---|
| 85 | the processor. |
|---|
| 86 | The LM78 can be interfaced to a system via the ISA bus and/or the |
|---|
| 87 | SMBus. |
|---|
| 88 | Most other sensor chips have comparable functionality. Each supported |
|---|
| 89 | chip should be documented somewhere in the doc/chips directory. |
|---|
| 90 | |
|---|
| 91 | |
|---|
| 92 | 1.2 Where do I find out more about any of these LMxx chips? |
|---|
| 93 | |
|---|
| 94 | National Semiconductor has an excellent documentation resource on their |
|---|
| 95 | web-site. Complete specs on the LMxx chips talked about in this FAQ can be |
|---|
| 96 | downloaded at: |
|---|
| 97 | http://www.national.com |
|---|
| 98 | The Winbond documentation for Wxxxxxxx chips can be found at: |
|---|
| 99 | http://www.winbond.com.tw/produ/perso7.htm |
|---|
| 100 | The Genesys Logic documentation for GLxxxxx chips is regrettably not freely |
|---|
| 101 | downloadable. |
|---|
| 102 | Please see the file doc/useful_addresses.html for hyperlinks to |
|---|
| 103 | the available documentation. |
|---|
| 104 | |
|---|
| 105 | |
|---|
| 106 | 2 How are these sensors read? |
|---|
| 107 | |
|---|
| 108 | The LM78 and most other chips sensors are usually read through |
|---|
| 109 | the ISA bus. Our code looks for the presence of the chips on the ISA bus |
|---|
| 110 | and attempts to read the sensors from it by reading their different internal |
|---|
| 111 | registers. |
|---|
| 112 | Many chips have also, or even exclusively, an SMBus interface, |
|---|
| 113 | explained more in detail below. |
|---|
| 114 | |
|---|
| 115 | |
|---|
| 116 | 2.1 What is the SMBus? And the I2C bus? |
|---|
| 117 | |
|---|
| 118 | The SMBus is the "System Management Bus". More specifically, it is a |
|---|
| 119 | 2-wire, low-speed serial communication bus used for basic health monitoring |
|---|
| 120 | and hardware management. It is a specific implementation of the more |
|---|
| 121 | general I2C (pronunciation: I-squared-C) bus. In fact, both I2C devices |
|---|
| 122 | and SMBus devices may be connected to the same (I2C) bus. |
|---|
| 123 | The SMBus (or I2C bus) starts at the host controller, used for |
|---|
| 124 | starting transactions on the SMBus. From the host interface, the |
|---|
| 125 | devices communicated with are the 'slave' devices. Each slave device has a |
|---|
| 126 | unique 7-bit address in which the host must refer to it with. |
|---|
| 127 | For each supported SMBus host, there is a separate kernel module |
|---|
| 128 | which implements the communication protocol with the host. Some SMBus hosts |
|---|
| 129 | really operate on the SMBus level; these hosts can not cope with pure I2C |
|---|
| 130 | devices. Other hosts are in fact I2C hosts: in this case, we implement |
|---|
| 131 | the SMBus protocol in terms of I2C operations. But these hosts can also |
|---|
| 132 | talk to pure I2C devices. |
|---|
| 133 | |
|---|
| 134 | |
|---|
| 135 | 2.2 What sensors does the Pentium 2 (and P6) have? |
|---|
| 136 | |
|---|
| 137 | The Pentium 2 'boxed' processor usually has an LM75 very close to the |
|---|
| 138 | base of the box. It can be read through the SMBus to report the approximate |
|---|
| 139 | temperature of the processor. The processor also contains an internal |
|---|
| 140 | temperature sensor (of low accuracy) used as a fail-safe to disable the |
|---|
| 141 | processor in case it gets much too hot (usually around 130 degrees C). And, |
|---|
| 142 | the Pentium 2 also has a hard-wired signal (VID lines) on it's SEC (single |
|---|
| 143 | edge connector) which indicates what power supply is required to operate the |
|---|
| 144 | processor. |
|---|
| 145 | Apparently, the SEC connector has connectivity to the thermal sensor |
|---|
| 146 | for devices like the Analog Devices ADM1021 which has circuitry for |
|---|
| 147 | converting the sense information into temperature degrees. The Winbond |
|---|
| 148 | chip's external temperature lines may be compatible with the P2 temperature |
|---|
