Merge tag 'perf-tools-for-v5.17-2022-01-22' of...

Merge tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull more perf tools updates from Arnaldo Carvalho de Melo:

 - Fix printing 'phys_addr' in 'perf script'.

 - Fix failure to add events with 'perf probe' in ppc64 due to not
   removing leading dot (ppc64 ABIv1).

 - Fix cpu_map__item() python binding building.

 - Support event alias in form foo-bar-baz, add pmu-events and
   parse-event tests for it.

 - No need to setup affinities when starting a workload or attaching to
   a pid.

 - Use path__join() to compose a path instead of ad-hoc snprintf()
   equivalent.

 - Override attr->sample_period for non-libpfm4 events.

 - Use libperf cpumap APIs instead of accessing the internal state
   directly.

 - Sync x86 arch prctl headers and files changed by the new
   set_mempolicy_home_node syscall with the kernel sources.

 - Remove duplicate include in cpumap.h.

 - Remove redundant err variable.

* tag 'perf-tools-for-v5.17-2022-01-22' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
  perf tools: Remove redundant err variable
  perf test: Add parse-events test for aliases with hyphens
  perf test: Add pmu-events test for aliases with hyphens
  perf parse-events: Support event alias in form foo-bar-baz
  perf evsel: Override attr->sample_period for non-libpfm4 events
  perf cpumap: Remove duplicate include in cpumap.h
  perf cpumap: Migrate to libperf cpumap api
  perf python: Fix cpu_map__item() building
  perf script: Fix printing 'phys_addr' failure issue
  tools headers UAPI: Sync files changed by new set_mempolicy_home_node syscall
  tools headers UAPI: Sync x86 arch prctl headers with the kernel sources
  perf machine: Use path__join() to compose a path instead of snprintf(dir, '/', filename)
  perf evlist: No need to setup affinities when disabling events for pid targets
  perf evlist: No need to setup affinities when enabling events for pid targets
  perf stat: No need to setup affinities when starting a workload
  perf affinity: Allow passing a NULL arg to affinity__cleanup()
  perf probe: Fix ppc64 'perf probe add events failed' case

tools/arch/x86/include/uapi/asm/prctl.h

@@ -13,6 +13,8 @@
 #define ARCH_GET_XCOMP_SUPP		0x1021
 #define ARCH_GET_XCOMP_PERM		0x1022
 #define ARCH_REQ_XCOMP_PERM		0x1023
+#define ARCH_GET_XCOMP_GUEST_PERM	0x1024
+#define ARCH_REQ_XCOMP_GUEST_PERM	0x1025
 
 #define ARCH_MAP_VDSO_X32		0x2001
 #define ARCH_MAP_VDSO_32		0x2002

tools/include/uapi/asm-generic/unistd.h

@@ -883,8 +883,11 @@ __SYSCALL(__NR_process_mrelease, sys_process_mrelease)
 #define __NR_futex_waitv 449
 __SYSCALL(__NR_futex_waitv, sys_futex_waitv)
 
+#define __NR_set_mempolicy_home_node 450
+__SYSCALL(__NR_set_mempolicy_home_node, sys_set_mempolicy_home_node)
+
 #undef __NR_syscalls
-#define __NR_syscalls 450
+#define __NR_syscalls 451
 
 /*
  * 32 bit systems traditionally used different

tools/lib/perf/evsel.c

@@ -141,7 +141,7 @@ int perf_evsel__open(struct perf_evsel *evsel, struct perf_cpu_map *cpus,
 	}
 
 	if (evsel->fd == NULL &&
-	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
+	    perf_evsel__alloc_fd(evsel, perf_cpu_map__nr(cpus), threads->nr) < 0)
 		return -ENOMEM;
 
 	perf_cpu_map__for_each_cpu(cpu, idx, cpus) {
@@ -384,7 +384,7 @@ int perf_evsel__apply_filter(struct perf_evsel *evsel, const char *filter)
 {
 	int err = 0, i;
 
-	for (i = 0; i < evsel->cpus->nr && !err; i++)
+	for (i = 0; i < perf_cpu_map__nr(evsel->cpus) && !err; i++)
 		err = perf_evsel__run_ioctl(evsel,
 				     PERF_EVENT_IOC_SET_FILTER,
 				     (void *)filter, i);

tools/perf/arch/mips/entry/syscalls/syscall_n64.tbl

@@ -364,3 +364,4 @@
 # 447 reserved for memfd_secret
 448	n64	process_mrelease		sys_process_mrelease
 449	n64	futex_waitv			sys_futex_waitv
+450	common	set_mempolicy_home_node		sys_set_mempolicy_home_node

tools/perf/arch/powerpc/entry/syscalls/syscall.tbl

@@ -529,3 +529,4 @@
 # 447 reserved for memfd_secret
 448	common	process_mrelease		sys_process_mrelease
 449	common  futex_waitv                     sys_futex_waitv
+450 	nospu	set_mempolicy_home_node		sys_set_mempolicy_home_node

tools/perf/arch/s390/entry/syscalls/syscall.tbl

@@ -452,3 +452,4 @@
 # 447 reserved for memfd_secret
 448  common	process_mrelease	sys_process_mrelease		sys_process_mrelease
 449  common	futex_waitv		sys_futex_waitv			sys_futex_waitv
+450  common	set_mempolicy_home_node	sys_set_mempolicy_home_node	sys_set_mempolicy_home_node

tools/perf/arch/x86/entry/syscalls/syscall_64.tbl

@@ -371,6 +371,7 @@
 447	common	memfd_secret		sys_memfd_secret
 448	common	process_mrelease	sys_process_mrelease
 449	common	futex_waitv		sys_futex_waitv
+450	common	set_mempolicy_home_node	sys_set_mempolicy_home_node
 
 #
 # Due to a historical design error, certain syscalls are numbered differently

tools/perf/bench/epoll-ctl.c

@@ -333,7 +333,7 @@ int bench_epoll_ctl(int argc, const char **argv)
 
 	/* default to the number of CPUs */
 	if (!nthreads)
-		nthreads = cpu->nr;
+		nthreads = perf_cpu_map__nr(cpu);
 
 	worker = calloc(nthreads, sizeof(*worker));
 	if (!worker)

tools/perf/bench/epoll-wait.c

@@ -452,7 +452,7 @@ int bench_epoll_wait(int argc, const char **argv)
 
 	/* default to the number of CPUs and leave one for the writer pthread */
 	if (!nthreads)
-		nthreads = cpu->nr - 1;
+		nthreads = perf_cpu_map__nr(cpu) - 1;
 
 	worker = calloc(nthreads, sizeof(*worker));
 	if (!worker) {

tools/perf/bench/evlist-open-close.c

@@ -71,7 +71,7 @@ static int evlist__count_evsel_fds(struct evlist *evlist)
 	int cnt = 0;
 
 	evlist__for_each_entry(evlist, evsel)
-		cnt += evsel->core.threads->nr * evsel->core.cpus->nr;
+		cnt += evsel->core.threads->nr * perf_cpu_map__nr(evsel->core.cpus);
 
 	return cnt;
 }
@@ -151,7 +151,7 @@ static int bench_evlist_open_close__run(char *evstr)
 
 	init_stats(&time_stats);
 
