On Mon March 26 2007 3:02 am, Stephane Eranian wrote:
> > I can see why one might want to use the 16-bit counters if it doubles the
> > number of events available. However, going from 4 32-bit counters to 8
> > 16-bit counters is going to possibly cause a lot more interrupts. Double
> > the number of interrupt sources and each interrupt at 2^16 times original
> > the frequency.
>
> I would go with 4 32-bit counters. Yet I am wondering if they do not have
> event restrictions which would make this difficult to hardcode. For
> instance, if event A can only be measured on counter C which is only
> "visible" in the 8-counter configuration.

The Cell PMU shouldn't have any restrictions like what you're describing. Each 
of the eight counters can connect to any of the PMU input-bus bits (which 
carry the event signals).

However, I now remember one situation that requires the first 32-bit counter 
(pm0) to be used as two 16-bit counters (pm0 and pm4). There are bits in one 
of the "global" control registers that allow the counters to be started when 
pm0 overflows, and/or to be stopped when pm4 overflows.

-- 
Kevin Corry
[EMAIL PROTECTED]
http://www.ibm.com/linux/
_______________________________________________
perfmon mailing list
[email protected]
http://www.hpl.hp.com/hosted/linux/mail-archives/perfmon/

Reply via email to