| 149 | lines as well. |
|---|
| 150 | The P6 (Pentium-Pro) may have an LM75 in or just under the socket. |
|---|
| 151 | P6's also have VID lines. |
|---|
| 152 | Pentiums and Pentium w/ MMX do not have VID lines, and sometimes have |
|---|
| 153 | LM75's under the sockets (depends on the mainboard, and how 'modern' the |
|---|
| 154 | mainboard is). |
|---|
| 155 | The P2 Xeon will be the first Intel processor to include the SMBus |
|---|
| 156 | interface on the P2 Xeon SEC. |
|---|
| 157 | |
|---|
| 158 | |
|---|
| 159 | 2.3 How often are the sensor values updated? |
|---|
| 160 | |
|---|
| 161 | The LM78, and most other sensor chips like it, reads its sensors one |
|---|
| 162 | by one. A complete scanning sweep will take about 1.5 seconds. The LM78 stops |
|---|
| 163 | readings sensors if you try to access it, so if you access it very often |
|---|
| 164 | (by reading sensor values; writing new limits is safe) it will not find the |
|---|
| 165 | time to update its sensor values at all! Fortunately, the kernel module takes |
|---|
| 166 | care not to do this, and only reads new values each 1.5 seconds. If you |
|---|
| 167 | read the values again, you will get the 'old' values again. |
|---|
| 168 | |
|---|
| 169 | |
|---|
| 170 | 2.4 How are alarms triggered? |
|---|
| 171 | |
|---|
| 172 | It is possible to monitor each sensor and let an alarm go off if |
|---|
| 173 | it crosses some pre-determined limits. There are two sorts of interrupts |
|---|
| 174 | which can be generated by sensor chips if this happens (it depends a bit on |
|---|
| 175 | the actual chip if both are supported; the LM80, for example, has only |
|---|
| 176 | IRQ interrupts): IRQ interrupts and SMI interrupts. IRQ stands for |
|---|
| 177 | Interrupt Request and are the interrupt lines you can find in /proc/interrupts. |
|---|
| 178 | SMI stands for System Management Interrupt, and is a special interrupt which |
|---|
| 179 | puts the processor in a secure environment independent of any other things |
|---|
| 180 | running. SMI is currently not supported by the Linux kernel. IRQs are |
|---|
| 181 | supported, of course. |
|---|
| 182 | Even if no interrupt is generated, some bits in a status register |
|---|
| 183 | will be set until the register is read the next time. If the alarm condition |
|---|
| 184 | persists after that, the bits will be set on the next scanning sweep, etc. |
|---|
| 185 | At this moment, interrupts are not supported. |
|---|
| 186 | |
|---|
| 187 | |
|---|
| 188 | 3 Installation and management |
|---|
| 189 | |
|---|
| 190 | 3.1 Why so many modules, and how do I cope with them? |
|---|
| 191 | |
|---|
| 192 | We tried to make this package as modular as possible. This makes it |
|---|
| 193 | easy to add new drivers, and unused drivers will take no precious kernel |
|---|
| 194 | space. On the other hand, it can be a bit confusing at first. |
|---|
| 195 | It is strongly suggested you read doc/modules, and follow the |
|---|
| 196 | installation described there. It will allow you to use simple modprobe |
|---|
| 197 | commands to load the modules, instead of the more low-level insmod calls. |
|---|
| 198 | This file also describes how you can load the modules automatically at |
|---|
| 199 | boot-time. |
|---|
| 200 | |
|---|
| 201 | |
|---|
| 202 | 3.2 How do I know which chips I own? |
|---|
| 203 | |
|---|
| 204 | By now, chip detection is fairly good. That means that it is |
|---|
| 205 | relatively harmless to insert more chip drivers than you need. This approach |
|---|
| 206 | can still lead to problems, though. |
|---|
| 207 | In the past, you had to muddle your way through and just tried |
|---|
| 208 | several drivers. But starting with 2.3.0, we have an excellent program |
|---|
| 209 | that scans all your hardware. It is called 'sensors-detect' and is |
|---|
| 210 | installed when you call 'make install'. Just execute this script, and |
|---|
| 211 | it will tell you exactly what you want to know. |
|---|
| 212 | |
|---|
