perf stat: Move zero_per_pkg into counter process code
authorJiri Olsa <jolsa@kernel.org>
Fri, 26 Jun 2015 09:29:21 +0000 (11:29 +0200)
committerArnaldo Carvalho de Melo <acme@redhat.com>
Fri, 26 Jun 2015 14:49:45 +0000 (11:49 -0300)
Moving zero_per_pkg into counter process code,
to make the reading path free of processing logic.

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1435310967-14570-17-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/builtin-stat.c

index 74ac92baa2bdfd52a0fc954e164ef7e09a224a1d..b7c184bd559ccc22cb1f65772199bf86b96daf8e 100644 (file)
@@ -285,6 +285,9 @@ static int process_counter(struct perf_evsel *counter)
 
        aggr->val = aggr->ena = aggr->run = 0;
 
+       if (counter->per_pkg)
+               zero_per_pkg(counter);
+
        ret = process_counter_maps(counter);
        if (ret)
                return ret;
@@ -328,9 +331,6 @@ static int read_counter(struct perf_evsel *counter)
        if (counter->system_wide)
                nthreads = 1;
 
-       if (counter->per_pkg)
-               zero_per_pkg(counter);
-
        for (thread = 0; thread < nthreads; thread++) {
                for (cpu = 0; cpu < ncpus; cpu++) {
                        struct perf_counts_values *count;