tools/perf/builtin-stat.c

/*
 * builtin-stat.c
 *
 * Builtin stat command: Give a precise performance counters summary
 * overview about any workload, CPU or specific PID.
 *
 * Sample output:

   $ perf stat ~/hackbench 10
   Time: 0.104

    Performance counter stats for '/home/mingo/hackbench':

       1255.538611  task clock ticks     #      10.143 CPU utilization factor
             54011  context switches     #       0.043 M/sec
               385  CPU migrations       #       0.000 M/sec
             17755  pagefaults           #       0.014 M/sec
        3808323185  CPU cycles           #    3033.219 M/sec
        1575111190  instructions         #    1254.530 M/sec
          17367895  cache references     #      13.833 M/sec
           7674421  cache misses         #       6.112 M/sec

    Wall-clock time elapsed:   123.786620 msecs

 *
 * Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
 *
 * Improvements and fixes by:
 *
 *   Arjan van de Ven <arjan@linux.intel.com>
 *   Yanmin Zhang <yanmin.zhang@intel.com>
 *   Wu Fengguang <fengguang.wu@intel.com>
 *   Mike Galbraith <efault@gmx.de>
 *   Paul Mackerras <paulus@samba.org>
 *   Jaswinder Singh Rajput <jaswinder@kernel.org>
 *
 * Released under the GPL v2. (and only v2, not any later version)
 */

#include "perf.h"
#include "builtin.h"
#include "util/util.h"
#include "util/parse-options.h"
#include "util/parse-events.h"

#include <sys/prctl.h>
#include <math.h>

static struct perf_counter_attr default_attrs[] = {

  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK	},
  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES},
  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS	},
  { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS	},

  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES	},
  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS	},
  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES},
  { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES	},

};

#define MAX_RUN			100

static int			system_wide			=  0;
static int			verbose				=  0;
static int			nr_cpus				=  0;
static int			run_idx				=  0;

static int			run_count			=  1;
static int			inherit				=  1;
static int			scale				=  1;
static int			target_pid			= -1;
static int			null_run			=  0;

static int			fd[MAX_NR_CPUS][MAX_COUNTERS];

static u64			runtime_nsecs[MAX_RUN];
static u64			walltime_nsecs[MAX_RUN];
static u64			runtime_cycles[MAX_RUN];

static u64			event_res[MAX_RUN][MAX_COUNTERS][3];
static u64			event_scaled[MAX_RUN][MAX_COUNTERS];

static u64			event_res_avg[MAX_COUNTERS][3];
static u64			event_res_noise[MAX_COUNTERS][3];

static u64			event_scaled_avg[MAX_COUNTERS];

static u64			runtime_nsecs_avg;
static u64			runtime_nsecs_noise;

static u64			walltime_nsecs_avg;
static u64			walltime_nsecs_noise;

static u64			runtime_cycles_avg;
static u64			runtime_cycles_noise;

#define ERR_PERF_OPEN \
"Error: counter %d, sys_perf_counter_open() syscall returned with %d (%s)\n"

static void create_perf_stat_counter(int counter, int pid)
{
	struct perf_counter_attr *attr = attrs + counter;

	if (scale)
		attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
				    PERF_FORMAT_TOTAL_TIME_RUNNING;

	if (system_wide) {
		int cpu;
		for (cpu = 0; cpu < nr_cpus; cpu++) {
			fd[cpu][counter] = sys_perf_counter_open(attr, -1, cpu, -1, 0);
			if (fd[cpu][counter] < 0 && verbose)
				fprintf(stderr, ERR_PERF_OPEN, counter,
					fd[cpu][counter], strerror(errno));
		}
	} else {
		attr->inherit	= inherit;
		attr->disabled	= 1;

		fd[0][counter] = sys_perf_counter_open(attr, pid, -1, -1, 0);
		if (fd[0][counter] < 0 && verbose)
			fprintf(stderr, ERR_PERF_OPEN, counter,
				fd[0][counter], strerror(errno));
	}
}

