treewide: refresh kmalloc_obj() conversionsgrafted

This is another run of the Coccinelle script for converting kmalloc()
family of allocations to kmalloc_obj() via the existing rules in
scripts/coccinelle/api/kmalloc_objs.cocci

This catches both the set of kmalloc() uses added since the first
kmalloc_obj() conversions in v7.0 and adds a large group missed in the
first pass due to Coccinelle not interacting well with the cleanup.h
scoped_...() family of macros[1]. I worked around this with spatch's
"--macro-file" argument to a file with all the scoped_...() macros mapped
to Coccinelle's YACFE_ITERATOR[2] as that was the closest viable control
flow indicator I could find.

Build tested allmodconfig on x86, arm64, arm, loongarch, mips, powerpc,
riscv, and s390 with no new warnings.

Link: https://lore.kernel.org/lkml/202609021314.8A9C0B8@keescook/ [1]
Link: https://github.com/coccinelle/coccinelle/blob/master/standard.h [2]
Signed-off-by: Kees Cook <kees+treewide@kernel.org>

1 files changed, 3103 insertions, 0 deletions

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
new file mode 100644
index 000000000..c50cf230f
--- /dev/null
+++ b/tools/perf/builtin-stat.c
@@ -0,0 +1,3103 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * 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: 0.118
+
+  Performance counter stats for './hackbench 10':
+
+       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) 2008-2011, 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>
+ */
+
+#include "builtin.h"
+#include "util/cgroup.h"
+#include <subcmd/parse-options.h>
+#include "util/parse-events.h"
+#include "util/pmus.h"
+#include "util/pmu.h"
+#include "util/tool_pmu.h"
+#include "util/event.h"
+#include "util/evlist.h"
+#include "util/evsel.h"
+#include "util/debug.h"
+#include "util/color.h"
+#include "util/stat.h"
+#include "util/header.h"
+#include "util/cpumap.h"
+#include "util/thread_map.h"
+#include "util/counts.h"
+#include "util/topdown.h"
+#include "util/session.h"
+#include "util/tool.h"
+#include "util/string2.h"
+#include "util/metricgroup.h"
+#include "util/synthetic-events.h"
+#include "util/target.h"
+#include "util/time-utils.h"
+#include "util/top.h"
+#include "util/affinity.h"
+#include "util/pfm.h"
+#include "util/bpf_counter.h"
+#include "util/iostat.h"
+#include "util/util.h"
+#include "util/intel-tpebs.h"
+#include "asm/bug.h"
+
+#include <linux/list_sort.h>
+#include <linux/time64.h>
+#include <linux/zalloc.h>
+#include <api/fs/fs.h>
+#include <errno.h>
+#include <signal.h>
+#include <stdlib.h>
+#include <sys/prctl.h>
+#include <inttypes.h>
+#include <locale.h>
+#include <math.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/wait.h>
+#include <unistd.h>
+#include <sys/time.h>
+#include <sys/resource.h>
+#include <linux/err.h>
+
+#include <linux/ctype.h>
+#include <perf/evlist.h>
+#include <internal/threadmap.h>
+
+#ifdef HAVE_BPF_SKEL
+#include "util/bpf_skel/bperf_cgroup.h"
+#endif
+
+#define DEFAULT_SEPARATOR	" "
+#define FREEZE_ON_SMI_PATH	"bus/event_source/devices/cpu/freeze_on_smi"
+
+struct rusage_stats {
+	struct stats ru_utime_usec_stat;
+	struct stats ru_stime_usec_stat;
+};
+
+static void print_counters(struct timespec *ts, int argc, const char **argv);
+
+static struct evlist	*evsel_list;
+static struct parse_events_option_args parse_events_option_args = {
+	.evlistp = &evsel_list,
+};
+
+static bool all_counters_use_bpf = true;
+
+static struct target target;
+
+static volatile sig_atomic_t	child_pid			= -1;
+static int			detailed_run			=  0;
+static bool			transaction_run;
+static bool			topdown_run			= false;
+static bool			smi_cost			= false;
+static bool			smi_reset			= false;
+static int			big_num_opt			=  -1;
+static const char		*pre_cmd			= NULL;
+static const char		*post_cmd			= NULL;
+static bool			sync_run			= false;
+static bool			forever				= false;
+static bool			force_metric_only		= false;
+static struct timespec		ref_time;
+static bool			append_file;
+static bool			interval_count;
+static const char		*output_name;
+static int			output_fd;
+static char			*metrics;
+static struct rusage_stats	ru_stats;
+
+struct perf_stat {
+	bool			 record;
+	struct perf_data	 data;
+	struct perf_session	*session;
+	u64			 bytes_written;
+	struct perf_tool	 tool;
+	bool			 maps_allocated;
+	struct perf_cpu_map	*cpus;
+	struct perf_thread_map *threads;
+	enum aggr_mode		 aggr_mode;
+	u32			 aggr_level;
+};
+
+static struct perf_stat		perf_stat;
+#define STAT_RECORD		perf_stat.record
+
+static volatile sig_atomic_t done = 0;
+
+/* Options set from the command line. */
+struct opt_aggr_mode {
+	bool node, socket, die, cluster, cache, core, thread, no_aggr;
+};
+
+/* Turn command line option into most generic aggregation mode setting. */
+static enum aggr_mode opt_aggr_mode_to_aggr_mode(const struct opt_aggr_mode *opt_mode)
+{
+	enum aggr_mode mode = AGGR_GLOBAL;
+
+	if (opt_mode->node)
+		mode = AGGR_NODE;
+	if (opt_mode->socket)
+		mode = AGGR_SOCKET;
+	if (opt_mode->die)
+		mode = AGGR_DIE;
+	if (opt_mode->cluster)
+		mode = AGGR_CLUSTER;
+	if (opt_mode->cache)
+		mode = AGGR_CACHE;
+	if (opt_mode->core)
+		mode = AGGR_CORE;
+	if (opt_mode->thread)
+		mode = AGGR_THREAD;
+	if (opt_mode->no_aggr)
+		mode = AGGR_NONE;
+	return mode;
+}
+
+static void evlist__check_cpu_maps(struct evlist *evlist)
+{
+	struct evsel *evsel, *warned_leader = NULL;
+
+	evlist__for_each_entry(evlist, evsel) {
+		struct evsel *leader = evsel__leader(evsel);
+
+		/* Check that leader matches cpus with each member. */
+		if (leader == evsel)
+			continue;
+		if (perf_cpu_map__equal(leader->core.cpus, evsel->core.cpus))
+			continue;
+
+		/* If there's mismatch disable the group and warn user. */
+		if (warned_leader != leader) {
+			char buf[200];
+
+			pr_warning("WARNING: grouped events cpus do not match.\n"
+				"Events with CPUs not matching the leader will "
+				"be removed from the group.\n");
+			evsel__group_desc(leader, buf, sizeof(buf));
+			pr_warning("  %s\n", buf);
+			warned_leader = leader;
+		}
+		if (verbose > 0) {
+			char buf[200];
+
+			cpu_map__snprint(leader->core.cpus, buf, sizeof(buf));
+			pr_warning("     %s: %s\n", leader->name, buf);
+			cpu_map__snprint(evsel->core.cpus, buf, sizeof(buf));
+			pr_warning("     %s: %s\n", evsel->name, buf);
+		}
+
+		evsel__remove_from_group(evsel, leader);
+	}
+}
+
+static inline void diff_timespec(struct timespec *r, struct timespec *a,
+				 struct timespec *b)
+{
+	r->tv_sec = a->tv_sec - b->tv_sec;
+	if (a->tv_nsec < b->tv_nsec) {
+		r->tv_nsec = a->tv_nsec + NSEC_PER_SEC - b->tv_nsec;
+		r->tv_sec--;
+	} else {
+		r->tv_nsec = a->tv_nsec - b->tv_nsec ;
+	}
+}
+
+static void perf_stat__reset_stats(void)
+{
+	evlist__reset_stats(evsel_list);
+	memset(stat_config.walltime_nsecs_stats, 0, sizeof(*stat_config.walltime_nsecs_stats));
+}
+
+static int process_synthesized_event(const struct perf_tool *tool __maybe_unused,
+				     union perf_event *event,
+				     struct perf_sample *sample __maybe_unused,
+				     struct machine *machine __maybe_unused)
+{
+	if (perf_data__write(&perf_stat.data, event, event->header.size) < 0) {
+		pr_err("failed to write perf data, error: %m\n");
+		return -1;
+	}
+
+	perf_stat.bytes_written += event->header.size;
+	return 0;
+}
+
+static int write_stat_round_event(u64 tm, u64 type)
+{
+	return perf_event__synthesize_stat_round(NULL, tm, type,
+						 process_synthesized_event,
+						 NULL);
+}
+
+#define WRITE_STAT_ROUND_EVENT(time, interval) \
+	write_stat_round_event(time, PERF_STAT_ROUND_TYPE__ ## interval)
+
+#define SID(e, x, y) xyarray__entry(e->core.sample_id, x, y)
+
+static int evsel__write_stat_event(struct evsel *counter, int cpu_map_idx, u32 thread,
+				   struct perf_counts_values *count)
+{
+	struct perf_sample_id *sid = SID(counter, cpu_map_idx, thread);
+	struct perf_cpu cpu = perf_cpu_map__cpu(evsel__cpus(counter), cpu_map_idx);
+
+	return perf_event__synthesize_stat(NULL, cpu, thread, sid->id, count,
+					   process_synthesized_event, NULL);
+}
+
+static int read_single_counter(struct evsel *counter, int cpu_map_idx, int thread)
+{
+	int err = evsel__read_counter(counter, cpu_map_idx, thread);
+
+	/*
+	 * Reading user and system time will fail when the process
+	 * terminates. Use the wait4 values in that case.
+	 */
+	if (err && cpu_map_idx == 0 &&
+	    (evsel__tool_event(counter) == TOOL_PMU__EVENT_USER_TIME ||
+	     evsel__tool_event(counter) == TOOL_PMU__EVENT_SYSTEM_TIME)) {
+		struct perf_counts_values *count =
+			perf_counts(counter->counts, cpu_map_idx, thread);
+		struct perf_counts_values *old_count = NULL;
+		u64 val;
+
+		if (counter->prev_raw_counts)
+			old_count = perf_counts(counter->prev_raw_counts, cpu_map_idx, thread);
+
+		if (evsel__tool_event(counter) == TOOL_PMU__EVENT_USER_TIME)
+			val = ru_stats.ru_utime_usec_stat.mean;
+		else
+			val = ru_stats.ru_stime_usec_stat.mean;
+
+		count->val = val;
+		if (old_count) {
+			count->run = old_count->run + 1;
+			count->ena = old_count->ena + 1;
+		} else {
+			count->run++;
+			count->ena++;
+		}
+		return 0;
+	}
+	return err;
+}
+
+/*
+ * Read out the results of a single counter:
+ * do not aggregate counts across CPUs in system-wide mode
+ */
+static int read_counter_cpu(struct evsel *counter, int cpu_map_idx)
+{
+	int nthreads = perf_thread_map__nr(evlist__core(evsel_list)->threads);
+	int thread;
+
+	if (!counter->supported)
+		return -ENOENT;
+
+	for (thread = 0; thread < nthreads; thread++) {
+		struct perf_counts_values *count;
+
+		count = perf_counts(counter->counts, cpu_map_idx, thread);
+
+		/*
+		 * The leader's group read loads data into its group members
+		 * (via evsel__read_counter()) and sets their count->loaded.
+		 */
+		if (!perf_counts__is_loaded(counter->counts, cpu_map_idx, thread) &&
+		    read_single_counter(counter, cpu_map_idx, thread)) {
+			counter->counts->scaled = -1;
+			perf_counts(counter->counts, cpu_map_idx, thread)->ena = 0;
+			perf_counts(counter->counts, cpu_map_idx, thread)->run = 0;
+			return -1;
+		}
+
+		perf_counts__set_loaded(counter->counts, cpu_map_idx, thread, false);
+
+		if (STAT_RECORD) {
+			if (evsel__write_stat_event(counter, cpu_map_idx, thread, count)) {
+				pr_err("failed to write stat event\n");
+				return -1;
+			}
+		}
+
+		if (verbose > 1) {
+			fprintf(stat_config.output,
+				"%s: %d: %" PRIu64 " %" PRIu64 " %" PRIu64 "\n",
+					evsel__name(counter),
+					perf_cpu_map__cpu(evsel__cpus(counter),
+							  cpu_map_idx).cpu,
+					count->val, count->ena, count->run);
+		}
+	}
+
+	return 0;
+}
+
+static int read_counters_with_affinity(void)
+{
+	struct evlist_cpu_iterator evlist_cpu_itr;
+
+	if (all_counters_use_bpf)
+		return 0;
+
+	evlist__for_each_cpu(evlist_cpu_itr, evsel_list) {
+		struct evsel *counter = evlist_cpu_itr.evsel;
+
+		if (evsel__is_bpf(counter))
+			continue;
+
+		if (evsel__is_tool(counter))
+			continue;
+
+		if (!counter->err)
+			counter->err = read_counter_cpu(counter, evlist_cpu_itr.cpu_map_idx);
+	}
+
+	return 0;
+}
+
+static int read_bpf_map_counters(void)
+{
+	struct evsel *counter;
+	int err;
+
+	evlist__for_each_entry(evsel_list, counter) {
+		if (!evsel__is_bpf(counter))
+			continue;
+
+		err = bpf_counter__read(counter);
+		if (err)
+			return err;
+	}
+	return 0;
+}
+
+static int read_tool_counters(void)
+{
+	struct evsel *counter;
+
+	evlist__for_each_entry(evsel_list, counter) {
+		unsigned int idx;
+
+		if (!evsel__is_tool(counter))
+			continue;
+
+		perf_cpu_map__for_each_idx(idx, counter->core.cpus) {
+			if (!counter->err)
+				counter->err = read_counter_cpu(counter, idx);
+		}
+	}
+	return 0;
+}
+
+static int read_counters(void)
+{
+	int ret;
+
+	if (stat_config.stop_read_counter)
+		return 0;
+
+	// Read all BPF counters first.