| 213 | |
|---|
| 214 | 3.3 Which modules should I insert? |
|---|
| 215 | |
|---|
| 216 | You need one module for each adapter you own; if there are |
|---|
| 217 | chips on the ISA bus, this includes i2c-isa.o. Also, you need one module |
|---|
| 218 | for each type of chip you own. That's all. On my computer, I could use the |
|---|
| 219 | following line: |
|---|
| 220 | modprobe i2c-isa && modprobe i2c-piix4 && \ |
|---|
| 221 | modprobe lm78 && modprobe lm75 && modprobe i2c-dev |
|---|
| 222 | |
|---|
| 223 | |
|---|
| 224 | 3.4 Do I need the configuration file? |
|---|
| 225 | |
|---|
| 226 | Usually, you do. It tells how to translate the values the chip |
|---|
| 227 | measures to real-world values. This is especially important for voltage |
|---|
| 228 | inputs. The default configuration file should usually do the trick. |
|---|
| 229 | It is automatically installed as /etc/sensors.conf, but it will not |
|---|
| 230 | overwrite any existing file with that name. |
|---|
| 231 | |
|---|
| 232 | |
|---|
| 233 | 3.5 What about the 'No such file or directory' warnings when I compile it? |
|---|
| 234 | |
|---|
| 235 | Don't worry about them. The dependency files (which tell which |
|---|
| 236 | files should be recompiled when certain files change) are created |
|---|
| 237 | dynamically. They are not distributed with the package. The `make' program |
|---|
| 238 | notices they are not there, and warns about that - and the first thing |
|---|
| 239 | it will do is generate them. So all is well. |
|---|
| 240 | |
|---|
| 241 | 3.6 I get all kinds of weird compilation errors? |
|---|
| 242 | |
|---|
| 243 | Check that the correct i2c header files are used. Depending on |
|---|
| 244 | how you installed, they should be under either /usr/local/include or |
|---|
| 245 | /usr/src/linux*/include. Try to edit the Makefile for the other setting. |
|---|
| 246 | |
|---|
| 247 | |
|---|
| 248 | 3.7 It still does not compile or patch! |
|---|
| 249 | Have you installed a recent version of the i2c package? Remember, |
|---|
| 250 | compilation is not enough, you also need to install it for the header |
|---|
| 251 | files to be found! |
|---|
| 252 | If you want to patch the kernel, you will have to apply the i2c |
|---|
| 253 | patches first! |
|---|
| 254 | |
|---|
| 255 | |
|---|
| 256 | 4 Problems |
|---|
| 257 | |
|---|
| 258 | 4.1 Why do my fans report exactly half/double their values by your code |
|---|
| 259 | compared to the BIOS? |
|---|
| 260 | |
|---|
| 261 | The problem with much of the sensor data is that it is impossible to |
|---|
| 262 | properly interpret some of the readings without knowing what the hardware |
|---|
| 263 | configuration is. Some fans report one 'tick' each rotation, some report |
|---|
| 264 | two 'ticks' each rotation. It is easy to resolve this through the |
|---|
| 265 | configuration file: |
|---|
| 266 | chip lm78-* # Or whatever chip this relates to |
|---|
| 267 | compute fan1 2*@,@/2 # Copy for each fan present |
|---|
| 268 | |
|---|
| 269 | |
|---|
| 270 | 4.2 Why do my two LM75's report "-48 degrees"? |
|---|
| 271 | |
|---|
| 272 | For starters, those aren't LM75's. Your mainboard actually has the |
|---|
| 273 | Winbond W83781D which emulates two LM75's, but many systems which use the |
|---|
| 274 | Winbond chip (such as the Asus P2B) don't have the thermo-resisters connected |
|---|
| 275 | to the chip resulting in these strange -48 degree readings. |
|---|
| 276 | If you have an Asus P2B and want more information on adding thermal |
|---|
| 277 | sensing capability, check out: |
|---|
| 278 | http://ultimatepc.fsn.net/techinfo/p2bthermistor/p2bthermistor.htm |
|---|
| 279 | In upcoming versions, you will be able to disable non-interesting |
|---|
| 280 | readings. |
|---|
| 281 | |
|---|
| 282 | |
|---|
| 283 | 4.3 Why do I have two Vcore readings, I have only one processor! |
|---|
| 284 | |
|---|
| 285 | The LM78 family has seven voltage sensors. The default way of |
|---|