-	printf("  Number of cpus:\t%d\n", evlist->core.cpus->nr);
+	printf("  Number of cpus:\t%d\n", perf_cpu_map__nr(evlist->core.cpus));
 	printf("  Number of threads:\t%d\n", evlist->core.threads->nr);
 	printf("  Number of events:\t%d (%d fds)\n",
 		evlist->core.nr_entries, evlist__count_evsel_fds(evlist));

tools/perf/bench/futex-hash.c

@@ -150,7 +150,7 @@ int bench_futex_hash(int argc, const char **argv)
 	}
 
 	if (!params.nthreads) /* default to the number of CPUs */
-		params.nthreads = cpu->nr;
+		params.nthreads = perf_cpu_map__nr(cpu);
 
 	worker = calloc(params.nthreads, sizeof(*worker));
 	if (!worker)

tools/perf/bench/futex-lock-pi.c

@@ -173,7 +173,7 @@ int bench_futex_lock_pi(int argc, const char **argv)
 	}
 
 	if (!params.nthreads)
-		params.nthreads = cpu->nr;
+		params.nthreads = perf_cpu_map__nr(cpu);
 
 	worker = calloc(params.nthreads, sizeof(*worker));
 	if (!worker)

tools/perf/bench/futex-requeue.c

@@ -175,7 +175,7 @@ int bench_futex_requeue(int argc, const char **argv)
 	}
 
 	if (!params.nthreads)
-		params.nthreads = cpu->nr;
+		params.nthreads = perf_cpu_map__nr(cpu);
 
 	worker = calloc(params.nthreads, sizeof(*worker));
 	if (!worker)

tools/perf/bench/futex-wake-parallel.c

@@ -252,7 +252,7 @@ int bench_futex_wake_parallel(int argc, const char **argv)
 		err(EXIT_FAILURE, "calloc");
 
 	if (!params.nthreads)
-		params.nthreads = cpu->nr;
+		params.nthreads = perf_cpu_map__nr(cpu);
 
 	/* some sanity checks */
 	if (params.nwakes > params.nthreads ||

tools/perf/bench/futex-wake.c

@@ -151,7 +151,7 @@ int bench_futex_wake(int argc, const char **argv)
 	}
 
 	if (!params.nthreads)
-		params.nthreads = cpu->nr;
+		params.nthreads = perf_cpu_map__nr(cpu);
 
 	worker = calloc(params.nthreads, sizeof(*worker));
 	if (!worker)

tools/perf/builtin-ftrace.c

@@ -281,7 +281,7 @@ static int set_tracing_cpumask(struct perf_cpu_map *cpumap)
 	int ret;
 	int last_cpu;
 
-	last_cpu = perf_cpu_map__cpu(cpumap, cpumap->nr - 1).cpu;
+	last_cpu = perf_cpu_map__cpu(cpumap, perf_cpu_map__nr(cpumap) - 1).cpu;
 	mask_size = last_cpu / 4 + 2; /* one more byte for EOS */
 	mask_size += last_cpu / 32; /* ',' is needed for every 32th cpus */
 

tools/perf/builtin-inject.c

@@ -535,12 +535,9 @@ static int perf_event__repipe_exit(struct perf_tool *tool,
 static int perf_event__repipe_tracing_data(struct perf_session *session,
 					   union perf_event *event)
 {
-	int err;
-
 	perf_event__repipe_synth(session->tool, event);
-	err = perf_event__process_tracing_data(session, event);
 
-	return err;
+	return perf_event__process_tracing_data(session, event);
 }
 
 static int dso__read_build_id(struct dso *dso)

tools/perf/builtin-script.c

@@ -515,7 +515,7 @@ static int evsel__check_attr(struct evsel *evsel, struct perf_session *session)
 		return -EINVAL;
 
 	if (PRINT_FIELD(PHYS_ADDR) &&
-	    evsel__check_stype(evsel, PERF_SAMPLE_PHYS_ADDR, "PHYS_ADDR", PERF_OUTPUT_PHYS_ADDR))
+	    evsel__do_check_stype(evsel, PERF_SAMPLE_PHYS_ADDR, "PHYS_ADDR", PERF_OUTPUT_PHYS_ADDR, allow_user_set))
 		return -EINVAL;
 
 	if (PRINT_FIELD(DATA_PAGE_SIZE) &&

tools/perf/builtin-stat.c

@@ -230,11 +230,12 @@ static bool cpus_map_matched(struct evsel *a, struct evsel *b)
 	if (!a->core.cpus || !b->core.cpus)
 		return false;
 
-	if (a->core.cpus->nr != b->core.cpus->nr)
+	if (perf_cpu_map__nr(a->core.cpus) != perf_cpu_map__nr(b->core.cpus))
 		return false;
 
-	for (int i = 0; i < a->core.cpus->nr; i++) {
-		if (a->core.cpus->map[i].cpu != b->core.cpus->map[i].cpu)
+	for (int i = 0; i < perf_cpu_map__nr(a->core.cpus); i++) {
+		if (perf_cpu_map__cpu(a->core.cpus, i).cpu !=
+		    perf_cpu_map__cpu(b->core.cpus, i).cpu)
 			return false;
 	}
 
@@ -788,7 +789,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 	const bool forks = (argc > 0);
 	bool is_pipe = STAT_RECORD ? perf_stat.data.is_pipe : false;
 	struct evlist_cpu_iterator evlist_cpu_itr;
-	struct affinity affinity;
+	struct affinity saved_affinity, *affinity = NULL;
 	int err;
 	bool second_pass = false;
 
@@ -803,8 +804,11 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 	if (group)
 		evlist__set_leader(evsel_list);
 
-	if (affinity__setup(&affinity) < 0)
+	if (!cpu_map__is_dummy(evsel_list->core.cpus)) {
+		if (affinity__setup(&saved_affinity) < 0)
 			return -1;
+		affinity = &saved_affinity;
+	}
 
 	evlist__for_each_entry(evsel_list, counter) {
 		if (bpf_counter__load(counter, &target))
@@ -813,7 +817,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 			all_counters_use_bpf = false;
 	}
 
-	evlist__for_each_cpu(evlist_cpu_itr, evsel_list, &affinity) {
+	evlist__for_each_cpu(evlist_cpu_itr, evsel_list, affinity) {
 		counter = evlist_cpu_itr.evsel;
 
