Re: [PATCH][RFC] P4/Xeon Thermal LVT support

Maciej W. Rozycki (macro@ds2.pg.gda.pl)
Tue, 2 Apr 2002 13:32:23 +0200 (MET DST)


On Sat, 30 Mar 2002, Pavel Machek wrote:

> Overheat is not neccessarily hardware failure.

It is the failing condition. It need not be a result of a hardware
failure.

> You see, I have a notebook. I put pen in it to stop the fan. Hardware
> is pretty much okay, but, well, pen does not allow fan to spin.

A fan blockage is a hardware failure as well. Regardless of the reason.
It certainly isn't a software failure.

> What's the best behaviour? Throttle is okay.

Sure that is a way to protect the CPU but it may fail if the reason is
not heat emitted by the CPU.

> And now, you have fire at server room. All cpus throtle, then are
> burn. Does it matter if they throttled? No.

But it matters if an operator got warned before (that is what I remarked
originally). The operator may be in a distant location. Or he may be
nearby and be able to act to stop the fire once he gets a message.

> So it seems to me throttle is always right answer.

Sure it is a way to try to recover, if hardware provides it, but it's
completely orthogonal to the question whether to report a thermal problem
or not and at which priority.

-- 
+  Maciej W. Rozycki, Technical University of Gdansk, Poland   +
+--------------------------------------------------------------+
+        e-mail: macro@ds2.pg.gda.pl, PGP key available        +

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/