Skip to contentSkip to conversation

perf: Add --pmu-filter option for filtering PMUs

PATCH Applied Qinxin Xia 3 participants 8 messages Mar 10 – Mar 27 perf events

lore
Share & tools
lore

Qinxin Xia

author
#0 lore raw

v2 cover perf: Add --pmu-filter option for filtering PMUs

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

    19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
    19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
    19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
        75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
        18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
    19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
      19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

     6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
        50,548,465      cpa_p0_wr_dat
         7,552,182      cpa_p0_rd_dat_64b
                 0      cpa_p0_rd_dat_32b
       6.234139320 seconds time elapsed

wildcard match:
[root@localhost ~]# ./perf stat --pmu-filter hisi_sicl[2,8]_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

    18,505,025,031      hisi_sicl2_cpa0/cpa_cycles/      #     0.42 cpa_p0_avg_bw
        80,007,944      hisi_sicl2_cpa0/cpa_p0_wr_dat/
        42,014,906      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
    18,504,957,351      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
      18.504994880 seconds time elapsed

Changes since V1:
- Address the comments from Namhyung, the description of the event filter in commit info is omitted, and wildcard matching is supported for pmu-filter.
- Link: https://lore.kernel.org/all/[email protected]/

Changes since RFC:
- Address the comments from Ian, replace '--uncorepmu' with '--pmu-filter'.
- Link: https://lore.kernel.org/linux-perf-users/[email protected]/

Qinxin Xia (1):
perf: Add --pmu-filter option for filtering PMUs

 tools/perf/Documentation/perf-stat.txt |  4 ++++
 tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
 tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
 tools/perf/util/parse-events.c         |  2 +-
 4 files changed, 37 insertions(+), 6 deletions(-)
Signature · 1 line
2.33.0
author
#1 lore raw

PATCH v2 1/1 perf: Add --pmu-filter option for filtering PMUs

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

    19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
    19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
    19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
        75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
        18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
    19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
      19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

     6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
        50,548,465      cpa_p0_wr_dat
         7,552,182      cpa_p0_rd_dat_64b
                 0      cpa_p0_rd_dat_32b
       6.234139320 seconds time elapsed
---
 tools/perf/Documentation/perf-stat.txt |  4 ++++
 tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
 tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
 tools/perf/util/parse-events.c         |  2 +-
 4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
 Only enable events on applying cpu with this type for hybrid platform
 (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
 EXAMPLES
 --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
 	return 0;
 }
 
+static int parse_pmu_filter(const struct option *opt,
+			   const char *str,
+			   int unset __maybe_unused)
+{
+	struct evlist *evlist = *(struct evlist **)opt->value;
+
+	if (!list_empty(&evlist->core.entries)) {
+		fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+		return -1;
+	}
+
+	parse_events_option_args.pmu_filter = str;
+	return 0;
+}
+
 static int parse_cache_level(const struct option *opt,
 			     const char *str,
 			     int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
 			"Only enable events on applying cpu with this type "
 			"for hybrid platform (e.g. core or atom)",
 			parse_cputype),
+		OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+			"Only enable events on applying pmu with specified "
+			"for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+			parse_pmu_filter),
 #ifdef HAVE_LIBPFM
 		OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
 			"libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
 				      const char *metric_or_groups)
 {
 	const char *pm_pmu = pm->pmu ?: "cpu";
+	struct perf_pmu *perf_pmu = NULL;
 
-	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+	if (pm->pmu)
+		perf_pmu = perf_pmus__find(pm->pmu);
+
+	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+	   (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
 		return false;
 
 	return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
 static int parse_ids(bool metric_no_merge, bool fake_pmu,
 		     struct expr_parse_ctx *ids, const char *modifier,
 		     bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-		     struct evlist **out_evlist)
+		     struct evlist **out_evlist,
+		     const char *filter_pmu)
 {
 	struct parse_events_error parse_error;
 	struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
 	}
 	pr_debug("Parsing metric events '%s'\n", events.buf);
 	parse_events_error__init(&parse_error);
-	ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+	ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
 			     &parse_error, fake_pmu, /*warn_if_reordered=*/false,
 			     /*fake_tp=*/false);
 	if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
 					/*modifier=*/NULL,
 					/*group_events=*/false,
 					tool_events,
-					&combined_evlist);
+					&combined_evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
 		}
 		if (combined)
 			expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
 		}
 		if (!metric_evlist) {
 			ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-					m->group_events, tool_events, &m->evlist);
+					m->group_events, tool_events, &m->evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
 			if (ret)
 				goto out;
 
tools/perf/util/parse-events.c +1 -1
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
 	if (parse_state->pmu_filter == NULL)
 		return false;
 
