7 files changed, 231 insertions, 4 deletions

diff --git a/tools/perf/bench/Build b/tools/perf/bench/Build
index 6b6155a8ad09..0f158dc8139b 100644
--- a/tools/perf/bench/Build
+++ b/tools/perf/bench/Build
@@ -15,6 +15,7 @@ perf-y += find-bit-bench.o
 perf-y += inject-buildid.o
 perf-y += evlist-open-close.o
 perf-y += breakpoint.o
+perf-y += pmu-scan.o
 
 perf-$(CONFIG_X86_64) += mem-memcpy-x86-64-asm.o
 perf-$(CONFIG_X86_64) += mem-memset-x86-64-asm.o
diff --git a/tools/perf/bench/bench.h b/tools/perf/bench/bench.h
index e43893151a3e..0d2b65976212 100644
--- a/tools/perf/bench/bench.h
+++ b/tools/perf/bench/bench.h
@@ -23,6 +23,7 @@ int bench_sched_messaging(int argc, const char **argv);
 int bench_sched_pipe(int argc, const char **argv);
 int bench_syscall_basic(int argc, const char **argv);
 int bench_syscall_getpgid(int argc, const char **argv);
+int bench_syscall_fork(int argc, const char **argv);
 int bench_syscall_execve(int argc, const char **argv);
 int bench_mem_memcpy(int argc, const char **argv);
 int bench_mem_memset(int argc, const char **argv);
@@ -41,6 +42,7 @@ int bench_inject_build_id(int argc, const char **argv);
 int bench_evlist_open_close(int argc, const char **argv);
 int bench_breakpoint_thread(int argc, const char **argv);
 int bench_breakpoint_enable(int argc, const char **argv);
+int bench_pmu_scan(int argc, const char **argv);
 
 #define BENCH_FORMAT_DEFAULT_STR	"default"
 #define BENCH_FORMAT_DEFAULT		0
diff --git a/tools/perf/bench/find-bit-bench.c b/tools/perf/bench/find-bit-bench.c
index d103c3136983..7e25b0e413f6 100644
--- a/tools/perf/bench/find-bit-bench.c
+++ b/tools/perf/bench/find-bit-bench.c
@@ -61,7 +61,6 @@ static int do_for_each_set_bit(unsigned int num_bits)
 	double time_average, time_stddev;
 	unsigned int bit, i, j;
 	unsigned int set_bits, skip;
-	unsigned int old;
 
 	init_stats(&fb_time_stats);
 	init_stats(&tb_time_stats);
@@ -73,7 +72,10 @@ static int do_for_each_set_bit(unsigned int num_bits)
 			__set_bit(i, to_test);
 
 		for (i = 0; i < outer_iterations; i++) {
-			old = accumulator;
+#ifndef NDEBUG
+			unsigned int old = accumulator;
+#endif
+
 			gettimeofday(&start, NULL);
 			for (j = 0; j < inner_iterations; j++) {
 				for_each_set_bit(bit, to_test, num_bits)
@@ -85,7 +87,9 @@ static int do_for_each_set_bit(unsigned int num_bits)
 			runtime_us = diff.tv_sec * USEC_PER_SEC + diff.tv_usec;
 			update_stats(&fb_time_stats, runtime_us);
 
+#ifndef NDEBUG
 			old = accumulator;
+#endif
 			gettimeofday(&start, NULL);
 			for (j = 0; j < inner_iterations; j++) {
 				for (bit = 0; bit < num_bits; bit++) {
diff --git a/tools/perf/bench/inject-buildid.c b/tools/perf/bench/inject-buildid.c
index 4561bda0ce6a..49331743c743 100644
--- a/tools/perf/bench/inject-buildid.c
+++ b/tools/perf/bench/inject-buildid.c
@@ -12,6 +12,7 @@
 #include <linux/time64.h>
 #include <linux/list.h>
 #include <linux/err.h>
+#include <linux/zalloc.h>
 #include <internal/lib.h>
 #include <subcmd/parse-options.h>
 
@@ -122,7 +123,7 @@ static void release_dso(void)
 	for (i = 0; i < nr_dsos; i++) {
 		struct bench_dso *dso = &dsos[i];
 
