*
* Sample output:
- $ perf stat ~/hackbench 10
- Time: 0.104
+ $ perf stat ./hackbench 10
- Performance counter stats for '/home/mingo/hackbench':
+ Time: 0.118
- 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
+ Performance counter stats for './hackbench 10':
- Wall-clock time elapsed: 123.786620 msecs
+ 1708.761321 task-clock # 11.037 CPUs utilized
+ 41,190 context-switches # 0.024 M/sec
+ 6,735 CPU-migrations # 0.004 M/sec
+ 17,318 page-faults # 0.010 M/sec
+ 5,205,202,243 cycles # 3.046 GHz
+ 3,856,436,920 stalled-cycles-frontend # 74.09% frontend cycles idle
+ 1,600,790,871 stalled-cycles-backend # 30.75% backend cycles idle
+ 2,603,501,247 instructions # 0.50 insns per cycle
+ # 1.48 stalled cycles per insn
+ 484,357,498 branches # 283.455 M/sec
+ 6,388,934 branch-misses # 1.32% of all branches
+
+ 0.154822978 seconds time elapsed
*
- * Copyright (C) 2008, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
+ * Copyright (C) 2008-2011, Red Hat Inc, Ingo Molnar <mingo@redhat.com>
*
* Improvements and fixes by:
*
#include "util/evlist.h"
#include "util/evsel.h"
#include "util/debug.h"
+#include "util/color.h"
#include "util/header.h"
#include "util/cpumap.h"
#include "util/thread.h"
#include <locale.h>
#define DEFAULT_SEPARATOR " "
+#define CNTR_NOT_SUPPORTED "<not supported>"
+#define CNTR_NOT_COUNTED "<not counted>"
static struct perf_event_attr default_attrs[] = {
{ .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_STALLED_CYCLES },
+ { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_FRONTEND },
+ { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES_BACKEND },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES },
};
+/*
+ * Detailed stats (-d), covering the L1 and last level data caches:
+ */
+static struct perf_event_attr detailed_attrs[] = {
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1D << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1D << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_LL << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_LL << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+};
+
+/*
+ * Very detailed stats (-d -d), covering the instruction cache and the TLB caches:
+ */
+static struct perf_event_attr very_detailed_attrs[] = {
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1I << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1I << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_DTLB << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_DTLB << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_ITLB << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_ITLB << 0 |
+ (PERF_COUNT_HW_CACHE_OP_READ << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+
+};
+
+/*
+ * Very, very detailed stats (-d -d -d), adding prefetch events:
+ */
+static struct perf_event_attr very_very_detailed_attrs[] = {
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1D << 0 |
+ (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_ACCESS << 16) },
+
+ { .type = PERF_TYPE_HW_CACHE,
+ .config =
+ PERF_COUNT_HW_CACHE_L1D << 0 |
+ (PERF_COUNT_HW_CACHE_OP_PREFETCH << 8) |
+ (PERF_COUNT_HW_CACHE_RESULT_MISS << 16) },
+};
+
+
+
struct perf_evlist *evsel_list;
static bool system_wide = false;
static bool no_inherit = false;
static bool scale = true;
static bool no_aggr = false;
