[ltp] Tracking down a EC firmware bug (fan reading)

[Henrique de Moraes Holschuh]

On Tue, 03 Oct 2006, Enrico Tröger wrote:
> The fan wasn't running when I checked the state. But then I made
> cat /dev/zero | bzip2 >/dev/null
> to produce some heat and checked again, and the status was still
> "disabled" while the fan was running.

That means you *do* have the EC bug.

> Oops, after reading this and after reading again your about your
> three-step-patch I know that I hadn't a fan-control patched kernel.
> After applying the fan-control patch, the status after a cold boot is
> "enabled" at level 7.

After a cold boot, a ThinkPad is usually at "auto" level, not level 7...
unless it is *very* hot (this is something Shem warned me over private
mail), but that's unlikely.

Stock ibm-acpi doesn't know about levels, so it reports the fan as
"disabled" if it is not reporting that it is in "auto" level.

> So, my Thinkpad probably doesn't have the bug? Sorry for confusion.

It does have the bug.

> I guess, your thermal sensor patch is completely independent of this
> behaviour?

Yes, it is completely independent.  It doesn't care about fan control at
all, it just reads the thermal sensors.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh