LKML Archive on lore.kernel.org
 help / color / Atom feed
From: James Clark <james.clark@arm.com>
To: linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org,
	jolsa@redhat.com, namhyung@kernel.org
Cc: james.clark@arm.com, john.garry@huawei.com,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Thomas Richter <tmricht@linux.ibm.com>
Subject: [PATCH v6 04/12] perf tools: Replace aggregation ID with a struct
Date: Thu, 26 Nov 2020 16:13:20 +0200
Message-ID: <20201126141328.6509-5-james.clark@arm.com> (raw)
In-Reply-To: <20201126141328.6509-1-james.clark@arm.com>

Replace all occurences of the usage of int with the new struct
cpu_aggr_id.

No functional changes.

Signed-off-by: James Clark <james.clark@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: John Garry <john.garry@huawei.com>
---
 tools/perf/builtin-stat.c      |  76 +++++++++++++----------
 tools/perf/tests/topology.c    |  17 +++---
 tools/perf/util/cpumap.c       |  82 ++++++++++++++-----------
 tools/perf/util/cpumap.h       |  10 +--
 tools/perf/util/stat-display.c | 108 +++++++++++++++++++--------------
 tools/perf/util/stat.c         |   2 +-
 tools/perf/util/stat.h         |   5 +-
 7 files changed, 173 insertions(+), 127 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index f15b2f8aa14d..f10c67a26472 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -1188,65 +1188,67 @@ static struct option stat_options[] = {
 	OPT_END()
 };
 
-static int perf_stat__get_socket(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id perf_stat__get_socket(struct perf_stat_config *config __maybe_unused,
 				 struct perf_cpu_map *map, int cpu)
 {
 	return cpu_map__get_socket(map, cpu, NULL);
 }
 
-static int perf_stat__get_die(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id perf_stat__get_die(struct perf_stat_config *config __maybe_unused,
 			      struct perf_cpu_map *map, int cpu)
 {
 	return cpu_map__get_die(map, cpu, NULL);
 }
 
-static int perf_stat__get_core(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id perf_stat__get_core(struct perf_stat_config *config __maybe_unused,
 			       struct perf_cpu_map *map, int cpu)
 {
 	return cpu_map__get_core(map, cpu, NULL);
 }
 
-static int perf_stat__get_node(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id 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,
+static struct aggr_cpu_id perf_stat__get_aggr(struct perf_stat_config *config,
 			       aggr_get_id_t get_id, struct perf_cpu_map *map, int idx)
 {
 	int cpu;
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
 
 	if (idx >= map->nr)
-		return -1;
+		return id;
 
 	cpu = map->map[idx];
 
 	if (config->cpus_aggr_map->map[cpu] == -1)
-		config->cpus_aggr_map->map[cpu] = get_id(config, map, idx);
+		config->cpus_aggr_map->map[cpu] = get_id(config, map, idx).id;
 
-	return config->cpus_aggr_map->map[cpu];
+	id.id = config->cpus_aggr_map->map[cpu];
+	return id;
 }
 
-static int perf_stat__get_socket_cached(struct perf_stat_config *config,
+static struct aggr_cpu_id perf_stat__get_socket_cached(struct perf_stat_config *config,
 					struct perf_cpu_map *map, int idx)
 {
 	return perf_stat__get_aggr(config, perf_stat__get_socket, map, idx);
 }
 
-static int perf_stat__get_die_cached(struct perf_stat_config *config,
+static struct aggr_cpu_id perf_stat__get_die_cached(struct perf_stat_config *config,
 					struct perf_cpu_map *map, int idx)
 {
 	return perf_stat__get_aggr(config, perf_stat__get_die, map, idx);
 }
 
-static int perf_stat__get_core_cached(struct perf_stat_config *config,
+static struct aggr_cpu_id perf_stat__get_core_cached(struct perf_stat_config *config,
 				      struct perf_cpu_map *map, int idx)
 {
 	return perf_stat__get_aggr(config, perf_stat__get_core, map, idx);
 }
 
-static int perf_stat__get_node_cached(struct perf_stat_config *config,
+static struct aggr_cpu_id 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);
@@ -1347,18 +1349,23 @@ static inline int perf_env__get_cpu(struct perf_env *env, struct perf_cpu_map *m
 	return cpu;
 }
 
-static int perf_env__get_socket(struct perf_cpu_map *map, int idx, void *data)
+static struct aggr_cpu_id perf_env__get_socket(struct perf_cpu_map *map, int idx, void *data)
 {
 	struct perf_env *env = data;
 	int cpu = perf_env__get_cpu(env, map, idx);
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
+
+	if (cpu != -1)
+		id.id = env->cpu[cpu].socket_id;
 
