16 files changed, 802 insertions, 122 deletions

diff --git a/tools/lib/perf/Documentation/libperf.txt b/tools/lib/perf/Documentation/libperf.txt
index 59aabdd3cabf..576ecc5fc312 100644
--- a/tools/lib/perf/Documentation/libperf.txt
+++ b/tools/lib/perf/Documentation/libperf.txt
@@ -210,6 +210,9 @@ SYNOPSIS
   struct perf_record_time_conv;
   struct perf_record_header_feature;
   struct perf_record_compressed;
+  struct perf_record_compressed2;
+  struct perf_record_schedstat_cpu;
+  struct perf_record_schedstat_domain;
 --
 
 DESCRIPTION
diff --git a/tools/lib/perf/Makefile b/tools/lib/perf/Makefile
index 3a9b2140aa04..32301a1d8f0c 100644
--- a/tools/lib/perf/Makefile
+++ b/tools/lib/perf/Makefile
@@ -39,28 +39,8 @@ libdir = $(prefix)/$(libdir_relative)
 libdir_SQ = $(subst ','\'',$(libdir))
 libdir_relative_SQ = $(subst ','\'',$(libdir_relative))
 
-ifeq ("$(origin V)", "command line")
-  VERBOSE = $(V)
-endif
-ifndef VERBOSE
-  VERBOSE = 0
-endif
-
-ifeq ($(VERBOSE),1)
-  Q =
-else
-  Q = @
-endif
-
 TEST_ARGS := $(if $(V),-v)
 
-# Set compile option CFLAGS
-ifdef EXTRA_CFLAGS
-  CFLAGS := $(EXTRA_CFLAGS)
-else
-  CFLAGS := -g -Wall
-endif
-
 INCLUDES = \
 -I$(srctree)/tools/lib/perf/include \
 -I$(srctree)/tools/lib/ \
@@ -70,11 +50,12 @@ INCLUDES = \
 -I$(srctree)/tools/include/uapi
 
 # Append required CFLAGS
-override CFLAGS += $(EXTRA_WARNINGS)
-override CFLAGS += -Werror -Wall
+override CFLAGS := $(INCLUDES) $(CFLAGS)
+override CFLAGS += -g -Werror -Wall
 override CFLAGS += -fPIC
-override CFLAGS += $(INCLUDES)
 override CFLAGS += -fvisibility=hidden
+override CFLAGS += $(EXTRA_WARNINGS)
+override CFLAGS += $(EXTRA_CFLAGS)
 
 all:
 
@@ -188,6 +169,7 @@ install_lib: libs
 		cp -fpR $(LIBPERF_ALL) $(DESTDIR)$(libdir_SQ)
 
 HDRS := bpf_perf.h core.h cpumap.h threadmap.h evlist.h evsel.h event.h mmap.h
+HDRS += schedstat-v15.h schedstat-v16.h schedstat-v17.h
 INTERNAL_HDRS := cpumap.h evlist.h evsel.h lib.h mmap.h rc_check.h threadmap.h xyarray.h
 
 INSTALL_HDRS_PFX := $(DESTDIR)$(prefix)/include/perf
diff --git a/tools/lib/perf/cpumap.c b/tools/lib/perf/cpumap.c
index fcc47214062a..e51b0490ad57 100644
--- a/tools/lib/perf/cpumap.c
+++ b/tools/lib/perf/cpumap.c
@@ -15,12 +15,12 @@
 
 #define MAX_NR_CPUS 4096
 
-void perf_cpu_map__set_nr(struct perf_cpu_map *map, int nr_cpus)
+void perf_cpu_map__set_nr(struct perf_cpu_map *map, unsigned int nr_cpus)
 {
 	RC_CHK_ACCESS(map)->nr = nr_cpus;
 }
 
-struct perf_cpu_map *perf_cpu_map__alloc(int nr_cpus)
+struct perf_cpu_map *perf_cpu_map__alloc(unsigned int nr_cpus)
 {
 	RC_STRUCT(perf_cpu_map) *cpus;
 	struct perf_cpu_map *result;
@@ -78,7 +78,7 @@ void perf_cpu_map__put(struct perf_cpu_map *map)
 static struct perf_cpu_map *cpu_map__new_sysconf(void)
 {
 	struct perf_cpu_map *cpus;
-	int nr_cpus, nr_cpus_conf;
+	long nr_cpus, nr_cpus_conf;
 