| 286 | connecting them is used in the configuration file. This includes a VCore2, |
|---|
| 287 | even if you do not have one. You can easily edit the configuration file |
|---|
| 288 | to give it another name, or (in upcoming versions) to make this reading |
|---|
| 289 | disappear. |
|---|
| 290 | Note that Vcore2 is usually the same as Vcore on motherboards which |
|---|
| 291 | only support one processor. It is possible, though, that is monitors something |
|---|
| 292 | else, so you should not be too surprised if the values are completely |
|---|
| 293 | different. |
|---|
| 294 | |
|---|
| 295 | |
|---|
| 296 | 4.4 How do those ALARMS work? The current value is within range but there |
|---|
| 297 | is still an ALARM warning! |
|---|
| 298 | |
|---|
| 299 | An ALARM will go off when a minimum or maximum limit is crossed. It |
|---|
| 300 | will stay there until the next internal update - which will be the next time |
|---|
| 301 | you read these values, but not within (usually) 1.5 seconds since the last |
|---|
| 302 | update. After that, it will only stay if the current value is out of range, |
|---|
| 303 | though this too depends a bit on the kind of chip. See the specific chip |
|---|
| 304 | documentation in doc/chips for more information. |
|---|
| 305 | |
|---|
| 306 | |
|---|
| 307 | 4.5 My voltage readings seem to drift a bit. Is something wrong with my power |
|---|
| 308 | supply? |
|---|
| 309 | |
|---|
| 310 | No, probably not. If your motherboard heats up a bit, the sensed |
|---|
| 311 | voltages will drift a bit. If your power supply is loaded (because a disk |
|---|
| 312 | gets going, for example), the voltages may get a bit lower. As long as they |
|---|
| 313 | stay within a sensible range (say 10% of the expected value), there is no |
|---|
| 314 | reason to worry. |
|---|
| 315 | |
|---|
| 316 | |
|---|
| 317 | 4.6 Some measurements are way out of range. What happened? |
|---|
| 318 | |
|---|
| 319 | Each module tries to set limits to sensible values on initialization, |
|---|
| 320 | but a module does not know how a chip is actually connected. This is |
|---|
| 321 | described in the configuration file, which is not read by kernel modules. |
|---|
| 322 | So limits can be strange, if the chip is connected in a non-standard way. |
|---|
| 323 | Readings can also be strange; there are several reasons for this. |
|---|
| 324 | Temperature sensors, for example, can simply not be present, even though |
|---|
| 325 | the chip supports them. Also, it can be that the input is used in a |
|---|
| 326 | non-standard way. You can use the configuration file to describe how this |
|---|
| 327 | measurement should be interpreted; see the comments the example file for |
|---|
| 328 | more information. |
|---|
| 329 | |
|---|
| 330 | |
|---|
| 331 | 4.7 What are VID lines? |
|---|
| 332 | |
|---|
| 333 | These describe the voltage your processors use. This is only supported |
|---|
| 334 | for Pentium 2 and newer processors, and even then they are not always |
|---|
| 335 | correctly connected to the sensor chip, so the readings may be out of |
|---|
| 336 | range. A value of +3.5 V is especially suspect. |
|---|
| 337 | |
|---|
| 338 | |
|---|
| 339 | 4.8 I try to read sensor values several times a second, but it seems to be |
|---|
| 340 | updated only each second or so. Why? |
|---|
| 341 | |
|---|
| 342 | If we would read the registers more often, it would not find the |
|---|
| 343 | time to update them. So we only update our readings once each 1.5 seconds |
|---|
| 344 | (the actual delay is chip-specific; for some chips, it may not be needed |
|---|
| 345 | at all). |
|---|
| 346 | |
|---|
| 347 | |
|---|
| 348 | 4.9 It sometimes seems to take a fraction of a seconds before I see the |
|---|
| 349 | sensor reading results. Why? |
|---|
| 350 | |
|---|
| 351 | ISA bus access is fast, but SMBus access is really slow. If you have |
|---|
| 352 | a lot of sensors, it just takes a lot of time to access them. Fortunately, |
|---|