-	return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+	return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
 }
 
 static int parse_events_add_pmu(struct parse_events_state *parse_state,
Signature · 1 line
2.33.0
maintainer
#2 lore raw
Quoted 29 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

    19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
    19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
    19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
        75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
        18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
    19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                 0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                 0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
      19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

Quoted 6 lines
     6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
        50,548,465      cpa_p0_wr_dat
         7,552,182      cpa_p0_rd_dat_64b
                 0      cpa_p0_rd_dat_32b
       6.234139320 seconds time elapsed

Ian, can you please take a look at this?

Thanks,
Namhyung

Quoted 135 lines
---
 tools/perf/Documentation/perf-stat.txt |  4 ++++
 tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
 tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
 tools/perf/util/parse-events.c         |  2 +-
 4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
 Only enable events on applying cpu with this type for hybrid platform
 (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
 EXAMPLES
 --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
 	return 0;
 }
 
+static int parse_pmu_filter(const struct option *opt,
+			   const char *str,
+			   int unset __maybe_unused)
+{
+	struct evlist *evlist = *(struct evlist **)opt->value;
+
+	if (!list_empty(&evlist->core.entries)) {
+		fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+		return -1;
+	}
+
+	parse_events_option_args.pmu_filter = str;
+	return 0;
+}
+
 static int parse_cache_level(const struct option *opt,
 			     const char *str,
 			     int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
 			"Only enable events on applying cpu with this type "
 			"for hybrid platform (e.g. core or atom)",
 			parse_cputype),
+		OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+			"Only enable events on applying pmu with specified "
+			"for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+			parse_pmu_filter),
 #ifdef HAVE_LIBPFM
 		OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
 			"libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
 				      const char *metric_or_groups)
 {
 	const char *pm_pmu = pm->pmu ?: "cpu";
+	struct perf_pmu *perf_pmu = NULL;
 
-	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+	if (pm->pmu)
+		perf_pmu = perf_pmus__find(pm->pmu);
+
+	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+	   (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
 		return false;
 
 	return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
 static int parse_ids(bool metric_no_merge, bool fake_pmu,
 		     struct expr_parse_ctx *ids, const char *modifier,
 		     bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-		     struct evlist **out_evlist)
+		     struct evlist **out_evlist,
+		     const char *filter_pmu)
 {
 	struct parse_events_error parse_error;
 	struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
 	}
 	pr_debug("Parsing metric events '%s'\n", events.buf);
 	parse_events_error__init(&parse_error);
-	ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+	ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
 			     &parse_error, fake_pmu, /*warn_if_reordered=*/false,
 			     /*fake_tp=*/false);
 	if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
 					/*modifier=*/NULL,
 					/*group_events=*/false,
 					tool_events,
-					&combined_evlist);
+					&combined_evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
 		}
 		if (combined)
 			expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
 		}
 		if (!metric_evlist) {
 			ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-					m->group_events, tool_events, &m->evlist);
+					m->group_events, tool_events, &m->evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
 			if (ret)
 				goto out;
 
tools/perf/util/parse-events.c +1 -1
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
 	if (parse_state->pmu_filter == NULL)
 		return false;
 
-	return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+	return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
 }
 
 static int parse_events_add_pmu(struct parse_events_state *parse_state,
Signature · 1 line
2.33.0
author
#3 lore raw
Quoted 169 lines · Namhyung Kim
Quoted 29 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

     19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
     19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
     19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
         75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
         18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
     19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
       19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

Quoted 6 lines
      6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
         50,548,465      cpa_p0_wr_dat
          7,552,182      cpa_p0_rd_dat_64b
                  0      cpa_p0_rd_dat_32b
        6.234139320 seconds time elapsed

Ian, can you please take a look at this?

