Re: [PATCH] perf_events: improve x86 event scheduling (v5)

[Frederic Weisbecker <fweisbec@gmail.com>]

On Mon, Jan 18, 2010 at 03:37:01PM +0100, Peter Zijlstra wrote:
> On Mon, 2010-01-18 at 15:20 +0100, Frederic Weisbecker wrote:
> > 
> > > Then there's still the question of having events of multiple hw pmus in
> > > a single group, I'd be perfectly fine with saying that's not allowed,
> > > what to others think?
> > 
> > 
> > I guess we need that. It can be insteresting to couple
> > hardware counters with memory accesses...or whatever.
> 
> That really depends on how easy it is to correlate events from the
> various pmus. This case could indeed do that, but the core vs uncore
> tihng is a lot less clear.



Not sure what you both mean by this core VS uncore thing :)
Is it about hardware counters that apply to single hardware threads
or shared among them inside a same core?


 
> > Perf stat combines cache miss counting with page faults,
> > cpu clock counters.
> 
> perf stat also doesn't use groups and it still works quite nicely.


Ah? I thought it does.


 
> > We shouldn't limit such possibilities for technical/cleanliness
> > reasons. We should rather adapt.
> 
> Maybe, I'm not a very big fan of groups myself, but they are clearly
> useful within a pmu, measuring cache misses through total-access for
> example, but the use between pmus is questionable.


Cross pmu, these seem to only make sense for non pinned groups.
If you want two non-pinned counters to be paired and not randomly
and separately scheduled.

For other cases, indeed I'm not sure it is useful :)