[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20210111084328.GG3241@leoy-ThinkPad-X240s>
Date: Mon, 11 Jan 2021 16:43:28 +0800
From: Leo Yan <leo.yan@...aro.org>
To: Namhyung Kim <namhyung@...nel.org>
Cc: Arnaldo Carvalho de Melo <acme@...nel.org>,
Jiri Olsa <jolsa@...hat.com>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>,
Mark Rutland <mark.rutland@....com>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Andi Kleen <ak@...ux.intel.com>,
Ian Rogers <irogers@...gle.com>,
Kan Liang <kan.liang@...ux.intel.com>,
Joe Mario <jmario@...hat.com>, David Ahern <dsahern@...il.com>,
Don Zickus <dzickus@...hat.com>, Al Grant <Al.Grant@....com>,
James Clark <james.clark@....com>,
linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 06/11] perf c2c: Refactor display filter macro
On Wed, Jan 06, 2021 at 04:47:00PM +0900, Namhyung Kim wrote:
> On Sun, Dec 13, 2020 at 10:39 PM Leo Yan <leo.yan@...aro.org> wrote:
> >
> > When sort on the respective metrics (lcl_hitm, rmt_hitm, tot_hitm),
> > macro FILTER_HITM is to filter out the cache line entries if its
> > overhead is less than 1%.
> >
> > This patch is to refactor macro FILTER_HITM. It uses more gernal name
> > FILTER_DISPLAY to replace the old name; and refines its parameter,
> > rather than passing field name for the data structure, it changes to
> > pass the cache line's statistic value and the sum value, this is more
> > flexsible, e.g. if consider to extend for sorting on all load hits
> > which combines multiple fields from structure c2c_stats.
>
> As it doesn't use field names anymore, I think it's better to change it to
> a static function.
Agreed. Will do.
Thanks,
Leo
> >
> > Signed-off-by: Leo Yan <leo.yan@...aro.org>
> > ---
> > tools/perf/builtin-c2c.c | 17 ++++++++++-------
> > 1 file changed, 10 insertions(+), 7 deletions(-)
> >
> > diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c
> > index 5cd30c083d6c..f11c3c84bb2b 100644
> > --- a/tools/perf/builtin-c2c.c
> > +++ b/tools/perf/builtin-c2c.c
> > @@ -2151,24 +2151,27 @@ static bool he__display(struct hist_entry *he, struct c2c_stats *stats)
> >
> > c2c_he = container_of(he, struct c2c_hist_entry, he);
> >
> > -#define FILTER_HITM(__h) \
> > - if (stats->__h) { \
> > - ld_dist = ((double)c2c_he->stats.__h / stats->__h); \
> > +#define FILTER_DISPLAY(val, sum) \
> > +{ \
> > + if ((sum)) { \
> > + ld_dist = ((double)(val) / (sum)); \
> > if (ld_dist < DISPLAY_LINE_LIMIT) \
> > he->filtered = HIST_FILTER__C2C; \
> > } else { \
> > he->filtered = HIST_FILTER__C2C; \
> > - }
> > + } \
> > +}
> >
> > switch (c2c.display) {
> > case DISPLAY_LCL:
> > - FILTER_HITM(lcl_hitm);
> > + FILTER_DISPLAY(c2c_he->stats.lcl_hitm, stats->lcl_hitm);
> > break;
> > case DISPLAY_RMT:
> > - FILTER_HITM(rmt_hitm);
> > + FILTER_DISPLAY(c2c_he->stats.rmt_hitm, stats->rmt_hitm);
> > break;
> > case DISPLAY_TOT:
> > - FILTER_HITM(tot_hitm);
> > + FILTER_DISPLAY(c2c_he->stats.tot_hitm, stats->tot_hitm);
> > + break;
> > default:
> > break;
> > }
> > --
> > 2.17.1
> >
Powered by blists - more mailing lists