-	return cpu == -1 ? -1 : env->cpu[cpu].socket_id;
+	return id;
 }
 
-static int perf_env__get_die(struct perf_cpu_map *map, int idx, void *data)
+static struct aggr_cpu_id perf_env__get_die(struct perf_cpu_map *map, int idx, void *data)
 {
 	struct perf_env *env = data;
-	int die_id = -1, cpu = perf_env__get_cpu(env, map, idx);
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
+	int cpu = perf_env__get_cpu(env, map, idx);
 
 	if (cpu != -1) {
 		/*
@@ -1368,21 +1375,22 @@ static int perf_env__get_die(struct perf_cpu_map *map, int idx, void *data)
 		 * socket + die id
 		 */
 		if (WARN_ONCE(env->cpu[cpu].socket_id >> 8, "The socket id number is too big.\n"))
-			return -1;
+			return cpu_map__empty_aggr_cpu_id();
 
 		if (WARN_ONCE(env->cpu[cpu].die_id >> 8, "The die id number is too big.\n"))
-			return -1;
+			return cpu_map__empty_aggr_cpu_id();
 
-		die_id = (env->cpu[cpu].socket_id << 8) | (env->cpu[cpu].die_id & 0xff);
+		id.id = (env->cpu[cpu].socket_id << 8) | (env->cpu[cpu].die_id & 0xff);
 	}
 
-	return die_id;
+	return id;
 }
 
-static int perf_env__get_core(struct perf_cpu_map *map, int idx, void *data)
+static struct aggr_cpu_id perf_env__get_core(struct perf_cpu_map *map, int idx, void *data)
 {
 	struct perf_env *env = data;
-	int core = -1, cpu = perf_env__get_cpu(env, map, idx);
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
+	int cpu = perf_env__get_cpu(env, map, idx);
 
 	if (cpu != -1) {
 		/*
@@ -1393,27 +1401,29 @@ static int perf_env__get_core(struct perf_cpu_map *map, int idx, void *data)
 		 * socket + die id + core id
 		 */
 		if (WARN_ONCE(env->cpu[cpu].socket_id >> 8, "The socket id number is too big.\n"))
-			return -1;
+			return cpu_map__empty_aggr_cpu_id();
 
 		if (WARN_ONCE(env->cpu[cpu].die_id >> 8, "The die id number is too big.\n"))
-			return -1;
+			return cpu_map__empty_aggr_cpu_id();
 
 		if (WARN_ONCE(env->cpu[cpu].core_id >> 16, "The core id number is too big.\n"))
-			return -1;
+			return cpu_map__empty_aggr_cpu_id();
 
-		core = (env->cpu[cpu].socket_id << 24) |
+		id.id = (env->cpu[cpu].socket_id << 24) |
 		       (env->cpu[cpu].die_id << 16) |
 		       (env->cpu[cpu].core_id & 0xffff);
 	}
 
-	return core;
+	return id;
 }
 
-static int perf_env__get_node(struct perf_cpu_map *map, int idx, void *data)
+static struct aggr_cpu_id perf_env__get_node(struct perf_cpu_map *map, int idx, void *data)
 {
 	int cpu = perf_env__get_cpu(data, map, idx);
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
 
-	return perf_env__numa_node(data, cpu);
+	id.id = perf_env__numa_node(data, cpu);
+	return id;
 }
 
 static int perf_env__build_socket_map(struct perf_env *env, struct perf_cpu_map *cpus,
@@ -1440,24 +1450,24 @@ static int perf_env__build_node_map(struct perf_env *env, struct perf_cpu_map *c
 	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,
+static struct aggr_cpu_id perf_stat__get_socket_file(struct perf_stat_config *config __maybe_unused,
 				      struct perf_cpu_map *map, int idx)
 {
 	return perf_env__get_socket(map, idx, &perf_stat.session->header.env);
 }
-static int perf_stat__get_die_file(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id perf_stat__get_die_file(struct perf_stat_config *config __maybe_unused,
 				   struct perf_cpu_map *map, int idx)
 {
 	return perf_env__get_die(map, idx, &perf_stat.session->header.env);
 }
 
-static int perf_stat__get_core_file(struct perf_stat_config *config __maybe_unused,
+static struct aggr_cpu_id perf_stat__get_core_file(struct perf_stat_config *config __maybe_unused,
 				    struct perf_cpu_map *map, int idx)
 {
 	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,
+static struct aggr_cpu_id 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);
diff --git a/tools/perf/tests/topology.c b/tools/perf/tests/topology.c
index 8228a1de7ac8..8dab1b4f323f 100644
--- a/tools/perf/tests/topology.c
+++ b/tools/perf/tests/topology.c
@@ -64,7 +64,8 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 		.path = path,
 		.mode = PERF_DATA_MODE_READ,
 	};
-	int i, id;
+	int i;
+	struct aggr_cpu_id id;
 
 	session = perf_session__new(&data, false, NULL);
 	TEST_ASSERT_VAL("can't get session", !IS_ERR(session));
@@ -110,14 +111,14 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 	for (i = 0; i < map->nr; i++) {
 		id = cpu_map__get_core(map, i, NULL);
 		TEST_ASSERT_VAL("Core map - Core ID doesn't match",
-			session->header.env.cpu[map->map[i]].core_id == cpu_map__id_to_cpu(id));
+			session->header.env.cpu[map->map[i]].core_id == cpu_map__id_to_cpu(id.id));
 