+	ret = read_bpf_map_counters();
+	if (ret)
+		return ret;
+
+	// Read non-BPF and non-tool counters next.
+	ret = read_counters_with_affinity();
+	if (ret)
+		return ret;
+
+	// Read the tool counters last. This way the duration_time counter
+	// should always be greater than any other counter's enabled time.
+	return read_tool_counters();
+}
+
+static void process_counters(void)
+{
+	struct evsel *counter;
+
+	evlist__for_each_entry(evsel_list, counter) {
+		if (counter->err)
+			pr_debug("failed to read counter %s\n", counter->name);
+		if (counter->err == 0 && perf_stat_process_counter(&stat_config, counter))
+			pr_warning("failed to process counter %s\n", counter->name);
+		counter->err = 0;
+	}
+
+	perf_stat_merge_counters(&stat_config, evsel_list);
+	perf_stat_process_percore(&stat_config, evsel_list);
+}
+
+static void process_interval(void)
+{
+	struct timespec ts, rs;
+
+	clock_gettime(CLOCK_MONOTONIC, &ts);
+	diff_timespec(&rs, &ts, &ref_time);
+
+	evlist__reset_aggr_stats(evsel_list);
+
+	if (read_counters() == 0)
+		process_counters();
+
+	if (STAT_RECORD) {
+		if (WRITE_STAT_ROUND_EVENT(rs.tv_sec * NSEC_PER_SEC + rs.tv_nsec, INTERVAL))
+			pr_err("failed to write stat round event\n");
+	}
+
+	init_stats(stat_config.walltime_nsecs_stats);
+	update_stats(stat_config.walltime_nsecs_stats, stat_config.interval * 1000000ULL);
+	print_counters(&rs, 0, NULL);
+}
+
+static bool handle_interval(unsigned int interval, int *times)
+{
+	if (interval) {
+		process_interval();
+		if (interval_count && !(--(*times)))
+			return true;
+	}
+	return false;
+}
+
+static int enable_counters(void)
+{
+	struct evsel *evsel;
+	int err;
+
+	evlist__for_each_entry(evsel_list, evsel) {
+		if (!evsel__is_bpf(evsel))
+			continue;
+
+		err = bpf_counter__enable(evsel);
+		if (err)
+			return err;
+	}
+
+	if (!target__enable_on_exec(&target)) {
+		if (!all_counters_use_bpf)
+			evlist__enable(evsel_list);
+	}
+	return 0;
+}
+
+static void disable_counters(void)
+{
+	struct evsel *counter;
+
+	/*
+	 * If we don't have tracee (attaching to task or cpu), counters may
+	 * still be running. To get accurate group ratios, we must stop groups
+	 * from counting before reading their constituent counters.
+	 */
+	if (!target__none(&target)) {
+		evlist__for_each_entry(evsel_list, counter)
+			bpf_counter__disable(counter);
+		if (!all_counters_use_bpf)
+			evlist__disable(evsel_list);
+	}
+}
+
+static volatile sig_atomic_t workload_exec_errno;
+
+/*
+ * evlist__prepare_workload will send a SIGUSR1
+ * if the fork fails, since we asked by setting its
+ * want_signal to true.
+ */
+static void workload_exec_failed_signal(int signo __maybe_unused, siginfo_t *info,
+					void *ucontext __maybe_unused)
+{
+	workload_exec_errno = info->si_value.sival_int;
+}
+
+static bool evsel__should_store_id(struct evsel *counter)
+{
+	return STAT_RECORD || counter->core.attr.read_format & PERF_FORMAT_ID;
+}
+
+static bool is_target_alive(struct target *_target,
+			    struct perf_thread_map *threads)
+{
+	struct stat st;
+	int i;
+
+	if (!target__has_task(_target))
+		return true;
+
+	for (i = 0; i < threads->nr; i++) {
+		char path[PATH_MAX];
+
+		scnprintf(path, PATH_MAX, "%s/%d", procfs__mountpoint(),
+			  threads->map[i].pid);
+
+		if (!stat(path, &st))
+			return true;
+	}
+
+	return false;
+}
+
+static void process_evlist(struct evlist *evlist, unsigned int interval)
+{
+	enum evlist_ctl_cmd cmd = EVLIST_CTL_CMD_UNSUPPORTED;
+
+	if (evlist__ctlfd_process(evlist, &cmd) > 0) {
+		switch (cmd) {
+		case EVLIST_CTL_CMD_ENABLE:
+			fallthrough;
+		case EVLIST_CTL_CMD_DISABLE:
+			if (interval)
+				process_interval();
+			break;
+		case EVLIST_CTL_CMD_SNAPSHOT:
+		case EVLIST_CTL_CMD_ACK:
+		case EVLIST_CTL_CMD_UNSUPPORTED:
+		case EVLIST_CTL_CMD_EVLIST:
+		case EVLIST_CTL_CMD_STOP:
+		case EVLIST_CTL_CMD_PING:
+		default:
+			break;
+		}
+	}
+}
+
+static void compute_tts(struct timespec *time_start, struct timespec *time_stop,
+			int *time_to_sleep)
+{
+	int tts = *time_to_sleep;
+	struct timespec time_diff;
+
+	diff_timespec(&time_diff, time_stop, time_start);
+
+	tts -= time_diff.tv_sec * MSEC_PER_SEC +
+	       time_diff.tv_nsec / NSEC_PER_MSEC;
+
+	if (tts < 0)
+		tts = 0;
+
+	*time_to_sleep = tts;
+}
+
+static int dispatch_events(bool forks, int timeout, int interval, int *times)
+{
+	int child_exited = 0, status = 0;
+	int time_to_sleep, sleep_time;
+	struct timespec time_start, time_stop;
+
+	if (interval)
+		sleep_time = interval;
+	else if (timeout)
+		sleep_time = timeout;
+	else
+		sleep_time = 1000;
+
+	time_to_sleep = sleep_time;
+
+	while (!done) {
+		if (forks) {
+			child_exited = waitpid(child_pid, &status, WNOHANG);
+		} else {
+			child_exited = !is_target_alive(&target,
+							evlist__core(evsel_list)->threads) ? 1 : 0;
+		}
+		if (child_exited)
+			break;
+
+		clock_gettime(CLOCK_MONOTONIC, &time_start);
+		if (!(evlist__poll(evsel_list, time_to_sleep) > 0)) { /* poll timeout or EINTR */
+			if (timeout || handle_interval(interval, times))
+				break;
+			time_to_sleep = sleep_time;
+		} else { /* fd revent */
+			process_evlist(evsel_list, interval);
+			clock_gettime(CLOCK_MONOTONIC, &time_stop);
+			compute_tts(&time_start, &time_stop, &time_to_sleep);
+		}
+	}
+
+	return status;
+}
+
+enum counter_recovery {
+	COUNTER_SKIP,
+	COUNTER_RETRY,
+};
+
+static enum counter_recovery stat_handle_error(struct evsel *counter, int err)
+{
+	char msg[BUFSIZ];
+
+	assert(!counter->supported);
+
+	/*
+	 * PPC returns ENXIO for HW counters until 2.6.37
+	 * (behavior changed with commit b0a873e).
+	 */
+	if (err == EINVAL || err == ENOSYS || err == ENOENT || err == ENXIO) {
+		if (verbose > 0) {
+			evsel__open_strerror(counter, &target, err, msg, sizeof(msg));
+			ui__warning("%s event is not supported by the kernel.\n%s\n",
+				    evsel__name(counter), msg);
+		}
+		return COUNTER_SKIP;
+	}
+	if (evsel__fallback(counter, &target, err, msg, sizeof(msg))) {
+		if (verbose > 0)
+			ui__warning("%s\n", msg);
+		counter->supported = true;
+		return COUNTER_RETRY;
+	}
+	if (target__has_per_thread(&target) && err != EOPNOTSUPP &&
+	    evlist__core(evsel_list)->threads &&
+	    evlist__core(evsel_list)->threads->err_thread != -1) {
+		/*
+		 * For global --per-thread case, skip current
+		 * error thread.
+		 */
+		if (!thread_map__remove(evlist__core(evsel_list)->threads,
+					evlist__core(evsel_list)->threads->err_thread)) {
+			evlist__core(evsel_list)->threads->err_thread = -1;
+			counter->supported = true;
+			return COUNTER_RETRY;
+		}
+	}
+	if (verbose > 0) {
+		evsel__open_strerror(counter, &target, err, msg, sizeof(msg));
+		ui__warning(err == EOPNOTSUPP
+			? "%s event is not supported by the kernel.\n%s\n"
+			: "skipping event %s that kernel failed to open.\n%s\n",
+			evsel__name(counter), msg);
+	}
+	return COUNTER_SKIP;
+}
+
+static int create_perf_stat_counter(struct evsel *evsel,
+				    struct perf_stat_config *config,
+				    int cpu_map_idx)
+{
+	struct perf_event_attr *attr = &evsel->core.attr;
+	struct evsel *leader = evsel__leader(evsel);
+
+	/* Reset supported flag as creating a stat counter is retried. */
+	attr->read_format = PERF_FORMAT_TOTAL_TIME_ENABLED |
+			    PERF_FORMAT_TOTAL_TIME_RUNNING;
+
+	/*
+	 * The event is part of non trivial group, let's enable
+	 * the group read (for leader) and ID retrieval for all
+	 * members.
+	 */
+	if (leader->core.nr_members > 1)
+		attr->read_format |= PERF_FORMAT_ID|PERF_FORMAT_GROUP;
+
+	attr->inherit = !config->no_inherit && list_empty(&evsel->bpf_counter_list);
+
+	/*
+	 * Some events get initialized with sample_(period/type) set,
+	 * like tracepoints. Clear it up for counting.
+	 */
+	attr->sample_period = 0;
+
+	if (config->identifier)
+		attr->sample_type = PERF_SAMPLE_IDENTIFIER;
+
+	if (config->all_user) {
+		attr->exclude_kernel = 1;
+		attr->exclude_user   = 0;
+	}
+
+	if (config->all_kernel) {
+		attr->exclude_kernel = 0;
+		attr->exclude_user   = 1;
+	}
+
+	/*
+	 * Disabling all counters initially, they will be enabled
+	 * either manually by us or by kernel via enable_on_exec
+	 * set later.
+	 */
+	if (evsel__is_group_leader(evsel)) {
+		attr->disabled = 1;
+
+		if (target__enable_on_exec(&target))
+			attr->enable_on_exec = 1;
+	}
+
+	return evsel__open_per_cpu_and_thread(evsel, evsel__cpus(evsel), cpu_map_idx,
+					      evsel->core.threads);
+}
+
+static void update_rusage_stats(const struct rusage *rusage)
+{
+	const u64 us_to_ns = 1000;
+	const u64 s_to_ns = 1000000000;
+
+	update_stats(&ru_stats.ru_utime_usec_stat,
+		(rusage->ru_utime.tv_usec * us_to_ns + rusage->ru_utime.tv_sec * s_to_ns));
+	update_stats(&ru_stats.ru_stime_usec_stat,
+		(rusage->ru_stime.tv_usec * us_to_ns + rusage->ru_stime.tv_sec * s_to_ns));
+}
+
+static int __run_perf_stat(int argc, const char **argv, int run_idx)
+{
+	int interval = stat_config.interval;
+	int times = stat_config.times;
+	int timeout = stat_config.timeout;
+	char msg[BUFSIZ];
+	unsigned long long t0, t1;
+	struct evsel *counter;
+	size_t l;
+	int status = 0;
+	const bool forks = (argc > 0);
+	bool is_pipe = STAT_RECORD ? perf_stat.data.is_pipe : false;
+	struct evlist_cpu_iterator evlist_cpu_itr;
+	int err, open_err = 0;
+	bool second_pass = false, has_supported_counters;
+
+	if (forks) {
+		if (evlist__prepare_workload(evsel_list, &target, argv, is_pipe,
+					     workload_exec_failed_signal) < 0) {
+			perror("failed to prepare workload");
+			return -1;
+		}
+		child_pid = evlist__workload_pid(evsel_list);
+	}
+
+	evlist__for_each_entry(evsel_list, counter) {
+		counter->reset_group = false;
+		if (bpf_counter__load(counter, &target)) {
+			err = -1;
+			goto err_out;
+		}
+		if (!(evsel__is_bperf(counter)))
+			all_counters_use_bpf = false;
+	}
+
+	evlist__reset_aggr_stats(evsel_list);
+
+	/*
+	 * bperf calls evsel__open_per_cpu() in bperf__load(), so
+	 * no need to call it again here.
