Skip to contentSkip to conversation

perf stat: Pass fewer metric arguments

PATCH Superseded Ian Rogers 2 participants 4 messages Feb 6, 2024 – Feb 21, 2024 perf events

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Ian Rogers

author reviewer
#0 lore raw

PATCH v2 1/3 perf stat: Pass fewer metric arguments

Pass metric_expr and evsel rather than specific variables from the struct, thereby reducing the number of arguments. This will enable later fixes.

To reduce the size of the diff, local variables are added to match the previous parameter names. This isn't done in the case of "name" as evsel->name is more intention revealing. A whitespace issue is also addressed.

---
 tools/perf/util/stat-shadow.c | 38 +++++++++++++++++------------------
 1 file changed, 18 insertions(+), 20 deletions(-)
tools/perf/util/stat-shadow.c +18 -20
index e31426167852..b3a25659b9e6 100644
@@ -355,11 +355,12 @@ static void print_nsecs(struct perf_stat_config *config,
 		print_metric(config, ctxp, NULL, NULL, "CPUs utilized", 0);
 }
 
-static int prepare_metric(struct evsel **metric_events,
-			  struct metric_ref *metric_refs,
+static int prepare_metric(const struct metric_expr *mexp,
 			  struct expr_parse_ctx *pctx,
 			  int aggr_idx)
 {
+	struct evsel * const *metric_events = mexp->metric_events;
+	struct metric_ref *metric_refs = mexp->metric_refs;
 	int i;
 
 	for (i = 0; metric_events[i]; i++) {
@@ -403,7 +404,7 @@ static int prepare_metric(struct evsel **metric_events,
 			if (!aggr)
 				break;
 
-                        if (!metric_events[i]->supported) {
+			if (!metric_events[i]->supported) {
 				/*
 				 * Not supported events will have a count of 0,
 				 * which can be confusing in a
@@ -441,18 +442,18 @@ static int prepare_metric(struct evsel **metric_events,
 }
 
 static void generic_metric(struct perf_stat_config *config,
-			   const char *metric_expr,
-			   const char *metric_threshold,
-			   struct evsel **metric_events,
-			   struct metric_ref *metric_refs,
-			   char *name,
-			   const char *metric_name,
-			   const char *metric_unit,
-			   int runtime,
+			   struct metric_expr *mexp,
+			   struct evsel *evsel,
 			   int aggr_idx,
 			   struct perf_stat_output_ctx *out)
 {
 	print_metric_t print_metric = out->print_metric;
+	const char *metric_name = mexp->metric_name;
+	const char *metric_expr = mexp->metric_expr;
+	const char *metric_threshold = mexp->metric_threshold;
+	const char *metric_unit = mexp->metric_unit;
+	struct evsel * const *metric_events = mexp->metric_events;
+	int runtime = mexp->runtime;
 	struct expr_parse_ctx *pctx;
 	double ratio, scale, threshold;
 	int i;
@@ -467,7 +468,7 @@ static void generic_metric(struct perf_stat_config *config,
 		pctx->sctx.user_requested_cpu_list = strdup(config->user_requested_cpu_list);
 	pctx->sctx.runtime = runtime;
 	pctx->sctx.system_wide = config->system_wide;
-	i = prepare_metric(metric_events, metric_refs, pctx, aggr_idx);
+	i = prepare_metric(mexp, pctx, aggr_idx);
 	if (i < 0) {
 		expr__ctx_free(pctx);
 		return;
@@ -502,18 +503,18 @@ static void generic_metric(struct perf_stat_config *config,
 				print_metric(config, ctxp, color, "%8.2f",
 					metric_name ?
 					metric_name :
-					out->force_header ?  name : "",
+					out->force_header ?  evsel->name : "",
 					ratio);
 			}
 		} else {
 			print_metric(config, ctxp, color, /*unit=*/NULL,
 				     out->force_header ?
-				     (metric_name ? metric_name : name) : "", 0);
+				     (metric_name ?: evsel->name) : "", 0);
 		}
 	} else {
 		print_metric(config, ctxp, color, /*unit=*/NULL,
 			     out->force_header ?
-			     (metric_name ? metric_name : name) : "", 0);
+			     (metric_name ?: evsel->name) : "", 0);
 	}
 
 	expr__ctx_free(pctx);
@@ -528,7 +529,7 @@ double test_generic_metric(struct metric_expr *mexp, int aggr_idx)
 	if (!pctx)
 		return NAN;
 
