Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>
Date:   Thu, 16 Apr 2020 11:57:04 -0300
To:     Adrian Hunter <adrian.hunter@intel.com>
Cc:     Jiri Olsa <jolsa@redhat.com>, Andi Kleen <ak@linux.intel.com>,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 10/16] perf tools: Add support for synthesized sample type
Message-ID: <20200416145704.GA1842@kernel.org>
References: <20200401101613.6201-1-adrian.hunter@intel.com>
 <20200401101613.6201-11-adrian.hunter@intel.com>
 <20200416145426.GA31666@kernel.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200416145426.GA31666@kernel.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Em Thu, Apr 16, 2020 at 11:54:26AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Wed, Apr 01, 2020 at 01:16:07PM +0300, Adrian Hunter escreveu:
> > For reporting purposes, an evsel sample can have a callchain synthesized
> > from AUX area data. Add support for keeping track of synthesized sample
> > types. Note, the recorded sample_type cannot be changed because it is
> > needed to continue to parse events.
> > 
> > Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> > ---
> >  tools/perf/util/evsel.c |  2 +-
> >  tools/perf/util/evsel.h | 15 ++++++++++++++-
> >  2 files changed, 15 insertions(+), 2 deletions(-)
> > 
> > diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
> > index eb880efbce16..60e6cd49dee3 100644
> > --- a/tools/perf/util/evsel.c
> > +++ b/tools/perf/util/evsel.c
> > @@ -2136,7 +2136,7 @@ int perf_evsel__parse_sample(struct evsel *evsel, union perf_event *event,
> >  		}
> >  	}
> >  
> > -	if (evsel__has_callchain(evsel)) {
> > +	if (type & PERF_SAMPLE_CALLCHAIN) {
> >  		const u64 max_callchain_nr = UINT64_MAX / sizeof(u64);
> 
> This ends up looking unrelated, I had to go and look at the source to
> see that this is just a simplification, i.e. earlier in this function
> (perf_evsel__parse_sample) we have:
> 
>         u64 type = evsel->core.attr.sample_type;
> 
> So the above doesn't change anything, good, but slowed reviewing a bit,
> please consider next time to have this in a separate patch, I'll do it
> this time.

I've added this as the cset comment:

Using 'type' variable for checking for callchains is equivalent to using
evsel__has_callchain(evsel) and is how the other PERF_SAMPLE_ bits are checked
in this function, so use it to be consistent.
 
> Thanks,
> 
> - Arnaldo
>   
> >  		OVERFLOW_CHECK_u64(array);
> > diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
> > index 53187c501ee8..e64ed4202cab 100644
> > --- a/tools/perf/util/evsel.h
> > +++ b/tools/perf/util/evsel.h
> > @@ -104,6 +104,14 @@ struct evsel {
> >  		perf_evsel__sb_cb_t	*cb;
> >  		void			*data;
> >  	} side_band;
> > +	/*
> > +	 * For reporting purposes, an evsel sample can have a callchain
> > +	 * synthesized from AUX area data. Keep track of synthesized sample
> > +	 * types here. Note, the recorded sample_type cannot be changed because
> > +	 * it is needed to continue to parse events.
> > +	 * See also evsel__has_callchain().
> > +	 */
> > +	__u64			synth_sample_type;
> >  };
> >  
> >  struct perf_missing_features {
> > @@ -398,7 +406,12 @@ static inline bool perf_evsel__has_branch_hw_idx(const struct evsel *evsel)
> >  
> >  static inline bool evsel__has_callchain(const struct evsel *evsel)
> >  {
> > -	return (evsel->core.attr.sample_type & PERF_SAMPLE_CALLCHAIN) != 0;
> > +	/*
> > +	 * For reporting purposes, an evsel sample can have a recorded callchain
> > +	 * or a callchain synthesized from AUX area data.
> > +	 */
> > +	return evsel->core.attr.sample_type & PERF_SAMPLE_CALLCHAIN ||
> > +	       evsel->synth_sample_type & PERF_SAMPLE_CALLCHAIN;
> >  }
> >  
> >  struct perf_env *perf_evsel__env(struct evsel *evsel);
> > -- 
> > 2.17.1
> > 
> 
> -- 
> 
> - Arnaldo

-- 

- Arnaldo