+	 */
+	if (!target.use_bpf) {
+		evlist__for_each_cpu(evlist_cpu_itr, evsel_list) {
+			counter = evlist_cpu_itr.evsel;
+
+			if (counter->reset_group || !counter->supported)
+				continue;
+			if (evsel__is_bperf(counter))
+				continue;
+
+			while (true) {
+				if (create_perf_stat_counter(counter, &stat_config,
+							      evlist_cpu_itr.cpu_map_idx) == 0)
+					break;
+
+				open_err = errno;
+				/*
+				 * Weak group failed. We cannot just undo this
+				 * here because earlier CPUs might be in group
+				 * mode, and the kernel doesn't support mixing
+				 * group and non group reads. Defer it to later.
+				 * Don't close here because we're in the wrong
+				 * affinity.
+				 */
+				if ((open_err == EINVAL || open_err == EBADF) &&
+					evsel__leader(counter) != counter &&
+					counter->weak_group) {
+					evlist__reset_weak_group(evsel_list, counter, false);
+					assert(counter->reset_group);
+					counter->supported = true;
+					second_pass = true;
+					break;
+				}
+
+				if (stat_handle_error(counter, open_err) != COUNTER_RETRY)
+					break;
+			}
+		}
+	}
+	if (second_pass) {
+		/*
+		 * Now redo all the weak group after closing them,
+		 * and also close errored counters.
+		 */
+
+		/* First close errored or weak retry */
+		evlist__for_each_cpu(evlist_cpu_itr, evsel_list) {
+			counter = evlist_cpu_itr.evsel;
+
+			if (!counter->reset_group && counter->supported)
+				continue;
+
+			perf_evsel__close_cpu(&counter->core, evlist_cpu_itr.cpu_map_idx);
+		}
+		/* Now reopen weak */
+		evlist__for_each_cpu(evlist_cpu_itr, evsel_list) {
+			counter = evlist_cpu_itr.evsel;
+
+			if (!counter->reset_group)
+				continue;
+
+			while (true) {
+				pr_debug2("reopening weak %s\n", evsel__name(counter));
+				if (create_perf_stat_counter(counter, &stat_config,
+							     evlist_cpu_itr.cpu_map_idx) == 0) {
+					evlist_cpu_iterator__exit(&evlist_cpu_itr);
+					break;
+				}
+				open_err = errno;
+				if (stat_handle_error(counter, open_err) != COUNTER_RETRY) {
+					evlist_cpu_iterator__exit(&evlist_cpu_itr);
+					break;
+				}
+			}
+		}
+	}
+
+	has_supported_counters = false;
+	evlist__for_each_entry(evsel_list, counter) {
+		if (!counter->supported) {
+			perf_evsel__free_fd(&counter->core);
+			continue;
+		}
+		has_supported_counters = true;
+
+		l = strlen(counter->unit);
+		if (l > stat_config.unit_width)
+			stat_config.unit_width = l;
+
+		if (evsel__should_store_id(counter) &&
+		    evsel__store_ids(counter, evsel_list)) {
+			err = -1;
+			goto err_out;
+		}
+	}
+	if (!has_supported_counters && !stat_config.null_run) {
+		if (open_err) {
+			evsel__open_strerror(evlist__first(evsel_list), &target, open_err,
+					     msg, sizeof(msg));
+		}
+		ui__error("No supported events found.\n%s\n", msg);
+
+		if (child_pid != -1)
+			kill(child_pid, SIGTERM);
+		err = -1;
+		goto err_out;
+	}
+
+	if (evlist__apply_filters(evsel_list, &counter, &target)) {
+		pr_err("failed to set filter \"%s\" on event %s: %m\n",
+			counter->filter, evsel__name(counter));
+		return -1;
+	}
+
+	if (STAT_RECORD) {
+		int fd = perf_data__fd(&perf_stat.data);
+
+		if (is_pipe) {
+			err = perf_header__write_pipe(perf_data__fd(&perf_stat.data));
+		} else {
+			err = perf_session__write_header(perf_stat.session, evsel_list,
+							 fd, false);
+		}
+
+		if (err < 0)
+			goto err_out;
+
+		err = perf_event__synthesize_stat_events(&stat_config, NULL, evsel_list,
+							 process_synthesized_event, is_pipe);
+		if (err < 0)
+			goto err_out;
+
+	}
+
+	if (target.initial_delay) {
+		pr_info(EVLIST_DISABLED_MSG);
+	} else {
+		err = enable_counters();
+		if (err) {
+			err = -1;
+			goto err_out;
+		}
+	}
+
+	/* Exec the command, if any */
+	if (forks)
+		evlist__start_workload(evsel_list);
+
+	if (target.initial_delay > 0) {
+		usleep(target.initial_delay * USEC_PER_MSEC);
+		err = enable_counters();
+		if (err) {
+			err = -1;
+			goto err_out;
+		}
+
+		pr_info(EVLIST_ENABLED_MSG);
+	}
+
+	t0 = rdclock();
+	clock_gettime(CLOCK_MONOTONIC, &ref_time);
+
+	if (forks) {
+		if (interval || timeout || evlist__ctlfd_initialized(evsel_list))
+			status = dispatch_events(forks, timeout, interval, &times);
+		if (child_pid != -1) {
+			if (timeout)
+				kill(child_pid, SIGTERM);
+			wait4(child_pid, &status, 0, &stat_config.ru_data);
+		}
+
+		if (workload_exec_errno) {
+			errno = workload_exec_errno;
+			pr_err("Workload failed: %m\n");
+			err = -1;
+			goto err_out;
+		}
+
+		if (WIFSIGNALED(status)) {
+			/*
+			 * We want to indicate failure to stop a repeat run,
+			 * hence negative. We want the value to be the exit code
+			 * of perf, which for termination by a signal is 128
+			 * plus the signal number.
+			 */
+			err = 0 - (128 + WTERMSIG(status));
+			psignal(WTERMSIG(status), argv[0]);
+		} else {
+			err = WEXITSTATUS(status);
+		}
+	} else {
+		err = dispatch_events(forks, timeout, interval, &times);
+	}
+
+	disable_counters();
+
+	t1 = rdclock();
+
+	if (stat_config.walltime_run_table)
+		stat_config.walltime_run[run_idx] = t1 - t0;
+
+	if (interval && stat_config.summary) {
+		stat_config.interval = 0;
+		stat_config.stop_read_counter = true;
+		init_stats(stat_config.walltime_nsecs_stats);
+		update_stats(stat_config.walltime_nsecs_stats, t1 - t0);
+
+		evlist__copy_prev_raw_counts(evsel_list);
+		evlist__reset_prev_raw_counts(evsel_list);
+		evlist__reset_aggr_stats(evsel_list);
+	} else {
+		update_stats(stat_config.walltime_nsecs_stats, t1 - t0);
+		update_rusage_stats(&stat_config.ru_data);
+	}
+
+	/*
+	 * Closing a group leader splits the group, and as we only disable
+	 * group leaders, results in remaining events becoming enabled. To
+	 * avoid arbitrary skew, we must read all counters before closing any
+	 * group leaders.
+	 */
+	if (read_counters() == 0)
+		process_counters();
+
+	/*
+	 * We need to keep evsel_list alive, because it's processed
+	 * later the evsel_list will be closed after.
+	 */
+	if (!STAT_RECORD)
+		evlist__close(evsel_list);
+
+	return err;
+
+err_out:
+	if (forks)
+		evlist__cancel_workload(evsel_list);
+
+	return err;
+}
+
+/*
+ * Returns -1 for fatal errors which signifies to not continue
+ * when in repeat mode.
+ *
+ * Returns < -1 error codes when stat record is used. These
+ * result in the stat information being displayed, but writing
+ * to the file fails and is non fatal.
+ */
+static int run_perf_stat(int argc, const char **argv, int run_idx)
+{
+	int ret;
+
+	if (pre_cmd) {
+		ret = system(pre_cmd);
+		if (ret)
+			return ret;
+	}
+
+	if (sync_run)
+		sync();
+
+	ret = __run_perf_stat(argc, argv, run_idx);
+	if (ret)
+		return ret;
+
+	if (post_cmd) {
+		ret = system(post_cmd);
+		if (ret)
+			return ret;
+	}
+
+	return ret;
+}
+
+static void print_counters(struct timespec *ts, int argc, const char **argv)
+{
+	/* Do not print anything if we record to the pipe. */
+	if (STAT_RECORD && perf_stat.data.is_pipe)
+		return;
+	if (quiet)
+		return;
+
+	evlist__print_counters(evsel_list, &stat_config, &target, ts, argc, argv);
+}
+
+static volatile sig_atomic_t signr = -1;
+
+static void skip_signal(int signo)
+{
+	if ((child_pid == -1) || stat_config.interval)
+		done = 1;
+
+	signr = signo;
+	/*
+	 * render child_pid harmless
+	 * won't send SIGTERM to a random
+	 * process in case of race condition
+	 * and fast PID recycling
+	 */
+	child_pid = -1;
+}
+
+static void sig_atexit(void)
+{
+	sigset_t set, oset;
+
+	/*
+	 * avoid race condition with SIGCHLD handler
+	 * in skip_signal() which is modifying child_pid
+	 * goal is to avoid send SIGTERM to a random
+	 * process
+	 */
+	sigemptyset(&set);
+	sigaddset(&set, SIGCHLD);
+	sigprocmask(SIG_BLOCK, &set, &oset);
+
+	if (child_pid != -1)
+		kill(child_pid, SIGTERM);
+
+	sigprocmask(SIG_SETMASK, &oset, NULL);
+
+	if (signr == -1)
+		return;
+
+	signal(signr, SIG_DFL);
+	kill(getpid(), signr);
+}
+
+static int stat__set_big_num(const struct option *opt __maybe_unused,
+			     const char *s __maybe_unused, int unset)
+{
+	big_num_opt = unset ? 0 : 1;
+	perf_stat__set_big_num(!unset);
+	return 0;
+}
+
+static int enable_metric_only(const struct option *opt __maybe_unused,
+			      const char *s __maybe_unused, int unset)
+{
+	force_metric_only = true;
+	stat_config.metric_only = !unset;
+	return 0;
+}
+
+static int append_metric_groups(const struct option *opt __maybe_unused,
+			       const char *str,
+			       int unset __maybe_unused)
+{
+	if (metrics) {
+		char *tmp;
+
+		if (asprintf(&tmp, "%s,%s", metrics, str) < 0)
+			return -ENOMEM;
+		free(metrics);
+		metrics = tmp;
+	} else {
+		metrics = strdup(str);
+		if (!metrics)
+			return -ENOMEM;
+	}
+	return 0;
+}
+
+static int parse_control_option(const struct option *opt,
+				const char *str,
+				int unset __maybe_unused)
+{
+	struct perf_stat_config *config = opt->value;
+
+	return evlist__parse_control(str, &config->ctl_fd, &config->ctl_fd_ack, &config->ctl_fd_close);
+}
+
+static int parse_stat_cgroups(const struct option *opt,
+			      const char *str, int unset)
+{
+	if (stat_config.cgroup_list) {
+		pr_err("--cgroup and --for-each-cgroup cannot be used together\n");
+		return -1;
+	}
+
+	return parse_cgroups(opt, str, unset);
+}
+
+static int parse_cputype(const struct option *opt,
+			     const char *str,
+			     int unset __maybe_unused)
+{
+	const struct perf_pmu *pmu;
+	struct evlist *evlist = *(struct evlist **)opt->value;
+
+	if (!list_empty(&evlist__core(evlist)->entries)) {
+		fprintf(stderr, "Must define cputype before events/metrics\n");
+		return -1;
+	}
+
+	pmu = perf_pmus__pmu_for_pmu_filter(str);
+	if (!pmu) {
+		fprintf(stderr, "--cputype %s is not supported!\n", str);
+		return -1;
+	}
+	parse_events_option_args.pmu_filter = pmu->name;
+	parse_events_option_args.cputype_filter = true;
+
+	return 0;
+}
+
+static int parse_pmu_filter(const struct option *opt,
+			   const char *str,
+			   int unset __maybe_unused)
+{
+	struct evlist *evlist = *(struct evlist **)opt->value;
+
+	if (!list_empty(&evlist__core(evlist)->entries)) {
+		fprintf(stderr, "Must define pmu-filter before events/metrics\n");
+		return -1;
+	}
+
+	parse_events_option_args.pmu_filter = str;
+	parse_events_option_args.cputype_filter = false;
+	return 0;
+}
+
+static int parse_cache_level(const struct option *opt,
+			     const char *str,
+			     int unset __maybe_unused)
+{
+	int level;
+	bool *per_cache = opt->value;
+	u32 *aggr_level = opt->data;
+
+	/*
+	 * If no string is specified, aggregate based on the topology of
+	 * Last Level Cache (LLC). Since the LLC level can change from
+	 * architecture to architecture, set level greater than
+	 * MAX_CACHE_LVL which will be interpreted as LLC.
+	 */
+	if (str == NULL) {
+		level = MAX_CACHE_LVL + 1;
+		goto out;
+	}
+
+	/*
+	 * The format to specify cache level is LX or lX where X is the
+	 * cache level.