-static pid_t target_pid = -1;
-static pid_t target_tid = -1;
+static const char *target_pid;
+static const char *target_tid;
static pid_t child_pid = -1;
static bool null_run = false;
+static int detailed_run = 0;
+static bool sync_run = false;
static bool big_num = true;
static int big_num_opt = -1;
static const char *cpu_list;
static const char *csv_sep = NULL;
static bool csv_output = false;
+static bool group = false;
+static const char *output_name = NULL;
+static FILE *output = NULL;
+static int output_fd;
static volatile int done = 0;
*/
static double stddev_stats(struct stats *stats)
{
- double variance = stats->M2 / (stats->n - 1);
- double variance_mean = variance / stats->n;
+ double variance, variance_mean;
+
+ if (!stats->n)
+ return 0.0;
+
+ variance = stats->M2 / (stats->n - 1);
+ variance_mean = variance / stats->n;
return sqrt(variance_mean);
}
struct stats runtime_nsecs_stats[MAX_NR_CPUS];
struct stats runtime_cycles_stats[MAX_NR_CPUS];
-struct stats runtime_stalled_cycles_stats[MAX_NR_CPUS];
+struct stats runtime_stalled_cycles_front_stats[MAX_NR_CPUS];
+struct stats runtime_stalled_cycles_back_stats[MAX_NR_CPUS];
struct stats runtime_branches_stats[MAX_NR_CPUS];
struct stats runtime_cacherefs_stats[MAX_NR_CPUS];
+struct stats runtime_l1_dcache_stats[MAX_NR_CPUS];
+struct stats runtime_l1_icache_stats[MAX_NR_CPUS];
+struct stats runtime_ll_cache_stats[MAX_NR_CPUS];
+struct stats runtime_itlb_cache_stats[MAX_NR_CPUS];
+struct stats runtime_dtlb_cache_stats[MAX_NR_CPUS];
struct stats walltime_nsecs_stats;
-static int create_perf_stat_counter(struct perf_evsel *evsel)
+static int create_perf_stat_counter(struct perf_evsel *evsel,
+ struct perf_evsel *first)
{
struct perf_event_attr *attr = &evsel->attr;
+ struct xyarray *group_fd = NULL;
+ bool exclude_guest_missing = false;
+ int ret;
+
+ if (group && evsel != first)
+ group_fd = first->fd;
if (scale)
attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
attr->inherit = !no_inherit;
- if (system_wide)
- return perf_evsel__open_per_cpu(evsel, evsel_list->cpus, false);
+retry:
+ if (exclude_guest_missing)
+ evsel->attr.exclude_guest = evsel->attr.exclude_host = 0;
- if (target_pid == -1 && target_tid == -1) {
+ if (system_wide) {
+ ret = perf_evsel__open_per_cpu(evsel, evsel_list->cpus,
+ group, group_fd);
+ if (ret)
+ goto check_ret;
+ return 0;
+ }
+
+ if (!target_pid && !target_tid && (!group || evsel == first)) {
attr->disabled = 1;
attr->enable_on_exec = 1;
}
- return perf_evsel__open_per_thread(evsel, evsel_list->threads, false);
+ ret = perf_evsel__open_per_thread(evsel, evsel_list->threads,
+ group, group_fd);
+ if (!ret)
+ return 0;
+ /* fall through */
+check_ret:
+ if (ret && errno == EINVAL) {
+ if (!exclude_guest_missing &&
+ (evsel->attr.exclude_guest || evsel->attr.exclude_host)) {
+ pr_debug("Old kernel, cannot exclude "
+ "guest or host samples.\n");
+ exclude_guest_missing = true;
+ goto retry;
+ }
+ }
+ return ret;
}
/*
update_stats(&runtime_nsecs_stats[0], count[0]);
else if (perf_evsel__match(counter, HARDWARE, HW_CPU_CYCLES))
update_stats(&runtime_cycles_stats[0], count[0]);
- else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES))
- update_stats(&runtime_stalled_cycles_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_FRONTEND))