Thanks,
Namhyung

Quoted 127 lines
---
  tools/perf/Documentation/perf-stat.txt |  4 ++++
  tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
  tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
  tools/perf/util/parse-events.c         |  2 +-
  4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
  Only enable events on applying cpu with this type for hybrid platform
  (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
  EXAMPLES
  --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
  	return 0;
  }
 
+static int parse_pmu_filter(const struct option *opt,
+			   const char *str,
+			   int unset __maybe_unused)
+{
+	struct evlist *evlist = *(struct evlist **)opt->value;
+
+	if (!list_empty(&evlist->core.entries)) {
+		fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+		return -1;
+	}
+
+	parse_events_option_args.pmu_filter = str;
+	return 0;
+}
+
  static int parse_cache_level(const struct option *opt,
  			     const char *str,
  			     int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
  			"Only enable events on applying cpu with this type "
  			"for hybrid platform (e.g. core or atom)",
  			parse_cputype),
+		OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+			"Only enable events on applying pmu with specified "
+			"for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+			parse_pmu_filter),
  #ifdef HAVE_LIBPFM
  		OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
  			"libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
  				      const char *metric_or_groups)
  {
  	const char *pm_pmu = pm->pmu ?: "cpu";
+	struct perf_pmu *perf_pmu = NULL;
 
-	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+	if (pm->pmu)
+		perf_pmu = perf_pmus__find(pm->pmu);
+
+	if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+	   (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
  		return false;
 
  	return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
  static int parse_ids(bool metric_no_merge, bool fake_pmu,
  		     struct expr_parse_ctx *ids, const char *modifier,
  		     bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-		     struct evlist **out_evlist)
+		     struct evlist **out_evlist,
+		     const char *filter_pmu)
  {
  	struct parse_events_error parse_error;
  	struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
  	}
  	pr_debug("Parsing metric events '%s'\n", events.buf);
  	parse_events_error__init(&parse_error);
-	ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+	ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
  			     &parse_error, fake_pmu, /*warn_if_reordered=*/false,
  			     /*fake_tp=*/false);
  	if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
  					/*modifier=*/NULL,
  					/*group_events=*/false,
  					tool_events,
-					&combined_evlist);
+					&combined_evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
  		}
  		if (combined)
  			expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
  		}
  		if (!metric_evlist) {
  			ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-					m->group_events, tool_events, &m->evlist);
+					m->group_events, tool_events, &m->evlist,
+					(pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
  			if (ret)
  				goto out;
 
tools/perf/util/parse-events.c +0 -0
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
  	if (parse_state->pmu_filter == NULL)
  		return false;
 
Quoted 9 lines
Quoted 8 lines
tools/perf/util/parse-events.c +1 -1
-	return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+	return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
  }
 
  static int parse_events_add_pmu(struct parse_events_state *parse_state,
Signature · 1 line
2.33.0

Gentle ping.

Signature · 2 lines
Thanks,
Qinxin
reviewer Reviewed-by
#4 lore raw
Quoted 40 lines · Qinxin Xia
Quoted 37 lines · Namhyung Kim
Quoted 35 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

     19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
     19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
     19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
         75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
         18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
     19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                  0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                  0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                  0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
       19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

      6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
         50,548,465      cpa_p0_wr_dat
          7,552,182      cpa_p0_rd_dat_64b
                  0      cpa_p0_rd_dat_32b
        6.234139320 seconds time elapsed
Quoted 7 lines
Quoted 7 lines

Ian, can you please take a look at this?

Thanks,
Namhyung

Quoted 2 lines

Sorry for the delay.

Thanks,
Ian

Quoted 140 lines
Quoted 125 lines
Quoted 125 lines
---
  tools/perf/Documentation/perf-stat.txt |  4 ++++
  tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
  tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
  tools/perf/util/parse-events.c         |  2 +-
  4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
  Only enable events on applying cpu with this type for hybrid platform
  (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
  EXAMPLES
  --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
     return 0;
  }
 
+static int parse_pmu_filter(const struct option *opt,
+                       const char *str,
+                       int unset __maybe_unused)
+{
+    struct evlist *evlist = *(struct evlist **)opt->value;
+
+    if (!list_empty(&evlist->core.entries)) {
+            fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+            return -1;
+    }
+
+    parse_events_option_args.pmu_filter = str;
+    return 0;
+}
+
  static int parse_cache_level(const struct option *opt,
                          const char *str,
                          int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
                     "Only enable events on applying cpu with this type "
                     "for hybrid platform (e.g. core or atom)",
                     parse_cputype),
+            OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+                    "Only enable events on applying pmu with specified "
+                    "for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+                    parse_pmu_filter),
  #ifdef HAVE_LIBPFM
             OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
                     "libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
                                   const char *metric_or_groups)
  {
     const char *pm_pmu = pm->pmu ?: "cpu";
+    struct perf_pmu *perf_pmu = NULL;
 