 		/*
@@ -869,7 +873,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 		 */
 
 		/* First close errored or weak retry */
-		evlist__for_each_cpu(evlist_cpu_itr, evsel_list, &affinity) {
+		evlist__for_each_cpu(evlist_cpu_itr, evsel_list, affinity) {
 			counter = evlist_cpu_itr.evsel;
 
 			if (!counter->reset_group && !counter->errored)
@@ -878,7 +882,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 			perf_evsel__close_cpu(&counter->core, evlist_cpu_itr.cpu_map_idx);
 		}
 		/* Now reopen weak */
-		evlist__for_each_cpu(evlist_cpu_itr, evsel_list, &affinity) {
+		evlist__for_each_cpu(evlist_cpu_itr, evsel_list, affinity) {
 			counter = evlist_cpu_itr.evsel;
 
 			if (!counter->reset_group && !counter->errored)
@@ -904,7 +908,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
 			counter->supported = true;
 		}
 	}
-	affinity__cleanup(&affinity);
+	affinity__cleanup(affinity);
 
 	evlist__for_each_entry(evsel_list, counter) {
 		if (!counter->supported) {

tools/perf/pmu-events/arch/test/test_soc/cpu/uncore.json

@@ -18,6 +18,22 @@
 	    "Invert": "0",
 	    "EdgeDetect": "0"
   },
+  {
+	    "Unit": "CBO",
+	    "EventCode": "0xE0",
+	    "UMask": "0x00",
+	    "EventName": "event-hyphen",
+	    "BriefDescription": "UNC_CBO_HYPHEN",
+	    "PublicDescription": "UNC_CBO_HYPHEN"
+  },
+  {
+	    "Unit": "CBO",
+	    "EventCode": "0xC0",
+	    "UMask": "0x00",
+	    "EventName": "event-two-hyph",
+	    "BriefDescription": "UNC_CBO_TWO_HYPH",
+	    "PublicDescription": "UNC_CBO_TWO_HYPH"
+  },
   {
 	    "EventCode": "0x7",
 	    "EventName": "uncore_hisi_l3c.rd_hit_cpipe",

tools/perf/tests/bitmap.c

@@ -17,8 +17,8 @@ static unsigned long *get_bitmap(const char *str, int nbits)
 	bm = bitmap_zalloc(nbits);
 
 	if (map && bm) {
-		for (i = 0; i < map->nr; i++)
-			set_bit(map->map[i].cpu, bm);
+		for (i = 0; i < perf_cpu_map__nr(map); i++)
+			set_bit(perf_cpu_map__cpu(map, i).cpu, bm);
 	}
 
 	if (map)

tools/perf/tests/event_update.c

@@ -75,10 +75,10 @@ static int process_event_cpus(struct perf_tool *tool __maybe_unused,
 
 	TEST_ASSERT_VAL("wrong id", ev->id == 123);
 	TEST_ASSERT_VAL("wrong type", ev->type == PERF_EVENT_UPDATE__CPUS);
-	TEST_ASSERT_VAL("wrong cpus", map->nr == 3);
-	TEST_ASSERT_VAL("wrong cpus", map->map[0].cpu == 1);
-	TEST_ASSERT_VAL("wrong cpus", map->map[1].cpu == 2);
-	TEST_ASSERT_VAL("wrong cpus", map->map[2].cpu == 3);
+	TEST_ASSERT_VAL("wrong cpus", perf_cpu_map__nr(map) == 3);
+	TEST_ASSERT_VAL("wrong cpus", perf_cpu_map__cpu(map, 0).cpu == 1);
+	TEST_ASSERT_VAL("wrong cpus", perf_cpu_map__cpu(map, 1).cpu == 2);
+	TEST_ASSERT_VAL("wrong cpus", perf_cpu_map__cpu(map, 2).cpu == 3);
 	perf_cpu_map__put(map);
 	return 0;
 }

tools/perf/tests/mem2node.c

@@ -25,14 +25,15 @@ static unsigned long *get_bitmap(const char *str, int nbits)
 {
 	struct perf_cpu_map *map = perf_cpu_map__new(str);
 	unsigned long *bm = NULL;
-	int i;
 
 	bm = bitmap_zalloc(nbits);
 
 	if (map && bm) {
-		for (i = 0; i < map->nr; i++) {
-			set_bit(map->map[i].cpu, bm);
-		}
+		struct perf_cpu cpu;
+		int i;
+
+		perf_cpu_map__for_each_cpu(cpu, i, map)
+			set_bit(cpu.cpu, bm);
 	}
 
 	if (map)

tools/perf/tests/mmap-basic.c

@@ -59,11 +59,12 @@ static int test__basic_mmap(struct test_suite *test __maybe_unused, int subtest
 	}
 
 	CPU_ZERO(&cpu_set);
-	CPU_SET(cpus->map[0].cpu, &cpu_set);
+	CPU_SET(perf_cpu_map__cpu(cpus, 0).cpu, &cpu_set);
 	sched_setaffinity(0, sizeof(cpu_set), &cpu_set);
 	if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0) {
 		pr_debug("sched_setaffinity() failed on CPU %d: %s ",
-			 cpus->map[0].cpu, str_error_r(errno, sbuf, sizeof(sbuf)));
+			 perf_cpu_map__cpu(cpus, 0).cpu,
+			 str_error_r(errno, sbuf, sizeof(sbuf)));
 		goto out_free_cpus;
 	}
 

tools/perf/tests/parse-events.c

@@ -2069,6 +2069,31 @@ static int test_event(struct evlist_test *e)
 	return ret;
 }
 
+static int test_event_fake_pmu(const char *str)
+{
+	struct parse_events_error err;
+	struct evlist *evlist;
+	int ret;
+
+	evlist = evlist__new();
+	if (!evlist)
+		return -ENOMEM;
+
+	parse_events_error__init(&err);
+	perf_pmu__test_parse_init();
+	ret = __parse_events(evlist, str, &err, &perf_pmu__fake);
+	if (ret) {
+		pr_debug("failed to parse event '%s', err %d, str '%s'\n",
+			 str, ret, err.str);
+		parse_events_error__print(&err, str);
+	}
+
+	parse_events_error__exit(&err);
+	evlist__delete(evlist);
+
+	return ret;
+}
+
 static int test_events(struct evlist_test *events, unsigned cnt)
 {
 	int ret1, ret2 = 0;
@@ -2276,6 +2301,26 @@ static int test_pmu_events_alias(char *event, char *alias)
 	return test_event(&e);
 }
 
+static int test_pmu_events_alias2(void)
+{
+	static const char events[][30] = {
+			"event-hyphen",
+			"event-two-hyph",
+	};
+	unsigned long i;
+	int ret = 0;
+
+	for (i = 0; i < ARRAY_SIZE(events); i++) {
+		ret = test_event_fake_pmu(&events[i][0]);
+		if (ret) {
+			pr_err("check_parse_fake %s failed\n", &events[i][0]);
+			break;
+		}
+	}
+
+	return ret;
+}
+
 static int test__parse_events(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
 {
 	int ret1, ret2 = 0;
@@ -2313,6 +2358,10 @@ do {							\
 			return ret;
 	}
 
+	ret1 = test_pmu_events_alias2();
+	if (!ret2)
+		ret2 = ret1;
+
 	ret1 = test_terms(test__terms, ARRAY_SIZE(test__terms));
 	if (!ret2)
 		ret2 = ret1;

tools/perf/tests/pmu-events.c

@@ -143,6 +143,34 @@ static const struct perf_pmu_test_event unc_cbo_xsnp_response_miss_eviction = {
 	.matching_pmu = "uncore_cbox_0",
 };
 