 		TEST_ASSERT_VAL("Core map - Socket ID doesn't match",
 			session->header.env.cpu[map->map[i]].socket_id ==
-				cpu_map__id_to_socket(id));
+				cpu_map__id_to_socket(id.id));
 
 		TEST_ASSERT_VAL("Core map - Die ID doesn't match",
-			session->header.env.cpu[map->map[i]].die_id == cpu_map__id_to_die(id));
+			session->header.env.cpu[map->map[i]].die_id == cpu_map__id_to_die(id.id));
 	}
 
 	// Test that die ID contains socket and die
@@ -125,25 +126,25 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 		id = cpu_map__get_die(map, i, NULL);
 		TEST_ASSERT_VAL("Die map - Socket ID doesn't match",
 			session->header.env.cpu[map->map[i]].socket_id ==
-				cpu_map__id_to_socket(id << 16));
+				cpu_map__id_to_socket(id.id << 16));
 
 		TEST_ASSERT_VAL("Die map - Die ID doesn't match",
 			session->header.env.cpu[map->map[i]].die_id ==
-				cpu_map__id_to_die(id << 16));
+				cpu_map__id_to_die(id.id << 16));
 	}
 
 	// Test that socket ID contains only socket
 	for (i = 0; i < map->nr; i++) {
 		id = cpu_map__get_socket(map, i, NULL);
 		TEST_ASSERT_VAL("Socket map - Socket ID doesn't match",
-			session->header.env.cpu[map->map[i]].socket_id == id);
+			session->header.env.cpu[map->map[i]].socket_id == id.id);
 	}
 
 	// Test that node ID contains only node
 	for (i = 0; i < map->nr; i++) {
 		id = cpu_map__get_node(map, i, NULL);
 		TEST_ASSERT_VAL("Node map - Node ID doesn't match",
-			cpu__get_node(map->map[i]) == id);
+			cpu__get_node(map->map[i]) == id.id);
 	}
 	perf_session__delete(session);
 
diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c
index 8624948b4f1d..e05a12bde073 100644
--- a/tools/perf/util/cpumap.c
+++ b/tools/perf/util/cpumap.c
@@ -111,30 +111,37 @@ int cpu_map__get_socket_id(int cpu)
 	return ret ?: value;
 }
 
-int cpu_map__get_socket(struct perf_cpu_map *map, int idx, void *data __maybe_unused)
+struct aggr_cpu_id cpu_map__get_socket(struct perf_cpu_map *map, int idx,
+					void *data __maybe_unused)
 {
 	int cpu;
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
 
 	if (idx > map->nr)
-		return -1;
+		return id;
 
 	cpu = map->map[idx];
 
-	return cpu_map__get_socket_id(cpu);
+	id.id = cpu_map__get_socket_id(cpu);
+	return id;
 }
 
-static int cmp_ids(const void *a, const void *b)
+static int cmp_aggr_cpu_id(const void *a_pointer, const void *b_pointer)
 {
-	return *(int *)a - *(int *)b;
+	struct aggr_cpu_id *a = (struct aggr_cpu_id *)a_pointer;
+	struct aggr_cpu_id *b = (struct aggr_cpu_id *)b_pointer;
+
+	return a->id - b->id;
 }
 
 int cpu_map__build_map(struct perf_cpu_map *cpus, struct perf_cpu_map **res,
-		       int (*f)(struct perf_cpu_map *map, int cpu, void *data),
+		       struct aggr_cpu_id (*f)(struct perf_cpu_map *map, int cpu, void *data),
 		       void *data)
 {
 	int nr = cpus->nr;
 	struct perf_cpu_map *c = perf_cpu_map__empty_new(nr);
-	int cpu, s1, s2;
+	int cpu, s2;
+	struct aggr_cpu_id s1;
 
 	if (!c)
 		return -1;
@@ -145,16 +152,16 @@ int cpu_map__build_map(struct perf_cpu_map *cpus, struct perf_cpu_map **res,
 	for (cpu = 0; cpu < nr; cpu++) {
 		s1 = f(cpus, cpu, data);
 		for (s2 = 0; s2 < c->nr; s2++) {
-			if (s1 == c->map[s2])
+			if (s1.id == c->map[s2])
 				break;
 		}
 		if (s2 == c->nr) {
-			c->map[c->nr] = s1;
+			c->map[c->nr] = s1.id;
 			c->nr++;
 		}
 	}
 	/* ensure we process id in increasing order */
-	qsort(c->map, c->nr, sizeof(int), cmp_ids);
+	qsort(c->map, c->nr, sizeof(struct aggr_cpu_id), cmp_aggr_cpu_id);
 