+ update_stats(&runtime_stalled_cycles_front_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HARDWARE, HW_STALLED_CYCLES_BACKEND))
+ update_stats(&runtime_stalled_cycles_back_stats[0], count[0]);
else if (perf_evsel__match(counter, HARDWARE, HW_BRANCH_INSTRUCTIONS))
update_stats(&runtime_branches_stats[0], count[0]);
else if (perf_evsel__match(counter, HARDWARE, HW_CACHE_REFERENCES))
update_stats(&runtime_cacherefs_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1D))
+ update_stats(&runtime_l1_dcache_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_L1I))
+ update_stats(&runtime_l1_icache_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_LL))
+ update_stats(&runtime_ll_cache_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_DTLB))
+ update_stats(&runtime_dtlb_cache_stats[0], count[0]);
+ else if (perf_evsel__match(counter, HW_CACHE, HW_CACHE_ITLB))
+ update_stats(&runtime_itlb_cache_stats[0], count[0]);
}
/*
update_stats(&ps->res_stats[i], count[i]);
if (verbose) {
- fprintf(stderr, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
+ fprintf(output, "%s: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
event_name(counter), count[0], count[1], count[2]);
}
static int run_perf_stat(int argc __used, const char **argv)
{
unsigned long long t0, t1;
- struct perf_evsel *counter;
+ struct perf_evsel *counter, *first;
int status = 0;
int child_ready_pipe[2], go_pipe[2];
const bool forks = (argc > 0);
exit(-1);
}
- if (target_tid == -1 && target_pid == -1 && !system_wide)
+ if (!target_tid && !target_pid && !system_wide)
evsel_list->threads->map[0] = child_pid;
/*
close(child_ready_pipe[0]);
}
+ first = list_entry(evsel_list->entries.next, struct perf_evsel, node);
+
list_for_each_entry(counter, &evsel_list->entries, node) {
- if (create_perf_stat_counter(counter) < 0) {
- if (errno == -EPERM || errno == -EACCES) {
+ if (create_perf_stat_counter(counter, first) < 0) {
+ /*
+ * PPC returns ENXIO for HW counters until 2.6.37
+ * (behavior changed with commit b0a873e).
+ */
+ if (errno == EINVAL || errno == ENOSYS ||
+ errno == ENOENT || errno == EOPNOTSUPP ||
+ errno == ENXIO) {
+ if (verbose)
+ ui__warning("%s event is not supported by the kernel.\n",
+ event_name(counter));
+ counter->supported = false;
+ continue;
+ }
+
+ if (errno == EPERM || errno == EACCES) {
error("You may not have permission to collect %sstats.\n"
"\t Consider tweaking"
" /proc/sys/kernel/perf_event_paranoid or running as root.",
system_wide ? "system-wide " : "");
- } else if (errno == ENOENT) {
- error("%s event is not supported. ", event_name(counter));
} else {
error("open_counter returned with %d (%s). "
"/bin/dmesg may provide additional information.\n",
die("Not all events could be opened.\n");
return -1;
}
+ counter->supported = true;
}
if (perf_evlist__set_filters(evsel_list)) {
if (forks) {
close(go_pipe[1]);
wait(&status);
+ if (WIFSIGNALED(status))
+ psignal(WTERMSIG(status), argv[0]);
} else {
while(!done) sleep(1);
}
return WEXITSTATUS(status);
}
+static void print_noise_pct(double total, double avg)
+{
+ double pct = 0.0;
+
+ if (avg)
+ pct = 100.0*total/avg;
+
+ if (csv_output)
+ fprintf(output, "%s%.2f%%", csv_sep, pct);
+ else if (pct)
+ fprintf(output, " ( +-%6.2f%% )", pct);
+}
+
static void print_noise(struct perf_evsel *evsel, double avg)