+static const struct perf_pmu_test_event uncore_hyphen = {
+	.event = {
+		.name = "event-hyphen",
+		.event = "umask=0x00,event=0xe0",
+		.desc = "Unit: uncore_cbox UNC_CBO_HYPHEN",
+		.topic = "uncore",
+		.long_desc = "UNC_CBO_HYPHEN",
+		.pmu = "uncore_cbox",
+	},
+	.alias_str = "umask=0,event=0xe0",
+	.alias_long_desc = "UNC_CBO_HYPHEN",
+	.matching_pmu = "uncore_cbox_0",
+};
+
+static const struct perf_pmu_test_event uncore_two_hyph = {
+	.event = {
+		.name = "event-two-hyph",
+		.event = "umask=0x00,event=0xc0",
+		.desc = "Unit: uncore_cbox UNC_CBO_TWO_HYPH",
+		.topic = "uncore",
+		.long_desc = "UNC_CBO_TWO_HYPH",
+		.pmu = "uncore_cbox",
+	},
+	.alias_str = "umask=0,event=0xc0",
+	.alias_long_desc = "UNC_CBO_TWO_HYPH",
+	.matching_pmu = "uncore_cbox_0",
+};
+
 static const struct perf_pmu_test_event uncore_hisi_l3c_rd_hit_cpipe = {
 	.event = {
 		.name = "uncore_hisi_l3c.rd_hit_cpipe",
@@ -188,6 +216,8 @@ static const struct perf_pmu_test_event uncore_imc_cache_hits = {
 static const struct perf_pmu_test_event *uncore_events[] = {
 	&uncore_hisi_ddrc_flux_wcmd,
 	&unc_cbo_xsnp_response_miss_eviction,
+	&uncore_hyphen,
+	&uncore_two_hyph,
 	&uncore_hisi_l3c_rd_hit_cpipe,
 	&uncore_imc_free_running_cache_miss,
 	&uncore_imc_cache_hits,
@@ -654,6 +684,8 @@ static struct perf_pmu_test_pmu test_pmus[] = {
 		},
 		.aliases = {
 			&unc_cbo_xsnp_response_miss_eviction,
+			&uncore_hyphen,
+			&uncore_two_hyph,
 		},
 	},
 	{

tools/perf/tests/topology.c

@@ -122,44 +122,48 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 	}
 
 	// Test that CPU ID contains socket, die, core and CPU
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		id = aggr_cpu_id__cpu(perf_cpu_map__cpu(map, i), NULL);
-		TEST_ASSERT_VAL("Cpu map - CPU ID doesn't match", map->map[i].cpu == id.cpu.cpu);
+		TEST_ASSERT_VAL("Cpu map - CPU ID doesn't match",
+				perf_cpu_map__cpu(map, i).cpu == id.cpu.cpu);
 
 		TEST_ASSERT_VAL("Cpu map - Core ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].core_id == id.core);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].core_id == id.core);
 		TEST_ASSERT_VAL("Cpu map - Socket ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].socket_id == id.socket);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id ==
+			id.socket);
 
 		TEST_ASSERT_VAL("Cpu map - Die ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].die_id == id.die);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die);
 		TEST_ASSERT_VAL("Cpu map - Node ID is set", id.node == -1);
 		TEST_ASSERT_VAL("Cpu map - Thread is set", id.thread == -1);
 	}
 
 	// Test that core ID contains socket, die and core
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		id = aggr_cpu_id__core(perf_cpu_map__cpu(map, i), NULL);
 		TEST_ASSERT_VAL("Core map - Core ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].core_id == id.core);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].core_id == id.core);
 
 		TEST_ASSERT_VAL("Core map - Socket ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].socket_id == id.socket);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id ==
+			id.socket);
 
 		TEST_ASSERT_VAL("Core map - Die ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].die_id == id.die);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die);
 		TEST_ASSERT_VAL("Core map - Node ID is set", id.node == -1);
 		TEST_ASSERT_VAL("Core map - Thread is set", id.thread == -1);
 	}
 
 	// Test that die ID contains socket and die
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		id = aggr_cpu_id__die(perf_cpu_map__cpu(map, i), NULL);
 		TEST_ASSERT_VAL("Die map - Socket ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].socket_id == id.socket);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id ==
+			id.socket);
 
 		TEST_ASSERT_VAL("Die map - Die ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].die_id == id.die);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].die_id == id.die);
 
 		TEST_ASSERT_VAL("Die map - Node ID is set", id.node == -1);
 		TEST_ASSERT_VAL("Die map - Core is set", id.core == -1);
@@ -168,10 +172,11 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 	}
 
 	// Test that socket ID contains only socket
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		id = aggr_cpu_id__socket(perf_cpu_map__cpu(map, i), NULL);
 		TEST_ASSERT_VAL("Socket map - Socket ID doesn't match",
-			session->header.env.cpu[map->map[i].cpu].socket_id == id.socket);
+			session->header.env.cpu[perf_cpu_map__cpu(map, i).cpu].socket_id ==
+			id.socket);
 
 		TEST_ASSERT_VAL("Socket map - Node ID is set", id.node == -1);
 		TEST_ASSERT_VAL("Socket map - Die ID is set", id.die == -1);
@@ -181,10 +186,10 @@ static int check_cpu_topology(char *path, struct perf_cpu_map *map)
 	}
 
 	// Test that node ID contains only node
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		id = aggr_cpu_id__node(perf_cpu_map__cpu(map, i), NULL);
 		TEST_ASSERT_VAL("Node map - Node ID doesn't match",
-			cpu__get_node(map->map[i]) == id.node);
+				cpu__get_node(perf_cpu_map__cpu(map, i)) == id.node);
 		TEST_ASSERT_VAL("Node map - Socket is set", id.socket == -1);
 		TEST_ASSERT_VAL("Node map - Die ID is set", id.die == -1);
 		TEST_ASSERT_VAL("Node map - Core is set", id.core == -1);

tools/perf/util/affinity.c

@@ -62,7 +62,7 @@ void affinity__set(struct affinity *a, int cpu)
 	clear_bit(cpu, a->sched_cpus);
 }
 
-void affinity__cleanup(struct affinity *a)
+static void __affinity__cleanup(struct affinity *a)
 {
 	int cpu_set_size = get_cpu_set_size();
 
@@ -71,3 +71,9 @@ void affinity__cleanup(struct affinity *a)
 	zfree(&a->sched_cpus);
 	zfree(&a->orig_cpus);
 }
+
+void affinity__cleanup(struct affinity *a)
+{
+	if (a != NULL)
+		__affinity__cleanup(a);
+}

tools/perf/util/auxtrace.c

@@ -174,7 +174,7 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp,
 	mp->idx = idx;
 
 	if (per_cpu) {
-		mp->cpu = evlist->core.cpus->map[idx];
+		mp->cpu = perf_cpu_map__cpu(evlist->core.cpus, idx);
 		if (evlist->core.threads)
 			mp->tid = perf_thread_map__pid(evlist->core.threads, 0);
 		else

tools/perf/util/counts.c

@@ -61,7 +61,7 @@ int evsel__alloc_counts(struct evsel *evsel)
 	struct perf_cpu_map *cpus = evsel__cpus(evsel);
 	int nthreads = perf_thread_map__nr(evsel->core.threads);
 
