Spurious thermal shutdowns on Dell Studio 1557

[Bartosz Fabianowski]

> You should hear the fan come on at 55C, and run faster at 71C, as
> active cooling is also on.  Is that happening?

Temperature never drops below 55°C and so the fan is always running. As 
far as memory serves me, this has been the case since day one. I have 
not noticed any difference in fan noise at different temperature levels. 
Considering how hot this machine is running, I would not be surprised if 
the fan was always in its "high" setting.

> If the fan is running properly, have you checked that the airways
> aren't blocked by dust and/or fluff?

The laptop has feet of ~5mm height. I have extended that to about 1cm by 
placing the computer on four stacks of post-it notes. This, again, has 
been the case since day one. I added a few more millimeters of post-it 
notes today and cleaned out any dust I could find. This might have 
improved things slightly but temperature still exceeds 85°C when 
compiling larger ports.

> Maybe Dell don't trust non-Windows OS to take care of passive
> cooling? Could be worth seeing if those values come up the same if
> you spoof some type of Windows as your OS, as detailed in the ACPI
> debugging section?

Yes, this is definitely something I will try. I thought that all the 
magic would be present in the DSDT. But now that I think about it, of 
course the BIOS could be changing the values read from the GNVS table 
based on OS. I will try spoofing Vista or Windows 7 and report back my 
findings.

- Bartosz