-	if (prepare_metric(mexp->metric_events, mexp->metric_refs, pctx, aggr_idx) < 0)
+	if (prepare_metric(mexp, pctx, aggr_idx) < 0)
 		goto out;
 
 	if (expr__parse(&ratio, pctx, mexp->metric_expr))
@@ -630,10 +631,7 @@ void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config,
 
 		if ((*num)++ > 0)
 			out->new_line(config, ctxp);
-		generic_metric(config, mexp->metric_expr, mexp->metric_threshold,
-			       mexp->metric_events, mexp->metric_refs, evsel->name,
-			       mexp->metric_name, mexp->metric_unit, mexp->runtime,
-			       aggr_idx, out);
+		generic_metric(config, mexp, evsel, aggr_idx, out);
 	}
 
 	return NULL;
Signature · 1 line
2.43.0.594.gd9cf4e227d-goog
author reviewer
#1 lore raw

PATCH v2 2/3 perf metrics: Compute unmerged uncore metrics individually

When merging counts from multiple uncore PMUs the metric is only computed for the metric leader. When merging/aggregation is disabled, prior to this patch just the leader's metric would be computed. Fix this by computing the metric for each PMU.

On a SkylakeX:
Before:

```
$ perf stat -A -M memory_bandwidth_total -a sleep 1

 Performance counter stats for 'system wide':

CPU0               82,217      UNC_M_CAS_COUNT.RD [uncore_imc_0] #      9.2 MB/s  memory_bandwidth_total
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_0] #      0.0 MB/s  memory_bandwidth_total
CPU0               61,395      UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU0                    0      UNC_M_CAS_COUNT.RD [uncore_imc_1]
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_1]
CPU0                    0      UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU0               81,570      UNC_M_CAS_COUNT.RD [uncore_imc_2]
CPU18             113,886      UNC_M_CAS_COUNT.RD [uncore_imc_2]
CPU0               62,330      UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU18              66,942      UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU0               75,489      UNC_M_CAS_COUNT.RD [uncore_imc_3]
CPU18              27,958      UNC_M_CAS_COUNT.RD [uncore_imc_3]
CPU0               55,864      UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU18              38,727      UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU0                    0      UNC_M_CAS_COUNT.RD [uncore_imc_4]
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_4]
CPU0                    0      UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU0               75,423      UNC_M_CAS_COUNT.RD [uncore_imc_5]
CPU18             104,527      UNC_M_CAS_COUNT.RD [uncore_imc_5]
CPU0               57,596      UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU18              56,777      UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU0        1,003,440,851 ns   duration_time

       1.003440851 seconds time elapsed
```

