1 |
>> After upgrading from 2.6.28 to 2.6.31, I noticed my CPU temperatures |
2 |
>> are reported a full 20C hotter. If I load the old kernel, the |
3 |
>> reported temperatures drops back down to normal. Has anyone else seen |
4 |
>> this? |
5 |
> |
6 |
> If you're using coretemp as sensor, the temps are always off (the coretemp |
7 |
> sensor of Intel chips is not accurate, not by any stretch of the |
8 |
> imagination.) It only reports the distance to the CPU's maximum thermal |
9 |
> junction, which then the coretemp driver *tries* to translate into a |
10 |
> temperature, but the result is wrong since the value reported by the CPU is |
11 |
> not accurate to start with (it only gets accurate as you approach the max |
12 |
> value). That maximum value is totally undocumented for desktop CPUs (the |
13 |
> docs Intel provided recently are wrong.) |
14 |
> |
15 |
> You should use your mainboard's sensors instead for accurate values. |
16 |
|
17 |
I'm actually using k8temp. Do you think it is susceptible to the same |
18 |
problems you're talking about? I also have an ACPI sensor available |
19 |
named THRM. Should that one be more accurate? |
20 |
|
21 |
BTW, another system of mine (Dell laptop) only seems to have available |
22 |
coretemp or an ACPI sensor which reports values like 46960 mWh. Am I |
23 |
totally out of luck with that one? |
24 |
|
25 |
So, In the end, it's fairly impossible to monitor a CPU's actual |
26 |
temperature in order to keep it below the published maximum? |
27 |
|
28 |
- Grant |