{
struct perf_stat *ps;
return;
ps = evsel->priv;
- fprintf(stderr, " ( +- %7.3f%% )",
- 100 * stddev_stats(&ps->res_stats[0]) / avg);
+ print_noise_pct(stddev_stats(&ps->res_stats[0]), avg);
}
static void nsec_printout(int cpu, struct perf_evsel *evsel, double avg)
{
double msecs = avg / 1e6;
char cpustr[16] = { '\0', };
- const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-24s";
+ const char *fmt = csv_output ? "%s%.6f%s%s" : "%s%18.6f%s%-25s";
if (no_aggr)
sprintf(cpustr, "CPU%*d%s",
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep);
- fprintf(stderr, fmt, cpustr, msecs, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, msecs, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
if (perf_evsel__match(evsel, SOFTWARE, SW_TASK_CLOCK))
- fprintf(stderr, " # %8.3f CPUs utilized ", avg / avg_stats(&walltime_nsecs_stats));
+ fprintf(output, " # %8.3f CPUs utilized ",
+ avg / avg_stats(&walltime_nsecs_stats));
+ else
+ fprintf(output, " ");
+}
+
+/* used for get_ratio_color() */
+enum grc_type {
+ GRC_STALLED_CYCLES_FE,
+ GRC_STALLED_CYCLES_BE,
+ GRC_CACHE_MISSES,
+ GRC_MAX_NR
+};
+
+static const char *get_ratio_color(enum grc_type type, double ratio)
+{
+ static const double grc_table[GRC_MAX_NR][3] = {
+ [GRC_STALLED_CYCLES_FE] = { 50.0, 30.0, 10.0 },
+ [GRC_STALLED_CYCLES_BE] = { 75.0, 50.0, 20.0 },
+ [GRC_CACHE_MISSES] = { 20.0, 10.0, 5.0 },
+ };
+ const char *color = PERF_COLOR_NORMAL;
+
+ if (ratio > grc_table[type][0])
+ color = PERF_COLOR_RED;
+ else if (ratio > grc_table[type][1])
+ color = PERF_COLOR_MAGENTA;
+ else if (ratio > grc_table[type][2])
+ color = PERF_COLOR_YELLOW;
+
+ return color;
+}
+
+static void print_stalled_cycles_frontend(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_cycles_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_STALLED_CYCLES_FE, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " frontend cycles idle ");
+}
+
+static void print_stalled_cycles_backend(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_cycles_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_STALLED_CYCLES_BE, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " backend cycles idle ");
+}
+
+static void print_branch_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_branches_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all branches ");
+}
+
+static void print_l1_dcache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_l1_dcache_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-dcache hits ");
+}
+
+static void print_l1_icache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_l1_icache_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all L1-icache hits ");
+}
+
+static void print_dtlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_dtlb_cache_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all dTLB cache hits ");
+}
+
+static void print_itlb_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_itlb_cache_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all iTLB cache hits ");
+}
+
+static void print_ll_cache_misses(int cpu, struct perf_evsel *evsel __used, double avg)
+{
+ double total, ratio = 0.0;
+ const char *color;
+
+ total = avg_stats(&runtime_ll_cache_stats[cpu]);
+
+ if (total)
+ ratio = avg / total * 100.0;
+
