Re: [PATCH 2/6] perf: Support branch events logging
From: Peter Zijlstra
Date: Fri Apr 14 2023 - 18:01:29 EST
On Fri, Apr 14, 2023 at 04:34:45PM -0400, Liang, Kan wrote:
> > I never understood the whole order thing. What was it trying to do?
>
> Let's say we have three events with the LBR event logging feature as below.
> perf record -e branches,branches,instructions:ppp -j any,event
>
> The counter 0 will be assigned to instructions:ppp, since the PDist is
> only supported on GP 0 & 1.
> The count 1 & 2 will be assigned to the other two branches.
>
> If branches occurs 1 time and instructions occurs 3 times in a LBR
> block, the LBR_INFO will have 0b010111 (counter order).
>
> But as you can see from the perf command, the first event is actually
> branches. Without the event IDs information, perf tool will interpret
> that branches 3 branches 1 and instructions:ppp 1. That's wrong.
>
> If there are multiple users, the situation becomes even worse.
But this makes no sense what so ever; in this case you have no control
over what events you'll actually get in your LBR. Could be none of the
events you're interested in end up in 0-3 but instead land on 4-7.
> >> But if we have two groups with LBR event, the order information is still
> >> required. Why we still want to group things?
> >
> > Why would you need that; what is that whole order nonsense about?
> >
> > {e1, e2, e3, e4}, {e5, e6, e7, e8} with e1 and e5 both having LBR on
> > just works no?
> >
> > Since they have LBR and that extra sample flag they all get a 0-3
> > constraint.
> >
> > Since both e1 and e5 use LBR, they're mutually exclusive, either e1 or
> > e5 group runs.
>
> It's possible that someone pins an event using LBR, and set more than 4
> events for logging, e0:D,{e1, e2},{e3, e4},{e5, e6}. If so, those events
> could do multiplexing. Without the event IDs information, perf tool has
> no idea how to interpret the information.
Yeah, don't do this. There is no guarantee what so ever you'll get any
of those events in the 0-3 range.
You really *must* make then a group such that perf knows what events to
associated with the LBR event and constain them to the 0-3 range of
PMCs.
If you want multiplexing, simply create multiple groups with an LBR
event in them.