-	evsel->counts = perf_counts__new(cpus ? cpus->nr : 1, nthreads);
+	evsel->counts = perf_counts__new(perf_cpu_map__nr(cpus), nthreads);
 	return evsel->counts != NULL ? 0 : -ENOMEM;
 }
 

tools/perf/util/cpumap.h

@@ -4,7 +4,6 @@
 
 #include <stdbool.h>
 #include <stdio.h>
-#include <stdbool.h>
 #include <internal/cpumap.h>
 #include <perf/cpumap.h>
 
@@ -57,7 +56,7 @@ struct perf_cpu cpu__max_present_cpu(void);
  */
 static inline bool cpu_map__is_dummy(struct perf_cpu_map *cpus)
 {
-	return cpus->nr == 1 && cpus->map[0].cpu == -1;
+	return perf_cpu_map__nr(cpus) == 1 && perf_cpu_map__cpu(cpus, 0).cpu == -1;
 }
 
 /**

tools/perf/util/cputopo.c

@@ -325,7 +325,7 @@ struct numa_topology *numa_topology__new(void)
 	if (!node_map)
 		goto out;
 
-	nr = (u32) node_map->nr;
+	nr = (u32) perf_cpu_map__nr(node_map);
 
 	tp = zalloc(sizeof(*tp) + sizeof(tp->nodes[0])*nr);
 	if (!tp)
@@ -334,7 +334,7 @@ struct numa_topology *numa_topology__new(void)
 	tp->nr = nr;
 
 	for (i = 0; i < nr; i++) {
-		if (load_numa_node(&tp->nodes[i], node_map->map[i].cpu)) {
+		if (load_numa_node(&tp->nodes[i], perf_cpu_map__cpu(node_map, i).cpu)) {
 			numa_topology__delete(tp);
 			tp = NULL;
 			break;

tools/perf/util/evlist-hybrid.c

@@ -124,22 +124,23 @@ int evlist__fix_hybrid_cpus(struct evlist *evlist, const char *cpu_list)
 
 		events_nr++;
 
-		if (matched_cpus->nr > 0 && (unmatched_cpus->nr > 0 ||
-		    matched_cpus->nr < cpus->nr ||
-		    matched_cpus->nr < pmu->cpus->nr)) {
+		if (perf_cpu_map__nr(matched_cpus) > 0 &&
+		    (perf_cpu_map__nr(unmatched_cpus) > 0 ||
+		     perf_cpu_map__nr(matched_cpus) < perf_cpu_map__nr(cpus) ||
+		     perf_cpu_map__nr(matched_cpus) < perf_cpu_map__nr(pmu->cpus))) {
 			perf_cpu_map__put(evsel->core.cpus);
 			perf_cpu_map__put(evsel->core.own_cpus);
 			evsel->core.cpus = perf_cpu_map__get(matched_cpus);
 			evsel->core.own_cpus = perf_cpu_map__get(matched_cpus);
 
-			if (unmatched_cpus->nr > 0) {
+			if (perf_cpu_map__nr(unmatched_cpus) > 0) {
 				cpu_map__snprint(matched_cpus, buf1, sizeof(buf1));
 				pr_warning("WARNING: use %s in '%s' for '%s', skip other cpus in list.\n",
 					   buf1, pmu->name, evsel->name);
 			}
 		}
 
-		if (matched_cpus->nr == 0) {
+		if (perf_cpu_map__nr(matched_cpus) == 0) {
 			evlist__remove(evlist, evsel);
 			evsel__delete(evsel);
 

tools/perf/util/evlist.c

@@ -430,15 +430,19 @@ static void __evlist__disable(struct evlist *evlist, char *evsel_name)
 {
 	struct evsel *pos;
 	struct evlist_cpu_iterator evlist_cpu_itr;
-	struct affinity affinity;
+	struct affinity saved_affinity, *affinity = NULL;
 	bool has_imm = false;
 
-	if (affinity__setup(&affinity) < 0)
+	// See explanation in evlist__close()
+	if (!cpu_map__is_dummy(evlist->core.cpus)) {
+		if (affinity__setup(&saved_affinity) < 0)
 			return;
+		affinity = &saved_affinity;
+	}
 
 	/* Disable 'immediate' events last */
 	for (int imm = 0; imm <= 1; imm++) {
-		evlist__for_each_cpu(evlist_cpu_itr, evlist, &affinity) {
+		evlist__for_each_cpu(evlist_cpu_itr, evlist, affinity) {
 			pos = evlist_cpu_itr.evsel;
 			if (evsel__strcmp(pos, evsel_name))
 				continue;
@@ -454,7 +458,7 @@ static void __evlist__disable(struct evlist *evlist, char *evsel_name)
 			break;
 	}
 
-	affinity__cleanup(&affinity);
+	affinity__cleanup(affinity);
 	evlist__for_each_entry(evlist, pos) {
 		if (evsel__strcmp(pos, evsel_name))
 			continue;
@@ -487,12 +491,16 @@ static void __evlist__enable(struct evlist *evlist, char *evsel_name)
 {
 	struct evsel *pos;
 	struct evlist_cpu_iterator evlist_cpu_itr;
-	struct affinity affinity;
+	struct affinity saved_affinity, *affinity = NULL;
 
-	if (affinity__setup(&affinity) < 0)
+	// See explanation in evlist__close()
+	if (!cpu_map__is_dummy(evlist->core.cpus)) {
+		if (affinity__setup(&saved_affinity) < 0)
 			return;
+		affinity = &saved_affinity;
+	}
 
-	evlist__for_each_cpu(evlist_cpu_itr, evlist, &affinity) {
+	evlist__for_each_cpu(evlist_cpu_itr, evlist, affinity) {
 		pos = evlist_cpu_itr.evsel;
 		if (evsel__strcmp(pos, evsel_name))
 			continue;
@@ -500,7 +508,7 @@ static void __evlist__enable(struct evlist *evlist, char *evsel_name)
 			continue;
 		evsel__enable_cpu(pos, evlist_cpu_itr.cpu_map_idx);
 	}
-	affinity__cleanup(&affinity);
+	affinity__cleanup(affinity);
 	evlist__for_each_entry(evlist, pos) {
 		if (evsel__strcmp(pos, evsel_name))
 			continue;

tools/perf/util/evsel.c

@@ -1064,6 +1064,17 @@ void __weak arch_evsel__fixup_new_cycles(struct perf_event_attr *attr __maybe_un
 {
 }
 