+ color = get_ratio_color(GRC_CACHE_MISSES, ratio);
+
+ fprintf(output, " # ");
+ color_fprintf(output, color, "%6.2f%%", ratio);
+ fprintf(output, " of all LL-cache hits ");
}
static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (csv_output)
fmt = "%s%.0f%s%s";
else if (big_num)
- fmt = "%s%'18.0f%s%-24s";
+ fmt = "%s%'18.0f%s%-25s";
else
- fmt = "%s%18.0f%s%-24s";
+ fmt = "%s%18.0f%s%-25s";
if (no_aggr)
sprintf(cpustr, "CPU%*d%s",
else
cpu = 0;
- fprintf(stderr, fmt, cpustr, avg, csv_sep, event_name(evsel));
+ fprintf(output, fmt, cpustr, avg, csv_sep, event_name(evsel));
if (evsel->cgrp)
- fprintf(stderr, "%s%s", csv_sep, evsel->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, evsel->cgrp->name);
if (csv_output)
return;
if (total)
ratio = avg / total;
- fprintf(stderr, " # %4.2f insns per cycle", ratio);
+ fprintf(output, " # %5.2f insns per cycle ", ratio);
- total = avg_stats(&runtime_stalled_cycles_stats[cpu]);
+ total = avg_stats(&runtime_stalled_cycles_front_stats[cpu]);
+ total = max(total, avg_stats(&runtime_stalled_cycles_back_stats[cpu]));
if (total && avg) {
ratio = total / avg;
- fprintf(stderr, "\n # %4.2f stalled cycles per insn", ratio);
+ fprintf(output, "\n # %5.2f stalled cycles per insn", ratio);
}
} else if (perf_evsel__match(evsel, HARDWARE, HW_BRANCH_MISSES) &&
runtime_branches_stats[cpu].n != 0) {
- total = avg_stats(&runtime_branches_stats[cpu]);
-
- if (total)
- ratio = avg * 100 / total;
-
- fprintf(stderr, " # %5.2f %% of all branches ", ratio);
-
+ print_branch_misses(cpu, evsel, avg);
+ } else if (
+ evsel->attr.type == PERF_TYPE_HW_CACHE &&
+ evsel->attr.config == ( PERF_COUNT_HW_CACHE_L1D |
+ ((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+ ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+ runtime_l1_dcache_stats[cpu].n != 0) {
+ print_l1_dcache_misses(cpu, evsel, avg);
+ } else if (
+ evsel->attr.type == PERF_TYPE_HW_CACHE &&
+ evsel->attr.config == ( PERF_COUNT_HW_CACHE_L1I |
+ ((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+ ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+ runtime_l1_icache_stats[cpu].n != 0) {
+ print_l1_icache_misses(cpu, evsel, avg);
+ } else if (
+ evsel->attr.type == PERF_TYPE_HW_CACHE &&
+ evsel->attr.config == ( PERF_COUNT_HW_CACHE_DTLB |
+ ((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+ ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+ runtime_dtlb_cache_stats[cpu].n != 0) {
+ print_dtlb_cache_misses(cpu, evsel, avg);
+ } else if (
+ evsel->attr.type == PERF_TYPE_HW_CACHE &&
+ evsel->attr.config == ( PERF_COUNT_HW_CACHE_ITLB |
+ ((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+ ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+ runtime_itlb_cache_stats[cpu].n != 0) {
+ print_itlb_cache_misses(cpu, evsel, avg);
+ } else if (
+ evsel->attr.type == PERF_TYPE_HW_CACHE &&
+ evsel->attr.config == ( PERF_COUNT_HW_CACHE_LL |
+ ((PERF_COUNT_HW_CACHE_OP_READ) << 8) |
+ ((PERF_COUNT_HW_CACHE_RESULT_MISS) << 16)) &&
+ runtime_ll_cache_stats[cpu].n != 0) {
+ print_ll_cache_misses(cpu, evsel, avg);
} else if (perf_evsel__match(evsel, HARDWARE, HW_CACHE_MISSES) &&
runtime_cacherefs_stats[cpu].n != 0) {
total = avg_stats(&runtime_cacherefs_stats[cpu]);
if (total)
ratio = avg * 100 / total;
- fprintf(stderr, " # %8.3f %% of all cache refs ", ratio);
-
- } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES)) {
- total = avg_stats(&runtime_cycles_stats[cpu]);
+ fprintf(output, " # %8.3f %% of all cache refs ", ratio);
- if (total)
- ratio = avg / total * 100.0;
-
- fprintf(stderr, " # %5.2f%% of all cycles are idle ", ratio);
+ } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_FRONTEND)) {
+ print_stalled_cycles_frontend(cpu, evsel, avg);
+ } else if (perf_evsel__match(evsel, HARDWARE, HW_STALLED_CYCLES_BACKEND)) {
+ print_stalled_cycles_backend(cpu, evsel, avg);
} else if (perf_evsel__match(evsel, HARDWARE, HW_CPU_CYCLES)) {
total = avg_stats(&runtime_nsecs_stats[cpu]);
if (total)
ratio = 1.0 * avg / total;
- fprintf(stderr, " # %8.3f GHz ", ratio);
+ fprintf(output, " # %8.3f GHz ", ratio);
} else if (runtime_nsecs_stats[cpu].n != 0) {
+ char unit = 'M';
+
total = avg_stats(&runtime_nsecs_stats[cpu]);
if (total)
ratio = 1000.0 * avg / total;
+ if (ratio < 0.001) {
+ ratio *= 1000;
+ unit = 'K';
+ }
- fprintf(stderr, " # %8.3f M/sec ", ratio);
+ fprintf(output, " # %8.3f %c/sec ", ratio, unit);
+ } else {
+ fprintf(output, " ");
}
}
int scaled = counter->counts->scaled;
if (scaled == -1) {
- fprintf(stderr, "%*s%s%*s",
+ fprintf(output, "%*s%s%*s",
csv_output ? 0 : 18,
- "<not counted>",
+ counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
csv_sep,
csv_output ? 0 : -24,
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s", csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
else
abs_printout(-1, counter, avg);
+ print_noise(counter, avg);
+
if (csv_output) {
- fputc('\n', stderr);
+ fputc('\n', output);
return;
}
- print_noise(counter, avg);
-
if (scaled) {
double avg_enabled, avg_running;
avg_enabled = avg_stats(&ps->res_stats[1]);
avg_running = avg_stats(&ps->res_stats[2]);
- fprintf(stderr, " (scaled from %.2f%%)",
- 100 * avg_running / avg_enabled);
+ fprintf(output, " [%5.2f%%]", 100 * avg_running / avg_enabled);
}
- fprintf(stderr, "\n");
+ fprintf(output, "\n");
}
/*
ena = counter->counts->cpu[cpu].ena;
run = counter->counts->cpu[cpu].run;
if (run == 0 || ena == 0) {
- fprintf(stderr, "CPU%*d%s%*s%s%*s",
+ fprintf(output, "CPU%*d%s%*s%s%*s",
csv_output ? 0 : -4,
evsel_list->cpus->map[cpu], csv_sep,
csv_output ? 0 : 18,
- "<not counted>", csv_sep,
+ counter->supported ? CNTR_NOT_COUNTED : CNTR_NOT_SUPPORTED,
+ csv_sep,
csv_output ? 0 : -24,
event_name(counter));
if (counter->cgrp)
- fprintf(stderr, "%s%s", csv_sep, counter->cgrp->name);
+ fprintf(output, "%s%s",
+ csv_sep, counter->cgrp->name);
- fputc('\n', stderr);
+ fputc('\n', output);
continue;
}
if (!csv_output) {
print_noise(counter, 1.0);
- if (run != ena) {
- fprintf(stderr, " (scaled from %.2f%%)",
+ if (run != ena)
+ fprintf(output, " (%.2f%%)",
100.0 * run / ena);
- }
}
- fputc('\n', stderr);
+ fputc('\n', output);
}
}
fflush(stdout);
if (!csv_output) {
- fprintf(stderr, "\n");
- fprintf(stderr, " Performance counter stats for ");
- if(target_pid == -1 && target_tid == -1) {
- fprintf(stderr, "\'%s", argv[0]);
+ fprintf(output, "\n");
+ fprintf(output, " Performance counter stats for ");
+ if (!target_pid && !target_tid) {
+ fprintf(output, "\'%s", argv[0]);
for (i = 1; i < argc; i++)
- fprintf(stderr, " %s", argv[i]);
- } else if (target_pid != -1)
- fprintf(stderr, "process id \'%d", target_pid);
