perf stat: Add --per-node agregation support

Adding new --per-node option to aggregate counts per NUMA
nodes for system-wide mode measurements.

You can specify --per-node in live mode:

  # perf stat  -a -I 1000 -e cycles --per-node
  #           time node   cpus             counts unit events
       1.000542550 N0       20          6,202,097      cycles
       1.000542550 N1       20            639,559      cycles
       2.002040063 N0       20          7,412,495      cycles
       2.002040063 N1       20          2,185,577      cycles
       3.003451699 N0       20          6,508,917      cycles
       3.003451699 N1       20            765,607      cycles
  ...

Or in the record/report stat session:

  # perf stat record -a -I 1000 -e cycles
  #           time             counts unit events
       1.000536937         10,008,468      cycles
       2.002090152          9,578,539      cycles
       3.003625233          7,647,869      cycles
       4.005135036          7,032,086      cycles
  ^C     4.340902364          3,923,893      cycles

  # perf stat report --per-node
  #           time node   cpus             counts unit events
       1.000536937 N0       20          9,355,086      cycles
       1.000536937 N1       20            653,382      cycles
       2.002090152 N0       20          7,712,838      cycles
       2.002090152 N1       20          1,865,701      cycles
       3.003625233 N0       20          6,604,441      cycles
       3.003625233 N1       20          1,043,428      cycles
       4.005135036 N0       20          6,350,522      cycles
       4.005135036 N1       20            681,564      cycles
       4.340902364 N0       20          3,403,188      cycles
       4.340902364 N1       20            520,705      cycles

Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Joe Mario <jmario@redhat.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20190904073415.723-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

1 files changed, 52 insertions, 0 deletions

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index c88d4e118409..5964e808d73d 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -792,6 +792,8 @@ static struct option stat_options[] = {
 		     "aggregate counts per physical processor core", AGGR_CORE),
 	OPT_SET_UINT(0, "per-thread", &stat_config.aggr_mode,
 		     "aggregate counts per thread", AGGR_THREAD),
+	OPT_SET_UINT(0, "per-node", &stat_config.aggr_mode,
+		     "aggregate counts per numa node", AGGR_NODE),
 	OPT_UINTEGER('D', "delay", &stat_config.initial_delay,
 		     "ms to wait before starting measurement after program start"),
 	OPT_CALLBACK_NOOPT(0, "metric-only", &stat_config.metric_only, NULL,
@@ -830,6 +832,12 @@ static int perf_stat__get_core(struct perf_stat_config *config __maybe_unused,
 	return cpu_map__get_core(map, cpu, NULL);
 }
 
+static int perf_stat__get_node(struct perf_stat_config *config __maybe_unused,
+			       struct perf_cpu_map *map, int cpu)
+{
+	return cpu_map__get_node(map, cpu, NULL);
+}
+
 static int perf_stat__get_aggr(struct perf_stat_config *config,
 			       aggr_get_id_t get_id, struct perf_cpu_map *map, int idx)
 {
@@ -864,6 +872,12 @@ static int perf_stat__get_core_cached(struct perf_stat_config *config,
 	return perf_stat__get_aggr(config, perf_stat__get_core, map, idx);
 }
 
+static int perf_stat__get_node_cached(struct perf_stat_config *config,
+				      struct perf_cpu_map *map, int idx)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_node, map, idx);
+}
+
 static bool term_percore_set(void)
 {
 	struct evsel *counter;
@@ -902,6 +916,13 @@ static int perf_stat_init_aggr_mode(void)
 		}
 		stat_config.aggr_get_id = perf_stat__get_core_cached;
 		break;
+	case AGGR_NODE:
+		if (cpu_map__build_node_map(evsel_list->core.cpus, &stat_config.aggr_map)) {
+			perror("cannot build core map");
+			return -1;
+		}
+		stat_config.aggr_get_id = perf_stat__get_node_cached;
+		break;
 	case AGGR_NONE:
 		if (term_percore_set()) {
 			if (cpu_map__build_core_map(evsel_list->core.cpus,
@@ -1014,6 +1035,13 @@ static int perf_env__get_core(struct perf_cpu_map *map, int idx, void *data)
 	return core;
 }
 
+static int perf_env__get_node(struct perf_cpu_map *map, int idx, void *data)
+{
+	int cpu = perf_env__get_cpu(data, map, idx);
+
+	return perf_env__numa_node(data, cpu);
+}
+
 static int perf_env__build_socket_map(struct perf_env *env, struct perf_cpu_map *cpus,
 				      struct perf_cpu_map **sockp)
 {
@@ -1032,6 +1060,12 @@ static int perf_env__build_core_map(struct perf_env *env, struct perf_cpu_map *c
 	return cpu_map__build_map(cpus, corep, perf_env__get_core, env);
 }
 
+static int perf_env__build_node_map(struct perf_env *env, struct perf_cpu_map *cpus,
+				    struct perf_cpu_map **nodep)
+{
+	return cpu_map__build_map(cpus, nodep, perf_env__get_node, env);
+}
+
 static int perf_stat__get_socket_file(struct perf_stat_config *config __maybe_unused,
 				      struct perf_cpu_map *map, int idx)
 {
@@ -1049,6 +1083,12 @@ static int perf_stat__get_core_file(struct perf_stat_config *config __maybe_unus
 	return perf_env__get_core(map, idx, &perf_stat.session->header.env);
 }
 
+static int perf_stat__get_node_file(struct perf_stat_config *config __maybe_unused,
+				    struct perf_cpu_map *map, int idx)
+{
+	return perf_env__get_node(map, idx, &perf_stat.session->header.env);
+}
+
 static int perf_stat_init_aggr_mode_file(struct perf_stat *st)
 {
 	struct perf_env *env = &st->session->header.env;
@@ -1075,6 +1115,13 @@ static int perf_stat_init_aggr_mode_file(struct perf_stat *st)
 		}
 		stat_config.aggr_get_id = perf_stat__get_core_file;
 		break;
+	case AGGR_NODE:
+		if (perf_env__build_node_map(env, evsel_list->core.cpus, &stat_config.aggr_map)) {
+			perror("cannot build core map");
+			return -1;
+		}
+		stat_config.aggr_get_id = perf_stat__get_node_file;
+		break;
 	case AGGR_NONE:
 	case AGGR_GLOBAL:
 	case AGGR_THREAD:
@@ -1622,6 +1669,8 @@ static int __cmd_report(int argc, const char **argv)
 		     "aggregate counts per processor die", AGGR_DIE),
 	OPT_SET_UINT(0, "per-core", &perf_stat.aggr_mode,
 		     "aggregate counts per physical processor core", AGGR_CORE),
+	OPT_SET_UINT(0, "per-node", &perf_stat.aggr_mode,
+		     "aggregate counts per numa node", AGGR_NODE),
 	OPT_SET_UINT('A', "no-aggr", &perf_stat.aggr_mode,
 		     "disable CPU count aggregation", AGGR_NONE),
 	OPT_END()
@@ -1896,6 +1945,9 @@ int cmd_stat(int argc, const char **argv)
 		}
 	}
 
+	if (stat_config.aggr_mode == AGGR_NODE)
+		cpu__setup_cpunode_map();
+
 	if (stat_config.times && interval)
 		interval_count = true;
 	else if (stat_config.times && !interval) {