+static void evsel__set_default_freq_period(struct record_opts *opts,
+					   struct perf_event_attr *attr)
+{
+	if (opts->freq) {
+		attr->freq = 1;
+		attr->sample_freq = opts->freq;
+	} else {
+		attr->sample_period = opts->default_interval;
+	}
+}
+
 /*
  * The enable_on_exec/disabled value strategy:
  *
@@ -1130,14 +1141,12 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts,
 	 * We default some events to have a default interval. But keep
 	 * it a weak assumption overridable by the user.
 	 */
-	if (!attr->sample_period) {
-		if (opts->freq) {
-			attr->freq		= 1;
-			attr->sample_freq	= opts->freq;
-		} else {
-			attr->sample_period = opts->default_interval;
-		}
-	}
+	if ((evsel->is_libpfm_event && !attr->sample_period) ||
+	    (!evsel->is_libpfm_event && (!attr->sample_period ||
+					 opts->user_freq != UINT_MAX ||
+					 opts->user_interval != ULLONG_MAX)))
+		evsel__set_default_freq_period(opts, attr);
+
 	/*
 	 * If attr->freq was set (here or earlier), ask for period
 	 * to be sampled.
@@ -1782,7 +1791,7 @@ static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus,
 		nthreads = threads->nr;
 
 	if (evsel->core.fd == NULL &&
-	    perf_evsel__alloc_fd(&evsel->core, cpus->nr, nthreads) < 0)
+	    perf_evsel__alloc_fd(&evsel->core, perf_cpu_map__nr(cpus), nthreads) < 0)
 		return -ENOMEM;
 
 	evsel->open_flags = PERF_FLAG_FD_CLOEXEC;
@@ -2020,9 +2029,10 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus,
 			test_attr__ready();
 
 			pr_debug2_peo("sys_perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx",
-				pid, cpus->map[idx].cpu, group_fd, evsel->open_flags);
+				pid, perf_cpu_map__cpu(cpus, idx).cpu, group_fd, evsel->open_flags);
 
-			fd = sys_perf_event_open(&evsel->core.attr, pid, cpus->map[idx].cpu,
+			fd = sys_perf_event_open(&evsel->core.attr, pid,
+						perf_cpu_map__cpu(cpus, idx).cpu,
 						group_fd, evsel->open_flags);
 
 			FD(evsel, idx, thread) = fd;
@@ -2038,7 +2048,8 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus,
 			bpf_counter__install_pe(evsel, idx, fd);
 
 			if (unlikely(test_attr__enabled)) {
-				test_attr__open(&evsel->core.attr, pid, cpus->map[idx],
+				test_attr__open(&evsel->core.attr, pid,
+						perf_cpu_map__cpu(cpus, idx),
 						fd, group_fd, evsel->open_flags);
 			}
 
@@ -2079,7 +2090,8 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus,
 	if (evsel__precise_ip_fallback(evsel))
 		goto retry_open;
 
-	if (evsel__ignore_missing_thread(evsel, cpus->nr, idx, threads, thread, err)) {
+	if (evsel__ignore_missing_thread(evsel, perf_cpu_map__nr(cpus),
+					 idx, threads, thread, err)) {
 		/* We just removed 1 thread, so lower the upper nthreads limit. */
 		nthreads--;
 
@@ -2119,7 +2131,7 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus,
 int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus,
 		struct perf_thread_map *threads)
 {
-	return evsel__open_cpu(evsel, cpus, threads, 0, cpus ? cpus->nr : 1);
+	return evsel__open_cpu(evsel, cpus, threads, 0, perf_cpu_map__nr(cpus));
 }
 
 void evsel__close(struct evsel *evsel)
@@ -2131,8 +2143,7 @@ void evsel__close(struct evsel *evsel)
 int evsel__open_per_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, int cpu_map_idx)
 {
 	if (cpu_map_idx == -1)
-		return evsel__open_cpu(evsel, cpus, NULL, 0,
-					cpus ? cpus->nr : 1);
+		return evsel__open_cpu(evsel, cpus, NULL, 0, perf_cpu_map__nr(cpus));
 
 	return evsel__open_cpu(evsel, cpus, NULL, cpu_map_idx, cpu_map_idx + 1);
 }
@@ -2982,7 +2993,7 @@ int evsel__store_ids(struct evsel *evsel, struct evlist *evlist)
 	struct perf_cpu_map *cpus = evsel->core.cpus;
 	struct perf_thread_map *threads = evsel->core.threads;
 
-	if (perf_evsel__alloc_id(&evsel->core, cpus->nr, threads->nr))
+	if (perf_evsel__alloc_id(&evsel->core, perf_cpu_map__nr(cpus), threads->nr))
 		return -ENOMEM;
 
 	return store_evsel_ids(evsel, evlist);

tools/perf/util/evsel.h

@@ -11,6 +11,7 @@
 #include <perf/evsel.h>
 #include "symbol_conf.h"
 #include <internal/cpumap.h>
+#include <perf/cpumap.h>
 
 struct bpf_object;
 struct cgroup;
@@ -191,7 +192,7 @@ static inline struct perf_cpu_map *evsel__cpus(struct evsel *evsel)
 
 static inline int evsel__nr_cpus(struct evsel *evsel)
 {
-	return evsel__cpus(evsel)->nr;
+	return perf_cpu_map__nr(evsel__cpus(evsel));
 }
 
 void evsel__compute_deltas(struct evsel *evsel, int cpu, int thread,

tools/perf/util/machine.c

@@ -16,6 +16,7 @@
 #include "map_symbol.h"
 #include "branch.h"
 #include "mem-events.h"
+#include "path.h"
 #include "srcline.h"
 #include "symbol.h"
 #include "sort.h"
@@ -1416,7 +1417,7 @@ static int maps__set_modules_path_dir(struct maps *maps, const char *dir_name, i
 		struct stat st;
 
 		/*sshfs might return bad dent->d_type, so we have to stat*/
-		snprintf(path, sizeof(path), "%s/%s", dir_name, dent->d_name);
+		path__join(path, sizeof(path), dir_name, dent->d_name);
 		if (stat(path, &st))
 			continue;
 

tools/perf/util/mmap.c

@@ -250,7 +250,7 @@ static void build_node_mask(int node, struct mmap_cpu_mask *mask)
 
 	nr_cpus = perf_cpu_map__nr(cpu_map);
 	for (idx = 0; idx < nr_cpus; idx++) {
-		cpu = cpu_map->map[idx]; /* map c index to online cpu index */
+		cpu = perf_cpu_map__cpu(cpu_map, idx); /* map c index to online cpu index */
 		if (cpu__get_node(cpu) == node)
 			set_bit(cpu.cpu, mask->bits);
 	}

tools/perf/util/parse-events.c

@@ -1697,6 +1697,15 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state,
 			}
 		}
 	}
+
+	if (parse_state->fake_pmu) {
+		if (!parse_events_add_pmu(parse_state, list, str, head,
+					  true, true)) {
+			pr_debug("%s -> %s/%s/\n", str, "fake_pmu", str);
+			ok++;
+		}
+	}
+
 out_err:
 	if (ok)
 		*listp = list;
@@ -2098,8 +2107,17 @@ static void perf_pmu__parse_init(void)
 	pmu = NULL;
 	while ((pmu = perf_pmu__scan(pmu)) != NULL) {
 		list_for_each_entry(alias, &pmu->aliases, list) {
-			if (strchr(alias->name, '-'))
+			char *tmp = strchr(alias->name, '-');
+
+			if (tmp) {
+				char *tmp2 = NULL;
+
+				tmp2 = strchr(tmp + 1, '-');
+				len++;
+				if (tmp2)
 					len++;
+			}
+
 			len++;
 		}
 	}
@@ -2119,8 +2137,20 @@ static void perf_pmu__parse_init(void)
 		list_for_each_entry(alias, &pmu->aliases, list) {
 			struct perf_pmu_event_symbol *p = perf_pmu_events_list + len;
 			char *tmp = strchr(alias->name, '-');
+			char *tmp2 = NULL;
 
