Otherwise load counting is an average. Without this change
duration_time in test_memory_bandwidth will alter its value if an
earlier test contains duration_time.
This patch fixes an issue that's introduced in the proposed patch:
https://lore.kernel.org/lkml/[email protected]/
in perf test "Parse and process metrics".
Signed-off-by: Ian Rogers <[email protected]>
---
tools/perf/tests/parse-metric.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/perf/tests/parse-metric.c b/tools/perf/tests/parse-metric.c
index 574b7e4efd3a..07b6f4ec024f 100644
--- a/tools/perf/tests/parse-metric.c
+++ b/tools/perf/tests/parse-metric.c
@@ -109,6 +109,7 @@ static void load_runtime_stat(struct runtime_stat *st, struct evlist *evlist,
struct evsel *evsel;
u64 count;
+ perf_stat__reset_shadow_stats();
evlist__for_each_entry(evlist, evsel) {
count = find_value(evsel->name, vals);
perf_stat__update_shadow_stats(evsel, count, 0, st);
--
2.34.0.rc2.393.gf8c9666880-goog
On Sun, Nov 28, 2021 at 12:58:10AM -0800, Ian Rogers wrote:
> Otherwise load counting is an average. Without this change
> duration_time in test_memory_bandwidth will alter its value if an
> earlier test contains duration_time.
>
> This patch fixes an issue that's introduced in the proposed patch:
> https://lore.kernel.org/lkml/[email protected]/
> in perf test "Parse and process metrics".
>
> Signed-off-by: Ian Rogers <[email protected]>
Acked-by: Jiri Olsa <[email protected]>
jirka
> ---
> tools/perf/tests/parse-metric.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/tools/perf/tests/parse-metric.c b/tools/perf/tests/parse-metric.c
> index 574b7e4efd3a..07b6f4ec024f 100644
> --- a/tools/perf/tests/parse-metric.c
> +++ b/tools/perf/tests/parse-metric.c
> @@ -109,6 +109,7 @@ static void load_runtime_stat(struct runtime_stat *st, struct evlist *evlist,
> struct evsel *evsel;
> u64 count;
>
> + perf_stat__reset_shadow_stats();
> evlist__for_each_entry(evlist, evsel) {
> count = find_value(evsel->name, vals);
> perf_stat__update_shadow_stats(evsel, count, 0, st);
> --
> 2.34.0.rc2.393.gf8c9666880-goog
>
Em Sun, Nov 28, 2021 at 05:29:29PM +0100, Jiri Olsa escreveu:
> On Sun, Nov 28, 2021 at 12:58:10AM -0800, Ian Rogers wrote:
> > Otherwise load counting is an average. Without this change
> > duration_time in test_memory_bandwidth will alter its value if an
> > earlier test contains duration_time.
> >
> > This patch fixes an issue that's introduced in the proposed patch:
> > https://lore.kernel.org/lkml/[email protected]/
> > in perf test "Parse and process metrics".
> >
> > Signed-off-by: Ian Rogers <[email protected]>
>
> Acked-by: Jiri Olsa <[email protected]>
Thanks, applied.
- Arnaldo
> jirka
>
> > ---
> > tools/perf/tests/parse-metric.c | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/tools/perf/tests/parse-metric.c b/tools/perf/tests/parse-metric.c
> > index 574b7e4efd3a..07b6f4ec024f 100644
> > --- a/tools/perf/tests/parse-metric.c
> > +++ b/tools/perf/tests/parse-metric.c
> > @@ -109,6 +109,7 @@ static void load_runtime_stat(struct runtime_stat *st, struct evlist *evlist,
> > struct evsel *evsel;
> > u64 count;
> >
> > + perf_stat__reset_shadow_stats();
> > evlist__for_each_entry(evlist, evsel) {
> > count = find_value(evsel->name, vals);
> > perf_stat__update_shadow_stats(evsel, count, 0, st);
> > --
> > 2.34.0.rc2.393.gf8c9666880-goog
> >
--
- Arnaldo