/*
 * Does the counter have nsecs as a unit?
 */
static inline int nsec_counter(int counter)
{
	if (attrs[counter].type != PERF_TYPE_SOFTWARE)
		return 0;

	if (attrs[counter].config == PERF_COUNT_SW_CPU_CLOCK)
		return 1;

	if (attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK)
		return 1;

	return 0;
}

/*
 * Read out the results of a single counter:
 */
static void read_counter(int counter)
{
	u64 *count, single_count[3];
	ssize_t res;
	int cpu, nv;
	int scaled;

	count = event_res[run_idx][counter];

	count[0] = count[1] = count[2] = 0;

	nv = scale ? 3 : 1;
	for (cpu = 0; cpu < nr_cpus; cpu++) {
		if (fd[cpu][counter] < 0)
			continue;

		res = read(fd[cpu][counter], single_count, nv * sizeof(u64));
		assert(res == nv * sizeof(u64));
		close(fd[cpu][counter]);
		fd[cpu][counter] = -1;

		count[0] += single_count[0];
		if (scale) {
			count[1] += single_count[1];
			count[2] += single_count[2];
		}
	}

	scaled = 0;
	if (scale) {
		if (count[2] == 0) {
			event_scaled[run_idx][counter] = -1;
			count[0] = 0;
			return;
		}

		if (count[2] < count[1]) {
			event_scaled[run_idx][counter] = 1;
			count[0] = (unsigned long long)
				((double)count[0] * count[1] / count[2] + 0.5);
		}
	}
	/*
	 * Save the full runtime - to allow normalization during printout:
	 */
	if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
		attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK)
		runtime_nsecs[run_idx] = count[0];
	if (attrs[counter].type == PERF_TYPE_HARDWARE &&
		attrs[counter].config == PERF_COUNT_HW_CPU_CYCLES)
		runtime_cycles[run_idx] = count[0];
}

static int run_perf_stat(int argc, const char **argv)
{
	unsigned long long t0, t1;
	int status = 0;
	int counter;
	int pid;
	int child_ready_pipe[2], go_pipe[2];
	char buf;

	if (!system_wide)
		nr_cpus = 1;

	if (pipe(child_ready_pipe) < 0 || pipe(go_pipe) < 0) {
		perror("failed to create pipes");
		exit(1);
	}

	if ((pid = fork()) < 0)
		perror("failed to fork");

	if (!pid) {
		close(child_ready_pipe[0]);
		close(go_pipe[1]);
		fcntl(go_pipe[0], F_SETFD, FD_CLOEXEC);

		/*
		 * Do a dummy execvp to get the PLT entry resolved,
		 * so we avoid the resolver overhead on the real
		 * execvp call.
		 */
		execvp("", (char **)argv);

		/*
		 * Tell the parent we're ready to go
		 */
		close(child_ready_pipe[1]);

		/*
		 * Wait until the parent tells us to go.
		 */
		read(go_pipe[0], &buf, 1);

		execvp(argv[0], (char **)argv);

		perror(argv[0]);
		exit(-1);
	}

	/*
	 * Wait for the child to be ready to exec.
	 */
	close(child_ready_pipe[1]);
	close(go_pipe[0]);
	read(child_ready_pipe[0], &buf, 1);
	close(child_ready_pipe[0]);

	for (counter = 0; counter < nr_counters; counter++)
		create_perf_stat_counter(counter, pid);

	/*
	 * Enable counters and exec the command:
	 */
	t0 = rdclock();
	prctl(PR_TASK_PERF_COUNTERS_ENABLE);

	close(go_pipe[1]);
	wait(&status);

	t1 = rdclock();

	walltime_nsecs[run_idx] = t1 - t0;

	for (counter = 0; counter < nr_counters; counter++)
		read_counter(counter);

	return WEXITSTATUS(status);
}

static void print_noise(u64 *count, u64 *noise)
{
	if (run_count > 1)
		fprintf(stderr, "   ( +- %7.3f%% )",
			(double)noise[0]/(count[0]+1)*100.0);
}

static void nsec_printout(int counter, u64 *count, u64 *noise)
{
	double msecs = (double)count[0] / 1000000;

	fprintf(stderr, " %14.6f  %-24s", msecs, event_name(counter));

	if (attrs[counter].type == PERF_TYPE_SOFTWARE &&
		attrs[counter].config == PERF_COUNT_SW_TASK_CLOCK) {