| 353 | this has almost no impact on the system as a whole, as another job can run |
|---|
| 354 | while we are waiting for the transaction to finish. |
|---|
| 355 | |
|---|
| 356 | |
|---|
| 357 | 4.10 Can I be alerted when an ALARM occurs? |
|---|
| 358 | |
|---|
| 359 | No, you can't; and it may well be never supported. |
|---|
| 360 | Almost no mainboard we have encountered have actually connected the |
|---|
| 361 | IRQ-out pin of sensor chips. That means that we could enable IRQ reporting, but |
|---|
| 362 | nothing would happen. Also, even if a motherboard has it connected, it is |
|---|
| 363 | unclear what interrupt number would be triggered. And IRQ lines are a scarce |
|---|
| 364 | facility, which means that almost nobody would be able to use it anyway. |
|---|
| 365 | The SMI interrupt is only available on a few types of chips. It is |
|---|
| 366 | really a very obscure way to handle interrupts, and supporting it under Linux |
|---|
| 367 | might be quite hard to do. |
|---|
| 368 | Your best bet would be to poll the alarm file with a user-land daemon |
|---|
| 369 | which alerts you if an alarm is raised. I am not aware of any program which |
|---|
| 370 | does the job, though you might want to examine one of the graphical monitor |
|---|
| 371 | programs under X, see doc/useful_addresses.html for addresses. |
|---|
| 372 | |
|---|
| 373 | |
|---|
| 374 | 4.11 SMBus transactions on my PIIX4 simply don't work (timeouts happen). Why? |
|---|
| 375 | |
|---|
| 376 | Some chips which mainboard makers connect to the SMBus are not SMBus |
|---|
| 377 | devices. An example is the 91xx clock generator chips. When read, these |
|---|
| 378 | devices can lock up the SMBus until the next hard reboot. This is because |
|---|
| 379 | they have a similar serial interface (like the I2C), but don't conform to |
|---|
| 380 | Intel's SMBus standard. |
|---|
| 381 | Why did they connect these devices to the SMBus if they aren't |
|---|
| 382 | compatible? Good question! :') Actually, these devices may support being |
|---|
| 383 | written to, but lock things up when they are read. |
|---|
| 384 | |
|---|
| 385 | |
|---|
| 386 | 4.12 My BIOS reports a much higher CPU temperature than your modules! |
|---|
| 387 | |
|---|
| 388 | We display the actual temperature of the sensor. This may not be the |
|---|
| 389 | temperature you are interested in, though. If a sensor should measure |
|---|
| 390 | the CPU temperature, it must be in thermal contact with it. In practice, |
|---|
| 391 | it is just somewhere near it. Your BIOS may correct for this (by adding, |
|---|
| 392 | for example, thirty degrees to the measured temperature). The correction |
|---|
| 393 | factor is regrettably different for each mainboard, so we can not do this |
|---|
| 394 | in the module itself. You can do it through the configuration file, though: |
|---|
| 395 | |
|---|
| 396 | chip lm75-*-49 # Or whatever chip this relates to |
|---|
| 397 | label temp "Processor" |
|---|
| 398 | compute temp @*1.2+13,(@-13)/1.2 # Or whatever formula |
|---|
| 399 | |
|---|
| 400 | |
|---|
| 401 | 4.13 I try to read the raw /proc files, but the values are strange?!? |
|---|
| 402 | |
|---|
| 403 | Remember, these values do not take the configuration file |
|---|
| 404 | 'compute' lines in account. This is especially obvious for voltage readings |
|---|
| 405 | (usually called in? or vin?). Use a program linked to libsensors (like |
|---|
| 406 | the provided 'sensors' program) instead. |
|---|
| 407 | |
|---|
| 408 | |
|---|
| 409 | 4.14 How do I set new limits? |
|---|
| 410 | |
|---|
| 411 | This is a bit intricate right now. In the near future, this can be |
|---|
| 412 | done through the 'sensors' program, which will take account of the |
|---|
| 413 | computations specified in the configuration file. At this moment, you must |
|---|
| 414 | also put the 'set' commands into the configuration file. This will get |
|---|
| 415 | easier in newer versions. |
|---|
| 416 | |
|---|
| 417 | |
|---|
