Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 1 | Kernel driver k10temp |
| 2 | ===================== |
| 3 | |
| 4 | Supported chips: |
| 5 | * AMD Family 10h processors: |
Clemens Ladisch | c5114a1 | 2010-01-10 20:52:34 +0100 | [diff] [blame] | 6 | Socket F: Quad-Core/Six-Core/Embedded Opteron (but see below) |
| 7 | Socket AM2+: Quad-Core Opteron, Phenom (II) X3/X4, Athlon X2 (but see below) |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 8 | Socket AM3: Quad-Core Opteron, Athlon/Phenom II X2/X3/X4, Sempron II |
| 9 | Socket S1G3: Athlon II, Sempron, Turion II |
| 10 | * AMD Family 11h processors: |
| 11 | Socket S1G2: Athlon (X2), Sempron (X2), Turion X2 (Ultra) |
Clemens Ladisch | af75d5b | 2011-07-03 13:32:54 +0200 | [diff] [blame] | 12 | * AMD Family 12h processors: "Llano" (E2/A4/A6/A8-Series) |
| 13 | * AMD Family 14h processors: "Brazos" (C/E/G/Z-Series) |
Andre Przywara | fbc729a | 2012-03-30 16:48:20 -0400 | [diff] [blame] | 14 | * AMD Family 15h processors: "Bulldozer" (FX-Series), "Trinity" |
Wei Hu | 30b146d | 2013-08-23 13:14:03 -0700 | [diff] [blame] | 15 | * AMD Family 16h processors: "Kabini" |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 16 | |
| 17 | Prefix: 'k10temp' |
| 18 | Addresses scanned: PCI space |
| 19 | Datasheets: |
| 20 | BIOS and Kernel Developer's Guide (BKDG) For AMD Family 10h Processors: |
| 21 | http://support.amd.com/us/Processor_TechDocs/31116.pdf |
| 22 | BIOS and Kernel Developer's Guide (BKDG) for AMD Family 11h Processors: |
| 23 | http://support.amd.com/us/Processor_TechDocs/41256.pdf |
Clemens Ladisch | af75d5b | 2011-07-03 13:32:54 +0200 | [diff] [blame] | 24 | BIOS and Kernel Developer's Guide (BKDG) for AMD Family 12h Processors: |
| 25 | http://support.amd.com/us/Processor_TechDocs/41131.pdf |
Clemens Ladisch | aa4790a | 2011-02-17 03:22:40 -0500 | [diff] [blame] | 26 | BIOS and Kernel Developer's Guide (BKDG) for AMD Family 14h Models 00h-0Fh Processors: |
| 27 | http://support.amd.com/us/Processor_TechDocs/43170.pdf |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 28 | Revision Guide for AMD Family 10h Processors: |
| 29 | http://support.amd.com/us/Processor_TechDocs/41322.pdf |
| 30 | Revision Guide for AMD Family 11h Processors: |
| 31 | http://support.amd.com/us/Processor_TechDocs/41788.pdf |
Clemens Ladisch | af75d5b | 2011-07-03 13:32:54 +0200 | [diff] [blame] | 32 | Revision Guide for AMD Family 12h Processors: |
| 33 | http://support.amd.com/us/Processor_TechDocs/44739.pdf |
Clemens Ladisch | aa4790a | 2011-02-17 03:22:40 -0500 | [diff] [blame] | 34 | Revision Guide for AMD Family 14h Models 00h-0Fh Processors: |
| 35 | http://support.amd.com/us/Processor_TechDocs/47534.pdf |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 36 | AMD Family 11h Processor Power and Thermal Data Sheet for Notebooks: |
| 37 | http://support.amd.com/us/Processor_TechDocs/43373.pdf |
| 38 | AMD Family 10h Server and Workstation Processor Power and Thermal Data Sheet: |
| 39 | http://support.amd.com/us/Processor_TechDocs/43374.pdf |
| 40 | AMD Family 10h Desktop Processor Power and Thermal Data Sheet: |
| 41 | http://support.amd.com/us/Processor_TechDocs/43375.pdf |
| 42 | |
| 43 | Author: Clemens Ladisch <clemens@ladisch.de> |
| 44 | |
| 45 | Description |
| 46 | ----------- |
| 47 | |
| 48 | This driver permits reading of the internal temperature sensor of AMD |
Andre Przywara | 9e58131 | 2011-05-25 20:43:31 +0200 | [diff] [blame] | 49 | Family 10h/11h/12h/14h/15h processors. |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 50 | |
Clemens Ladisch | c5114a1 | 2010-01-10 20:52:34 +0100 | [diff] [blame] | 51 | All these processors have a sensor, but on those for Socket F or AM2+, |
| 52 | the sensor may return inconsistent values (erratum 319). The driver |
| 53 | will refuse to load on these revisions unless you specify the "force=1" |
| 54 | module parameter. |
| 55 | |
| 56 | Due to technical reasons, the driver can detect only the mainboard's |
| 57 | socket type, not the processor's actual capabilities. Therefore, if you |
| 58 | are using an AM3 processor on an AM2+ mainboard, you can safely use the |
| 59 | "force=1" parameter. |
Clemens Ladisch | 3c57e89 | 2009-12-16 21:38:25 +0100 | [diff] [blame] | 60 | |
| 61 | There is one temperature measurement value, available as temp1_input in |
| 62 | sysfs. It is measured in degrees Celsius with a resolution of 1/8th degree. |
| 63 | Please note that it is defined as a relative value; to quote the AMD manual: |
| 64 | |
| 65 | Tctl is the processor temperature control value, used by the platform to |
| 66 | control cooling systems. Tctl is a non-physical temperature on an |
| 67 | arbitrary scale measured in degrees. It does _not_ represent an actual |
| 68 | physical temperature like die or case temperature. Instead, it specifies |
| 69 | the processor temperature relative to the point at which the system must |
| 70 | supply the maximum cooling for the processor's specified maximum case |
| 71 | temperature and maximum thermal power dissipation. |
| 72 | |
| 73 | The maximum value for Tctl is available in the file temp1_max. |
| 74 | |
| 75 | If the BIOS has enabled hardware temperature control, the threshold at |
| 76 | which the processor will throttle itself to avoid damage is available in |
| 77 | temp1_crit and temp1_crit_hyst. |