-    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+    if (pm->pmu)
+            perf_pmu = perf_pmus__find(pm->pmu);
+
+    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+       (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
             return false;
 
     return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
  static int parse_ids(bool metric_no_merge, bool fake_pmu,
                  struct expr_parse_ctx *ids, const char *modifier,
                  bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-                 struct evlist **out_evlist)
+                 struct evlist **out_evlist,
+                 const char *filter_pmu)
  {
     struct parse_events_error parse_error;
     struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
     }
     pr_debug("Parsing metric events '%s'\n", events.buf);
     parse_events_error__init(&parse_error);
-    ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+    ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
                          &parse_error, fake_pmu, /*warn_if_reordered=*/false,
                          /*fake_tp=*/false);
     if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
                                     /*modifier=*/NULL,
                                     /*group_events=*/false,
                                     tool_events,
-                                    &combined_evlist);
+                                    &combined_evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
             }
             if (combined)
                     expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
             }
             if (!metric_evlist) {
                     ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-                                    m->group_events, tool_events, &m->evlist);
+                                    m->group_events, tool_events, &m->evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
                     if (ret)
                             goto out;
 
tools/perf/util/parse-events.c +0 -0
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
     if (parse_state->pmu_filter == NULL)
             return false;
 
Quoted 9 lines
Quoted 8 lines
tools/perf/util/parse-events.c +1 -2
-    return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+    return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
  }
 
  static int parse_events_add_pmu(struct parse_events_state *parse_state,
--

2.33.0

Gentle ping.
--
Thanks,
Qinxin

author
#5 lore raw
Quoted 47 lines · Ian Rogers
Quoted 46 lines · Qinxin Xia
Quoted 43 lines · Namhyung Kim
Quoted 29 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

      19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
      19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
      19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
          75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
          18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
      19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
        19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

Quoted 6 lines
       6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
          50,548,465      cpa_p0_wr_dat
           7,552,182      cpa_p0_rd_dat_64b
                   0      cpa_p0_rd_dat_32b
         6.234139320 seconds time elapsed

Ian, can you please take a look at this?

Thanks,
Namhyung

Quoted 1 line
Quoted 9 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line

Sorry for the delay.

Thanks,
Ian

Hi Namhyung!

Ian has already provided Reviewed-by. Could you please apply this patch if everything looks good to you? :-)

Signature · 2 lines
Thanks,
Qinxin
Quoted 140 lines
Quoted 140 lines
Quoted 125 lines
Quoted 125 lines
---
   tools/perf/Documentation/perf-stat.txt |  4 ++++
   tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
   tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
   tools/perf/util/parse-events.c         |  2 +-
   4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
   Only enable events on applying cpu with this type for hybrid platform
   (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
   EXAMPLES
   --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
      return 0;
   }
 
+static int parse_pmu_filter(const struct option *opt,
+                       const char *str,
+                       int unset __maybe_unused)
+{
+    struct evlist *evlist = *(struct evlist **)opt->value;
+
+    if (!list_empty(&evlist->core.entries)) {
+            fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+            return -1;
+    }
+
+    parse_events_option_args.pmu_filter = str;
+    return 0;
+}
+
   static int parse_cache_level(const struct option *opt,
                           const char *str,
                           int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
                      "Only enable events on applying cpu with this type "
                      "for hybrid platform (e.g. core or atom)",
                      parse_cputype),
+            OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+                    "Only enable events on applying pmu with specified "
+                    "for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+                    parse_pmu_filter),
   #ifdef HAVE_LIBPFM
              OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
                      "libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
                                    const char *metric_or_groups)
   {
      const char *pm_pmu = pm->pmu ?: "cpu";
+    struct perf_pmu *perf_pmu = NULL;
 
-    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+    if (pm->pmu)
+            perf_pmu = perf_pmus__find(pm->pmu);
+
+    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+       (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
              return false;
 
      return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
   static int parse_ids(bool metric_no_merge, bool fake_pmu,
                   struct expr_parse_ctx *ids, const char *modifier,
                   bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-                 struct evlist **out_evlist)
+                 struct evlist **out_evlist,
+                 const char *filter_pmu)
   {
      struct parse_events_error parse_error;
      struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
      }
      pr_debug("Parsing metric events '%s'\n", events.buf);
      parse_events_error__init(&parse_error);
-    ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+    ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
                           &parse_error, fake_pmu, /*warn_if_reordered=*/false,
                           /*fake_tp=*/false);
      if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
                                      /*modifier=*/NULL,
                                      /*group_events=*/false,
                                      tool_events,
-                                    &combined_evlist);
+                                    &combined_evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
              }
              if (combined)
                      expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
              }
              if (!metric_evlist) {
                      ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-                                    m->group_events, tool_events, &m->evlist);
+                                    m->group_events, tool_events, &m->evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
                      if (ret)
                              goto out;
 
tools/perf/util/parse-events.c +0 -0
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
      if (parse_state->pmu_filter == NULL)
              return false;
 