 	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
 	if (nr_cpus < 0)
@@ -86,15 +86,13 @@ static struct perf_cpu_map *cpu_map__new_sysconf(void)
 
 	nr_cpus_conf = sysconf(_SC_NPROCESSORS_CONF);
 	if (nr_cpus != nr_cpus_conf) {
-		pr_warning("Number of online CPUs (%d) differs from the number configured (%d) the CPU map will only cover the first %d CPUs.",
+		pr_warning("Number of online CPUs (%ld) differs from the number configured (%ld) the CPU map will only cover the first %ld CPUs.",
 			nr_cpus, nr_cpus_conf, nr_cpus);
 	}
 
 	cpus = perf_cpu_map__alloc(nr_cpus);
 	if (cpus != NULL) {
-		int i;
-
-		for (i = 0; i < nr_cpus; ++i)
+		for (long i = 0; i < nr_cpus; ++i)
 			RC_CHK_ACCESS(cpus)->map[i].cpu = i;
 	}
 
@@ -132,23 +130,23 @@ static int cmp_cpu(const void *a, const void *b)
 	return cpu_a->cpu - cpu_b->cpu;
 }
 
-static struct perf_cpu __perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx)
+static struct perf_cpu __perf_cpu_map__cpu(const struct perf_cpu_map *cpus, unsigned int idx)
 {
 	return RC_CHK_ACCESS(cpus)->map[idx];
 }
 
-static struct perf_cpu_map *cpu_map__trim_new(int nr_cpus, const struct perf_cpu *tmp_cpus)
+static struct perf_cpu_map *cpu_map__trim_new(unsigned int nr_cpus, const struct perf_cpu *tmp_cpus)
 {
 	size_t payload_size = nr_cpus * sizeof(struct perf_cpu);
 	struct perf_cpu_map *cpus = perf_cpu_map__alloc(nr_cpus);
-	int i, j;
 
 	if (cpus != NULL) {
+		unsigned int j = 0;
+
 		memcpy(RC_CHK_ACCESS(cpus)->map, tmp_cpus, payload_size);
 		qsort(RC_CHK_ACCESS(cpus)->map, nr_cpus, sizeof(struct perf_cpu), cmp_cpu);
 		/* Remove dups */
-		j = 0;
-		for (i = 0; i < nr_cpus; i++) {
+		for (unsigned int i = 0; i < nr_cpus; i++) {
 			if (i == 0 ||
 			    __perf_cpu_map__cpu(cpus, i).cpu !=
 			    __perf_cpu_map__cpu(cpus, i - 1).cpu) {
@@ -167,9 +165,8 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list)
 	struct perf_cpu_map *cpus = NULL;
 	unsigned long start_cpu, end_cpu = 0;
 	char *p = NULL;
-	int i, nr_cpus = 0;
+	unsigned int nr_cpus = 0, max_entries = 0;
 	struct perf_cpu *tmp_cpus = NULL, *tmp;
-	int max_entries = 0;
 
 	if (!cpu_list)
 		return perf_cpu_map__new_online_cpus();
@@ -185,7 +182,7 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list)
 	while (isdigit(*cpu_list)) {
 		p = NULL;
 		start_cpu = strtoul(cpu_list, &p, 0);
-		if (start_cpu >= INT_MAX
+		if (start_cpu >= INT16_MAX
 		    || (*p != '\0' && *p != ',' && *p != '-' && *p != '\n'))
 			goto invalid;
 
@@ -194,7 +191,7 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list)
 			p = NULL;
 			end_cpu = strtoul(cpu_list, &p, 0);
 
-			if (end_cpu >= INT_MAX || (*p != '\0' && *p != ',' && *p != '\n'))
+			if (end_cpu >= INT16_MAX || (*p != '\0' && *p != ',' && *p != '\n'))
 				goto invalid;
 
 			if (end_cpu < start_cpu)
@@ -208,9 +205,10 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list)
 
 		for (; start_cpu <= end_cpu; start_cpu++) {
 			/* check for duplicates */
-			for (i = 0; i < nr_cpus; i++)
-				if (tmp_cpus[i].cpu == (int)start_cpu)
+			for (unsigned int i = 0; i < nr_cpus; i++) {
+				if (tmp_cpus[i].cpu == (int16_t)start_cpu)
 					goto invalid;
+			}
 
 			if (nr_cpus == max_entries) {
 				max_entries += max(end_cpu - start_cpu + 1, 16UL);
@@ -219,7 +217,7 @@ struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list)
 					goto invalid;
 				tmp_cpus = tmp;
 			}
-			tmp_cpus[nr_cpus++].cpu = (int)start_cpu;
+			tmp_cpus[nr_cpus++].cpu = (int16_t)start_cpu;
 		}
 		if (*p)
 			++p;
@@ -242,12 +240,22 @@ out:
 	return cpus;
 }
 