 	*res = c;
 	return 0;
@@ -167,23 +174,24 @@ int cpu_map__get_die_id(int cpu)
 	return ret ?: value;
 }
 
-int cpu_map__get_die(struct perf_cpu_map *map, int idx, void *data)
+struct aggr_cpu_id cpu_map__get_die(struct perf_cpu_map *map, int idx, void *data)
 {
-	int cpu, die_id, s;
+	int cpu, s;
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
 
 	if (idx > map->nr)
-		return -1;
+		return id;
 
 	cpu = map->map[idx];
 
-	die_id = cpu_map__get_die_id(cpu);
+	id.id = cpu_map__get_die_id(cpu);
 	/* There is no die_id on legacy system. */
-	if (die_id == -1)
-		die_id = 0;
+	if (id.id == -1)
+		id.id = 0;
 
-	s = cpu_map__get_socket(map, idx, data);
+	s = cpu_map__get_socket(map, idx, data).id;
 	if (s == -1)
-		return -1;
+		return cpu_map__empty_aggr_cpu_id();
 
 	/*
 	 * Encode socket in bit range 15:8
@@ -191,13 +199,14 @@ int cpu_map__get_die(struct perf_cpu_map *map, int idx, void *data)
 	 * we need a global id. So we combine
 	 * socket + die id
 	 */
-	if (WARN_ONCE(die_id >> 8, "The die id number is too big.\n"))
-		return -1;
+	if (WARN_ONCE(id.id >> 8, "The die id number is too big.\n"))
+		return cpu_map__empty_aggr_cpu_id();
 
 	if (WARN_ONCE(s >> 8, "The socket id number is too big.\n"))
-		return -1;
+		return cpu_map__empty_aggr_cpu_id();
 
-	return (s << 8) | (die_id & 0xff);
+	id.id = (s << 8) | (id.id & 0xff);
+	return id;
 }
 
 int cpu_map__get_core_id(int cpu)
@@ -211,21 +220,22 @@ int cpu_map__get_node_id(int cpu)
 	return cpu__get_node(cpu);
 }
 
-int cpu_map__get_core(struct perf_cpu_map *map, int idx, void *data)
+struct aggr_cpu_id cpu_map__get_core(struct perf_cpu_map *map, int idx, void *data)
 {
-	int cpu, s_die;
+	int cpu;
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
 
 	if (idx > map->nr)
-		return -1;
+		return id;
 
 	cpu = map->map[idx];
 
 	cpu = cpu_map__get_core_id(cpu);
 
-	/* s_die is the combination of socket + die id */
-	s_die = cpu_map__get_die(map, idx, data);
-	if (s_die == -1)
-		return -1;
+	/* cpu_map__get_die returns the combination of socket + die id */
+	id = cpu_map__get_die(map, idx, data);
+	if (cpu_map__aggr_cpu_id_is_empty(id))
+		return id;
 
 	/*
 	 * encode socket in bit range 31:24
@@ -235,17 +245,21 @@ int cpu_map__get_core(struct perf_cpu_map *map, int idx, void *data)
 	 * socket + die id + core id
 	 */
 	if (WARN_ONCE(cpu >> 16, "The core id number is too big.\n"))
-		return -1;
+		return cpu_map__empty_aggr_cpu_id();
 
-	return (s_die << 16) | (cpu & 0xffff);
+	id.id = (id.id << 16) | (cpu & 0xffff);
+	return id;
 }
 
-int cpu_map__get_node(struct perf_cpu_map *map, int idx, void *data __maybe_unused)
+struct aggr_cpu_id cpu_map__get_node(struct perf_cpu_map *map, int idx, void *data __maybe_unused)
 {
+	struct aggr_cpu_id id = cpu_map__empty_aggr_cpu_id();
+
 	if (idx < 0 || idx >= map->nr)
-		return -1;
+		return id;
 
-	return cpu_map__get_node_id(map->map[idx]);
+	id.id = cpu_map__get_node_id(map->map[idx]);
+	return id;
 }
 