+	 */
+	if (strlen(str) != 2 || (str[0] != 'l' && str[0] != 'L')) {
+		pr_err("Cache level must be of form L[1-%d], or l[1-%d]\n",
+		       MAX_CACHE_LVL,
+		       MAX_CACHE_LVL);
+		return -EINVAL;
+	}
+
+	level = atoi(&str[1]);
+	if (level < 1) {
+		pr_err("Cache level must be of form L[1-%d], or l[1-%d]\n",
+		       MAX_CACHE_LVL,
+		       MAX_CACHE_LVL);
+		return -EINVAL;
+	}
+
+	if (level > MAX_CACHE_LVL) {
+		pr_err("perf only supports max cache level of %d.\n"
+		       "Consider increasing MAX_CACHE_LVL\n", MAX_CACHE_LVL);
+		return -EINVAL;
+	}
+out:
+	*per_cache = true;
+	*aggr_level = level;
+	return 0;
+}
+
+/**
+ * Calculate the cache instance ID from the map in
+ * /sys/devices/system/cpu/cpuX/cache/indexY/shared_cpu_list
+ * Cache instance ID is the first CPU reported in the shared_cpu_list file.
+ */
+static int cpu__get_cache_id_from_map(struct perf_cpu cpu, char *map)
+{
+	int id;
+	struct perf_cpu_map *cpu_map = perf_cpu_map__new(map);
+
+	/*
+	 * If the map contains no CPU, consider the current CPU to
+	 * be the first online CPU in the cache domain else use the
+	 * first online CPU of the cache domain as the ID.
+	 */
+	id = perf_cpu_map__min(cpu_map).cpu;
+	if (id == -1)
+		id = cpu.cpu;
+
+	/* Free the perf_cpu_map used to find the cache ID */
+	perf_cpu_map__put(cpu_map);
+
+	return id;
+}
+
+/**
+ * cpu__get_cache_id - Returns 0 if successful in populating the
+ * cache level and cache id. Cache level is read from
+ * /sys/devices/system/cpu/cpuX/cache/indexY/level where as cache instance ID
+ * is the first CPU reported by
+ * /sys/devices/system/cpu/cpuX/cache/indexY/shared_cpu_list
+ */
+static int cpu__get_cache_details(struct perf_cpu cpu, struct perf_cache *cache)
+{
+	int ret = 0;
+	u32 cache_level = stat_config.aggr_level;
+	struct cpu_cache_level caches[MAX_CACHE_LVL];
+	u32 i = 0, caches_cnt = 0;
+
+	cache->cache_lvl = (cache_level > MAX_CACHE_LVL) ? 0 : cache_level;
+	cache->cache = -1;
+
+	ret = build_caches_for_cpu(cpu.cpu, caches, &caches_cnt);
+	if (ret) {
+		/*
+		 * If caches_cnt is not 0, cpu_cache_level data
+		 * was allocated when building the topology.
+		 * Free the allocated data before returning.
+		 */
+		if (caches_cnt)
+			goto free_caches;
+
+		return ret;
+	}
+
+	if (!caches_cnt)
+		return -1;
+
+	/*
+	 * Save the data for the highest level if no
+	 * level was specified by the user.
+	 */
+	if (cache_level > MAX_CACHE_LVL) {
+		int max_level_index = 0;
+
+		for (i = 1; i < caches_cnt; ++i) {
+			if (caches[i].level > caches[max_level_index].level)
+				max_level_index = i;
+		}
+
+		cache->cache_lvl = caches[max_level_index].level;
+		cache->cache = cpu__get_cache_id_from_map(cpu, caches[max_level_index].map);
+
+		/* Reset i to 0 to free entire caches[] */
+		i = 0;
+		goto free_caches;
+	}
+
+	for (i = 0; i < caches_cnt; ++i) {
+		if (caches[i].level == cache_level) {
+			cache->cache_lvl = cache_level;
+			cache->cache = cpu__get_cache_id_from_map(cpu, caches[i].map);
+		}
+
+		cpu_cache_level__free(&caches[i]);
+	}
+
+free_caches:
+	/*
+	 * Free all the allocated cpu_cache_level data.
+	 */
+	while (i < caches_cnt)
+		cpu_cache_level__free(&caches[i++]);
+
+	return ret;
+}
+
+/**
+ * aggr_cpu_id__cache - Create an aggr_cpu_id with cache instache ID, cache
+ * level, die and socket populated with the cache instache ID, cache level,
+ * die and socket for cpu. The function signature is compatible with
+ * aggr_cpu_id_get_t.
+ */
+static struct aggr_cpu_id aggr_cpu_id__cache(struct perf_cpu cpu, void *data)
+{
+	int ret;
+	struct aggr_cpu_id id;
+	struct perf_cache cache;
+
+	id = aggr_cpu_id__die(cpu, data);
+	if (aggr_cpu_id__is_empty(&id))
+		return id;
+
+	ret = cpu__get_cache_details(cpu, &cache);
+	if (ret)
+		return id;
+
+	id.cache_lvl = cache.cache_lvl;
+	id.cache = cache.cache;
+	return id;
+}
+
+static const char *const aggr_mode__string[] = {
+	[AGGR_CORE] = "core",
+	[AGGR_CACHE] = "cache",
+	[AGGR_CLUSTER] = "cluster",
+	[AGGR_DIE] = "die",
+	[AGGR_GLOBAL] = "global",
+	[AGGR_NODE] = "node",
+	[AGGR_NONE] = "none",
+	[AGGR_SOCKET] = "socket",
+	[AGGR_THREAD] = "thread",
+	[AGGR_UNSET] = "unset",
+};
+
+static struct aggr_cpu_id perf_stat__get_socket(struct perf_stat_config *config __maybe_unused,
+						struct perf_cpu cpu)
+{
+	return aggr_cpu_id__socket(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_die(struct perf_stat_config *config __maybe_unused,
+					     struct perf_cpu cpu)
+{
+	return aggr_cpu_id__die(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_cache_id(struct perf_stat_config *config __maybe_unused,
+						  struct perf_cpu cpu)
+{
+	return aggr_cpu_id__cache(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_cluster(struct perf_stat_config *config __maybe_unused,
+						 struct perf_cpu cpu)
+{
+	return aggr_cpu_id__cluster(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_core(struct perf_stat_config *config __maybe_unused,
+					      struct perf_cpu cpu)
+{
+	return aggr_cpu_id__core(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_node(struct perf_stat_config *config __maybe_unused,
+					      struct perf_cpu cpu)
+{
+	return aggr_cpu_id__node(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_global(struct perf_stat_config *config __maybe_unused,
+						struct perf_cpu cpu)
+{
+	return aggr_cpu_id__global(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_cpu(struct perf_stat_config *config __maybe_unused,
+					     struct perf_cpu cpu)
+{
+	return aggr_cpu_id__cpu(cpu, /*data=*/NULL);
+}
+
+static struct aggr_cpu_id perf_stat__get_aggr(struct perf_stat_config *config,
+					      aggr_get_id_t get_id, struct perf_cpu cpu)
+{
+	struct aggr_cpu_id id;
+
+	/* per-process mode - should use global aggr mode */
+	if (cpu.cpu == -1 || cpu.cpu >= config->cpus_aggr_map->nr)
+		return get_id(config, cpu);
+
+	if (aggr_cpu_id__is_empty(&config->cpus_aggr_map->map[cpu.cpu]))
+		config->cpus_aggr_map->map[cpu.cpu] = get_id(config, cpu);
+
+	id = config->cpus_aggr_map->map[cpu.cpu];
+	return id;
+}
+
+static struct aggr_cpu_id perf_stat__get_socket_cached(struct perf_stat_config *config,
+						       struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_socket, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_die_cached(struct perf_stat_config *config,
+						    struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_die, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_cluster_cached(struct perf_stat_config *config,
+							struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_cluster, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_cache_id_cached(struct perf_stat_config *config,
+							 struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_cache_id, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_core_cached(struct perf_stat_config *config,
+						     struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_core, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_node_cached(struct perf_stat_config *config,
+						     struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_node, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_global_cached(struct perf_stat_config *config,
+						       struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_global, cpu);
+}
+
+static struct aggr_cpu_id perf_stat__get_cpu_cached(struct perf_stat_config *config,
+						    struct perf_cpu cpu)
+{
+	return perf_stat__get_aggr(config, perf_stat__get_cpu, cpu);
+}
+
+static aggr_cpu_id_get_t aggr_mode__get_aggr(enum aggr_mode aggr_mode)
+{
+	switch (aggr_mode) {
+	case AGGR_SOCKET:
+		return aggr_cpu_id__socket;
+	case AGGR_DIE:
+		return aggr_cpu_id__die;
+	case AGGR_CLUSTER:
+		return aggr_cpu_id__cluster;
+	case AGGR_CACHE:
+		return aggr_cpu_id__cache;
+	case AGGR_CORE:
+		return aggr_cpu_id__core;
+	case AGGR_NODE:
+		return aggr_cpu_id__node;
+	case AGGR_NONE:
+		return aggr_cpu_id__cpu;
+	case AGGR_GLOBAL:
+		return aggr_cpu_id__global;
+	case AGGR_THREAD:
+	case AGGR_UNSET:
+	case AGGR_MAX:
+	default:
+		return NULL;
+	}
+}
+
+static aggr_get_id_t aggr_mode__get_id(enum aggr_mode aggr_mode)
+{
+	switch (aggr_mode) {
+	case AGGR_SOCKET:
+		return perf_stat__get_socket_cached;
+	case AGGR_DIE:
+		return perf_stat__get_die_cached;
+	case AGGR_CLUSTER:
+		return perf_stat__get_cluster_cached;
+	case AGGR_CACHE:
+		return perf_stat__get_cache_id_cached;
+	case AGGR_CORE:
+		return perf_stat__get_core_cached;
+	case AGGR_NODE:
+		return perf_stat__get_node_cached;
+	case AGGR_NONE:
+		return perf_stat__get_cpu_cached;
+	case AGGR_GLOBAL:
+		return perf_stat__get_global_cached;
+	case AGGR_THREAD:
+	case AGGR_UNSET:
+	case AGGR_MAX:
+	default:
+		return NULL;
+	}
+}
+
+static int perf_stat_init_aggr_mode(void)
+{
+	int nr;
+	aggr_cpu_id_get_t get_id = aggr_mode__get_aggr(stat_config.aggr_mode);
+
+	if (get_id) {
+		bool needs_sort = stat_config.aggr_mode != AGGR_NONE;
+		stat_config.aggr_map = cpu_aggr_map__new(
+			evlist__core(evsel_list)->user_requested_cpus,
+			get_id, /*data=*/NULL, needs_sort);
+		if (!stat_config.aggr_map) {
+			pr_err("cannot build %s map\n", aggr_mode__string[stat_config.aggr_mode]);
+			return -1;
+		}
+		stat_config.aggr_get_id = aggr_mode__get_id(stat_config.aggr_mode);
+	}
+
+	if (stat_config.aggr_mode == AGGR_THREAD) {
+		nr = perf_thread_map__nr(evlist__core(evsel_list)->threads);
+		stat_config.aggr_map = cpu_aggr_map__empty_new(nr);
+		if (stat_config.aggr_map == NULL)
+			return -ENOMEM;
+
+		for (int s = 0; s < nr; s++) {
+			struct aggr_cpu_id id = aggr_cpu_id__empty();
+
+			id.thread_idx = s;
+			stat_config.aggr_map->map[s] = id;
+		}
+		return 0;
+	}
+
+	/*
+	 * The evsel_list->cpus is the base we operate on,
+	 * taking the highest cpu number to be the size of
+	 * the aggregation translate cpumap.
+	 */
+	nr = perf_cpu_map__max(evlist__core(evsel_list)->all_cpus).cpu + 1;
+	stat_config.cpus_aggr_map = cpu_aggr_map__empty_new(nr);
+	return stat_config.cpus_aggr_map ? 0 : -ENOMEM;
+}
+
+static void cpu_aggr_map__delete(struct cpu_aggr_map *map)
+{
+	free(map);
+}
+
+static void perf_stat__exit_aggr_mode(void)
+{
+	cpu_aggr_map__delete(stat_config.aggr_map);
+	cpu_aggr_map__delete(stat_config.cpus_aggr_map);
+	stat_config.aggr_map = NULL;
+	stat_config.cpus_aggr_map = NULL;
+}
+
+static struct aggr_cpu_id perf_env__get_socket_aggr_by_cpu(struct perf_cpu cpu, void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo)
+		id.socket = topo->socket_id;
+
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_die_aggr_by_cpu(struct perf_cpu cpu, void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo) {
+		/*
+		 * die_id is relative to socket, so start
+		 * with the socket ID and then add die to
+		 * make a unique ID.
+		 */
+		id.socket = topo->socket_id;
+		id.die = topo->die_id;
+	}
+
+	return id;
+}
+
+static void perf_env__get_cache_id_for_cpu(struct perf_cpu cpu, struct perf_env *env,
+					   u32 cache_level, struct aggr_cpu_id *id)
+{
+	int i;
+	int caches_cnt = env->caches_cnt;
+	struct cpu_cache_level *caches = env->caches;
+
+	id->cache_lvl = (cache_level > MAX_CACHE_LVL) ? 0 : cache_level;
+	id->cache = -1;
+
+	if (!caches_cnt)
+		return;
+
+	for (i = caches_cnt - 1; i > -1; --i) {
+		struct perf_cpu_map *cpu_map;
+		int map_contains_cpu;
+
+		/*
+		 * If user has not specified a level, find the fist level with
+		 * the cpu in the map. Since building the map is expensive, do
+		 * this only if levels match.