-static int __perf_cpu_map__nr(const struct perf_cpu_map *cpus)
+struct perf_cpu_map *perf_cpu_map__new_int(int cpu)
+{
+	struct perf_cpu_map *cpus = perf_cpu_map__alloc(1);
+
+	if (cpus)
+		RC_CHK_ACCESS(cpus)->map[0].cpu = cpu;
+
+	return cpus;
+}
+
+static unsigned int __perf_cpu_map__nr(const struct perf_cpu_map *cpus)
 {
 	return RC_CHK_ACCESS(cpus)->nr;
 }
 
-struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx)
+struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, unsigned int idx)
 {
 	struct perf_cpu result = {
 		.cpu = -1
@@ -259,7 +267,7 @@ struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx)
 	return result;
 }
 
-int perf_cpu_map__nr(const struct perf_cpu_map *cpus)
+unsigned int perf_cpu_map__nr(const struct perf_cpu_map *cpus)
 {
 	return cpus ? __perf_cpu_map__nr(cpus) : 1;
 }
@@ -284,7 +292,7 @@ bool perf_cpu_map__is_empty(const struct perf_cpu_map *map)
 
 int perf_cpu_map__idx(const struct perf_cpu_map *cpus, struct perf_cpu cpu)
 {
-	int low, high;
+	unsigned int low, high;
 
 	if (!cpus)
 		return -1;
@@ -314,7 +322,7 @@ bool perf_cpu_map__has(const struct perf_cpu_map *cpus, struct perf_cpu cpu)
 
 bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, const struct perf_cpu_map *rhs)
 {
-	int nr;
+	unsigned int nr;
 
 	if (lhs == rhs)
 		return true;
@@ -326,7 +334,7 @@ bool perf_cpu_map__equal(const struct perf_cpu_map *lhs, const struct perf_cpu_m
 	if (nr != __perf_cpu_map__nr(rhs))
 		return false;
 
-	for (int idx = 0; idx < nr; idx++) {
+	for (unsigned int idx = 0; idx < nr; idx++) {
 		if (__perf_cpu_map__cpu(lhs, idx).cpu != __perf_cpu_map__cpu(rhs, idx).cpu)
 			return false;
 	}
@@ -343,7 +351,7 @@ struct perf_cpu perf_cpu_map__min(const struct perf_cpu_map *map)
 	struct perf_cpu cpu, result = {
 		.cpu = -1
 	};
-	int idx;
+	unsigned int idx;
 
 	perf_cpu_map__for_each_cpu_skip_any(cpu, idx, map) {
 		result = cpu;
@@ -358,10 +366,12 @@ struct perf_cpu perf_cpu_map__max(const struct perf_cpu_map *map)
 		.cpu = -1
 	};
 
-	// cpu_map__trim_new() qsort()s it, cpu_map__default_new() sorts it as well.
-	return __perf_cpu_map__nr(map) > 0
-		? __perf_cpu_map__cpu(map, __perf_cpu_map__nr(map) - 1)
-		: result;
+	if (!map)
+		return result;
+
+	// The CPUs are always sorted and nr is always > 0 as 0 length map is
+	// encoded as NULL.
+	return __perf_cpu_map__cpu(map, __perf_cpu_map__nr(map) - 1);
 }
 
 /** Is 'b' a subset of 'a'. */
@@ -372,7 +382,7 @@ bool perf_cpu_map__is_subset(const struct perf_cpu_map *a, const struct perf_cpu
 	if (!a || __perf_cpu_map__nr(b) > __perf_cpu_map__nr(a))
 		return false;
 