+ fprintf(output, " %s", argv[i]);
+ } else if (target_pid)
+ fprintf(output, "process id \'%s", target_pid);
else
- fprintf(stderr, "thread id \'%d", target_tid);
+ fprintf(output, "thread id \'%s", target_tid);
- fprintf(stderr, "\'");
+ fprintf(output, "\'");
if (run_count > 1)
- fprintf(stderr, " (%d runs)", run_count);
- fprintf(stderr, ":\n\n");
+ fprintf(output, " (%d runs)", run_count);
+ fprintf(output, ":\n\n");
}
if (no_aggr) {
}
if (!csv_output) {
- fprintf(stderr, "\n");
- fprintf(stderr, " %18.9f seconds time elapsed",
+ if (!null_run)
+ fprintf(output, "\n");
+ fprintf(output, " %17.9f seconds time elapsed",
avg_stats(&walltime_nsecs_stats)/1e9);
if (run_count > 1) {
- fprintf(stderr, " ( +-%5.2f%% )",
- 100*stddev_stats(&walltime_nsecs_stats) /
- avg_stats(&walltime_nsecs_stats));
+ fprintf(output, " ");
+ print_noise_pct(stddev_stats(&walltime_nsecs_stats),
+ avg_stats(&walltime_nsecs_stats));
}
- fprintf(stderr, "\n\n");
+ fprintf(output, "\n\n");
}
}
return 0;
}
+static bool append_file;
+
static const struct option options[] = {
OPT_CALLBACK('e', "event", &evsel_list, "event",
"event selector. use 'perf list' to list available events",
- parse_events),
+ parse_events_option),
OPT_CALLBACK(0, "filter", &evsel_list, "filter",
"event filter", parse_filter),
OPT_BOOLEAN('i', "no-inherit", &no_inherit,
"child tasks do not inherit counters"),
- OPT_INTEGER('p', "pid", &target_pid,
- "stat events on existing process id"),
- OPT_INTEGER('t', "tid", &target_tid,
- "stat events on existing thread id"),
+ OPT_STRING('p', "pid", &target_pid, "pid",
+ "stat events on existing process id"),
+ OPT_STRING('t', "tid", &target_tid, "tid",
+ "stat events on existing thread id"),
OPT_BOOLEAN('a', "all-cpus", &system_wide,
"system-wide collection from all CPUs"),
+ OPT_BOOLEAN('g', "group", &group,
+ "put the counters into a counter group"),
OPT_BOOLEAN('c', "scale", &scale,
"scale/normalize counters"),
OPT_INCR('v', "verbose", &verbose,
"repeat command and print average + stddev (max: 100)"),
OPT_BOOLEAN('n', "null", &null_run,
"null run - dont start any counters"),
+ OPT_INCR('d', "detailed", &detailed_run,
+ "detailed run - start a lot of events"),
+ OPT_BOOLEAN('S', "sync", &sync_run,
+ "call sync() before starting a run"),
OPT_CALLBACK_NOOPT('B', "big-num", NULL, NULL,
"print large numbers with thousands\' separators",
stat__set_big_num),
OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
"monitor event in cgroup name only",
parse_cgroups),
+ OPT_STRING('o', "output", &output_name, "file",
+ "output file name"),
+ OPT_BOOLEAN(0, "append", &append_file, "append to the output file"),
+ OPT_INTEGER(0, "log-fd", &output_fd,
+ "log output to fd, instead of stderr"),
OPT_END()
};
+/*
+ * Add default attributes, if there were no attributes specified or
+ * if -d/--detailed, -d -d or -d -d -d is used:
+ */
+static int add_default_attributes(void)
+{
+ /* Set attrs if no event is selected and !null_run: */
+ if (null_run)
+ return 0;
+
+ if (!evsel_list->nr_entries) {
+ if (perf_evlist__add_attrs_array(evsel_list, default_attrs) < 0)
+ return -1;
+ }
+
+ /* Detailed events get appended to the event list: */
+
+ if (detailed_run < 1)
+ return 0;
+
+ /* Append detailed run extra attributes: */
+ if (perf_evlist__add_attrs_array(evsel_list, detailed_attrs) < 0)