+		 */
+		if (cache_level <= MAX_CACHE_LVL && caches[i].level != cache_level)
+			continue;
+
+		cpu_map = perf_cpu_map__new(caches[i].map);
+		map_contains_cpu = perf_cpu_map__idx(cpu_map, cpu);
+		perf_cpu_map__put(cpu_map);
+
+		if (map_contains_cpu != -1) {
+			id->cache_lvl = caches[i].level;
+			id->cache = cpu__get_cache_id_from_map(cpu, caches[i].map);
+			return;
+		}
+	}
+}
+
+static struct aggr_cpu_id perf_env__get_cache_aggr_by_cpu(struct perf_cpu cpu,
+							  void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo) {
+		u32 cache_level = (perf_stat.aggr_level) ?: stat_config.aggr_level;
+
+		id.socket = topo->socket_id;
+		id.die = topo->die_id;
+		perf_env__get_cache_id_for_cpu(cpu, env, cache_level, &id);
+	}
+
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_cluster_aggr_by_cpu(struct perf_cpu cpu,
+							    void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo) {
+		id.socket = topo->socket_id;
+		id.die = topo->die_id;
+		id.cluster = topo->cluster_id;
+	}
+
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_core_aggr_by_cpu(struct perf_cpu cpu, void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo) {
+		/*
+		 * core_id is relative to socket, die and cluster, we need a
+		 * global id. So we set socket, die id, cluster id and core id.
+		 */
+		id.socket = topo->socket_id;
+		id.die = topo->die_id;
+		id.cluster = topo->cluster_id;
+		id.core = topo->core_id;
+	}
+
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_cpu_aggr_by_cpu(struct perf_cpu cpu, void *data)
+{
+	struct perf_env *env = data;
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+	struct cpu_topology_map *topo = perf_env__get_cpu_topology(env, cpu);
+
+	if (topo) {
+		/*
+		 * core_id is relative to socket and die,
+		 * we need a global id. So we set
+		 * socket, die id and core id
+		 */
+		id.socket = topo->socket_id;
+		id.die = topo->die_id;
+		id.core = topo->core_id;
+	}
+	id.cpu = cpu;
+
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_node_aggr_by_cpu(struct perf_cpu cpu, void *data)
+{
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+
+	id.node = perf_env__numa_node(data, cpu);
+	return id;
+}
+
+static struct aggr_cpu_id perf_env__get_global_aggr_by_cpu(struct perf_cpu cpu __maybe_unused,
+							   void *data __maybe_unused)
+{
+	struct aggr_cpu_id id = aggr_cpu_id__empty();
+
+	/* it always aggregates to the cpu 0 */
+	id.cpu = (struct perf_cpu){ .cpu = 0 };
+	return id;
+}
+
+static struct aggr_cpu_id perf_stat__get_socket_file(struct perf_stat_config *config __maybe_unused,
+						     struct perf_cpu cpu)
+{
+	return perf_env__get_socket_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+static struct aggr_cpu_id perf_stat__get_die_file(struct perf_stat_config *config __maybe_unused,
+						  struct perf_cpu cpu)
+{
+	return perf_env__get_die_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_cluster_file(struct perf_stat_config *config __maybe_unused,
+						      struct perf_cpu cpu)
+{
+	return perf_env__get_cluster_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_cache_file(struct perf_stat_config *config __maybe_unused,
+						    struct perf_cpu cpu)
+{
+	return perf_env__get_cache_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_core_file(struct perf_stat_config *config __maybe_unused,
+						   struct perf_cpu cpu)
+{
+	return perf_env__get_core_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_cpu_file(struct perf_stat_config *config __maybe_unused,
+						  struct perf_cpu cpu)
+{
+	return perf_env__get_cpu_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_node_file(struct perf_stat_config *config __maybe_unused,
+						   struct perf_cpu cpu)
+{
+	return perf_env__get_node_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static struct aggr_cpu_id perf_stat__get_global_file(struct perf_stat_config *config __maybe_unused,
+						     struct perf_cpu cpu)
+{
+	return perf_env__get_global_aggr_by_cpu(cpu, perf_session__env(perf_stat.session));
+}
+
+static aggr_cpu_id_get_t aggr_mode__get_aggr_file(enum aggr_mode aggr_mode)
+{
+	switch (aggr_mode) {
+	case AGGR_SOCKET:
+		return perf_env__get_socket_aggr_by_cpu;
+	case AGGR_DIE:
+		return perf_env__get_die_aggr_by_cpu;
+	case AGGR_CLUSTER:
+		return perf_env__get_cluster_aggr_by_cpu;
+	case AGGR_CACHE:
+		return perf_env__get_cache_aggr_by_cpu;
+	case AGGR_CORE:
+		return perf_env__get_core_aggr_by_cpu;
+	case AGGR_NODE:
+		return perf_env__get_node_aggr_by_cpu;
+	case AGGR_GLOBAL:
+		return perf_env__get_global_aggr_by_cpu;
+	case AGGR_NONE:
+		return perf_env__get_cpu_aggr_by_cpu;
+	case AGGR_THREAD:
+	case AGGR_UNSET:
+	case AGGR_MAX:
+	default:
+		return NULL;
+	}
+}
+
+static aggr_get_id_t aggr_mode__get_id_file(enum aggr_mode aggr_mode)
+{
+	switch (aggr_mode) {
+	case AGGR_SOCKET:
+		return perf_stat__get_socket_file;
+	case AGGR_DIE:
+		return perf_stat__get_die_file;
+	case AGGR_CLUSTER:
+		return perf_stat__get_cluster_file;
+	case AGGR_CACHE:
+		return perf_stat__get_cache_file;
+	case AGGR_CORE:
+		return perf_stat__get_core_file;
+	case AGGR_NODE:
+		return perf_stat__get_node_file;
+	case AGGR_GLOBAL:
+		return perf_stat__get_global_file;
+	case AGGR_NONE:
+		return perf_stat__get_cpu_file;
+	case AGGR_THREAD:
+	case AGGR_UNSET:
+	case AGGR_MAX:
+	default:
+		return NULL;
+	}
+}
+
+static int perf_stat_init_aggr_mode_file(struct perf_stat *st)
+{
+	struct perf_env *env = perf_session__env(st->session);
+	aggr_cpu_id_get_t get_id = aggr_mode__get_aggr_file(stat_config.aggr_mode);
+	bool needs_sort = stat_config.aggr_mode != AGGR_NONE;
+
+	if (stat_config.aggr_mode == AGGR_THREAD) {
+		int nr = perf_thread_map__nr(evlist__core(evsel_list)->threads);
+
+		stat_config.aggr_map = cpu_aggr_map__empty_new(nr);
+		if (stat_config.aggr_map == NULL)
+			return -ENOMEM;
+
+		for (int s = 0; s < nr; s++) {
+			struct aggr_cpu_id id = aggr_cpu_id__empty();
+
+			id.thread_idx = s;
+			stat_config.aggr_map->map[s] = id;
+		}
+		return 0;
+	}
+
+	if (!get_id)
+		return 0;
+
+	stat_config.aggr_map = cpu_aggr_map__new(evlist__core(evsel_list)->user_requested_cpus,
+						 get_id, env, needs_sort);
+	if (!stat_config.aggr_map) {
+		pr_err("cannot build %s map\n", aggr_mode__string[stat_config.aggr_mode]);
+		return -1;
+	}
+	stat_config.aggr_get_id = aggr_mode__get_id_file(stat_config.aggr_mode);
+	return 0;
+}
+
+static int default_evlist_evsel_cmp(void *priv __maybe_unused,
+				    const struct list_head *l,
+				    const struct list_head *r)
+{
+	const struct perf_evsel *lhs_core = container_of(l, struct perf_evsel, node);
+	const struct evsel *lhs = container_of(lhs_core, struct evsel, core);
+	const struct perf_evsel *rhs_core = container_of(r, struct perf_evsel, node);
+	const struct evsel *rhs = container_of(rhs_core, struct evsel, core);
+	const struct evsel *lhs_leader = evsel__leader(lhs);
+	const struct evsel *rhs_leader = evsel__leader(rhs);
+
+	if (lhs_leader == rhs_leader) {
+		/* Within the same group, respect the original order. */
+		return lhs_core->idx - rhs_core->idx;
+	}
+
+	/*
+	 * Compare using leader's attributes so that all members of a group
+	 * stay together. This ensures leaders are opened before their members.
+	 */
+
+	/* Sort default metrics evsels first, and default show events before those. */
+	if (lhs_leader->default_metricgroup != rhs_leader->default_metricgroup)
+		return lhs_leader->default_metricgroup ? -1 : 1;
+
+	if (lhs_leader->default_show_events != rhs_leader->default_show_events)
+		return lhs_leader->default_show_events ? -1 : 1;
+
+	/* Sort by PMU type (prefers legacy types first). */
+	if (lhs_leader->pmu != rhs_leader->pmu)
+		return lhs_leader->pmu->type - rhs_leader->pmu->type;
+
+	/* Sort by leader's name. */
+	return strcmp(evsel__name((struct evsel *)lhs_leader),
+		      evsel__name((struct evsel *)rhs_leader));
+}
+
+/*
+ * Add default events, if there were no attributes specified or
+ * if -d/--detailed, -d -d or -d -d -d is used:
+ */
+static int add_default_events(void)
+{
+	const char *pmu = parse_events_option_args.pmu_filter ?: "all";
+	struct parse_events_error err;
+	struct evlist *evlist = evlist__new();
+	struct evsel *evsel;
+	int ret = 0;
+
+	if (!evlist)
+		return -ENOMEM;
+
+	parse_events_error__init(&err);
+
+	/* Set attrs if no event is selected and !null_run: */
+	if (stat_config.null_run)
+		goto out;
+
+	if (transaction_run) {
+		/* Handle -T as -M transaction. Once platform specific metrics
+		 * support has been added to the json files, all architectures
+		 * will use this approach. To determine transaction support
+		 * on an architecture test for such a metric name.
+		 */
+		if (!metricgroup__has_metric_or_groups(pmu, "transaction")) {
+			pr_err("Missing transaction metrics\n");
+			ret = -1;
+			goto out;
+		}
+		ret = metricgroup__parse_groups(evlist, pmu,
+						parse_events_option_args.cputype_filter,
+						"transaction",
+						stat_config.metric_no_group,
+						stat_config.metric_no_merge,
+						stat_config.metric_no_threshold,
+						stat_config.user_requested_cpu_list,
+						stat_config.system_wide,
+						stat_config.hardware_aware_grouping);
+		goto out;
+	}
+
+	if (smi_cost) {
+		int smi;
+
+		if (sysfs__read_int(FREEZE_ON_SMI_PATH, &smi) < 0) {
+			pr_err("freeze_on_smi is not supported.\n");
+			ret = -1;
+			goto out;
+		}
+
+		if (!smi) {
+			if (sysfs__write_int(FREEZE_ON_SMI_PATH, 1) < 0) {
+				pr_err("Failed to set freeze_on_smi.\n");
+				ret = -1;
+				goto out;
+			}
+			smi_reset = true;
+		}
+
+		if (!metricgroup__has_metric_or_groups(pmu, "smi")) {
+			pr_err("Missing smi metrics\n");
+			ret = -1;
+			goto out;
+		}
+
+		if (!force_metric_only)
+			stat_config.metric_only = true;
+
+		ret = metricgroup__parse_groups(evlist, pmu,
+						parse_events_option_args.cputype_filter,
+						"smi",
+						stat_config.metric_no_group,
+						stat_config.metric_no_merge,
+						stat_config.metric_no_threshold,
+						stat_config.user_requested_cpu_list,
+						stat_config.system_wide,
+						stat_config.hardware_aware_grouping);
+		goto out;
+	}
+
+	if (topdown_run) {
+		unsigned int max_level = metricgroups__topdown_max_level();
+		char str[] = "TopdownL1";
+
+		if (!force_metric_only)
+			stat_config.metric_only = true;
+
+		if (!max_level) {
+			pr_err("Topdown requested but the topdown metric groups aren't present.\n"
+				"(See perf list the metric groups have names like TopdownL1)\n");
+			ret = -1;
+			goto out;
+		}
+		if (stat_config.topdown_level > max_level) {
+			pr_err("Invalid top-down metrics level. The max level is %u.\n", max_level);
+			ret = -1;
+			goto out;
+		} else if (!stat_config.topdown_level) {
+			stat_config.topdown_level = 1;
+		}
+		if (!stat_config.interval && !stat_config.metric_only) {
+			fprintf(stat_config.output,
+				"Topdown accuracy may decrease when measuring long periods.\n"
+				"Please print the result regularly, e.g. -I1000\n");
+		}
+		str[8] = stat_config.topdown_level + '0';
+		if (metricgroup__parse_groups(evlist,
+					      pmu, parse_events_option_args.cputype_filter, str,
+						/*metric_no_group=*/false,
+						/*metric_no_merge=*/false,
+						/*metric_no_threshold=*/true,
+						stat_config.user_requested_cpu_list,
+						stat_config.system_wide,
+						stat_config.hardware_aware_grouping) < 0) {
+			ret = -1;
+			goto out;
+		}
+	}
+
+	if (!stat_config.topdown_level)
+		stat_config.topdown_level = 1;
+
+	if (!evlist__nr_entries(evlist) && !evlist__nr_entries(evsel_list)) {
+		/*
+		 * Add Default metrics. To minimize multiplexing, don't request
+		 * threshold computation, but it will be computed if the events
+		 * are present.
