Re: [PATCH v1] perf stat: Don't display zero tool counts

[Andi Kleen]

On Mon, Aug 07, 2023 at 04:57:31PM -0300, Arnaldo Carvalho de Melo wrote:
> Em Mon, Aug 07, 2023 at 04:43:38PM -0300, Arnaldo Carvalho de Melo escreveu:
> > Em Mon, Aug 07, 2023 at 07:54:47AM -0700, Andi Kleen escreveu:
> > > On Thu, Aug 03, 2023 at 05:54:59PM -0300, Arnaldo Carvalho de Melo wrote:
> > > > Em Tue, Aug 01, 2023 at 01:54:52PM -0700, Ian Rogers escreveu:
> > > > > Skip zero counts for tool events.
> > > > > 
> > > > > Reported-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
> > > > 
> > > > Andi,
> > > > 
> > > > 	Have you tested this? Can we please have your Tested-by?
> > > 
> > > I thought I had sent it earlier?
> > > 
> > > Tested-by: Andi Kleen <ak@xxxxxxxxxxxxxxx>
> > 
> > Yeah, you did it, sorry, somehow I didn't notice.
> > 
> > Applying.
> 
> Would be good to have the original link with your report and to figure
> out the cset that introduced the problem, so that we could have a Fixes
> tag to help justifying getting this into 6.5.

Just bisected it. The original patch was below. Remarkably it had a "Fixes"
tag too)

My report was 

https://lore.kernel.org/linux-perf-users/ZMlrzcVrVi1lTDmn@tassilo/

commit b897613510890d6e92b6a276a20f6c3d96fe90e8
Author: Namhyung Kim <namhyung@xxxxxxxxxx>
Date:   Tue Dec 6 09:58:04 2022 -0800

    perf stat: Update event skip condition for system-wide per-thread mode and merged uncore and hybrid events

    In print_counter_aggrdata(), it skips some events that has no aggregate
    count.  It's actually for system-wide per-thread mode and merged uncore
    and hybrid events.

    Let's update the condition to check them explicitly.

    Fixes: 91f85f98da7ab8c3 ("perf stat: Display event stats using aggr counts")