-		free(dso->name);
+		zfree(&dso->name);
 	}
 	free(dsos);
 }
diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
index 9717c6c17433..1fbd7c947abc 100644
--- a/tools/perf/bench/numa.c
+++ b/tools/perf/bench/numa.c
@@ -847,7 +847,7 @@ static u64 do_work(u8 *__data, long bytes, int nr, int nr_max, int loop, u64 val
 
 	if (g->p.data_rand_walk) {
 		u32 lfsr = nr + loop + val;
-		int j;
+		long j;
 
 		for (i = 0; i < words/1024; i++) {
 			long start, end;
diff --git a/tools/perf/bench/pmu-scan.c b/tools/perf/bench/pmu-scan.c
new file mode 100644
index 000000000000..f0f007843bb8
--- /dev/null
+++ b/tools/perf/bench/pmu-scan.c
@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Benchmark scanning sysfs files for PMU information.
+ *
+ * Copyright 2023 Google LLC.
+ */
+#include <stdio.h>
+#include "bench.h"
+#include "util/debug.h"
+#include "util/pmu.h"
+#include "util/pmus.h"
+#include "util/stat.h"
+#include <linux/atomic.h>
+#include <linux/err.h>
+#include <linux/time64.h>
+#include <subcmd/parse-options.h>
+
+static unsigned int iterations = 100;
+
+struct pmu_scan_result {
+	char *name;
+	int nr_aliases;
+	int nr_formats;
+	int nr_caps;
+};
+
+static const struct option options[] = {
+	OPT_UINTEGER('i', "iterations", &iterations,
+		"Number of iterations used to compute average"),
+	OPT_END()
+};
+
+static const char *const bench_usage[] = {
+	"perf bench internals pmu-scan <options>",
+	NULL
+};
+
+static int nr_pmus;
+static struct pmu_scan_result *results;
+
+static int save_result(void)
+{
+	struct perf_pmu *pmu;
+	struct list_head *list;
+	struct pmu_scan_result *r;
+
+	perf_pmu__scan(NULL);
+
+	perf_pmus__for_each_pmu(pmu) {
+		r = realloc(results, (nr_pmus + 1) * sizeof(*r));
+		if (r == NULL)
+			return -ENOMEM;
+
+		results = r;
+		r = results + nr_pmus;
+
+		r->name = strdup(pmu->name);
+		r->nr_caps = pmu->nr_caps;
+
+		r->nr_aliases = 0;
+		list_for_each(list, &pmu->aliases)
+			r->nr_aliases++;
+
+		r->nr_formats = 0;
+		list_for_each(list, &pmu->format)
+			r->nr_formats++;
+
+		pr_debug("pmu[%d] name=%s, nr_caps=%d, nr_aliases=%d, nr_formats=%d\n",
+			nr_pmus, r->name, r->nr_caps, r->nr_aliases, r->nr_formats);
+		nr_pmus++;
+	}
+
+	perf_pmu__destroy();
+	return 0;
+}
+
+static int check_result(void)
+{
+	struct pmu_scan_result *r;
+	struct perf_pmu *pmu;
+	struct list_head *list;
+	int nr;
+
+	for (int i = 0; i < nr_pmus; i++) {
+		r = &results[i];
+		pmu = perf_pmu__find(r->name);
+		if (pmu == NULL) {
+			pr_err("Cannot find PMU %s\n", r->name);
+			return -1;
+		}
+
+		if (pmu->nr_caps != (u32)r->nr_caps) {
+			pr_err("Unmatched number of event caps in %s: expect %d vs got %d\n",
+				pmu->name, r->nr_caps, pmu->nr_caps);
+			return -1;
+		}
+
+		nr = 0;
+		list_for_each(list, &pmu->aliases)
+			nr++;
+		if (nr != r->nr_aliases) {
+			pr_err("Unmatched number of event aliases in %s: expect %d vs got %d\n",
+				pmu->name, r->nr_aliases, nr);
+			return -1;
+		}
+
+		nr = 0;
+		list_for_each(list, &pmu->format)
+			nr++;
+		if (nr != r->nr_formats) {
+			pr_err("Unmatched number of event formats in %s: expect %d vs got %d\n",
+				pmu->name, r->nr_formats, nr);
+			return -1;
+		}
+	}
+	return 0;
+}
+
+static void delete_result(void)
+{
+	for (int i = 0; i < nr_pmus; i++)
+		free(results[i].name);
+	free(results);
+
+	results = NULL;
+	nr_pmus = 0;
+}
+
+static int run_pmu_scan(void)
+{
+	struct stats stats;
+	struct timeval start, end, diff;
+	double time_average, time_stddev;
+	u64 runtime_us;
+	unsigned int i;
+	int ret;
+
+	init_stats(&stats);
+	pr_info("Computing performance of sysfs PMU event scan for %u times\n",
+		iterations);
+
+	if (save_result() < 0) {
+		pr_err("Failed to initialize PMU scan result\n");
+		return -1;
+	}
+
+	for (i = 0; i < iterations; i++) {
+		gettimeofday(&start, NULL);
+		perf_pmu__scan(NULL);
+		gettimeofday(&end, NULL);
+
+		timersub(&end, &start, &diff);
+		runtime_us = diff.tv_sec * USEC_PER_SEC + diff.tv_usec;
+		update_stats(&stats, runtime_us);
+
+		ret = check_result();
+		perf_pmu__destroy();
+		if (ret < 0)
+			break;
+	}
+
+	time_average = avg_stats(&stats);
+	time_stddev = stddev_stats(&stats);
+	pr_info("  Average PMU scanning took: %.3f usec (+- %.3f usec)\n",
+		time_average, time_stddev);
+
+	delete_result();
+	return 0;
+}
+
+int bench_pmu_scan(int argc, const char **argv)
+{
+	int err = 0;
+
+	argc = parse_options(argc, argv, options, bench_usage, 0);
+	if (argc) {
+		usage_with_options(bench_usage, options);
+		exit(EXIT_FAILURE);
+	}
+
+	err = run_pmu_scan();
+
+	return err;
+}
diff --git a/tools/perf/bench/syscall.c b/tools/perf/bench/syscall.c
index fe79f7f3091e..ea4dfc07cbd6 100644
--- a/tools/perf/bench/syscall.c
+++ b/tools/perf/bench/syscall.c
@@ -18,6 +18,10 @@
 #include <unistd.h>
 #include <stdlib.h>
 