-			if (tmp != NULL) {
+			if (tmp)
+				tmp2 = strchr(tmp + 1, '-');
+			if (tmp2) {
+				SET_SYMBOL(strndup(alias->name, tmp - alias->name),
+						PMU_EVENT_SYMBOL_PREFIX);
+				p++;
+				tmp++;
+				SET_SYMBOL(strndup(tmp, tmp2 - tmp), PMU_EVENT_SYMBOL_SUFFIX);
+				p++;
+				SET_SYMBOL(strdup(++tmp2), PMU_EVENT_SYMBOL_SUFFIX2);
+				len += 3;
+			} else if (tmp) {
 				SET_SYMBOL(strndup(alias->name, tmp - alias->name),
 						PMU_EVENT_SYMBOL_PREFIX);
 				p++;
@@ -2147,23 +2177,38 @@ static void perf_pmu__parse_init(void)
  */
 int perf_pmu__test_parse_init(void)
 {
-	struct perf_pmu_event_symbol *list;
+	struct perf_pmu_event_symbol *list, *tmp, symbols[] = {
+		{(char *)"read", PMU_EVENT_SYMBOL},
+		{(char *)"event", PMU_EVENT_SYMBOL_PREFIX},
+		{(char *)"two", PMU_EVENT_SYMBOL_SUFFIX},
+		{(char *)"hyphen", PMU_EVENT_SYMBOL_SUFFIX},
+		{(char *)"hyph", PMU_EVENT_SYMBOL_SUFFIX2},
+	};
+	unsigned long i, j;
 
-	list = malloc(sizeof(*list) * 1);
+	tmp = list = malloc(sizeof(*list) * ARRAY_SIZE(symbols));
 	if (!list)
 		return -ENOMEM;
 
-	list->type   = PMU_EVENT_SYMBOL;
-	list->symbol = strdup("read");
-
-	if (!list->symbol) {
-		free(list);
-		return -ENOMEM;
+	for (i = 0; i < ARRAY_SIZE(symbols); i++, tmp++) {
+		tmp->type = symbols[i].type;
+		tmp->symbol = strdup(symbols[i].symbol);
+		if (!list->symbol)
+			goto err_free;
 	}
 
 	perf_pmu_events_list = list;
-	perf_pmu_events_list_num = 1;
+	perf_pmu_events_list_num = ARRAY_SIZE(symbols);
+
+	qsort(perf_pmu_events_list, ARRAY_SIZE(symbols),
+	      sizeof(struct perf_pmu_event_symbol), comp_pmu);
 	return 0;
+
+err_free:
+	for (j = 0, tmp = list; j < i; j++, tmp++)
+		free(tmp->symbol);
+	free(list);
+	return -ENOMEM;
 }
 
 enum perf_pmu_event_symbol_type

tools/perf/util/parse-events.h

@@ -53,6 +53,7 @@ enum perf_pmu_event_symbol_type {
 	PMU_EVENT_SYMBOL,		/* normal style PMU event */
 	PMU_EVENT_SYMBOL_PREFIX,	/* prefix of pre-suf style event */
 	PMU_EVENT_SYMBOL_SUFFIX,	/* suffix of pre-suf style event */
+	PMU_EVENT_SYMBOL_SUFFIX2,	/* suffix of pre-suf2 style event */
 };
 
 struct perf_pmu_event_symbol {

tools/perf/util/parse-events.l

@@ -149,6 +149,8 @@ static int pmu_str_check(yyscan_t scanner, struct parse_events_state *parse_stat
 			return PE_PMU_EVENT_PRE;
 		case PMU_EVENT_SYMBOL_SUFFIX:
 			return PE_PMU_EVENT_SUF;
+		case PMU_EVENT_SYMBOL_SUFFIX2:
+			return PE_PMU_EVENT_SUF2;
 		case PMU_EVENT_SYMBOL:
 			return parse_state->fake_pmu
 				? PE_PMU_EVENT_FAKE : PE_KERNEL_PMU_EVENT;

tools/perf/util/parse-events.y

@@ -69,7 +69,7 @@ static void inc_group_count(struct list_head *list,
 %token PE_NAME_CACHE_TYPE PE_NAME_CACHE_OP_RESULT
 %token PE_PREFIX_MEM PE_PREFIX_RAW PE_PREFIX_GROUP
 %token PE_ERROR
-%token PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_KERNEL_PMU_EVENT PE_PMU_EVENT_FAKE
+%token PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_PMU_EVENT_SUF2 PE_KERNEL_PMU_EVENT PE_PMU_EVENT_FAKE
 %token PE_ARRAY_ALL PE_ARRAY_RANGE
 %token PE_DRV_CFG_TERM
 %type <num> PE_VALUE
@@ -87,7 +87,7 @@ static void inc_group_count(struct list_head *list,
 %type <str> PE_MODIFIER_EVENT
 %type <str> PE_MODIFIER_BP
 %type <str> PE_EVENT_NAME
-%type <str> PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_KERNEL_PMU_EVENT PE_PMU_EVENT_FAKE
+%type <str> PE_PMU_EVENT_PRE PE_PMU_EVENT_SUF PE_PMU_EVENT_SUF2 PE_KERNEL_PMU_EVENT PE_PMU_EVENT_FAKE
 %type <str> PE_DRV_CFG_TERM
 %type <str> event_pmu_name
 %destructor { free ($$); } <str>
@@ -372,6 +372,19 @@ PE_KERNEL_PMU_EVENT opt_pmu_config
 	$$ = list;
 }
 |
+PE_PMU_EVENT_PRE '-' PE_PMU_EVENT_SUF '-' PE_PMU_EVENT_SUF2 sep_dc
+{
+	struct list_head *list;
+	char pmu_name[128];
+	snprintf(pmu_name, sizeof(pmu_name), "%s-%s-%s", $1, $3, $5);
+	free($1);
+	free($3);
+	free($5);
+	if (parse_events_multi_pmu_add(_parse_state, pmu_name, NULL, &list) < 0)
+		YYABORT;
+	$$ = list;
+}
+|
 PE_PMU_EVENT_PRE '-' PE_PMU_EVENT_SUF sep_dc
 {
 	struct list_head *list;

tools/perf/util/perf_api_probe.c

@@ -67,7 +67,7 @@ static bool perf_probe_api(setup_probe_fn_t fn)
 	cpus = perf_cpu_map__new(NULL);
 	if (!cpus)
 		return false;
-	cpu = cpus->map[0];
+	cpu = perf_cpu_map__cpu(cpus, 0);
 	perf_cpu_map__put(cpus);
 
 	do {
@@ -144,7 +144,7 @@ bool perf_can_record_cpu_wide(void)
 	if (!cpus)
 		return false;
 