-	for (int i = 0, j = 0; i < __perf_cpu_map__nr(a); i++) {
+	for (unsigned int i = 0, j = 0; i < __perf_cpu_map__nr(a); i++) {
 		if (__perf_cpu_map__cpu(a, i).cpu > __perf_cpu_map__cpu(b, j).cpu)
 			return false;
 		if (__perf_cpu_map__cpu(a, i).cpu == __perf_cpu_map__cpu(b, j).cpu) {
@@ -398,8 +408,7 @@ bool perf_cpu_map__is_subset(const struct perf_cpu_map *a, const struct perf_cpu
 int perf_cpu_map__merge(struct perf_cpu_map **orig, struct perf_cpu_map *other)
 {
 	struct perf_cpu *tmp_cpus;
-	int tmp_len;
-	int i, j, k;
+	unsigned int tmp_len, i, j, k;
 	struct perf_cpu_map *merged;
 
 	if (perf_cpu_map__is_subset(*orig, other))
@@ -443,21 +452,33 @@ int perf_cpu_map__merge(struct perf_cpu_map **orig, struct perf_cpu_map *other)
 struct perf_cpu_map *perf_cpu_map__intersect(struct perf_cpu_map *orig,
 					     struct perf_cpu_map *other)
 {
-	struct perf_cpu *tmp_cpus;
-	int tmp_len;
-	int i, j, k;
-	struct perf_cpu_map *merged = NULL;
+	unsigned int i, j, k;
+	struct perf_cpu_map *merged;
 
 	if (perf_cpu_map__is_subset(other, orig))
 		return perf_cpu_map__get(orig);
 	if (perf_cpu_map__is_subset(orig, other))
 		return perf_cpu_map__get(other);
 
-	tmp_len = max(__perf_cpu_map__nr(orig), __perf_cpu_map__nr(other));
-	tmp_cpus = malloc(tmp_len * sizeof(struct perf_cpu));
-	if (!tmp_cpus)
+	i = j = k = 0;
+	while (i < __perf_cpu_map__nr(orig) && j < __perf_cpu_map__nr(other)) {
+		if (__perf_cpu_map__cpu(orig, i).cpu < __perf_cpu_map__cpu(other, j).cpu)
+			i++;
+		else if (__perf_cpu_map__cpu(orig, i).cpu > __perf_cpu_map__cpu(other, j).cpu)
+			j++;
+		else { /* CPUs match. */
+			i++;
+			j++;
+			k++;
+		}
+	}
+	if (k == 0) /* Maps are completely disjoint. */
 		return NULL;
 
+	merged = perf_cpu_map__alloc(k);
+	if (!merged)
+		return NULL;
+	/* Entries are added to merged in sorted order, so no need to sort again. */
 	i = j = k = 0;
 	while (i < __perf_cpu_map__nr(orig) && j < __perf_cpu_map__nr(other)) {
 		if (__perf_cpu_map__cpu(orig, i).cpu < __perf_cpu_map__cpu(other, j).cpu)
@@ -466,11 +487,8 @@ struct perf_cpu_map *perf_cpu_map__intersect(struct perf_cpu_map *orig,
 			j++;
 		else {
 			j++;
-			tmp_cpus[k++] = __perf_cpu_map__cpu(orig, i++);
+			RC_CHK_ACCESS(merged)->map[k++] = __perf_cpu_map__cpu(orig, i++);
 		}
 	}
-	if (k)
-		merged = cpu_map__trim_new(k, tmp_cpus);
-	free(tmp_cpus);
 	return merged;
 }
diff --git a/tools/lib/perf/evlist.c b/tools/lib/perf/evlist.c
index b1f4c8176b32..1f210dadd666 100644
--- a/tools/lib/perf/evlist.c
+++ b/tools/lib/perf/evlist.c
@@ -36,49 +36,110 @@ void perf_evlist__init(struct perf_evlist *evlist)
 static void __perf_evlist__propagate_maps(struct perf_evlist *evlist,
 					  struct perf_evsel *evsel)
 {
-	if (evsel->system_wide) {
-		/* System wide: set the cpu map of the evsel to all online CPUs. */
-		perf_cpu_map__put(evsel->cpus);
-		evsel->cpus = perf_cpu_map__new_online_cpus();
-	} else if (evlist->has_user_cpus && evsel->is_pmu_core) {
-		/*
-		 * User requested CPUs on a core PMU, ensure the requested CPUs
-		 * are valid by intersecting with those of the PMU.
-		 */
+	if (perf_cpu_map__is_empty(evsel->cpus)) {
+		if (perf_cpu_map__is_empty(evsel->pmu_cpus)) {
+			/*
+			 * Assume the unset PMU cpus were for a system-wide
+			 * event, like a software or tracepoint.
+			 */
+			evsel->pmu_cpus = perf_cpu_map__new_online_cpus();
+		}
+		if (evlist->has_user_cpus && !evsel->system_wide) {
+			/*
+			 * Use the user CPUs unless the evsel is set to be
+			 * system wide, such as the dummy event.
+			 */
+			evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
+		} else {
+			/*
+			 * System wide and other modes, assume the cpu map
+			 * should be set to all PMU CPUs.
+			 */
+			evsel->cpus = perf_cpu_map__get(evsel->pmu_cpus);
+		}
+	}
+	/*
+	 * Avoid "any CPU"(-1) for uncore and PMUs that require a CPU, even if
+	 * requested.
+	 */
+	if (evsel->requires_cpu && perf_cpu_map__has_any_cpu(evsel->cpus)) {
 		perf_cpu_map__put(evsel->cpus);
-		evsel->cpus = perf_cpu_map__intersect(evlist->user_requested_cpus, evsel->own_cpus);
+		evsel->cpus = perf_cpu_map__get(evsel->pmu_cpus);
+	}
 
