Re: [RFC PATCH v1 0/3] Scaled statistics using APERF/MPERF in x86

[Vaidyanathan Srinivasan]

* Arjan van de Ven <arjan@xxxxxxxxxxxxx> [2008-05-27 07:15:32]:

> On Tue, 27 May 2008 18:49:26 +0530
> > At any time we will have both the traditional utilisation value
> > relative to current CPU capacity, and scaled utilisation that is
> > relative to maximum CPU capacity.
> > 
> 
> this is where I raise a red flag *because the patch is not doing
> that* !!

Your concern is valid.  I assume you are objecting to the usefulness
and interpretation of the scaled metric and not to the fact that both
scaled and default non scaled metrics are independently available.
 
> sadly it gives a random metric which in some circumstances looks like
> it does that, but in reality it is not doing that.

I agree and I am interested in designing a metric that is useful in
most circumstances.  I am open to suggestions to cover the case of
power constraint and acceleration scenario.

Thanks,
Vaidy

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