 int cpu_map__build_socket_map(struct perf_cpu_map *cpus, struct perf_cpu_map **sockp)
diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h
index 1cdccc69cd4b..b8c2288a3f6d 100644
--- a/tools/perf/util/cpumap.h
+++ b/tools/perf/util/cpumap.h
@@ -19,13 +19,13 @@ size_t cpu_map__snprint(struct perf_cpu_map *map, char *buf, size_t size);
 size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size);
 size_t cpu_map__fprintf(struct perf_cpu_map *map, FILE *fp);
 int cpu_map__get_socket_id(int cpu);
-int cpu_map__get_socket(struct perf_cpu_map *map, int idx, void *data);
+struct aggr_cpu_id cpu_map__get_socket(struct perf_cpu_map *map, int idx, void *data);
 int cpu_map__get_die_id(int cpu);
-int cpu_map__get_die(struct perf_cpu_map *map, int idx, void *data);
+struct aggr_cpu_id cpu_map__get_die(struct perf_cpu_map *map, int idx, void *data);
 int cpu_map__get_core_id(int cpu);
-int cpu_map__get_core(struct perf_cpu_map *map, int idx, void *data);
+struct aggr_cpu_id cpu_map__get_core(struct perf_cpu_map *map, int idx, void *data);
 int cpu_map__get_node_id(int cpu);
-int cpu_map__get_node(struct perf_cpu_map *map, int idx, void *data);
+struct aggr_cpu_id  cpu_map__get_node(struct perf_cpu_map *map, int idx, void *data);
 int cpu_map__build_socket_map(struct perf_cpu_map *cpus, struct perf_cpu_map **sockp);
 int cpu_map__build_die_map(struct perf_cpu_map *cpus, struct perf_cpu_map **diep);
 int cpu_map__build_core_map(struct perf_cpu_map *cpus, struct perf_cpu_map **corep);
@@ -62,7 +62,7 @@ int cpu__max_present_cpu(void);
 int cpu__get_node(int cpu);
 
 int cpu_map__build_map(struct perf_cpu_map *cpus, struct perf_cpu_map **res,
-		       int (*f)(struct perf_cpu_map *map, int cpu, void *data),
+		       struct aggr_cpu_id (*f)(struct perf_cpu_map *map, int cpu, void *data),
 		       void *data);
 
 int cpu_map__cpu(struct perf_cpu_map *cpus, int idx);
diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c
index 4b57c0c07632..1a3d66329d73 100644
--- a/tools/perf/util/stat-display.c
+++ b/tools/perf/util/stat-display.c
@@ -68,15 +68,15 @@ static void print_cgroup(struct perf_stat_config *config, struct evsel *evsel)
 
 
 static void aggr_printout(struct perf_stat_config *config,
-			  struct evsel *evsel, int id, int nr)
+			  struct evsel *evsel, struct aggr_cpu_id id, int nr)
 {
 	switch (config->aggr_mode) {
 	case AGGR_CORE:
 		fprintf(config->output, "S%d-D%d-C%*d%s%*d%s",
-			cpu_map__id_to_socket(id),
-			cpu_map__id_to_die(id),
+			cpu_map__id_to_socket(id.id),
+			cpu_map__id_to_die(id.id),
 			config->csv_output ? 0 : -8,
-			cpu_map__id_to_cpu(id),
+			cpu_map__id_to_cpu(id.id),
 			config->csv_sep,
 			config->csv_output ? 0 : 4,
 			nr,
@@ -84,9 +84,9 @@ static void aggr_printout(struct perf_stat_config *config,
 		break;
 	case AGGR_DIE:
 		fprintf(config->output, "S%d-D%*d%s%*d%s",
-			cpu_map__id_to_socket(id << 16),
+			cpu_map__id_to_socket(id.id << 16),
 			config->csv_output ? 0 : -8,
-			cpu_map__id_to_die(id << 16),
+			cpu_map__id_to_die(id.id << 16),
 			config->csv_sep,
 			config->csv_output ? 0 : 4,
 			nr,
@@ -95,7 +95,7 @@ static void aggr_printout(struct perf_stat_config *config,
 	case AGGR_SOCKET:
 		fprintf(config->output, "S%*d%s%*d%s",
 			config->csv_output ? 0 : -5,
-			id,
+			id.id,
 			config->csv_sep,
 			config->csv_output ? 0 : 4,
 			nr,
@@ -104,7 +104,7 @@ static void aggr_printout(struct perf_stat_config *config,
 	case AGGR_NODE:
 		fprintf(config->output, "N%*d%s%*d%s",
 			config->csv_output ? 0 : -5,
-			id,
+			id.id,
 			config->csv_sep,
 			config->csv_output ? 0 : 4,
 			nr,
@@ -113,23 +113,23 @@ static void aggr_printout(struct perf_stat_config *config,
 	case AGGR_NONE:
 		if (evsel->percore && !config->percore_show_thread) {
 			fprintf(config->output, "S%d-D%d-C%*d%s",
-				cpu_map__id_to_socket(id),
-				cpu_map__id_to_die(id),
+				cpu_map__id_to_socket(id.id),
+				cpu_map__id_to_die(id.id),
 				config->csv_output ? 0 : -3,
-				cpu_map__id_to_cpu(id), config->csv_sep);
-		} else if (id > -1) {
+				cpu_map__id_to_cpu(id.id), config->csv_sep);
+		} else if (id.id > -1) {
 			fprintf(config->output, "CPU%*d%s",
 				config->csv_output ? 0 : -7,
-				evsel__cpus(evsel)->map[id],
+				evsel__cpus(evsel)->map[id.id],
 				config->csv_sep);
 		}
 		break;
 	case AGGR_THREAD:
 		fprintf(config->output, "%*s-%*d%s",
 			config->csv_output ? 0 : 16,
-			perf_thread_map__comm(evsel->core.threads, id),
+			perf_thread_map__comm(evsel->core.threads, id.id),
 			config->csv_output ? 0 : -8,
-			perf_thread_map__pid(evsel->core.threads, id),
+			perf_thread_map__pid(evsel->core.threads, id.id),
 			config->csv_sep);
 		break;
 	case AGGR_GLOBAL:
@@ -144,7 +144,8 @@ struct outstate {
 	bool newline;
 	const char *prefix;
 	int  nfields;
-	int  id, nr;
+	int  nr;
+	struct aggr_cpu_id id;
 	struct evsel *evsel;
 };
 
