Re: [RFC][PATCH 1/3] perf: Tighten (and fix) the grouping condition

[Peter Zijlstra <peterz@infradead.org>]

On Fri, Jan 23, 2015 at 03:22:57PM +0000, Mark Rutland wrote:
> > > It seems this would still allow you to group CPU-affine software and
> > > uncore events, which also doesn't make sense: the software events will
> > > count on a single CPU while the uncore events aren't really CPU-affine.
> > > 
> > > Which isn't anything against this patch, but probably something we
> > > should tighten up too.
> > 
> > Indeed, that would need a wee bit of extra infrastructure though; as we
> > cannot currently distinguish between regular cpuctx and uncore like
> > things.
> 
> Isn't the event->pmu->task_ctx_nr sufficient, as with how we identify
> software events?
> 
> Or am I making some completely bogus assumptions in the diff below?

>  	/*
> +	 * System-wide (A.K.A. "uncore") events cannot be associated with a
> +	 * particular CPU, and hence cannot be associated with a particular
> +	 * task either. It's non-sensical to group them with other event types,
> +	 * which are CPU or task bound.
> +	 */

So I think we want to allow grouping software events with say uncore
events; if you start them both out on the same 'cpu'
perf_pmu_migrate_context() would move the software event along with it.

The use case is for non-sampling uncores, where if you have a software
leader you can still get a periodic samples. Clearly looking at task
state or the like is pointless, but PERF_SAMPLE_READ is useful to record
values at regular intervals into the buffer.

But yes, I think ctx_nr might just do.