+		 */
+		const char *default_metricgroup_names[] = {
+			"Default", "Default2", "Default3", "Default4",
+		};
+
+		for (size_t i = 0; i < ARRAY_SIZE(default_metricgroup_names); i++) {
+			struct evlist *metric_evlist;
+
+			if (!metricgroup__has_metric_or_groups(pmu, default_metricgroup_names[i]))
+				continue;
+
+			if ((int)i > detailed_run)
+				break;
+
+			metric_evlist = evlist__new();
+			if (!metric_evlist) {
+				ret = -ENOMEM;
+				break;
+			}
+			if (metricgroup__parse_groups(metric_evlist, pmu,
+						      parse_events_option_args.cputype_filter,
+						      default_metricgroup_names[i],
+							/*metric_no_group=*/false,
+							/*metric_no_merge=*/false,
+							/*metric_no_threshold=*/true,
+							stat_config.user_requested_cpu_list,
+							stat_config.system_wide,
+							stat_config.hardware_aware_grouping) < 0) {
+				evlist__put(metric_evlist);
+				ret = -1;
+				break;
+			}
+
+			evlist__for_each_entry(metric_evlist, evsel)
+				evsel->default_metricgroup = true;
+
+			evlist__splice_list_tail(evlist, &evlist__core(metric_evlist)->entries);
+			metricgroup__copy_metric_events(evlist, /*cgrp=*/NULL,
+							evlist__metric_events(evlist),
+							evlist__metric_events(metric_evlist));
+			evlist__put(metric_evlist);
+		}
+		list_sort(/*priv=*/NULL, &evlist__core(evlist)->entries, default_evlist_evsel_cmp);
+
+	}
+out:
+	if (!ret) {
+		evlist__for_each_entry(evlist, evsel) {
+			/*
+			 * Make at least one event non-skippable so fatal errors are visible.
+			 * 'cycles' always used to be default and non-skippable, so use that.
+			 */
+			if (!evsel__match(evsel, HARDWARE, HW_CPU_CYCLES))
+				evsel->skippable = true;
+		}
+	}
+	parse_events_error__exit(&err);
+	evlist__splice_list_tail(evsel_list, &evlist__core(evlist)->entries);
+	metricgroup__copy_metric_events(evsel_list, /*cgrp=*/NULL,
+					evlist__metric_events(evsel_list),
+					evlist__metric_events(evlist));
+	evlist__put(evlist);
+	return ret;
+}
+
+static const char * const stat_record_usage[] = {
+	"perf stat record [<options>]",
+	NULL,
+};
+
+static void init_features(struct perf_session *session)
+{
+	int feat;
+
+	for (feat = HEADER_FIRST_FEATURE; feat < HEADER_LAST_FEATURE; feat++)
+		perf_header__set_feat(&session->header, feat);
+
+	perf_header__clear_feat(&session->header, HEADER_DIR_FORMAT);
+	perf_header__clear_feat(&session->header, HEADER_BUILD_ID);
+	perf_header__clear_feat(&session->header, HEADER_TRACING_DATA);
+	perf_header__clear_feat(&session->header, HEADER_BRANCH_STACK);
+	perf_header__clear_feat(&session->header, HEADER_AUXTRACE);
+}
+
+static int __cmd_record(const struct option stat_options[], struct opt_aggr_mode *opt_mode,
+			int argc, const char **argv)
+{
+	struct perf_session *session;
+	struct perf_data *data = &perf_stat.data;
+
+	argc = parse_options(argc, argv, stat_options, stat_record_usage,
+			     PARSE_OPT_STOP_AT_NON_OPTION);
+	stat_config.aggr_mode = opt_aggr_mode_to_aggr_mode(opt_mode);
+
+	if (output_name)
+		data->path = output_name;
+
+	if (stat_config.run_count != 1 || forever) {
+		pr_err("Cannot use -r option with perf stat record.\n");
+		return -1;
+	}
+
+	session = perf_session__new(data, NULL);
+	if (IS_ERR(session)) {
+		pr_err("Perf session creation failed\n");
+		return PTR_ERR(session);
+	}
+
+	init_features(session);
+
+	session->evlist   = evsel_list;
+	perf_stat.session = session;
+	perf_stat.record  = true;
+	return argc;
+}
+
+static int process_stat_round_event(const struct perf_tool *tool __maybe_unused,
+				    struct perf_session *session,
+				    union perf_event *event)
+{
+	struct perf_record_stat_round *stat_round = &event->stat_round;
+	struct timespec tsh, *ts = NULL;
+	struct perf_env *env = perf_session__env(session);
+	const char **argv = env->cmdline_argv;
+	int argc = env->nr_cmdline;
+
+	process_counters();
+
+	if (stat_round->type == PERF_STAT_ROUND_TYPE__FINAL)
+		update_stats(stat_config.walltime_nsecs_stats, stat_round->time);
+
+	if (stat_config.interval && stat_round->time) {
+		tsh.tv_sec  = stat_round->time / NSEC_PER_SEC;
+		tsh.tv_nsec = stat_round->time % NSEC_PER_SEC;
+		ts = &tsh;
+	}
+
+	print_counters(ts, argc, argv);
+	return 0;
+}
+
+static
+int process_stat_config_event(const struct perf_tool *tool,
+			      struct perf_session *session,
+			      union perf_event *event)
+{
+	struct perf_stat *st = container_of(tool, struct perf_stat, tool);
+
+	perf_event__read_stat_config(&stat_config, &event->stat_config);
+
+	if (perf_cpu_map__is_empty(st->cpus)) {
+		if (st->aggr_mode != AGGR_UNSET)
+			pr_warning("warning: processing task data, aggregation mode not set\n");
+	} else if (st->aggr_mode != AGGR_UNSET) {
+		stat_config.aggr_mode = st->aggr_mode;
+	}
+
+	if (perf_stat.data.is_pipe)
+		perf_stat_init_aggr_mode();
+	else
+		perf_stat_init_aggr_mode_file(st);
+
+	if (stat_config.aggr_map) {
+		int nr_aggr = stat_config.aggr_map->nr;
+
+		if (evlist__alloc_aggr_stats(session->evlist, nr_aggr) < 0) {
+			pr_err("cannot allocate aggr counts\n");
+			return -1;
+		}
+	}
+	return 0;
+}
+
+static int set_maps(struct perf_stat *st)
+{
+	if (!st->cpus || !st->threads)
+		return 0;
+
+	if (WARN_ONCE(st->maps_allocated, "stats double allocation\n"))
+		return -EINVAL;
+
+	perf_evlist__set_maps(evlist__core(evsel_list), st->cpus, st->threads);
+
+	if (evlist__alloc_stats(&stat_config, evsel_list, /*alloc_raw=*/true))
+		return -ENOMEM;
+
+	st->maps_allocated = true;
+	return 0;
+}
+
+static
+int process_thread_map_event(const struct perf_tool *tool,
+			     struct perf_session *session __maybe_unused,
+			     union perf_event *event)
+{
+	struct perf_stat *st = container_of(tool, struct perf_stat, tool);
+
+	if (st->threads) {
+		pr_warning("Extra thread map event, ignoring.\n");
+		return 0;
+	}
+
+	st->threads = thread_map__new_event(&event->thread_map);
+	if (!st->threads)
+		return -ENOMEM;
+
+	return set_maps(st);
+}
+
+static
+int process_cpu_map_event(const struct perf_tool *tool,
+			  struct perf_session *session __maybe_unused,
+			  union perf_event *event)
+{
+	struct perf_stat *st = container_of(tool, struct perf_stat, tool);
+	struct perf_cpu_map *cpus;
+
+	if (st->cpus) {
+		pr_warning("Extra cpu map event, ignoring.\n");
+		return 0;
+	}
+
+	cpus = cpu_map__new_data(&event->cpu_map.data);
+	if (!cpus)
+		return -ENOMEM;
+
+	st->cpus = cpus;
+	return set_maps(st);
+}
+
+static const char * const stat_report_usage[] = {
+	"perf stat report [<options>]",
+	NULL,
+};
+
+static struct perf_stat perf_stat = {
+	.aggr_mode	= AGGR_UNSET,
+	.aggr_level	= 0,
+};
+
+static int __cmd_report(int argc, const char **argv)
+{
+	struct perf_session *session;
+	struct opt_aggr_mode opt_mode = {};
+	const struct option options[] = {
+	OPT_STRING('i', "input", &input_name, "file", "input file name"),
+	OPT_BOOLEAN(0, "per-thread", &opt_mode.thread, "aggregate counts per thread"),
+	OPT_BOOLEAN(0, "per-socket", &opt_mode.socket,
+		    "aggregate counts per processor socket"),
+	OPT_BOOLEAN(0, "per-die", &opt_mode.die, "aggregate counts per processor die"),
+	OPT_BOOLEAN(0, "per-cluster", &opt_mode.cluster,
+		    "aggregate counts per processor cluster"),
+	OPT_CALLBACK_OPTARG(0, "per-cache", &opt_mode.cache, &perf_stat.aggr_level,
+			    "cache level", "aggregate count at this cache level (Default: LLC)",
+			    parse_cache_level),
+	OPT_BOOLEAN(0, "per-core", &opt_mode.core,
+		    "aggregate counts per physical processor core"),
+	OPT_BOOLEAN(0, "per-node", &opt_mode.node, "aggregate counts per numa node"),
+	OPT_BOOLEAN('A', "no-aggr", &opt_mode.no_aggr,
+		    "disable aggregation across CPUs or PMUs"),
+	OPT_END()
+	};
+	struct stat st;
+	int ret;
+
+	argc = parse_options(argc, argv, options, stat_report_usage, 0);
+
+	perf_stat.aggr_mode = opt_aggr_mode_to_aggr_mode(&opt_mode);
+	if (perf_stat.aggr_mode == AGGR_GLOBAL)
+		perf_stat.aggr_mode = AGGR_UNSET; /* No option found so leave unset. */
+
+	if (!input_name || !strlen(input_name)) {
+		if (!fstat(STDIN_FILENO, &st) && S_ISFIFO(st.st_mode))
+			input_name = "-";
+		else
+			input_name = "perf.data";
+	}
+
+	perf_stat.data.path = input_name;
+	perf_stat.data.mode = PERF_DATA_MODE_READ;
+
+	perf_tool__init(&perf_stat.tool, /*ordered_events=*/false);
+	perf_stat.tool.attr		= perf_event__process_attr;
+	perf_stat.tool.event_update	= perf_event__process_event_update;
+	perf_stat.tool.thread_map	= process_thread_map_event;
+	perf_stat.tool.cpu_map		= process_cpu_map_event;
+	perf_stat.tool.stat_config	= process_stat_config_event;
+	perf_stat.tool.stat		= perf_event__process_stat_event;
+	perf_stat.tool.stat_round	= process_stat_round_event;
+
+	session = perf_session__new(&perf_stat.data, &perf_stat.tool);
+	if (IS_ERR(session))
+		return PTR_ERR(session);
+
+	perf_stat.session  = session;
+	stat_config.output = stderr;
+	evlist__put(evsel_list);
+	evsel_list         = session->evlist;
+
+	ret = perf_session__process_events(session);
+	if (ret)
+		return ret;
+
+	perf_session__delete(session);
+	return 0;
+}
+
+static void setup_system_wide(int forks)
+{
+	/*
+	 * Make system wide (-a) the default target if
+	 * no target was specified and one of following
+	 * conditions is met:
+	 *
+	 *   - there's no workload specified
+	 *   - there is workload specified but all requested
+	 *     events are system wide events
+	 */
+	if (!target__none(&target))
+		return;
+
+	if (!forks)
+		target.system_wide = true;
+	else {
+		struct evsel *counter;
+
+		evlist__for_each_entry(evsel_list, counter) {
+			if (!counter->core.requires_cpu &&
+			    !evsel__name_is(counter, "duration_time")) {
+				return;
+			}
+		}
+
+		if (evlist__nr_entries(evsel_list))
+			target.system_wide = true;
+	}
+}
+
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+static int parse_tpebs_mode(const struct option *opt, const char *str,
+			    int unset __maybe_unused)
+{
+	enum tpebs_mode *mode = opt->value;
+
+	if (!strcasecmp("mean", str)) {
+		*mode = TPEBS_MODE__MEAN;
+		return 0;
+	}
+	if (!strcasecmp("min", str)) {
+		*mode = TPEBS_MODE__MIN;
+		return 0;
+	}
+	if (!strcasecmp("max", str)) {
+		*mode = TPEBS_MODE__MAX;
+		return 0;
+	}
+	if (!strcasecmp("last", str)) {
+		*mode = TPEBS_MODE__LAST;
+		return 0;
+	}
+	return -1;
+}
+#endif // HAVE_ARCH_X86_64_SUPPORT
+
+int cmd_stat(int argc, const char **argv)
+{
+	struct opt_aggr_mode opt_mode = {};