		if (walltime_nsecs_avg)
			fprintf(stderr, " # %10.3f CPUs ",
				(double)count[0] / (double)walltime_nsecs_avg);
	}
	print_noise(count, noise);
}

static void abs_printout(int counter, u64 *count, u64 *noise)
{
	fprintf(stderr, " %14Ld  %-24s", count[0], event_name(counter));

	if (runtime_cycles_avg &&
		attrs[counter].type == PERF_TYPE_HARDWARE &&
			attrs[counter].config == PERF_COUNT_HW_INSTRUCTIONS) {

		fprintf(stderr, " # %10.3f IPC  ",
			(double)count[0] / (double)runtime_cycles_avg);
	} else {
		if (runtime_nsecs_avg) {
			fprintf(stderr, " # %10.3f M/sec",
				(double)count[0]/runtime_nsecs_avg*1000.0);
		}
	}
	print_noise(count, noise);
}

/*
 * Print out the results of a single counter:
 */
static void print_counter(int counter)
{
	u64 *count, *noise;
	int scaled;

	count = event_res_avg[counter];
	noise = event_res_noise[counter];
	scaled = event_scaled_avg[counter];

	if (scaled == -1) {
		fprintf(stderr, " %14s  %-24s\n",
			"<not counted>", event_name(counter));
		return;
	}

	if (nsec_counter(counter))
		nsec_printout(counter, count, noise);
	else
		abs_printout(counter, count, noise);

	if (scaled)
		fprintf(stderr, "  (scaled from %.2f%%)",
			(double) count[2] / count[1] * 100);

	fprintf(stderr, "\n");
}

/*
 * normalize_noise noise values down to stddev:
 */
static void normalize_noise(u64 *val)
{
	double res;

	res = (double)*val / (run_count * sqrt((double)run_count));

	*val = (u64)res;
}

static void update_avg(const char *name, int idx, u64 *avg, u64 *val)
{
	*avg += *val;

	if (verbose > 1)
		fprintf(stderr, "debug: %20s[%d]: %Ld\n", name, idx, *val);
}
/*
 * Calculate the averages and noises:
 */
static void calc_avg(void)
{
	int i, j;

	if (verbose > 1)
		fprintf(stderr, "\n");

	for (i = 0; i < run_count; i++) {
		update_avg("runtime", 0, &runtime_nsecs_avg, runtime_nsecs + i);
		update_avg("walltime", 0, &walltime_nsecs_avg, walltime_nsecs + i);
		update_avg("runtime_cycles", 0, &runtime_cycles_avg, runtime_cycles + i);

		for (j = 0; j < nr_counters; j++) {
			update_avg("counter/0", j,
				event_res_avg[j]+0, event_res[i][j]+0);
			update_avg("counter/1", j,
				event_res_avg[j]+1, event_res[i][j]+1);
			update_avg("counter/2", j,
				event_res_avg[j]+2, event_res[i][j]+2);
			if (event_scaled[i][j] != -1)
				update_avg("scaled", j,
					event_scaled_avg + j, event_scaled[i]+j);
			else
				event_scaled_avg[j] = -1;
		}
	}
	runtime_nsecs_avg /= run_count;
	walltime_nsecs_avg /= run_count;
	runtime_cycles_avg /= run_count;

	for (j = 0; j < nr_counters; j++) {
		event_res_avg[j][0] /= run_count;
		event_res_avg[j][1] /= run_count;
		event_res_avg[j][2] /= run_count;
	}

	for (i = 0; i < run_count; i++) {
		runtime_nsecs_noise +=
			abs((s64)(runtime_nsecs[i] - runtime_nsecs_avg));
		walltime_nsecs_noise +=
			abs((s64)(walltime_nsecs[i] - walltime_nsecs_avg));
		runtime_cycles_noise +=
			abs((s64)(runtime_cycles[i] - runtime_cycles_avg));

		for (j = 0; j < nr_counters; j++) {
			event_res_noise[j][0] +=
				abs((s64)(event_res[i][j][0] - event_res_avg[j][0]));
			event_res_noise[j][1] +=
				abs((s64)(event_res[i][j][1] - event_res_avg[j][1]));
			event_res_noise[j][2] +=
				abs((s64)(event_res[i][j][2] - event_res_avg[j][2]));
		}
	}