-		/*
-		 * Empty cpu lists would eventually get opened as "any" so remove
-		 * genuinely empty ones before they're opened in the wrong place.
-		 */
-		if (perf_cpu_map__is_empty(evsel->cpus)) {
-			struct perf_evsel *next = perf_evlist__next(evlist, evsel);
-
-			perf_evlist__remove(evlist, evsel);
-			/* Keep idx contiguous */
-			if (next)
-				list_for_each_entry_from(next, &evlist->entries, node)
-					next->idx--;
+	/*
+	 * Globally requested CPUs replace user requested unless the evsel is
+	 * set to be system wide.
+	 */
+	if (evlist->has_user_cpus && !evsel->system_wide) {
+		assert(!perf_cpu_map__has_any_cpu(evlist->user_requested_cpus));
+		if (!perf_cpu_map__equal(evsel->cpus, evlist->user_requested_cpus)) {
+			perf_cpu_map__put(evsel->cpus);
+			evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
 		}
-	} else if (!evsel->own_cpus || evlist->has_user_cpus ||
-		(!evsel->requires_cpu && perf_cpu_map__has_any_cpu(evlist->user_requested_cpus))) {
-		/*
-		 * The PMU didn't specify a default cpu map, this isn't a core
-		 * event and the user requested CPUs or the evlist user
-		 * requested CPUs have the "any CPU" (aka dummy) CPU value. In
-		 * which case use the user requested CPUs rather than the PMU
-		 * ones.
-		 */
+	}
+
+	/* Ensure cpus only references valid PMU CPUs. */
+	if (!perf_cpu_map__has_any_cpu(evsel->cpus) &&
+	    !perf_cpu_map__is_subset(evsel->pmu_cpus, evsel->cpus)) {
+		struct perf_cpu_map *tmp = perf_cpu_map__intersect(evsel->pmu_cpus, evsel->cpus);
+
 		perf_cpu_map__put(evsel->cpus);
-		evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
-	} else if (evsel->cpus != evsel->own_cpus) {
-		/*
-		 * No user requested cpu map but the PMU cpu map doesn't match
-		 * the evsel's. Reset it back to the PMU cpu map.
-		 */
+		evsel->cpus = tmp;
+	}
+
+	/*
+	 * Was event requested on all the PMU's CPUs but the user requested is
+	 * any CPU (-1)? If so switch to using any CPU (-1) to reduce the number
+	 * of events.
+	 */
+	if (!evsel->system_wide &&
+	    !evsel->requires_cpu &&
+	    perf_cpu_map__equal(evsel->cpus, evsel->pmu_cpus) &&
+	    perf_cpu_map__has_any_cpu(evlist->user_requested_cpus)) {
 		perf_cpu_map__put(evsel->cpus);
-		evsel->cpus = perf_cpu_map__get(evsel->own_cpus);
+		evsel->cpus = perf_cpu_map__get(evlist->user_requested_cpus);
+	}
+
+	/*
+	 * Tool events may only read on the first CPU index to avoid double
+	 * counting things like duration_time. Make the evsel->cpus contain just
+	 * that single entry otherwise we may spend time changing affinity to
+	 * CPUs that just have tool events, etc.
+	 */
+	if (evsel->reads_only_on_cpu_idx0 && perf_cpu_map__nr(evsel->cpus) > 0) {
+		struct perf_cpu_map *srcs[3] = {
+			evlist->all_cpus,
+			evlist->user_requested_cpus,
+			evsel->pmu_cpus,
+		};
+		for (size_t i = 0; i < ARRAY_SIZE(srcs); i++) {
+			if (!srcs[i])
+				continue;
+
+			perf_cpu_map__put(evsel->cpus);
+			evsel->cpus = perf_cpu_map__new_int(perf_cpu_map__cpu(srcs[i], 0).cpu);
+			break;
+		}
+	}
+
+	/* Sanity check assert before the evsel is potentially removed. */
+	assert(!evsel->requires_cpu || !perf_cpu_map__has_any_cpu(evsel->cpus));
+
+	/*
+	 * Empty cpu lists would eventually get opened as "any" so remove
+	 * genuinely empty ones before they're opened in the wrong place.
+	 */
+	if (perf_cpu_map__is_empty(evsel->cpus)) {
+		struct perf_evsel *next = perf_evlist__next(evlist, evsel);
+
+		perf_evlist__remove(evlist, evsel);
+		/* Keep idx contiguous */
+		if (next)
+			list_for_each_entry_from(next, &evlist->entries, node)
+				next->idx--;
+
+		return;
 	}
 