-	cpu = cpus->map[0];
+	cpu = perf_cpu_map__cpu(cpus, 0);
 	perf_cpu_map__put(cpus);
 
 	fd = sys_perf_event_open(&attr, -1, cpu.cpu, -1, 0);

tools/perf/util/probe-event.c

@@ -3083,6 +3083,9 @@ static int find_probe_trace_events_from_map(struct perf_probe_event *pev,
 	for (j = 0; j < num_matched_functions; j++) {
 		sym = syms[j];
 
+		if (sym->type != STT_FUNC)
+			continue;
+
 		/* There can be duplicated symbols in the map */
 		for (i = 0; i < j; i++)
 			if (sym->start == syms[i]->start) {

tools/perf/util/python.c

@@ -638,17 +638,17 @@ static Py_ssize_t pyrf_cpu_map__length(PyObject *obj)
 {
 	struct pyrf_cpu_map *pcpus = (void *)obj;
 
-	return pcpus->cpus->nr;
+	return perf_cpu_map__nr(pcpus->cpus);
 }
 
 static PyObject *pyrf_cpu_map__item(PyObject *obj, Py_ssize_t i)
 {
 	struct pyrf_cpu_map *pcpus = (void *)obj;
 
-	if (i >= pcpus->cpus->nr)
+	if (i >= perf_cpu_map__nr(pcpus->cpus))
 		return NULL;
 
-	return Py_BuildValue("i", pcpus->cpus->map[i]);
+	return Py_BuildValue("i", perf_cpu_map__cpu(pcpus->cpus, i).cpu);
 }
 
 static PySequenceMethods pyrf_cpu_map__sequence_methods = {

tools/perf/util/record.c

@@ -106,7 +106,7 @@ void evlist__config(struct evlist *evlist, struct record_opts *opts, struct call
 	if (opts->group)
 		evlist__set_leader(evlist);
 
-	if (evlist->core.cpus->map[0].cpu < 0)
+	if (perf_cpu_map__cpu(evlist->core.cpus, 0).cpu < 0)
 		opts->no_inherit = true;
 
 	use_comm_exec = perf_can_comm_exec();
@@ -248,11 +248,11 @@ bool evlist__can_select_event(struct evlist *evlist, const char *str)
 		struct perf_cpu_map *cpus = perf_cpu_map__new(NULL);
 
 		if (cpus)
-			cpu =  cpus->map[0];
+			cpu =  perf_cpu_map__cpu(cpus, 0);
 
 		perf_cpu_map__put(cpus);
 	} else {
-		cpu = evlist->core.cpus->map[0];
+		cpu = perf_cpu_map__cpu(evlist->core.cpus, 0);
 	}
 
 	while (1) {

tools/perf/util/scripting-engines/trace-event-python.c

@@ -1607,8 +1607,8 @@ static void python_process_stat(struct perf_stat_config *config,
 	}
 
 	for (thread = 0; thread < threads->nr; thread++) {
-		for (cpu = 0; cpu < cpus->nr; cpu++) {
-			process_stat(counter, cpus->map[cpu],
+		for (cpu = 0; cpu < perf_cpu_map__nr(cpus); cpu++) {
+			process_stat(counter, perf_cpu_map__cpu(cpus, cpu),
 				     perf_thread_map__pid(threads, thread), tstamp,
 				     perf_counts(counter->counts, cpu, thread));
 		}

tools/perf/util/session.c

@@ -2537,8 +2537,8 @@ int perf_session__cpu_bitmap(struct perf_session *session,
 		return -1;
 	}
 
-	for (i = 0; i < map->nr; i++) {
-		struct perf_cpu cpu = map->map[i];
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
+		struct perf_cpu cpu = perf_cpu_map__cpu(map, i);
 
 		if (cpu.cpu >= nr_cpus) {
 			pr_err("Requested CPU %d too large. "

tools/perf/util/svghelper.c

@@ -734,8 +734,8 @@ static int str_to_bitmap(char *s, cpumask_t *b, int nr_cpus)
 	if (!m)
 		return -1;
 
-	for (i = 0; i < m->nr; i++) {
-		c = m->map[i];
+	for (i = 0; i < perf_cpu_map__nr(m); i++) {
+		c = perf_cpu_map__cpu(m, i);
 		if (c.cpu >= nr_cpus) {
 			ret = -1;
 			break;

tools/perf/util/synthetic-events.c

@@ -1186,12 +1186,12 @@ int perf_event__synthesize_thread_map2(struct perf_tool *tool,
 static void synthesize_cpus(struct cpu_map_entries *cpus,
 			    struct perf_cpu_map *map)
 {
-	int i;
+	int i, map_nr = perf_cpu_map__nr(map);
 
-	cpus->nr = map->nr;
+	cpus->nr = map_nr;
 
-	for (i = 0; i < map->nr; i++)
-		cpus->cpu[i] = map->map[i].cpu;
+	for (i = 0; i < map_nr; i++)
+		cpus->cpu[i] = perf_cpu_map__cpu(map, i).cpu;
 }
 
 static void synthesize_mask(struct perf_record_record_cpu_map *mask,
@@ -1202,13 +1202,13 @@ static void synthesize_mask(struct perf_record_record_cpu_map *mask,
 	mask->nr = BITS_TO_LONGS(max);
 	mask->long_size = sizeof(long);
 
-	for (i = 0; i < map->nr; i++)
-		set_bit(map->map[i].cpu, mask->mask);
+	for (i = 0; i < perf_cpu_map__nr(map); i++)
+		set_bit(perf_cpu_map__cpu(map, i).cpu, mask->mask);
 }
 
 static size_t cpus_size(struct perf_cpu_map *map)
 {
-	return sizeof(struct cpu_map_entries) + map->nr * sizeof(u16);
+	return sizeof(struct cpu_map_entries) + perf_cpu_map__nr(map) * sizeof(u16);
 }
 
 static size_t mask_size(struct perf_cpu_map *map, int *max)
@@ -1217,9 +1217,9 @@ static size_t mask_size(struct perf_cpu_map *map, int *max)
 
 	*max = 0;
 
-	for (i = 0; i < map->nr; i++) {
+	for (i = 0; i < perf_cpu_map__nr(map); i++) {
 		/* bit position of the cpu is + 1 */
-		int bit = map->map[i].cpu + 1;
+		int bit = perf_cpu_map__cpu(map, i).cpu + 1;
 
 		if (bit > *max)
 			*max = bit;

tools/perf/util/top.c

@@ -95,15 +95,15 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size)
 
 	if (target->cpu_list)
 		ret += SNPRINTF(bf + ret, size - ret, ", CPU%s: %s)",
-				top->evlist->core.cpus->nr > 1 ? "s" : "",
+				perf_cpu_map__nr(top->evlist->core.cpus) > 1 ? "s" : "",
 				target->cpu_list);
 	else {
 		if (target->tid)
 			ret += SNPRINTF(bf + ret, size - ret, ")");
 		else
 			ret += SNPRINTF(bf + ret, size - ret, ", %d CPU%s)",
-					top->evlist->core.cpus->nr,
-					top->evlist->core.cpus->nr > 1 ? "s" : "");
+					perf_cpu_map__nr(top->evlist->core.cpus),
+					perf_cpu_map__nr(top->evlist->core.cpus) > 1 ? "s" : "");
 	}
 
 	perf_top__reset_sample_counters(top);