	normalize_noise(&runtime_nsecs_noise);
	normalize_noise(&walltime_nsecs_noise);
	normalize_noise(&runtime_cycles_noise);

	for (j = 0; j < nr_counters; j++) {
		normalize_noise(&event_res_noise[j][0]);
		normalize_noise(&event_res_noise[j][1]);
		normalize_noise(&event_res_noise[j][2]);
	}
}

static void print_stat(int argc, const char **argv)
{
	int i, counter;

	calc_avg();

	fflush(stdout);

	fprintf(stderr, "\n");
	fprintf(stderr, " Performance counter stats for \'%s", argv[0]);

	for (i = 1; i < argc; i++)
		fprintf(stderr, " %s", argv[i]);

	fprintf(stderr, "\'");
	if (run_count > 1)
		fprintf(stderr, " (%d runs)", run_count);
	fprintf(stderr, ":\n\n");

	for (counter = 0; counter < nr_counters; counter++)
		print_counter(counter);

	fprintf(stderr, "\n");
	fprintf(stderr, " %14.9f  seconds time elapsed",
			(double)walltime_nsecs_avg/1e9);
	if (run_count > 1) {
		fprintf(stderr, "   ( +- %7.3f%% )",
			100.0*(double)walltime_nsecs_noise/(double)walltime_nsecs_avg);
	}
	fprintf(stderr, "\n\n");
}

static volatile int signr = -1;

static void skip_signal(int signo)
{
	signr = signo;
}

static void sig_atexit(void)
{
	if (signr == -1)
		return;

	signal(signr, SIG_DFL);
	kill(getpid(), signr);
}

static const char * const stat_usage[] = {
	"perf stat [<options>] <command>",
	NULL
};

static const struct option options[] = {
	OPT_CALLBACK('e', "event", NULL, "event",
		     "event selector. use 'perf list' to list available events",
		     parse_events),
	OPT_BOOLEAN('i', "inherit", &inherit,
		    "child tasks inherit counters"),
	OPT_INTEGER('p', "pid", &target_pid,
		    "stat events on existing pid"),
	OPT_BOOLEAN('a', "all-cpus", &system_wide,
		    "system-wide collection from all CPUs"),
	OPT_BOOLEAN('S', "scale", &scale,
		    "scale/normalize counters"),
	OPT_BOOLEAN('v', "verbose", &verbose,
		    "be more verbose (show counter open errors, etc)"),
	OPT_INTEGER('r', "repeat", &run_count,
		    "repeat command and print average + stddev (max: 100)"),
	OPT_BOOLEAN('n', "null", &null_run,
		    "null run - dont start any counters"),
	OPT_END()
};

int cmd_stat(int argc, const char **argv, const char *prefix)
{
	int status;

	argc = parse_options(argc, argv, options, stat_usage, 0);
	if (!argc)
		usage_with_options(stat_usage, options);
	if (run_count <= 0 || run_count > MAX_RUN)
		usage_with_options(stat_usage, options);

	/* Set attrs and nr_counters if no event is selected and !null_run */
	if (!null_run && !nr_counters) {
		memcpy(attrs, default_attrs, sizeof(default_attrs));
		nr_counters = ARRAY_SIZE(default_attrs);
	}

	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
	assert(nr_cpus <= MAX_NR_CPUS);
	assert(nr_cpus >= 0);

	/*
	 * We dont want to block the signals - that would cause
	 * child tasks to inherit that and Ctrl-C would not work.
	 * What we want is for Ctrl-C to work in the exec()-ed
	 * task, but being ignored by perf stat itself:
	 */
	atexit(sig_atexit);
	signal(SIGINT,  skip_signal);
	signal(SIGALRM, skip_signal);
	signal(SIGABRT, skip_signal);

	status = 0;
	for (run_idx = 0; run_idx < run_count; run_idx++) {
		if (run_count != 1 && verbose)
			fprintf(stderr, "[ perf stat: executing run #%d ... ]\n", run_idx + 1);
		status = run_perf_stat(argc, argv);
	}

	print_stat(argc, argv);

	return status;
}