After:

```
$ perf stat -A -M memory_bandwidth_total -a sleep 1

 Performance counter stats for 'system wide':

CPU0               88,968      UNC_M_CAS_COUNT.RD [uncore_imc_0] #      9.5 MB/s  memory_bandwidth_total
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_0] #      0.0 MB/s  memory_bandwidth_total
CPU0               59,498      UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_0]
CPU0                    0      UNC_M_CAS_COUNT.RD [uncore_imc_1] #      0.0 MB/s  memory_bandwidth_total
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_1] #      0.0 MB/s  memory_bandwidth_total
CPU0                    0      UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_1]
CPU0               88,635      UNC_M_CAS_COUNT.RD [uncore_imc_2] #      9.5 MB/s  memory_bandwidth_total
CPU18             117,975      UNC_M_CAS_COUNT.RD [uncore_imc_2] #     11.5 MB/s  memory_bandwidth_total
CPU0               60,829      UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU18              62,105      UNC_M_CAS_COUNT.WR [uncore_imc_2]
CPU0               82,238      UNC_M_CAS_COUNT.RD [uncore_imc_3] #      8.7 MB/s  memory_bandwidth_total
CPU18              22,906      UNC_M_CAS_COUNT.RD [uncore_imc_3] #      3.6 MB/s  memory_bandwidth_total
CPU0               53,959      UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU18              32,990      UNC_M_CAS_COUNT.WR [uncore_imc_3]
CPU0                    0      UNC_M_CAS_COUNT.RD [uncore_imc_4] #      0.0 MB/s  memory_bandwidth_total
CPU18                   0      UNC_M_CAS_COUNT.RD [uncore_imc_4] #      0.0 MB/s  memory_bandwidth_total
CPU0                    0      UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU18                   0      UNC_M_CAS_COUNT.WR [uncore_imc_4]
CPU0               83,595      UNC_M_CAS_COUNT.RD [uncore_imc_5] #      8.9 MB/s  memory_bandwidth_total
CPU18             110,151      UNC_M_CAS_COUNT.RD [uncore_imc_5] #     10.5 MB/s  memory_bandwidth_total
CPU0               56,540      UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU18              53,816      UNC_M_CAS_COUNT.WR [uncore_imc_5]
CPU0        1,003,353,416 ns   duration_time
```
---
 tools/perf/util/metricgroup.c |  2 ++
 tools/perf/util/stat-shadow.c | 31 +++++++++++++++++++++++++++----
 2 files changed, 29 insertions(+), 4 deletions(-)
tools/perf/util/metricgroup.c +2 -0
index ca3e0404f187..c33ffee837ca 100644
@@ -44,6 +44,8 @@ struct metric_event *metricgroup__lookup(struct rblist *metric_events,
 	if (!metric_events)
 		return NULL;
 
+	if (evsel->metric_leader)
+		me.evsel = evsel->metric_leader;
 	nd = rblist__find(metric_events, &me);
 	if (nd)
 		return container_of(nd, struct metric_event, nd);
tools/perf/util/stat-shadow.c +27 -4
index b3a25659b9e6..c745337f50a4 100644
@@ -356,6 +356,7 @@ static void print_nsecs(struct perf_stat_config *config,
 }
 
 static int prepare_metric(const struct metric_expr *mexp,
+			  const struct evsel *evsel,
 			  struct expr_parse_ctx *pctx,
 			  int aggr_idx)
 {
@@ -399,8 +400,29 @@ static int prepare_metric(const struct metric_expr *mexp,
 			source_count = 1;
 		} else {
 			struct perf_stat_evsel *ps = metric_events[i]->stats;
-			struct perf_stat_aggr *aggr = &ps->aggr[aggr_idx];
+			struct perf_stat_aggr *aggr;
 
+			/*
+			 * If there are multiple uncore PMUs and we're not
+			 * reading the leader's stats, determine the stats for
+			 * the appropriate uncore PMU.
+			 */
+			if (evsel && evsel->metric_leader &&
+			    evsel->pmu != evsel->metric_leader->pmu &&
+			    mexp->metric_events[i]->pmu == evsel->metric_leader->pmu) {
+				struct evsel *pos;
+
+				evlist__for_each_entry(evsel->evlist, pos) {
+					if (pos->pmu != evsel->pmu)
+						continue;
+					if (pos->metric_leader != mexp->metric_events[i])
+						continue;
+					ps = pos->stats;
+					source_count = 1;
+					break;
+				}
+			}
+			aggr = &ps->aggr[aggr_idx];
 			if (!aggr)
 				break;
 
@@ -421,7 +443,8 @@ static int prepare_metric(const struct metric_expr *mexp,
 				 * metric.
 				 */
 				val = aggr->counts.val * (1.0 / metric_events[i]->scale);
-				source_count = evsel__source_count(metric_events[i]);
+				if (!source_count)
+					source_count = evsel__source_count(metric_events[i]);
 			}
 		}
 		n = strdup(evsel__metric_id(metric_events[i]));
@@ -468,7 +491,7 @@ static void generic_metric(struct perf_stat_config *config,
 		pctx->sctx.user_requested_cpu_list = strdup(config->user_requested_cpu_list);
 	pctx->sctx.runtime = runtime;
 	pctx->sctx.system_wide = config->system_wide;
-	i = prepare_metric(mexp, pctx, aggr_idx);
+	i = prepare_metric(mexp, evsel, pctx, aggr_idx);
 	if (i < 0) {
 		expr__ctx_free(pctx);
 		return;
@@ -529,7 +552,7 @@ double test_generic_metric(struct metric_expr *mexp, int aggr_idx)
 	if (!pctx)
 		return NAN;
 