| 418 | 4.15 Some sensors are doubly detected? |
|---|
| 419 | |
|---|
| 420 | Yes, this is still a problem. It will partially solved in the future, |
|---|
| 421 | but it is really tough. Double detections can be caused by two things: |
|---|
| 422 | sensors can be detected to both the ISA and the SMBus (and if you have |
|---|
| 423 | loaded the approprate adapter drivers, it will be detected on both), and |
|---|
| 424 | some chips simulate other chips (the Winbond W83781D simulates LM75 chips |
|---|
| 425 | on the SMBus, for example). Remove the offending adapter or chip driver, or |
|---|
| 426 | run sensors-detect and add the insmod parameters it suggests. |
|---|
| 427 | |
|---|
| 428 | |
|---|
| 429 | 4.16 I ran sensors-detect, but now I get very strange readings?!? |
|---|
| 430 | |
|---|
| 431 | Your SMBus (PIIX4?) is probably crashed. There are some mainboards |
|---|
| 432 | which connect a clock chip to the SMBus. Unfortunately, this clock chip |
|---|
| 433 | hangs the PIIX4 if it is read (it is an I2C device, but not SMBus compatible). |
|---|
| 434 | We have found no way of solving this, except for rebooting your computer. |
|---|
| 435 | Next time when you run sensors-detect, you may want to exclude addresses |
|---|
| 436 | 0x69 and/or 0x6a, by entering 's' when you are asked whether you want to |
|---|
| 437 | scan the PIIX4. |
|---|
| 438 | |
|---|
| 439 | 5 How to ask for help: |
|---|
| 440 | |
|---|
| 441 | We are always willing to answer questions if things don't work out. |
|---|
| 442 | Please mail sensors@stimpy.netroedge.com, and not the individual authors, |
|---|
| 443 | unless you have something private to say. Especially, do not mail |
|---|
| 444 | Alexander Larsson, as he has left the development team. You can be assured |
|---|
| 445 | that any mail sent to sensors@stimpy.netroedge.com will arrive at anyone |
|---|
| 446 | who answered your email, so please do not CC: him again. |
|---|
| 447 | Instead of using email, you can also use the web-based support |
|---|
| 448 | area, at http://www.netroedge.com/~lm78/support.html. You will be helped |
|---|
| 449 | just as fast, and others may profit from the answer too. You will be |
|---|
| 450 | noticed automatically when your question has been answered. |
|---|
| 451 | |
|---|
| 452 | |
|---|
| 453 | 5.1 What to do if it won't insert? |
|---|
| 454 | |
|---|
| 455 | Always inspect the output of 'dmesg' and send it to us. Check |
|---|
| 456 | /proc/pci for your SMBus adapter, and send the output too. Check whether |
|---|
| 457 | you actually have a sensor chip - sometimes, it is only an upgrade option. |
|---|
| 458 | If your mainboard manual happens to mention the chip type, send it too. |
|---|
| 459 | Tell us what mainboard you own. |
|---|
| 460 | |
|---|
| 461 | |
|---|
| 462 | 5.2 What to do if it inserts, but nothing happens? |
|---|
| 463 | |
|---|
| 464 | Check /proc/pci to see whether you have a supported adapter. If it |
|---|
| 465 | is not there, we do not support it. Future version may do it. If you want, |
|---|
| 466 | you can send us an email telling you have a such-and-so mainboard, together |
|---|
| 467 | with the PCI output - that way, we can see what SMBus hosts are used most. |
|---|
| 468 | |
|---|
| 469 | |
|---|
| 470 | 5.3 What to do if I read only bogus information? |
|---|
| 471 | |
|---|
| 472 | It may be that this was a mis-detection: the chip may not be |
|---|
| 473 | present. If you are convinced there is something wrong, please send us |
|---|
| 474 | the usual information (see 5.1). |
|---|
| 475 | |
|---|
| 476 | |
|---|
| 477 | 5.4 What to do if you have other problems? |
|---|
| 478 | |
|---|
| 479 | Again, send the output of 'dmesg', /proc/sensors and /proc/pci. |
|---|
| 480 | |
|---|
| 481 | |
|---|
| 482 | 5.5 What if it just works like a charm? |
|---|
| 483 | |
|---|
| 484 | Drop us a mail if you feel like it, mentioning the mainboard and |
|---|
| 485 | detected chip type. That way, we have some positive feedback, too! |
