Re: [PATCH V2 1/6] perf,core: allow invalid context events to be part of sw/hw groups

[Andi Kleen <andi@firstfloor.org>]

> If you don't care about sampling and only care about totals, then you
> can just open the events concurrently *without* grouping them, as I
> stated previously.

perf record doesn't really support that. We need some group reader
that runs regularly. The best choice for the leader is a CPU sample event,
which also has the advantage that it won't disturb idle states.

> > > From my PoV that violates group semantics, because now the events aren't
> > > always counting at the same time (which would be the reason I grouped
> > > them in the first place).
> > 
> > You never use the absolute value, just differences. The differences 
> > effectively count only when the group runs.
> 
> Except that the uncore PMU is counting during the periods the CPU PMU is
> disabled (e.g. when it is being programmed, read, or written). There's a
> race there that you cannot solve given the two are indepedent agents.

We get useful information, like

"memory bandwidth during the group run time".

You're right formulas between different PMUs are problematic though.
But in most cases time relation is good enough.

-Andi

-- 
ak@linux.intel.com -- Speaking for myself only.