-	if (prepare_metric(mexp, pctx, aggr_idx) < 0)
+	if (prepare_metric(mexp, /*evsel=*/NULL, pctx, aggr_idx) < 0)
 		goto out;
 
 	if (expr__parse(&ratio, pctx, mexp->metric_expr))
Signature · 1 line
2.43.0.594.gd9cf4e227d-goog
author reviewer
#2 lore raw

PATCH v2 3/3 perf stat: Fix metric-only aggregation index

In reply to Ian Rogers

Aggregation index was being computed using the evsel's cpumap which may have a different (typically the same or fewer) entries.

Before:

```
$ perf stat --metric-only -A -M memory_bandwidth_total -a sleep 1

 Performance counter stats for 'system wide':

       MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total
CPU0                            12.8                           0.0                          12.9                          12.7                           0.0                          12.6
CPU1

       1.007806367 seconds time elapsed
```

After:

```
$ perf stat --metric-only -A -M memory_bandwidth_total -a sleep 1

 Performance counter stats for 'system wide':

       MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total MB/s  memory_bandwidth_total
CPU0                            15.4                           0.0                          15.3                          15.0                           0.0                          14.9
CPU18                            0.0                           0.0                          13.5                           5.2                           0.0                          11.9

       1.007858736 seconds time elapsed
```
---
 tools/perf/util/stat-display.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
tools/perf/util/stat-display.c +7 -2
index 8c61f8627ebc..ce830c6afdf2 100644
@@ -1126,11 +1126,16 @@ static void print_no_aggr_metric(struct perf_stat_config *config,
 			u64 ena, run, val;
 			double uval;
 			struct perf_stat_evsel *ps = counter->stats;
-			int aggr_idx = perf_cpu_map__idx(evsel__cpus(counter), cpu);
+			int aggr_idx = 0;
 
-			if (aggr_idx < 0)
+			if (!perf_cpu_map__has(evsel__cpus(counter), cpu))
 				continue;
 
+			cpu_aggr_map__for_each_idx(aggr_idx, config->aggr_map) {
+				if (config->aggr_map->map[aggr_idx].cpu.cpu == cpu.cpu)
+					break;
+			}
+
 			os->evsel = counter;
 			os->id = aggr_cpu_id__cpu(cpu, /*data=*/NULL);
 			if (first) {
Signature · 1 line
2.43.0.594.gd9cf4e227d-goog
maintainer Acked-by on v2 1/3
#3 lore raw
In reply to Ian Rogers
Quoted 11 lines · Ian Rogers

Pass metric_expr and evsel rather than specific variables from the struct, thereby reducing the number of arguments. This will enable later fixes.

To reduce the size of the diff, local variables are added to match the previous parameter names. This isn't done in the case of "name" as evsel->name is more intention revealing. A whitespace issue is also addressed.

This doesn't apply to perf-tools-next anymore. Can you please refresh? Anyway the change looks good to me now. For the series:

Thanks,
Namhyung

Quoted 114 lines
---
 tools/perf/util/stat-shadow.c | 38 +++++++++++++++++------------------
 1 file changed, 18 insertions(+), 20 deletions(-)
tools/perf/util/stat-shadow.c +18 -21
index e31426167852..b3a25659b9e6 100644
@@ -355,11 +355,12 @@ static void print_nsecs(struct perf_stat_config *config,
                print_metric(config, ctxp, NULL, NULL, "CPUs utilized", 0);
 }
 