+	bool affinity = true, affinity_set = false;
+	struct option stat_options[] = {
+		OPT_BOOLEAN('T', "transaction", &transaction_run,
+			"hardware transaction statistics"),
+		OPT_CALLBACK('e', "event", &parse_events_option_args, "event",
+			"event selector. use 'perf list' to list available events",
+			parse_events_option),
+		OPT_CALLBACK(0, "filter", &evsel_list, "filter",
+			"event filter", parse_filter),
+		OPT_BOOLEAN('i', "no-inherit", &stat_config.no_inherit,
+			"child tasks do not inherit counters"),
+		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"),
+#ifdef HAVE_BPF_SKEL
+		OPT_STRING('b', "bpf-prog", &target.bpf_str, "bpf-prog-id",
+			"stat events on existing bpf program id"),
+		OPT_BOOLEAN(0, "bpf-counters", &target.use_bpf,
+			"use bpf program to count events"),
+		OPT_STRING(0, "bpf-attr-map", &target.attr_map, "attr-map-path",
+			"path to perf_event_attr map"),
+#endif
+		OPT_BOOLEAN('a', "all-cpus", &target.system_wide,
+			"system-wide collection from all CPUs"),
+		OPT_BOOLEAN(0, "scale", &stat_config.scale,
+			"Use --no-scale to disable counter scaling for multiplexing"),
+		OPT_INCR('v', "verbose", &verbose,
+			"be more verbose (show counter open errors, etc)"),
+		OPT_INTEGER('r', "repeat", &stat_config.run_count,
+			"repeat command and print average + stddev (max: 100, forever: 0)"),
+		OPT_BOOLEAN(0, "table", &stat_config.walltime_run_table,
+			"display details about each run (only with -r option)"),
+		OPT_BOOLEAN('n', "null", &stat_config.null_run,
+			"null run - dont start any counters"),
+		OPT_BOOLEAN(0, "hide-zero-events", &stat_config.hide_zero,
+			"Do not show events with a zero count"),
+		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_STRING('C', "cpu", &target.cpu_list, "cpu",
+			"list of cpus to monitor in system-wide"),
+		OPT_BOOLEAN('A', "no-aggr", &opt_mode.no_aggr,
+			"disable aggregation across CPUs or PMUs"),
+		OPT_BOOLEAN(0, "no-merge", &opt_mode.no_aggr,
+			"disable aggregation the same as -A or -no-aggr"),
+		OPT_BOOLEAN(0, "hybrid-merge", &stat_config.hybrid_merge,
+			"Merge identical named hybrid events"),
+		OPT_STRING('x', "field-separator", &stat_config.csv_sep, "separator",
+			"print counts with custom separator"),
+		OPT_BOOLEAN('j', "json-output", &stat_config.json_output,
+			"print counts in JSON format"),
+		OPT_CALLBACK('G', "cgroup", &evsel_list, "name",
+			"monitor event in cgroup name only", parse_stat_cgroups),
+		OPT_STRING(0, "for-each-cgroup", &stat_config.cgroup_list, "name",
+			"expand events for each cgroup"),
+		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_STRING(0, "pre", &pre_cmd, "command",
+			"command to run prior to the measured command"),
+		OPT_STRING(0, "post", &post_cmd, "command",
+			"command to run after to the measured command"),
+		OPT_UINTEGER('I', "interval-print", &stat_config.interval,
+			"print counts at regular interval in ms "
+			"(overhead is possible for values <= 100ms)"),
+		OPT_INTEGER(0, "interval-count", &stat_config.times,
+			"print counts for fixed number of times"),
+		OPT_BOOLEAN(0, "interval-clear", &stat_config.interval_clear,
+			"clear screen in between new interval"),
+		OPT_UINTEGER(0, "timeout", &stat_config.timeout,
+			"stop workload and print counts after a timeout period in ms (>= 10ms)"),
+		OPT_BOOLEAN(0, "per-socket", &opt_mode.socket,
+			"aggregate counts per processor socket"),
+		OPT_BOOLEAN(0, "per-die", &opt_mode.die, "aggregate counts per processor die"),
+		OPT_BOOLEAN(0, "per-cluster", &opt_mode.cluster,
+			"aggregate counts per processor cluster"),
+		OPT_CALLBACK_OPTARG(0, "per-cache", &opt_mode.cache, &stat_config.aggr_level,
+				"cache level", "aggregate count at this cache level (Default: LLC)",
+				parse_cache_level),
+		OPT_BOOLEAN(0, "per-core", &opt_mode.core,
+			"aggregate counts per physical processor core"),
+		OPT_BOOLEAN(0, "per-thread", &opt_mode.thread, "aggregate counts per thread"),
+		OPT_BOOLEAN(0, "per-node", &opt_mode.node, "aggregate counts per numa node"),
+		OPT_INTEGER('D', "delay", &target.initial_delay,
+			"ms to wait before starting measurement after program start (-1: start with events disabled)"),
+		OPT_CALLBACK_NOOPT(0, "metric-only", &stat_config.metric_only, NULL,
+				"Only print computed metrics. No raw values", enable_metric_only),
+		OPT_BOOLEAN(0, "metric-no-group", &stat_config.metric_no_group,
+			"don't group metric events, impacts multiplexing"),
+		OPT_BOOLEAN(0, "metric-no-merge", &stat_config.metric_no_merge,
+			"don't try to share events between metrics in a group"),
+		OPT_BOOLEAN(0, "metric-no-threshold", &stat_config.metric_no_threshold,
+			"disable adding events for the metric threshold calculation"),
+		OPT_BOOLEAN(0, "topdown", &topdown_run,
+			"measure top-down statistics"),
+#ifdef HAVE_ARCH_X86_64_SUPPORT
+		OPT_BOOLEAN(0, "record-tpebs", &tpebs_recording,
+			"enable recording for tpebs when retire_latency required"),
+		OPT_CALLBACK(0, "tpebs-mode", &tpebs_mode, "tpebs-mode",
+			"Mode of TPEBS recording: mean, min or max",
+			parse_tpebs_mode),
+#endif
+		OPT_UINTEGER(0, "td-level", &stat_config.topdown_level,
+			"Set the metrics level for the top-down statistics (0: max level)"),
+		OPT_BOOLEAN(0, "smi-cost", &smi_cost,
+			"measure SMI cost"),
+		OPT_CALLBACK('M', "metrics", &evsel_list, "metric/metric group list",
+			"monitor specified metrics or metric groups (separated by ,)",
+			append_metric_groups),
+		OPT_BOOLEAN_FLAG(0, "all-kernel", &stat_config.all_kernel,
+				"Configure all used events to run in kernel space.",
+				PARSE_OPT_EXCLUSIVE),
+		OPT_BOOLEAN_FLAG(0, "all-user", &stat_config.all_user,
+				"Configure all used events to run in user space.",
+				PARSE_OPT_EXCLUSIVE),
+		OPT_BOOLEAN(0, "percore-show-thread", &stat_config.percore_show_thread,
+			"Use with 'percore' event qualifier to show the event "
+			"counts of one hardware thread by sum up total hardware "
+			"threads of same physical core"),
+		OPT_BOOLEAN(0, "summary", &stat_config.summary,
+			"print summary for interval mode"),
+		OPT_BOOLEAN(0, "no-csv-summary", &stat_config.no_csv_summary,
+			"don't print 'summary' for CSV summary output"),
+		OPT_BOOLEAN(0, "quiet", &quiet,
+			"don't print any output, messages or warnings (useful with record)"),
+		OPT_BOOLEAN_SET(0, "affinity", &affinity, &affinity_set,
+			"enable (default) or disable affinity optimizations to reduce IPIs"),
+		OPT_CALLBACK(0, "cputype", &evsel_list, "hybrid cpu type",
+			"Only enable events on applying cpu with this type "
+			"for hybrid platform (e.g. core or atom)",
+			parse_cputype),
+		OPT_CALLBACK(0, "pmu-filter", &evsel_list, "pmu",
+			"Only enable events on applying pmu with specified "
+			"for multiple pmus with same type(e.g. hisi_sicl2_cpa0 or hisi_sicl0_cpa0)",
+			parse_pmu_filter),
+#ifdef HAVE_LIBPFM
+		OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
+			"libpfm4 event selector. use 'perf list' to list available events",
+			parse_libpfm_events_option),
+#endif
+		OPT_CALLBACK(0, "control", &stat_config, "fd:ctl-fd[,ack-fd] or fifo:ctl-fifo[,ack-fifo]",
+			"Listen on ctl-fd descriptor for command to control measurement ('enable': enable events, 'disable': disable events).\n"
+			"\t\t\t  Optionally send control command completion ('ack\\n') to ack-fd descriptor.\n"
+			"\t\t\t  Alternatively, ctl-fifo / ack-fifo will be opened and used as ctl-fd / ack-fd.",
+			parse_control_option),
+		OPT_CALLBACK_OPTARG(0, "iostat", &evsel_list, &stat_config, "default",
+				"measure I/O performance metrics provided by arch/platform",
+				iostat_parse),
+		OPT_END()
+	};
+	const char * const stat_usage[] = {
+		"perf stat [<options>] [<command>]",
+		NULL
+	};
+	int status = -EINVAL, run_idx, err;
+	const char *mode;
+	FILE *output = stderr;
+	unsigned int interval, timeout;
+	const char * const stat_subcommands[] = { "record", "report" };
+	char errbuf[BUFSIZ];
+	struct evsel *counter;
+
+	setlocale(LC_ALL, "");
+
+	evsel_list = evlist__new();
+	if (evsel_list == NULL)
+		return -ENOMEM;
+
+	parse_events__shrink_config_terms();
+
+	/* String-parsing callback-based options would segfault when negated */
+	set_option_flag(stat_options, 'e', "event", PARSE_OPT_NONEG);
+	set_option_flag(stat_options, 'M', "metrics", PARSE_OPT_NONEG);
+	set_option_flag(stat_options, 'G', "cgroup", PARSE_OPT_NONEG);
+
+	argc = parse_options_subcommand(argc, argv, stat_options, stat_subcommands,
+					(const char **) stat_usage,
+					PARSE_OPT_STOP_AT_NON_OPTION);
+
+	stat_config.aggr_mode = opt_aggr_mode_to_aggr_mode(&opt_mode);
+
+	if (stat_config.csv_sep && stat_config.json_output) {
+		fprintf(stderr, "cannot use both --field-separator and --json-output\n");
+		parse_options_usage(stat_usage, stat_options, "x", 1);
+		parse_options_usage(NULL, stat_options, "j", 1);
+		goto out;
+	}
+
+	if (stat_config.csv_sep) {
+		stat_config.csv_output = true;
+		if (!strcmp(stat_config.csv_sep, "\\t"))
+			stat_config.csv_sep = "\t";
+	} else
+		stat_config.csv_sep = DEFAULT_SEPARATOR;
+
+	if (affinity_set)
+		evlist__set_no_affinity(evsel_list, !affinity);
+
+	if (argc && strlen(argv[0]) > 2 && strstarts("record", argv[0])) {
+		argc = __cmd_record(stat_options, &opt_mode, argc, argv);
+		if (argc < 0)
+			return -1;
+	} else if (argc && strlen(argv[0]) > 2 && strstarts("report", argv[0]))
+		return __cmd_report(argc, argv);
+
+	interval = stat_config.interval;
+	timeout = stat_config.timeout;
+
+	/*
+	 * For record command the -o is already taken care of.
+	 */
+	if (!STAT_RECORD && output_name && strcmp(output_name, "-"))
+		output = NULL;
+
+	if (output_name && output_fd) {
+		fprintf(stderr, "cannot use both --output and --log-fd\n");
+		parse_options_usage(stat_usage, stat_options, "o", 1);
+		parse_options_usage(NULL, stat_options, "log-fd", 0);
+		goto out;
+	}
+
+	if (stat_config.metric_only && stat_config.aggr_mode == AGGR_THREAD) {
+		fprintf(stderr, "--metric-only is not supported with --per-thread\n");
+		goto out;
+	}
+
+	if (stat_config.metric_only && stat_config.run_count > 1) {
+		fprintf(stderr, "--metric-only is not supported with -r\n");
+		goto out;
+	}
+
+	if (stat_config.csv_output || (stat_config.metric_only && stat_config.json_output)) {
+		/*
+		 * Current CSV and metric-only JSON output doesn't display the
+		 * metric threshold so don't compute it.