 	if (evsel->system_wide) {
@@ -94,12 +155,22 @@ static void __perf_evlist__propagate_maps(struct perf_evlist *evlist,
 
 static void perf_evlist__propagate_maps(struct perf_evlist *evlist)
 {
-	struct perf_evsel *evsel, *n;
-
 	evlist->needs_map_propagation = true;
 
-	list_for_each_entry_safe(evsel, n, &evlist->entries, node)
-		__perf_evlist__propagate_maps(evlist, evsel);
+	/* Clear the all_cpus set which will be merged into during propagation. */
+	perf_cpu_map__put(evlist->all_cpus);
+	evlist->all_cpus = NULL;
+
+	/* 2 rounds so that reads_only_on_cpu_idx0 benefit from knowing the other CPU maps. */
+	for (int round = 0; round < 2; round++) {
+		struct perf_evsel *evsel, *n;
+
+		list_for_each_entry_safe(evsel, n, &evlist->entries, node) {
+			if ((!evsel->reads_only_on_cpu_idx0 && round == 0) ||
+			    (evsel->reads_only_on_cpu_idx0 && round == 1))
+				__perf_evlist__propagate_maps(evlist, evsel);
+		}
+	}
 }
 
 void perf_evlist__add(struct perf_evlist *evlist,
diff --git a/tools/lib/perf/evsel.c b/tools/lib/perf/evsel.c
index c475319e2e41..f747c0bc692d 100644
--- a/tools/lib/perf/evsel.c
+++ b/tools/lib/perf/evsel.c
@@ -40,8 +40,19 @@ struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr)
 	return evsel;
 }
 
+void perf_evsel__exit(struct perf_evsel *evsel)
+{
+	assert(evsel->fd == NULL);  /* If not fds were not closed. */
+	assert(evsel->mmap == NULL); /* If not munmap wasn't called. */
+	assert(evsel->sample_id == NULL); /* If not free_id wasn't called. */
+	perf_cpu_map__put(evsel->cpus);
+	perf_cpu_map__put(evsel->pmu_cpus);
+	perf_thread_map__put(evsel->threads);
+}
+
 void perf_evsel__delete(struct perf_evsel *evsel)
 {
+	perf_evsel__exit(evsel);
 	free(evsel);
 }
 
@@ -116,7 +127,8 @@ int perf_evsel__open(struct perf_evsel *evsel, struct perf_cpu_map *cpus,
 		     struct perf_thread_map *threads)
 {
 	struct perf_cpu cpu;
-	int idx, thread, err = 0;
+	unsigned int idx;
+	int thread, err = 0;
 
 	if (cpus == NULL) {
 		static struct perf_cpu_map *empty_cpu_map;
@@ -449,7 +461,7 @@ int perf_evsel__enable_cpu(struct perf_evsel *evsel, int cpu_map_idx)
 int perf_evsel__enable_thread(struct perf_evsel *evsel, int thread)
 {
 	struct perf_cpu cpu __maybe_unused;
-	int idx;
+	unsigned int idx;
 	int err;
 
 	perf_cpu_map__for_each_cpu(cpu, idx, evsel->cpus) {
@@ -488,12 +500,13 @@ int perf_evsel__disable(struct perf_evsel *evsel)
 
 int perf_evsel__apply_filter(struct perf_evsel *evsel, const char *filter)
 {
-	int err = 0, i;
+	int err = 0;
 