-static int prepare_metric(struct evsel **metric_events,
-                         struct metric_ref *metric_refs,
+static int prepare_metric(const struct metric_expr *mexp,
                          struct expr_parse_ctx *pctx,
                          int aggr_idx)
 {
+       struct evsel * const *metric_events = mexp->metric_events;
+       struct metric_ref *metric_refs = mexp->metric_refs;
        int i;
 
        for (i = 0; metric_events[i]; i++) {
@@ -403,7 +404,7 @@ static int prepare_metric(struct evsel **metric_events,
                        if (!aggr)
                                break;
 
-                        if (!metric_events[i]->supported) {
+                       if (!metric_events[i]->supported) {
                                /*
                                 * Not supported events will have a count of 0,
                                 * which can be confusing in a
@@ -441,18 +442,18 @@ static int prepare_metric(struct evsel **metric_events,
 }
 
 static void generic_metric(struct perf_stat_config *config,
-                          const char *metric_expr,
-                          const char *metric_threshold,
-                          struct evsel **metric_events,
-                          struct metric_ref *metric_refs,
-                          char *name,
-                          const char *metric_name,
-                          const char *metric_unit,
-                          int runtime,
+                          struct metric_expr *mexp,
+                          struct evsel *evsel,
                           int aggr_idx,
                           struct perf_stat_output_ctx *out)
 {
        print_metric_t print_metric = out->print_metric;
+       const char *metric_name = mexp->metric_name;
+       const char *metric_expr = mexp->metric_expr;
+       const char *metric_threshold = mexp->metric_threshold;
+       const char *metric_unit = mexp->metric_unit;
+       struct evsel * const *metric_events = mexp->metric_events;
+       int runtime = mexp->runtime;
        struct expr_parse_ctx *pctx;
        double ratio, scale, threshold;
        int i;
@@ -467,7 +468,7 @@ static void generic_metric(struct perf_stat_config *config,
                pctx->sctx.user_requested_cpu_list = strdup(config->user_requested_cpu_list);
        pctx->sctx.runtime = runtime;
        pctx->sctx.system_wide = config->system_wide;
-       i = prepare_metric(metric_events, metric_refs, pctx, aggr_idx);
+       i = prepare_metric(mexp, pctx, aggr_idx);
        if (i < 0) {
                expr__ctx_free(pctx);
                return;
@@ -502,18 +503,18 @@ static void generic_metric(struct perf_stat_config *config,
                                print_metric(config, ctxp, color, "%8.2f",
                                        metric_name ?
                                        metric_name :
-                                       out->force_header ?  name : "",
+                                       out->force_header ?  evsel->name : "",
                                        ratio);
                        }
                } else {
                        print_metric(config, ctxp, color, /*unit=*/NULL,
                                     out->force_header ?
-                                    (metric_name ? metric_name : name) : "", 0);
+                                    (metric_name ?: evsel->name) : "", 0);
                }
        } else {
                print_metric(config, ctxp, color, /*unit=*/NULL,
                             out->force_header ?
-                            (metric_name ? metric_name : name) : "", 0);
+                            (metric_name ?: evsel->name) : "", 0);
        }
 
        expr__ctx_free(pctx);
@@ -528,7 +529,7 @@ double test_generic_metric(struct metric_expr *mexp, int aggr_idx)
        if (!pctx)
                return NAN;
 
-       if (prepare_metric(mexp->metric_events, mexp->metric_refs, pctx, aggr_idx) < 0)
+       if (prepare_metric(mexp, pctx, aggr_idx) < 0)
                goto out;
 
        if (expr__parse(&ratio, pctx, mexp->metric_expr))
@@ -630,10 +631,7 @@ void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config,
 
                if ((*num)++ > 0)
                        out->new_line(config, ctxp);
-               generic_metric(config, mexp->metric_expr, mexp->metric_threshold,
-                              mexp->metric_events, mexp->metric_refs, evsel->name,
-                              mexp->metric_name, mexp->metric_unit, mexp->runtime,
-                              aggr_idx, out);
+               generic_metric(config, mexp, evsel, aggr_idx, out);
        }
 
        return NULL;
--

2.43.0.594.gd9cf4e227d-goog