+		 */
+		stat_config.metric_no_threshold = true;
+	}
+
+	if (stat_config.walltime_run_table && stat_config.run_count <= 1) {
+		fprintf(stderr, "--table is only supported with -r\n");
+		parse_options_usage(stat_usage, stat_options, "r", 1);
+		parse_options_usage(NULL, stat_options, "table", 0);
+		goto out;
+	}
+
+	if (output_fd < 0) {
+		fprintf(stderr, "argument to --log-fd must be a > 0\n");
+		parse_options_usage(stat_usage, stat_options, "log-fd", 0);
+		goto out;
+	}
+
+	if (!output && !quiet) {
+		struct timespec tm;
+		mode = append_file ? "a" : "w";
+
+		output = fopen(output_name, mode);
+		if (!output) {
+			perror("failed to create output file");
+			status = -1;
+			goto out;
+		}
+		if (!stat_config.json_output) {
+			clock_gettime(CLOCK_REALTIME, &tm);
+			fprintf(output, "# started on %s\n", ctime(&tm.tv_sec));
+		}
+	} else if (output_fd > 0) {
+		mode = append_file ? "a" : "w";
+		output = fdopen(output_fd, mode);
+		if (!output) {
+			perror("Failed opening logfd");
+			status = -errno;
+			goto out;
+		}
+	}
+
+	if (stat_config.interval_clear && !isatty(fileno(output))) {
+		fprintf(stderr, "--interval-clear does not work with output\n");
+		parse_options_usage(stat_usage, stat_options, "o", 1);
+		parse_options_usage(NULL, stat_options, "log-fd", 0);
+		parse_options_usage(NULL, stat_options, "interval-clear", 0);
+		status = -1;
+		goto out;
+	}
+
+	stat_config.output = output;
+
+	/*
+	 * let the spreadsheet do the pretty-printing
+	 */
+	if (stat_config.csv_output) {
+		/* User explicitly passed -B? */
+		if (big_num_opt == 1) {
+			fprintf(stderr, "-B option not supported with -x\n");
+			parse_options_usage(stat_usage, stat_options, "B", 1);
+			parse_options_usage(NULL, stat_options, "x", 1);
+			goto out;
+		} else /* Nope, so disable big number formatting */
+			stat_config.big_num = false;
+	} else if (big_num_opt == 0) /* User passed --no-big-num */
+		stat_config.big_num = false;
+
+	target.inherit = !stat_config.no_inherit;
+	err = target__validate(&target);
+	if (err) {
+		target__strerror(&target, err, errbuf, BUFSIZ);
+		pr_warning("%s\n", errbuf);
+	}
+
+	setup_system_wide(argc);
+
+	/*
+	 * Display user/system times only for single
+	 * run and when there's specified tracee.
+	 */
+	if ((stat_config.run_count == 1) && target__none(&target))
+		stat_config.ru_display = true;
+
+	if (stat_config.run_count < 0) {
+		pr_err("Run count must be a positive number\n");
+		parse_options_usage(stat_usage, stat_options, "r", 1);
+		goto out;
+	} else if (stat_config.run_count == 0) {
+		forever = true;
+		stat_config.run_count = 1;
+	}
+
+	if (stat_config.walltime_run_table) {
+		stat_config.walltime_run = calloc(stat_config.run_count, sizeof(stat_config.walltime_run[0]));
+		if (!stat_config.walltime_run) {
+			pr_err("failed to setup -r option");
+			goto out;
+		}
+	}
+
+	if ((stat_config.aggr_mode == AGGR_THREAD) &&
+		!target__has_task(&target)) {
+		if (!target.system_wide || target.cpu_list) {
+			fprintf(stderr, "The --per-thread option is only "
+				"available when monitoring via -p -t -a "
+				"options or only --per-thread.\n");
+			parse_options_usage(NULL, stat_options, "p", 1);
+			parse_options_usage(NULL, stat_options, "t", 1);
+			goto out;
+		}
+	}
+
+	/*
+	 * no_aggr, cgroup are for system-wide only
+	 * --per-thread is aggregated per thread, we dont mix it with cpu mode
+	 */
+	if (((stat_config.aggr_mode != AGGR_GLOBAL &&
+	      stat_config.aggr_mode != AGGR_THREAD) ||
+	     (nr_cgroups || stat_config.cgroup_list)) &&
+	    !target__has_cpu(&target)) {
+		fprintf(stderr, "both cgroup and no-aggregation "
+			"modes only available in system-wide mode\n");
+
+		parse_options_usage(stat_usage, stat_options, "G", 1);
+		parse_options_usage(NULL, stat_options, "A", 1);
+		parse_options_usage(NULL, stat_options, "a", 1);
+		parse_options_usage(NULL, stat_options, "for-each-cgroup", 0);
+		goto out;
+	}
+
+	if (stat_config.iostat_run) {
+		status = iostat_prepare(&evsel_list, &stat_config);
+		if (status)
+			goto out;
+		if (iostat_mode == IOSTAT_LIST) {
+			iostat_list(evsel_list, &stat_config);
+			goto out;
+		} else if (verbose > 0)
+			iostat_list(evsel_list, &stat_config);
+		if (iostat_mode == IOSTAT_RUN && !target__has_cpu(&target))
+			target.system_wide = true;
+	}
+
+	if ((stat_config.aggr_mode == AGGR_THREAD) && (target.system_wide))
+		target.per_thread = true;
+
+	stat_config.system_wide = target.system_wide;
+	if (target.cpu_list) {
+		stat_config.user_requested_cpu_list = strdup(target.cpu_list);
+		if (!stat_config.user_requested_cpu_list) {
+			status = -ENOMEM;
+			goto out;
+		}
+	}
+
+	/*
+	 * Metric parsing needs to be delayed as metrics may optimize events
+	 * knowing the target is system-wide.
+	 */
+	if (metrics) {
+		const char *pmu = parse_events_option_args.pmu_filter ?: "all";
+		int ret = metricgroup__parse_groups(evsel_list, pmu,
+						    parse_events_option_args.cputype_filter,
+						    metrics,
+						stat_config.metric_no_group,
+						stat_config.metric_no_merge,
+						stat_config.metric_no_threshold,
+						stat_config.user_requested_cpu_list,
+						stat_config.system_wide,
+						stat_config.hardware_aware_grouping);
+
+		zfree(&metrics);
+		if (ret) {
+			status = ret;
+			goto out;
+		}
+	}
+
+	if (add_default_events())
+		goto out;
+
+	if (stat_config.cgroup_list) {
+		if (nr_cgroups > 0) {
+			pr_err("--cgroup and --for-each-cgroup cannot be used together\n");
+			parse_options_usage(stat_usage, stat_options, "G", 1);
+			parse_options_usage(NULL, stat_options, "for-each-cgroup", 0);
+			goto out;
+		}
+
+		if (evlist__expand_cgroup(evsel_list, stat_config.cgroup_list,
+					  !target.use_bpf) < 0) {
+			parse_options_usage(stat_usage, stat_options,
+					    "for-each-cgroup", 0);
+			goto out;
+		}
+	}
+#ifdef HAVE_BPF_SKEL
+	if (target.use_bpf && nr_cgroups &&
+	    (evlist__nr_entries(evsel_list) / nr_cgroups) > BPERF_CGROUP__MAX_EVENTS) {
+		pr_warning("Disabling BPF counters due to more events (%d) than the max (%d)\n",
+			   evlist__nr_entries(evsel_list) / nr_cgroups,
+			   BPERF_CGROUP__MAX_EVENTS);
+		target.use_bpf = false;
+	}
+#endif // HAVE_BPF_SKEL
+	evlist__warn_user_requested_cpus(evsel_list, target.cpu_list);
+
+	evlist__for_each_entry(evsel_list, counter) {
+		/*
+		 * Setup BPF counters to require CPUs as any(-1) isn't
+		 * supported. evlist__create_maps below will propagate this
+		 * information to the evsels. Note, evsel__is_bperf isn't yet
+		 * set up, and this change must happen early, so directly use
+		 * the bpf_counter variable and target information.
+		 */
+		if ((counter->bpf_counter || target.use_bpf) && !target__has_cpu(&target))
+			counter->core.requires_cpu = true;
+	}
+
+	if (evlist__create_maps(evsel_list, &target) < 0) {
+		if (target__has_task(&target)) {
+			pr_err("Problems finding threads of monitor\n");
+			parse_options_usage(stat_usage, stat_options, "p", 1);
+			parse_options_usage(NULL, stat_options, "t", 1);
+		} else if (target__has_cpu(&target)) {
+			perror("failed to parse CPUs map");
+			parse_options_usage(stat_usage, stat_options, "C", 1);
+			parse_options_usage(NULL, stat_options, "a", 1);
+		}
+		goto out;
+	}
+
+	evlist__check_cpu_maps(evsel_list);
+
+	/*
+	 * Initialize thread_map with comm names,
+	 * so we could print it out on output.
+	 */
+	if (stat_config.aggr_mode == AGGR_THREAD) {
+		thread_map__read_comms(evlist__core(evsel_list)->threads);
+	}
+
+	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) {
+		pr_err("interval-count option should be used together with "
+				"interval-print.\n");
+		parse_options_usage(stat_usage, stat_options, "interval-count", 0);
+		parse_options_usage(stat_usage, stat_options, "I", 1);
+		goto out;
+	}
+
+	if (timeout && timeout < 100) {
+		if (timeout < 10) {
+			pr_err("timeout must be >= 10ms.\n");
+			parse_options_usage(stat_usage, stat_options, "timeout", 0);
+			goto out;
+		} else
+			pr_warning("timeout < 100ms. "
+				   "The overhead percentage could be high in some cases. "
+				   "Please proceed with caution.\n");
+	}
+	if (timeout && interval) {
+		pr_err("timeout option is not supported with interval-print.\n");
+		parse_options_usage(stat_usage, stat_options, "timeout", 0);
+		parse_options_usage(stat_usage, stat_options, "I", 1);
+		goto out;
+	}
+
+	if (perf_stat_init_aggr_mode())
+		goto out;
+
+	if (evlist__alloc_stats(&stat_config, evsel_list, interval))
+		goto out;
+
+	/*
+	 * Set sample_type to PERF_SAMPLE_IDENTIFIER, which should be harmless
+	 * while avoiding that older tools show confusing messages.
+	 *
+	 * However for pipe sessions we need to keep it zero,
+	 * because script's perf_evsel__check_attr is triggered
+	 * by attr->sample_type != 0, and we can't run it on
+	 * stat sessions.
+	 */
+	stat_config.identifier = !(STAT_RECORD && perf_stat.data.is_pipe);
+
+	/*
+	 * 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);
+	if (!forever)
+		signal(SIGINT,  skip_signal);
+	signal(SIGCHLD, skip_signal);
+	signal(SIGALRM, skip_signal);
+	signal(SIGABRT, skip_signal);
+
+	if (evlist__initialize_ctlfd(evsel_list, stat_config.ctl_fd, stat_config.ctl_fd_ack))
+		goto out;
+
+	/* Enable ignoring missing threads when -p option is defined. */
+	evlist__first(evsel_list)->ignore_missing_thread = target.pid;
+	status = 0;
+	for (run_idx = 0; forever || run_idx < stat_config.run_count; run_idx++) {
+		if (stat_config.run_count != 1 && verbose > 0)
+			fprintf(output, "[ perf stat: executing run #%d ... ]\n",
+				run_idx + 1);
+
+		if (run_idx != 0)
+			evlist__reset_prev_raw_counts(evsel_list);
+
+		status = run_perf_stat(argc, argv, run_idx);
+		if (status < 0)
+			break;
+
+		if (forever && !interval) {
+			print_counters(NULL, argc, argv);
+			perf_stat__reset_stats();
+		}
+	}
+
+	if (!forever && status != -1 && (!interval || stat_config.summary)) {
+		if (stat_config.run_count > 1)
+			evlist__copy_res_stats(&stat_config, evsel_list);
+		print_counters(NULL, argc, argv);
+	}
+
+	evlist__finalize_ctlfd(evsel_list);
+
+	if (STAT_RECORD) {
+		/*
+		 * We synthesize the kernel mmap record just so that older tools
+		 * don't emit warnings about not being able to resolve symbols
+		 * due to /proc/sys/kernel/kptr_restrict settings and instead provide
+		 * a saner message about no samples being in the perf.data file.
+		 *
+		 * This also serves to suppress a warning about f_header.data.size == 0
+		 * in header.c at the moment 'perf stat record' gets introduced, which
+		 * is not really needed once we start adding the stat specific PERF_RECORD_
+		 * records, but the need to suppress the kptr_restrict messages in older
+		 * tools remain  -acme
+		 */
+		int fd = perf_data__fd(&perf_stat.data);
+
+		err = perf_event__synthesize_kernel_mmap((void *)&perf_stat,
+							 process_synthesized_event,
+							 &perf_stat.session->machines.host);
+		if (err) {
+			pr_warning("Couldn't synthesize the kernel mmap record, harmless, "
+				   "older tools may produce warnings about this file\n.");
+		}
+
+		if (!interval) {
+			if (WRITE_STAT_ROUND_EVENT(stat_config.walltime_nsecs_stats->max, FINAL))
+				pr_err("failed to write stat round event\n");
+		}
+
+		if (!perf_stat.data.is_pipe) {
+			perf_stat.session->header.data_size += perf_stat.bytes_written;
+			perf_session__write_header(perf_stat.session, evsel_list, fd, true);
+		}
+
+		evlist__close(evsel_list);
+		perf_session__delete(perf_stat.session);
+	}
+
+	perf_stat__exit_aggr_mode();
+	evlist__free_stats(evsel_list);
+out:
+	if (stat_config.iostat_run)
+		iostat_release(evsel_list);
+
+	zfree(&stat_config.walltime_run);
+	zfree(&stat_config.user_requested_cpu_list);
+
+	if (smi_cost && smi_reset)
+		sysfs__write_int(FREEZE_ON_SMI_PATH, 0);
+
+	evlist__put(evsel_list);
+
+	evlist__close_control(stat_config.ctl_fd, stat_config.ctl_fd_ack, &stat_config.ctl_fd_close);
+
+	/* Only the low byte of status becomes the exit code. */
+	return abs(status);
+}