@@ -319,7 +320,7 @@ static void print_metric_header(struct perf_stat_config *config,
 }
 
 static int first_shadow_cpu(struct perf_stat_config *config,
-			    struct evsel *evsel, int id)
+			    struct evsel *evsel, struct aggr_cpu_id id)
 {
 	struct evlist *evlist = evsel->evlist;
 	int i;
@@ -328,7 +329,7 @@ static int first_shadow_cpu(struct perf_stat_config *config,
 		return 0;
 
 	if (config->aggr_mode == AGGR_NONE)
-		return id;
+		return id.id;
 
 	if (config->aggr_mode == AGGR_GLOBAL)
 		return 0;
@@ -336,14 +337,17 @@ static int first_shadow_cpu(struct perf_stat_config *config,
 	for (i = 0; i < evsel__nr_cpus(evsel); i++) {
 		int cpu2 = evsel__cpus(evsel)->map[i];
 
-		if (config->aggr_get_id(config, evlist->core.cpus, cpu2) == id)
+		if (cpu_map__compare_aggr_cpu_id(
+					config->aggr_get_id(config, evlist->core.cpus, cpu2),
+					id)) {
 			return cpu2;
+		}
 	}
 	return 0;
 }
 
 static void abs_printout(struct perf_stat_config *config,
-			 int id, int nr, struct evsel *evsel, double avg)
+			 struct aggr_cpu_id id, int nr, struct evsel *evsel, double avg)
 {
 	FILE *output = config->output;
 	double sc =  evsel->scale;
@@ -396,7 +400,7 @@ static bool is_mixed_hw_group(struct evsel *counter)
 	return false;
 }
 
-static void printout(struct perf_stat_config *config, int id, int nr,
+static void printout(struct perf_stat_config *config, struct aggr_cpu_id id, int nr,
 		     struct evsel *counter, double uval,
 		     char *prefix, u64 run, u64 ena, double noise,
 		     struct runtime_stat *st)
@@ -499,17 +503,18 @@ static void printout(struct perf_stat_config *config, int id, int nr,
 static void aggr_update_shadow(struct perf_stat_config *config,
 			       struct evlist *evlist)
 {
-	int cpu, s2, id, s;
+	int cpu, s;
+	struct aggr_cpu_id s2, id;
 	u64 val;
 	struct evsel *counter;
 
 	for (s = 0; s < config->aggr_map->nr; s++) {
-		id = config->aggr_map->map[s];
+		id.id = config->aggr_map->map[s];
 		evlist__for_each_entry(evlist, counter) {
 			val = 0;
 			for (cpu = 0; cpu < evsel__nr_cpus(counter); cpu++) {
 				s2 = config->aggr_get_id(config, evlist->core.cpus, cpu);
-				if (s2 != id)
+				if (!cpu_map__compare_aggr_cpu_id(s2, id))
 					continue;
 				val += perf_counts(counter->counts, cpu, 0)->val;
 			}
@@ -587,7 +592,7 @@ static bool collect_data(struct perf_stat_config *config, struct evsel *counter,
 
 struct aggr_data {
 	u64 ena, run, val;
-	int id;
+	struct aggr_cpu_id id;
 	int nr;
 	int cpu;
 };
@@ -596,13 +601,14 @@ static void aggr_cb(struct perf_stat_config *config,
 		    struct evsel *counter, void *data, bool first)
 {
 	struct aggr_data *ad = data;
-	int cpu, s2;
+	int cpu;
+	struct aggr_cpu_id s2;
 
 	for (cpu = 0; cpu < evsel__nr_cpus(counter); cpu++) {
 		struct perf_counts_values *counts;
 
 		s2 = config->aggr_get_id(config, evsel__cpus(counter), cpu);
-		if (s2 != ad->id)
+		if (!cpu_map__compare_aggr_cpu_id(s2, ad->id))
 			continue;
 		if (first)
 			ad->nr++;
@@ -631,10 +637,11 @@ static void print_counter_aggrdata(struct perf_stat_config *config,
 	struct aggr_data ad;
 	FILE *output = config->output;
 	u64 ena, run, val;
-	int id, nr;
+	int nr;
+	struct aggr_cpu_id id;
 	double uval;
 