-	for (i = 0; i < perf_cpu_map__nr(evsel->cpus) && !err; i++)
+	for (unsigned int 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);
+	}
 	return err;
 }
 
diff --git a/tools/lib/perf/include/internal/cpumap.h b/tools/lib/perf/include/internal/cpumap.h
index e2be2d17c32b..c19678188b17 100644
--- a/tools/lib/perf/include/internal/cpumap.h
+++ b/tools/lib/perf/include/internal/cpumap.h
@@ -16,16 +16,16 @@
 DECLARE_RC_STRUCT(perf_cpu_map) {
 	refcount_t	refcnt;
 	/** Length of the map array. */
-	int		nr;
+	unsigned int	nr;
 	/** The CPU values. */
 	struct perf_cpu	map[];
 };
 
-struct perf_cpu_map *perf_cpu_map__alloc(int nr_cpus);
+struct perf_cpu_map *perf_cpu_map__alloc(unsigned int nr_cpus);
 int perf_cpu_map__idx(const struct perf_cpu_map *cpus, struct perf_cpu cpu);
 bool perf_cpu_map__is_subset(const struct perf_cpu_map *a, const struct perf_cpu_map *b);
 
-void perf_cpu_map__set_nr(struct perf_cpu_map *map, int nr_cpus);
+void perf_cpu_map__set_nr(struct perf_cpu_map *map, unsigned int nr_cpus);
 
 static inline refcount_t *perf_cpu_map__refcnt(struct perf_cpu_map *map)
 {
diff --git a/tools/lib/perf/include/internal/evsel.h b/tools/lib/perf/include/internal/evsel.h
index ea78defa77d0..b988034f1371 100644
--- a/tools/lib/perf/include/internal/evsel.h
+++ b/tools/lib/perf/include/internal/evsel.h
@@ -99,7 +99,7 @@ struct perf_evsel {
 	 * cpu map for opening the event on, for example, the first CPU on a
 	 * socket for an uncore event.
 	 */
-	struct perf_cpu_map	*own_cpus;
+	struct perf_cpu_map	*pmu_cpus;
 	struct perf_thread_map	*threads;
 	struct xyarray		*fd;
 	struct xyarray		*mmap;
@@ -128,11 +128,14 @@ struct perf_evsel {
 	bool			 requires_cpu;
 	/** Is the PMU for the event a core one? Effects the handling of own_cpus. */
 	bool			 is_pmu_core;
+	/** Does the evsel on read on the first CPU index such as tool time events? */
+	bool			 reads_only_on_cpu_idx0;
 	int			 idx;
 };
 
 void perf_evsel__init(struct perf_evsel *evsel, struct perf_event_attr *attr,
 		      int idx);
+void perf_evsel__exit(struct perf_evsel *evsel);
 int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads);
 void perf_evsel__close_fd(struct perf_evsel *evsel);
 void perf_evsel__free_fd(struct perf_evsel *evsel);
diff --git a/tools/lib/perf/include/perf/core.h b/tools/lib/perf/include/perf/core.h
index a3f6d68edad7..06cc132d88cf 100644
--- a/tools/lib/perf/include/perf/core.h
+++ b/tools/lib/perf/include/perf/core.h
@@ -5,7 +5,7 @@
 #include <stdarg.h>
 
 #ifndef LIBPERF_API
-#define LIBPERF_API __attribute__((visibility("default")))
+#define LIBPERF_API extern __attribute__((visibility("default")))
 #endif
 
 enum libperf_print_level {
diff --git a/tools/lib/perf/include/perf/cpumap.h b/tools/lib/perf/include/perf/cpumap.h
index 188a667babc6..a1dd25db65b6 100644
--- a/tools/lib/perf/include/perf/cpumap.h
+++ b/tools/lib/perf/include/perf/cpumap.h
@@ -4,10 +4,11 @@
 