Quoted 9 lines
Quoted 8 lines
tools/perf/util/parse-events.c +1 -2
-    return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+    return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
   }
 
   static int parse_events_add_pmu(struct parse_events_state *parse_state,
--

2.33.0

Gentle ping.
--
Thanks,
Qinxin

reviewer
#6 lore raw
Quoted 61 lines · Qinxin Xia
Quoted 47 lines · Ian Rogers
Quoted 46 lines · Qinxin Xia
Quoted 43 lines · Namhyung Kim
Quoted 29 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

      19,417,779,115      hisi_sicl0_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl0_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl0_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl0_cpa0/cpa_p0_rd_dat_32b/
      19,417,751,103      hisi_sicl10_cpa0/cpa_cycles/     #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl10_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl10_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl10_cpa0/cpa_p0_rd_dat_32b/
      19,417,730,679      hisi_sicl2_cpa0/cpa_cycles/      #     0.31 cpa_p0_avg_bw
          75,635,749      hisi_sicl2_cpa0/cpa_p0_wr_dat/
          18,520,640      hisi_sicl2_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl2_cpa0/cpa_p0_rd_dat_32b/
      19,417,674,227      hisi_sicl8_cpa0/cpa_cycles/      #     0.00 cpa_p0_avg_bw
                   0      hisi_sicl8_cpa0/cpa_p0_wr_dat/
                   0      hisi_sicl8_cpa0/cpa_p0_rd_dat_64b/
                   0      hisi_sicl8_cpa0/cpa_p0_rd_dat_32b/
        19.417734480 seconds time elapsed

[root@localhost tmp]# perf stat --pmu-filter hisi_sicl2_cpa0 -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

Quoted 6 lines
       6,234,093,559      cpa_cycles                       #     0.60 cpa_p0_avg_bw
          50,548,465      cpa_p0_wr_dat
           7,552,182      cpa_p0_rd_dat_64b
                   0      cpa_p0_rd_dat_32b
         6.234139320 seconds time elapsed

Ian, can you please take a look at this?

Thanks,
Namhyung

Quoted 1 line
Quoted 9 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line

Sorry for the delay.

Thanks,
Ian

Quoted 8 lines

Hi Namhyung!

Ian has already provided Reviewed-by. Could you please apply this patch if everything looks good to you? :-)

--
Thanks,
Qinxin

Hi Qinxin,

I see it queued up on tmp.perf-tools-next:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/log/?h=tmp.perf-tools-next
which is where patches are tested prior to landing in perf-tools-next :-)

Thanks,
Ian

Quoted 140 lines
Quoted 140 lines
Quoted 140 lines
Quoted 125 lines
Quoted 125 lines
---
   tools/perf/Documentation/perf-stat.txt |  4 ++++
   tools/perf/builtin-stat.c              | 19 +++++++++++++++++++
   tools/perf/util/metricgroup.c          | 18 +++++++++++++-----
   tools/perf/util/parse-events.c         |  2 +-
   4 files changed, 37 insertions(+), 6 deletions(-)
tools/perf/Documentation/perf-stat.txt +4 -0
index 7cccc3a847d1..b72a29c9223c 100644
@@ -578,6 +578,10 @@ $ perf config stat.no-csv-summary=true
   Only enable events on applying cpu with this type for hybrid platform
   (e.g. core or atom)"
 
+--pmu-filter::
+Only enable events on applying pmu with specified for multiple
+pmus with same type (e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)
+
   EXAMPLES
   --------
 
tools/perf/builtin-stat.c +19 -0
index 73c2ba7e3076..69c83a5beb71 100644
@@ -1214,6 +1214,21 @@ static int parse_cputype(const struct option *opt,
      return 0;
   }
 