-	ad.id = id = config->aggr_map->map[s];
+	ad.id.id = id.id = config->aggr_map->map[s];
 	ad.val = ad.ena = ad.run = 0;
 	ad.nr = 0;
 	if (!collect_data(config, counter, aggr_cb, &ad))
@@ -652,8 +659,12 @@ static void print_counter_aggrdata(struct perf_stat_config *config,
 		fprintf(output, "%s", prefix);
 
 	uval = val * counter->scale;
-	printout(config, cpu != -1 ? cpu : id, nr, counter, uval, prefix,
-		 run, ena, 1.0, &rt_stat);
+	if (cpu != -1) {
+		id = cpu_map__empty_aggr_cpu_id();
+		id.id = cpu;
+	}
+	printout(config, id, nr, counter, uval,
+		 prefix, run, ena, 1.0, &rt_stat);
 	if (!metric_only)
 		fputc('\n', output);
 }
@@ -731,7 +742,8 @@ static struct perf_aggr_thread_value *sort_aggr_thread(
 			continue;
 
 		buf[i].counter = counter;
-		buf[i].id = thread;
+		buf[i].id = cpu_map__empty_aggr_cpu_id();
+		buf[i].id.id = thread;
 		buf[i].uval = uval;
 		buf[i].val = val;
 		buf[i].run = run;
@@ -754,7 +766,8 @@ static void print_aggr_thread(struct perf_stat_config *config,
 	FILE *output = config->output;
 	int nthreads = perf_thread_map__nr(counter->core.threads);
 	int ncpus = perf_cpu_map__nr(counter->core.cpus);
-	int thread, sorted_threads, id;
+	int thread, sorted_threads;
+	struct aggr_cpu_id id;
 	struct perf_aggr_thread_value *buf;
 
 	buf = sort_aggr_thread(counter, nthreads, ncpus, &sorted_threads, _target);
@@ -771,7 +784,7 @@ static void print_aggr_thread(struct perf_stat_config *config,
 		if (config->stats)
 			printout(config, id, 0, buf[thread].counter, buf[thread].uval,
 				 prefix, buf[thread].run, buf[thread].ena, 1.0,
-				 &config->stats[id]);
+				 &config->stats[id.id]);
 		else
 			printout(config, id, 0, buf[thread].counter, buf[thread].uval,
 				 prefix, buf[thread].run, buf[thread].ena, 1.0,
@@ -817,8 +830,8 @@ static void print_counter_aggr(struct perf_stat_config *config,
 		fprintf(output, "%s", prefix);
 
 	uval = cd.avg * counter->scale;
-	printout(config, -1, 0, counter, uval, prefix, cd.avg_running, cd.avg_enabled,
-		 cd.avg, &rt_stat);
+	printout(config, cpu_map__empty_aggr_cpu_id(), 0, counter, uval, prefix, cd.avg_running,
+		 cd.avg_enabled, cd.avg, &rt_stat);
 	if (!metric_only)
 		fprintf(output, "\n");
 }
@@ -845,6 +858,7 @@ static void print_counter(struct perf_stat_config *config,
 	u64 ena, run, val;
 	double uval;
 	int cpu;
+	struct aggr_cpu_id id;
 
 	for (cpu = 0; cpu < evsel__nr_cpus(counter); cpu++) {
 		struct aggr_data ad = { .cpu = cpu };
@@ -859,8 +873,10 @@ static void print_counter(struct perf_stat_config *config,
 			fprintf(output, "%s", prefix);
 
 		uval = val * counter->scale;
-		printout(config, cpu, 0, counter, uval, prefix, run, ena, 1.0,
-			 &rt_stat);
+		id = cpu_map__empty_aggr_cpu_id();
+		id.id = cpu;
+		printout(config, id, 0, counter, uval, prefix,
+			 run, ena, 1.0, &rt_stat);
 
 		fputc('\n', output);
 	}
@@ -875,6 +891,7 @@ static void print_no_aggr_metric(struct perf_stat_config *config,
 	struct evsel *counter;
 	u64 ena, run, val;
 	double uval;
+	struct aggr_cpu_id id;
 
 	nrcpus = evlist->core.cpus->nr;
 	for (cpu = 0; cpu < nrcpus; cpu++) {
@@ -883,8 +900,10 @@ static void print_no_aggr_metric(struct perf_stat_config *config,
 		if (prefix)
 			fputs(prefix, config->output);
 		evlist__for_each_entry(evlist, counter) {
+			id = cpu_map__empty_aggr_cpu_id();
+			id.id = cpu;
 			if (first) {
-				aggr_printout(config, counter, cpu, 0);
+				aggr_printout(config, counter, id, 0);
 				first = false;
 			}
 			val = perf_counts(counter->counts, cpu, 0)->val;
@@ -892,8 +911,8 @@ static void print_no_aggr_metric(struct perf_stat_config *config,
 			run = perf_counts(counter->counts, cpu, 0)->run;
 