+#ifndef __NR_fork
+#define __NR_fork -1
+#endif
+
 #define LOOPS_DEFAULT 10000000
 static	int loops = LOOPS_DEFAULT;
 
@@ -31,6 +35,23 @@ static const char * const bench_syscall_usage[] = {
 	NULL
 };
 
+static void test_fork(void)
+{
+	pid_t pid = fork();
+
+	if (pid < 0) {
+		fprintf(stderr, "fork failed\n");
+		exit(1);
+	} else if (pid == 0) {
+		exit(0);
+	} else {
+		if (waitpid(pid, NULL, 0) < 0) {
+			fprintf(stderr, "waitpid failed\n");
+			exit(1);
+		}
+	}
+}
+
 static void test_execve(void)
 {
 	const char *pathname = "/bin/true";
@@ -71,6 +92,12 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
 		case __NR_getpgid:
 			getpgid(0);
 			break;
+		case __NR_fork:
+			test_fork();
+			/* Only loop 10000 times to save time */
+			if (i == 10000)
+				loops = 10000;
+			break;
 		case __NR_execve:
 			test_execve();
 			/* Only loop 10000 times to save time */
@@ -92,6 +119,9 @@ static int bench_syscall_common(int argc, const char **argv, int syscall)
 	case __NR_getpgid:
 		name = "getpgid()";
 		break;
+	case __NR_fork:
+		name = "fork()";
+		break;
 	case __NR_execve:
 		name = "execve()";
 		break;
@@ -143,6 +173,11 @@ int bench_syscall_getpgid(int argc, const char **argv)
 	return bench_syscall_common(argc, argv, __NR_getpgid);
 }
 
+int bench_syscall_fork(int argc, const char **argv)
+{
+	return bench_syscall_common(argc, argv, __NR_fork);
+}
+
 int bench_syscall_execve(int argc, const char **argv)
 {
 	return bench_syscall_common(argc, argv, __NR_execve);