+static int parse_pmu_filter(const struct option *opt,
+                       const char *str,
+                       int unset __maybe_unused)
+{
+    struct evlist *evlist = *(struct evlist **)opt->value;
+
+    if (!list_empty(&evlist->core.entries)) {
+            fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+            return -1;
+    }
+
+    parse_events_option_args.pmu_filter = str;
+    return 0;
+}
+
   static int parse_cache_level(const struct option *opt,
                           const char *str,
                           int unset __maybe_unused)
@@ -2561,6 +2576,10 @@ int cmd_stat(int argc, const char **argv)
                      "Only enable events on applying cpu with this type "
                      "for hybrid platform (e.g. core or atom)",
                      parse_cputype),
+            OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+                    "Only enable events on applying pmu with specified "
+                    "for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+                    parse_pmu_filter),
   #ifdef HAVE_LIBPFM
              OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
                      "libpfm4 event selector. use 'perf list' to list available events",
tools/perf/util/metricgroup.c +13 -5
index 46bf4dfeebc8..89b83ca38483 100644
@@ -387,8 +387,13 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p
                                    const char *metric_or_groups)
   {
      const char *pm_pmu = pm->pmu ?: "cpu";
+    struct perf_pmu *perf_pmu = NULL;
 
-    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu))
+    if (pm->pmu)
+            perf_pmu = perf_pmus__find(pm->pmu);
+
+    if (strcmp(pmu, "all") && strcmp(pm_pmu, pmu) &&
+       (perf_pmu && !perf_pmu__name_wildcard_match(perf_pmu, pmu)))
              return false;
 
      return match_metric_or_groups(pm->metric_group, metric_or_groups) ||
@@ -1259,7 +1264,8 @@ static int build_combined_expr_ctx(const struct list_head *metric_list,
   static int parse_ids(bool metric_no_merge, bool fake_pmu,
                   struct expr_parse_ctx *ids, const char *modifier,
                   bool group_events, const bool tool_events[TOOL_PMU__EVENT_MAX],
-                 struct evlist **out_evlist)
+                 struct evlist **out_evlist,
+                 const char *filter_pmu)
   {
      struct parse_events_error parse_error;
      struct evlist *parsed_evlist;
@@ -1313,7 +1319,7 @@ static int parse_ids(bool metric_no_merge, bool fake_pmu,
      }
      pr_debug("Parsing metric events '%s'\n", events.buf);
      parse_events_error__init(&parse_error);
-    ret = __parse_events(parsed_evlist, events.buf, /*pmu_filter=*/NULL,
+    ret = __parse_events(parsed_evlist, events.buf, filter_pmu,
                           &parse_error, fake_pmu, /*warn_if_reordered=*/false,
                           /*fake_tp=*/false);
      if (ret) {
@@ -1416,7 +1422,8 @@ static int parse_groups(struct evlist *perf_evlist,
                                      /*modifier=*/NULL,
                                      /*group_events=*/false,
                                      tool_events,
-                                    &combined_evlist);
+                                    &combined_evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
              }
              if (combined)
                      expr__ctx_free(combined);
@@ -1471,7 +1478,8 @@ static int parse_groups(struct evlist *perf_evlist,
              }
              if (!metric_evlist) {
                      ret = parse_ids(metric_no_merge, fake_pmu, m->pctx, m->modifier,
-                                    m->group_events, tool_events, &m->evlist);
+                                    m->group_events, tool_events, &m->evlist,
+                                    (pmu && strcmp(pmu, "all") == 0) ? NULL : pmu);
                      if (ret)
                              goto out;
 
tools/perf/util/parse-events.c +0 -0
index b9efb296bba5..789ff4d15e47 100644
@@ -429,7 +429,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
      if (parse_state->pmu_filter == NULL)
              return false;
 
Quoted 9 lines
Quoted 8 lines
tools/perf/util/parse-events.c +1 -2
-    return strcmp(parse_state->pmu_filter, pmu->name) != 0;
+    return perf_pmu__wildcard_match(pmu, parse_state->pmu_filter) == 0;
   }
 
   static int parse_events_add_pmu(struct parse_events_state *parse_state,
--

2.33.0

Gentle ping.
--
Thanks,
Qinxin

maintainer Applied
#7 lore raw
In reply to Qinxin Xia
Quoted 8 lines · Qinxin Xia

This patch adds a new --pmu-filter option to perf-stat command to allow filtering events on specific PMUs. This is useful when there are multiple PMUs with same type (e.g. hisi_sicl2_cpa0 and hisi_sicl0_cpa0).

[root@localhost tmp]# perf stat -M cpa_p0_avg_bw
Performance counter stats for 'system wide':

[...]

Applied to perf-tools-next, thanks!

Best regards,
Namhyung