From: Weilin Wang <weilin.wang@intel.com>
Add print out functions so that users could read retire latency values.
Example output:
Performance counter stats for 'system wide':
25,717 MEM_INST_RETIRED.SPLIT_STORES # 2.2 % tma_split_stores
28,365,080 CPU_CLK_UNHALTED.THREAD
24.00 MEM_INST_RETIRED.SPLIT_STORES:p # 96 4
2.054365083 seconds time elapsed
This output of retire latency data is in format:
<val> <event-name> # <sum> <count>.
Signed-off-by: Weilin Wang <weilin.wang@intel.com>
Reviewed-by: Ian Rogers <irogers@google.com>
---
tools/perf/util/stat-display.c | 69 ++++++++++++++++++++++++++++++++++
1 file changed, 69 insertions(+)
diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
index bfc1d705f437..c32d484b53a9 100644
--- a/tools/perf/util/stat-display.c
+++ b/tools/perf/util/stat-display.c
@@ -21,6 +21,7 @@
#include "iostat.h"
#include "pmu.h"
#include "pmus.h"
+#include "metricgroup.h"
#define CNTR_NOT_SUPPORTED "<not supported>"
#define CNTR_NOT_COUNTED "<not counted>"
@@ -34,6 +35,7 @@
#define COMM_LEN 16
#define PID_LEN 7
#define CPUS_LEN 4
+#define RETIRE_LEN 8
static int aggr_header_lens[] = {
[AGGR_CORE] = 18,
@@ -426,6 +428,71 @@ static void print_metric_std(struct perf_stat_config *config,
fprintf(out, " %-*s", METRIC_LEN - n - 1, unit);
}
+static void print_retire_lat_std(struct perf_stat_config *config,
+ struct outstate *os)
+{
+ FILE *out = os->fh;
+ bool newline = os->newline;
+ struct tpebs_retire_lat *t;
+ struct list_head *retire_lats = &config->tpebs_results;
+
+ list_for_each_entry(t, retire_lats, event.nd) {
+ if (newline)
+ do_new_line_std(config, os);
+ fprintf(out, "%'*.2f ", COUNTS_LEN, t->val);
+ /* For print alignment */
+ fprintf(out, "%-*s ", config->unit_width, "");
+ fprintf(out, "%-*s", EVNAME_LEN, t->event.name);
+ fprintf(out, " # ");
+ fprintf(out, "%*d %*ld\n", RETIRE_LEN, t->sum,
+ RETIRE_LEN, t->count);
+ }
+}
+
+static void print_retire_lat_csv(struct perf_stat_config *config,
+ struct outstate *os)
+{
+ FILE *out = os->fh;
+ struct tpebs_retire_lat *t;
+ struct list_head *retire_lats = &config->tpebs_results;
+ const char *sep = config->csv_sep;
+
+ list_for_each_entry(t, retire_lats, event.nd) {
+ fprintf(out, "%f%s%s%s%s%ld%s%d\n", t->val, sep, sep, t->event.name, sep,
+ t->count, sep, t->sum);
+ }
+}
+
+static void print_retire_lat_json(struct perf_stat_config *config,
+ struct outstate *os)
+{
+ FILE *out = os->fh;
+ struct tpebs_retire_lat *t;
+ struct list_head *retire_lats = &config->tpebs_results;
+
+ fprintf(out, "{");
+ list_for_each_entry(t, retire_lats, event.nd) {
+ fprintf(out, "\"retire_latency-value\" : \"%f\", ", t->val);
+ fprintf(out, "\"event-name\" : \"%s\"", t->event.name);
+ fprintf(out, "\"sample-counts\" : \"%ld\"", t->count);
+ fprintf(out, "\"retire_latency-sum\" : \"%d\"", t->sum);
+ }
+ fprintf(out, "}");
+}
+
+static void print_retire_lat(struct perf_stat_config *config,
+ struct outstate *os)
+{
+ if (!&config->tpebs_results)
+ return;
+ if (config->json_output)
+ print_retire_lat_json(config, os);
+ else if (config->csv_output)
+ print_retire_lat_csv(config, os);
+ else
+ print_retire_lat_std(config, os);
+}
+
static void new_line_csv(struct perf_stat_config *config, void *ctx)
{
struct outstate *os = ctx;
@@ -1609,6 +1676,8 @@ void evlist__print_counters(struct evlist *evlist, struct perf_stat_config *conf
break;
}
+ print_retire_lat(config, &os);
+
print_footer(config);
fflush(config->output);
--
2.43.0On Tue, Apr 2, 2024 at 2:46 PM <weilin.wang@intel.com> wrote:
>
> From: Weilin Wang <weilin.wang@intel.com>
>
> Add print out functions so that users could read retire latency values.
>
> Example output:
>
> Performance counter stats for 'system wide':
>
> 25,717 MEM_INST_RETIRED.SPLIT_STORES # 2.2 % tma_split_stores
> 28,365,080 CPU_CLK_UNHALTED.THREAD
> 24.00 MEM_INST_RETIRED.SPLIT_STORES:p # 96 4
>
> 2.054365083 seconds time elapsed
>
> This output of retire latency data is in format:
> <val> <event-name> # <sum> <count>.
Nop, please follow the perf stat output format. It'd be
<sum> <event-name> # ...
Hmm.. maybe you don't need this at all as it'll have the event
anyway (probably without the 'p' modifier like in the example
above).
Then I think you can just add a metric value saying the average
latency in the comment area.
Performance counter stats for 'system wide':
25,717 MEM_INST_RETIRED.SPLIT_STORES # 2.2 %
tma_split_stores
#
24.0 average latency
28,365,080 CPU_CLK_UNHALTED.THREAD
Thanks,
Namhyung
>
> Signed-off-by: Weilin Wang <weilin.wang@intel.com>
> Reviewed-by: Ian Rogers <irogers@google.com>
On Mon, Apr 22, 2024 at 12:55 PM Namhyung Kim <namhyung@kernel.org> wrote: > > On Tue, Apr 2, 2024 at 2:46 PM <weilin.wang@intel.com> wrote: > > > > From: Weilin Wang <weilin.wang@intel.com> > > > > Add print out functions so that users could read retire latency values. > > > > Example output: > > > > Performance counter stats for 'system wide': > > > > 25,717 MEM_INST_RETIRED.SPLIT_STORES # 2.2 % tma_split_stores > > 28,365,080 CPU_CLK_UNHALTED.THREAD > > 24.00 MEM_INST_RETIRED.SPLIT_STORES:p # 96 4 > > > > 2.054365083 seconds time elapsed > > > > This output of retire latency data is in format: > > <val> <event-name> # <sum> <count>. > > Nop, please follow the perf stat output format. It'd be > > <sum> <event-name> # ... > > Hmm.. maybe you don't need this at all as it'll have the event > anyway (probably without the 'p' modifier like in the example > above). > > Then I think you can just add a metric value saying the average > latency in the comment area. > > Performance counter stats for 'system wide': > > 25,717 MEM_INST_RETIRED.SPLIT_STORES # 2.2 % > tma_split_stores > # > 24.0 average latency Sorry for the line wrapping. They should be in the same line. Thanks, Namhyung
© 2016 - 2026 Red Hat, Inc.