On Thu, Jun 16, 2016 at 01:38:13PM +0800, Huang Rui wrote:
> On Wed, Jun 15, 2016 at 09:13:59PM -0400, Vince Weaver wrote:
> >
> > 2. Unless I'm misunderstanding things, the code seems to be accumulating
> > Power. (see chunk below) Power is an instantaneous measurement, it
> > makes no sense to add values. If you use 5W for 1ms and 10W for
> > 1ms, the average power across the 2ms interval is not 15W.
> >
> > You can add energy, but not power.
> >
> > > + delta *= cpu_pwr_sample_ratio * 1000;
> > > + tdelta = new_ptsc - prev_ptsc;
> > > +
> > > + do_div(delta, tdelta);
> > > + local64_add(delta, &event->count);
> >
>
> You're right. Nice catch! The average power is per compute unit. We
> cannot add the power simplely for each processor/package.
>
> So here, the average power per package should be (delta1 + delta2 + ... +
> deltaN)/(tdelta_avg).
> I will work out a fix. Thanks to point out.
>
After considering carefully, the original method should be OK.
AMD nomenclature for CMT systems:
[node 0] -> [Compute Unit 0] -> [Compute Unit Core 0] -> Linux CPU 0
-> [Compute Unit Core 1] -> Linux CPU 1
-> [Compute Unit 1] -> [Compute Unit Core 0] -> Linux CPU 2
-> [Compute Unit Core 1] -> Linux CPU 3
The deltaN is power per compute unit. Current one package has two CUs.
In the *same* interval, CU0's power is 10W, CU1's power is 15W. The
package (CU0 + CU1) power should be 15W, right? Because the interval
is the same.
Q = Q1 + Q2. P = Q/t = (Q1 + Q2)/t = Q1/t + Q2/t = P1 + P2.
Is that clear?
Thanks,
Rui
