[ltp] [PATCH] In-kernel automatic fan control

[Paul RIVIER]

>Oops, you're right of course!
>
>I'll combine this fix with latest version of ibm-acpi (0.12a?) and
>repost patch if you want. 
>
>I also have some thoughts on using fan-speed control feature with
>kautofand. Do you think there could be any advantage in using more
>finegrained fan speed control (than current on/off)?
>  
>
I think that speed control is a smarter way to manage the fan.
It is probably better for it's life time and reduce thermal oscillation 
in components.
But for properly configured laptop, there is not need to trigger at its
maximum speed if not under heavy load.
By and large, my point of view is :
1) lower values of the hysterisis values (max 50, min 44 for exemple)
2) over 50 : fan with speed control
3) if in any case even at low speed, the fan can lower temps under 44°C, 
stop it.

Anything else : as we discussed 3 months ago, the default configuration 
for X
does not include DynamicClock. With this configuration and under certain
conditions, the GPU can outcome 70 °C while the cpu is under 50°C.
There is fortunatly no incident reported yet about it, but it is theoricaly
possible to burn the GPU before the cpu reach 65°C, so with fan off.
Do not forget that in your patch :)

Paul

>Personally I doubt there will be any advantage. Since our criterion of
>optimality is amount of noise produced by fan (integrated over time).
>And this cost function has very high value even for "level 1". In other
>words it is better to use high fan speed for shorter periods of time.
>
>  
>