 			uval = val * counter->scale;
-			printout(config, cpu, 0, counter, uval, prefix, run, ena, 1.0,
-				 &rt_stat);
+			printout(config, id, 0, counter, uval, prefix,
+				 run, ena, 1.0, &rt_stat);
 		}
 		fputc('\n', config->output);
 	}
@@ -1143,14 +1162,15 @@ static void print_footer(struct perf_stat_config *config)
 static void print_percore_thread(struct perf_stat_config *config,
 				 struct evsel *counter, char *prefix)
 {
-	int s, s2, id;
+	int s;
+	struct aggr_cpu_id s2, id;
 	bool first = true;
 
 	for (int i = 0; i < evsel__nr_cpus(counter); i++) {
 		s2 = config->aggr_get_id(config, evsel__cpus(counter), i);
 		for (s = 0; s < config->aggr_map->nr; s++) {
-			id = config->aggr_map->map[s];
-			if (s2 == id)
+			id.id = config->aggr_map->map[s];
+			if (cpu_map__compare_aggr_cpu_id(s2, id))
 				break;
 		}
 
diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
index bd0decd6d753..70c1634f4d62 100644
--- a/tools/perf/util/stat.c
+++ b/tools/perf/util/stat.c
@@ -313,7 +313,7 @@ static int check_per_pkg(struct evsel *counter,
 	if (!(vals->run && vals->ena))
 		return 0;
 
-	s = cpu_map__get_socket(cpus, cpu, NULL);
+	s = cpu_map__get_socket(cpus, cpu, NULL).id;
 	if (s < 0)
 		return -1;
 
diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h
index 05adf8165025..888a78bef014 100644
--- a/tools/perf/util/stat.h
+++ b/tools/perf/util/stat.h
@@ -6,6 +6,7 @@
 #include <stdio.h>
 #include <sys/types.h>
 #include <sys/resource.h>
+#include "cpumap.h"
 #include "rblist.h"
 
 struct perf_cpu_map;
@@ -99,7 +100,7 @@ struct runtime_stat {
 	struct rblist value_list;
 };
 
-typedef int (*aggr_get_id_t)(struct perf_stat_config *config,
+typedef struct aggr_cpu_id (*aggr_get_id_t)(struct perf_stat_config *config,
 			     struct perf_cpu_map *m, int cpu);
 
 struct perf_stat_config {
@@ -170,7 +171,7 @@ struct evlist;
 
 struct perf_aggr_thread_value {
 	struct evsel *counter;
-	int id;
+	struct aggr_cpu_id id;
 	double uval;
 	u64 val;
 	u64 run;
-- 
2.28.0


  parent reply index

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-26 14:13 [PATCH v6 00/12] perf tools: fix perf stat with large socket IDs James Clark
2020-11-26 14:13 ` [PATCH v6 01/12] perf tools: Improve topology test James Clark
2020-11-26 14:13 ` [PATCH v6 02/12] perf tools: Use allocator for perf_cpu_map James Clark
2020-11-26 14:13 ` [PATCH v6 03/12] perf tools: Add new struct for cpu aggregation James Clark
2020-11-26 14:13 ` James Clark [this message]
2020-11-26 14:13 ` [PATCH v6 05/12] perf tools: add new map type for aggregation James Clark
2020-11-26 14:13 ` [PATCH v6 06/12] perf tools: drop in cpu_aggr_map struct James Clark
2020-11-26 14:13 ` [PATCH v6 07/12] perf tools: Start using cpu_aggr_id in map James Clark
2020-11-26 14:13 ` [PATCH v6 08/12] perf tools: Add separate node member James Clark
2020-11-26 14:13 ` [PATCH v6 09/12] perf tools: Add separate socket member James Clark
2020-11-26 14:13 ` [PATCH v6 10/12] perf tools: Add separate die member James Clark
2020-11-26 14:13 ` [PATCH v6 11/12] perf tools: Add separate core member James Clark
2020-11-26 14:13 ` [PATCH v6 12/12] perf tools: Add separate thread member James Clark
2020-11-30 14:33 ` [PATCH v6 00/12] perf tools: fix perf stat with large socket IDs Namhyung Kim
2020-12-03 15:39 ` Jiri Olsa
2020-12-04 11:48   ` John Garry
2020-12-23 22:17     ` Jiri Olsa
2020-12-24 12:30       ` Arnaldo Carvalho de Melo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20201126141328.6509-5-james.clark@arm.com \
    --to=james.clark@arm.com \
    --cc=acme@kernel.org \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=john.garry@huawei.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tmricht@linux.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

LKML Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git
	git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git
	git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git
	git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git
	git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git
	git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git
	git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git
	git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git
	git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git
	git clone --mirror https://lore.kernel.org/lkml/9 lkml/git/9.git
	git clone --mirror https://lore.kernel.org/lkml/10 lkml/git/10.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \
		linux-kernel@vger.kernel.org
	public-inbox-index lkml

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git