|---|
| 486 | |
|---|
| 487 | |
|---|
| 488 | 6 Version 1 specific questions |
|---|
| 489 | |
|---|
| 490 | 6.1 My manufacturer swears that my mainboard has an SMBus, but your code |
|---|
| 491 | reports that it can't find it. What's wrong? |
|---|
| 492 | |
|---|
| 493 | Currently, our code only assumes that an SMBus exists if it originates |
|---|
| 494 | from the Intel PIIX4 (82371AB). If your computer doesn't have one, or if your |
|---|
| 495 | SMBus originates from a different SMBus 'host', then you are out of luck. :'( |
|---|
| 496 | Our experience is, though, that most machines have a PIIX4, and that it is |
|---|
| 497 | where the SMBus is hosted. |
|---|
| 498 | |
|---|
| 499 | Regarding the VIA chip set(s): |
|---|
| 500 | Right now, the SMBus code depends on the Intel PIIX4 chip to handle |
|---|
| 501 | SMBus transactions. The VIA chip set is NOT supported at this time because |
|---|
| 502 | it forces much of the SMBus protocol to be implemented by software. |
|---|
| 503 | To implement the SMBus correctly with the VIA chip, it needs to be |
|---|
| 504 | written at a low level to be quick. A more attractive alternative is to |
|---|
| 505 | use the Bios SMBus interface (not always available nor standard). |
|---|
| 506 | |
|---|
| 507 | Version 2 supports the VIA chipset, and will support other chipsets. |
|---|
| 508 | Version 1 never will. |
|---|
| 509 | |
|---|
| 510 | |
|---|
| 511 | 6.2 The modules won't load, saying 'SMBus not detected'. |
|---|
| 512 | |
|---|
| 513 | This should no longer be an issue in 1.4.10 and later; it will |
|---|
| 514 | continue loading, but it won't support SMBus-connected devices, of course. |
|---|
| 515 | |
|---|
| 516 | |
|---|
| 517 | 6.3 I try to read /proc/sensors, and I get a "No sensor data yet (try again in |
|---|
| 518 | a few moments)" message. Why? |
|---|
| 519 | |
|---|
| 520 | It takes about 1.5 seconds for the LM78 to update all its sensor |
|---|
| 521 | values. If we would try to read it before it finished that, you would get |
|---|
| 522 | old garbage instead. So you have to wait 1.5 seconds after the module is |
|---|
| 523 | inserted before you can access /proc/sensors. |
|---|
| 524 | Module versions 1.3.7 and later let the process sleep if it tries |
|---|
| 525 | to access sensor data right after insertion time, and do not display this |
|---|
| 526 | message anymore. |
|---|
| 527 | |
|---|
| 528 | |
|---|
| 529 | 6.4 On my Dell, a LM80 is detected, but all readings are 0! |
|---|
| 530 | |
|---|
| 531 | This is a bug we have only observed on Dell computers. There is |
|---|
| 532 | probably a problem with the way the SMBus is accessed; but it is not yet |
|---|
| 533 | clear whether the problem is in our code or with the Dells. |
|---|
| 534 | There are very probably no sensor chips at all on your computer; |
|---|
| 535 | but until somebody contacts Dell about this and tells us the results, we |
|---|
| 536 | can not be sure. |
|---|
| 537 | |
|---|
| 538 | |
|---|
| 539 | ----------- |
|---|
| 540 | Rev 2.2 (Frodo) Corrections for lm_sensors 2.4, 19990920 |
|---|
| 541 | Rev 2.1 (Frodo) Corrections for lm_sensors 2.2, 19990112 |
|---|
| 542 | Rev 2.0 (Frodo) Major revision for lm_sensors 2.1, 19981229 |
|---|
| 543 | Rev 1.10 (Frodo) Modified 3.8, updated some other things, 19980924 |
|---|
| 544 | Rev 1.9 (Frodo) Added 3.15, 19980906 |
|---|
| 545 | Rev 1.8 (Frodo) Added 3.14, 19980905 |
|---|
| 546 | Rev 1.7 (Phil) Added 3.13 and some other minor changes, 19980901 |
|---|
| 547 | Rev 1.6 (Frodo) Added 4, 4.1, 4.2, 4.3, 4.4, 4.5, 19980901 |
|---|
| 548 | Rev 1.5 (Frodo) Added 2.3, 2.4, 3.9, 3.10, 3.11, 19980826 |
|---|
| 549 | Rev 1.4 (Frodo) Added some more Winbond information, and 3.5-3.8, 19980817 |
|---|
| 550 | Rev 1.3 Added info on the Winbond chip, 19980816 |
|---|
| 551 | Rev 1.2 Adapation by Frodo Looijaard, 19980810 |
|---|
| 552 | Rev 1.1 Modifications by Philip Edelbrock, 19980809 |
|---|
| 553 | Rev 1.0 Written by Philip Edelbrock, 19980803 |
|---|