+ return -1;
+
+ if (detailed_run < 2)
+ return 0;
+
+ /* Append very detailed run extra attributes: */
+ if (perf_evlist__add_attrs_array(evsel_list, very_detailed_attrs) < 0)
+ return -1;
+
+ if (detailed_run < 3)
+ return 0;
+
+ /* Append very, very detailed run extra attributes: */
+ return perf_evlist__add_attrs_array(evsel_list, very_very_detailed_attrs);
+}
+
int cmd_stat(int argc, const char **argv, const char *prefix __used)
{
struct perf_evsel *pos;
int status = -ENOMEM;
+ const char *mode;
setlocale(LC_ALL, "");
argc = parse_options(argc, argv, options, stat_usage,
PARSE_OPT_STOP_AT_NON_OPTION);
- if (csv_sep)
+ output = stderr;
+ if (output_name && strcmp(output_name, "-"))
+ output = NULL;
+
+ if (output_name && output_fd) {
+ fprintf(stderr, "cannot use both --output and --log-fd\n");
+ usage_with_options(stat_usage, options);
+ }
+ if (!output) {
+ struct timespec tm;
+ mode = append_file ? "a" : "w";
+
+ output = fopen(output_name, mode);
+ if (!output) {
+ perror("failed to create output file");
+ exit(-1);
+ }
+ clock_gettime(CLOCK_REALTIME, &tm);
+ fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
+ } else if (output_fd != 2) {
+ mode = append_file ? "a" : "w";
+ output = fdopen(output_fd, mode);
+ if (!output) {
+ perror("Failed opening logfd");
+ return -errno;
+ }
+ }
+
+ if (csv_sep) {
csv_output = true;
- else
+ if (!strcmp(csv_sep, "\\t"))
+ csv_sep = "\t";
+ } else
csv_sep = DEFAULT_SEPARATOR;
/*
* let the spreadsheet do the pretty-printing
*/
if (csv_output) {
- /* User explicitely passed -B? */
+ /* User explicitly passed -B? */
if (big_num_opt == 1) {
fprintf(stderr, "-B option not supported with -x\n");
usage_with_options(stat_usage, options);
} else if (big_num_opt == 0) /* User passed --no-big-num */
big_num = false;
- if (!argc && target_pid == -1 && target_tid == -1)
+ if (!argc && !target_pid && !target_tid)
usage_with_options(stat_usage, options);
if (run_count <= 0)
usage_with_options(stat_usage, options);
usage_with_options(stat_usage, options);
}
- /* Set attrs and nr_counters if no event is selected and !null_run */
- if (!null_run && !evsel_list->nr_entries) {
- size_t c;
+ if (add_default_attributes())
+ goto out;
- for (c = 0; c < ARRAY_SIZE(default_attrs); ++c) {
- pos = perf_evsel__new(&default_attrs[c], c);
- if (pos == NULL)
- goto out;
- perf_evlist__add(evsel_list, pos);
- }
- }
-
- if (target_pid != -1)
+ if (target_pid)
target_tid = target_pid;
- evsel_list->threads = thread_map__new(target_pid, target_tid);
+ evsel_list->threads = thread_map__new_str(target_pid,
+ target_tid, UINT_MAX);
if (evsel_list->threads == NULL) {
pr_err("Problems finding threads of monitor\n");
usage_with_options(stat_usage, options);
list_for_each_entry(pos, &evsel_list->entries, node) {
if (perf_evsel__alloc_stat_priv(pos) < 0 ||
- perf_evsel__alloc_counts(pos, evsel_list->cpus->nr) < 0 ||
- perf_evsel__alloc_fd(pos, evsel_list->cpus->nr, evsel_list->threads->nr) < 0)
+ perf_evsel__alloc_counts(pos, evsel_list->cpus->nr) < 0)
goto out_free_fd;
}
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);
+ fprintf(output, "[ perf stat: executing run #%d ... ]\n",
+ run_idx + 1);
+
+ if (sync_run)
+ sync();
+
status = run_perf_stat(argc, argv);
}
projects / linux-flexiantxendom0-3.2.10.git / blobdiff