R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 1 | Kernel driver lm78 |
| 2 | ================== |
| 3 | |
| 4 | Supported chips: |
Jean Delvare | 27fe048 | 2005-07-27 21:30:16 +0200 | [diff] [blame] | 5 | * National Semiconductor LM78 / LM78-J |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 6 | Prefix: 'lm78' |
Jean Delvare | 6722fea | 2007-10-07 12:25:46 +0200 | [diff] [blame] | 7 | Addresses scanned: I2C 0x28 - 0x2f, ISA 0x290 (8 I/O ports) |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 8 | Datasheet: Publicly available at the National Semiconductor website |
| 9 | http://www.national.com/ |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 10 | * National Semiconductor LM79 |
| 11 | Prefix: 'lm79' |
Jean Delvare | 6722fea | 2007-10-07 12:25:46 +0200 | [diff] [blame] | 12 | Addresses scanned: I2C 0x28 - 0x2f, ISA 0x290 (8 I/O ports) |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 13 | Datasheet: Publicly available at the National Semiconductor website |
| 14 | http://www.national.com/ |
| 15 | |
Jean Delvare | 156e2d1 | 2011-07-25 21:46:11 +0200 | [diff] [blame] | 16 | Authors: Frodo Looijaard <frodol@dds.nl> |
Jean Delvare | 7c81c60 | 2014-01-29 20:40:08 +0100 | [diff] [blame] | 17 | Jean Delvare <jdelvare@suse.de> |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 18 | |
| 19 | Description |
| 20 | ----------- |
| 21 | |
| 22 | This driver implements support for the National Semiconductor LM78, LM78-J |
| 23 | and LM79. They are described as 'Microprocessor System Hardware Monitors'. |
| 24 | |
| 25 | There is almost no difference between the three supported chips. Functionally, |
| 26 | the LM78 and LM78-J are exactly identical. The LM79 has one more VID line, |
| 27 | which is used to report the lower voltages newer Pentium processors use. |
| 28 | From here on, LM7* means either of these three types. |
| 29 | |
| 30 | The LM7* implements one temperature sensor, three fan rotation speed sensors, |
| 31 | seven voltage sensors, VID lines, alarms, and some miscellaneous stuff. |
| 32 | |
| 33 | Temperatures are measured in degrees Celsius. An alarm is triggered once |
| 34 | when the Overtemperature Shutdown limit is crossed; it is triggered again |
| 35 | as soon as it drops below the Hysteresis value. A more useful behavior |
| 36 | can be found by setting the Hysteresis value to +127 degrees Celsius; in |
| 37 | this case, alarms are issued during all the time when the actual temperature |
| 38 | is above the Overtemperature Shutdown value. Measurements are guaranteed |
| 39 | between -55 and +125 degrees, with a resolution of 1 degree. |
| 40 | |
| 41 | Fan rotation speeds are reported in RPM (rotations per minute). An alarm is |
| 42 | triggered if the rotation speed has dropped below a programmable limit. Fan |
| 43 | readings can be divided by a programmable divider (1, 2, 4 or 8) to give |
| 44 | the readings more range or accuracy. Not all RPM values can accurately be |
| 45 | represented, so some rounding is done. With a divider of 2, the lowest |
| 46 | representable value is around 2600 RPM. |
| 47 | |
| 48 | Voltage sensors (also known as IN sensors) report their values in volts. |
| 49 | An alarm is triggered if the voltage has crossed a programmable minimum |
| 50 | or maximum limit. Note that minimum in this case always means 'closest to |
| 51 | zero'; this is important for negative voltage measurements. All voltage |
| 52 | inputs can measure voltages between 0 and 4.08 volts, with a resolution |
| 53 | of 0.016 volt. |
| 54 | |
| 55 | The VID lines encode the core voltage value: the voltage level your processor |
| 56 | should work with. This is hardcoded by the mainboard and/or processor itself. |
| 57 | It is a value in volts. When it is unconnected, you will often find the |
| 58 | value 3.50 V here. |
| 59 | |
R.Marek@sh.cvut.cz | 7f15b66 | 2005-05-26 12:42:19 +0000 | [diff] [blame] | 60 | If an alarm triggers, it will remain triggered until the hardware register |
| 61 | is read at least once. This means that the cause for the alarm may |
| 62 | already have disappeared! Note that in the current implementation, all |
| 63 | hardware registers are read whenever any data is read (unless it is less |
| 64 | than 1.5 seconds since the last update). This means that you can easily |
| 65 | miss once-only alarms. |
| 66 | |
| 67 | The LM7* only updates its values each 1.5 seconds; reading it more often |
| 68 | will do no harm, but will return 'old' values. |