 #include <perf/core.h>
 #include <stdbool.h>
+#include <stdint.h>
 
 /** A wrapper around a CPU to avoid confusion with the perf_cpu_map's map's indices. */
 struct perf_cpu {
-	int cpu;
+	int16_t cpu;
 };
 
 struct perf_cache {
@@ -36,6 +37,8 @@ LIBPERF_API struct perf_cpu_map *perf_cpu_map__new_online_cpus(void);
  *                     perf_cpu_map__new_online_cpus is returned.
  */
 LIBPERF_API struct perf_cpu_map *perf_cpu_map__new(const char *cpu_list);
+/** perf_cpu_map__new_int - create a map with the one given cpu. */
+LIBPERF_API struct perf_cpu_map *perf_cpu_map__new_int(int cpu);
 LIBPERF_API struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map);
 LIBPERF_API int perf_cpu_map__merge(struct perf_cpu_map **orig,
 				    struct perf_cpu_map *other);
@@ -46,7 +49,7 @@ LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map);
  * perf_cpu_map__cpu - get the CPU value at the given index. Returns -1 if index
  *                     is invalid.
  */
-LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx);
+LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, unsigned int idx);
 /**
  * perf_cpu_map__nr - for an empty map returns 1, as perf_cpu_map__cpu returns a
  *                    cpu of -1 for an invalid index, this makes an empty map
@@ -54,7 +57,7 @@ LIBPERF_API struct perf_cpu perf_cpu_map__cpu(const struct perf_cpu_map *cpus, i
  *                    the result is the number CPUs in the map plus one if the
  *                    "any CPU"/dummy value is present.
  */
-LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus);
+LIBPERF_API unsigned int perf_cpu_map__nr(const struct perf_cpu_map *cpus);
 /**
  * perf_cpu_map__has_any_cpu_or_is_empty - is map either empty or has the "any CPU"/dummy value.
  */
diff --git a/tools/lib/perf/include/perf/event.h b/tools/lib/perf/include/perf/event.h
index 37bb7771d914..9043dc72b5d6 100644
--- a/tools/lib/perf/include/perf/event.h
+++ b/tools/lib/perf/include/perf/event.h
@@ -151,6 +151,18 @@ struct perf_record_switch {
 	__u32			 next_prev_tid;
 };
 
+struct perf_record_callchain_deferred {
+	struct perf_event_header header;
+	/*
+	 * This is to match kernel and (deferred) user stacks together.
+	 * The kernel part will be in the sample callchain array after
+	 * the PERF_CONTEXT_USER_DEFERRED entry.
+	 */
+	__u64			 cookie;
+	__u64			 nr;
+	__u64			 ips[];
+};
+
 struct perf_record_header_attr {
 	struct perf_event_header header;
 	struct perf_event_attr	 attr;
@@ -291,6 +303,7 @@ struct perf_record_header_event_type {
 struct perf_record_header_tracing_data {
 	struct perf_event_header header;
 	__u32			 size;
+	__u32			 pad;
 };
 
 #define PERF_RECORD_MISC_BUILD_ID_SIZE (1 << 15)
@@ -457,6 +470,97 @@ struct perf_record_compressed {
 	char			 data[];
 };
 
+/*
+ * `header.size` includes the padding we are going to add while writing the record.
+ * `data_size` only includes the size of `data[]` itself.
+ */
+struct perf_record_compressed2 {
+	struct perf_event_header header;
+	__u64			 data_size;
+	char			 data[];
+};
+
+#define BPF_METADATA_KEY_LEN   64
+#define BPF_METADATA_VALUE_LEN 256
+#define BPF_PROG_NAME_LEN      KSYM_NAME_LEN
+
+struct perf_record_bpf_metadata_entry {
+	char key[BPF_METADATA_KEY_LEN];
+	char value[BPF_METADATA_VALUE_LEN];
+};
+
+struct perf_record_bpf_metadata {
+	struct perf_event_header	      header;
+	char				      prog_name[BPF_PROG_NAME_LEN];
+	__u64				      nr_entries;
+	struct perf_record_bpf_metadata_entry entries[];
+};
+
+struct perf_record_schedstat_cpu_v15 {
+#define CPU_FIELD(_type, _name, _desc, _format, _is_pct, _pct_of, _ver)		_type _name
+#include "schedstat-v15.h"
+#undef CPU_FIELD
+};
+
+struct perf_record_schedstat_cpu_v16 {
+#define CPU_FIELD(_type, _name, _desc, _format, _is_pct, _pct_of, _ver)		_type _name
+#include "schedstat-v16.h"
+#undef CPU_FIELD
+};
+
+struct perf_record_schedstat_cpu_v17 {
+#define CPU_FIELD(_type, _name, _desc, _format, _is_pct, _pct_of, _ver)		_type _name
+#include "schedstat-v17.h"
+#undef CPU_FIELD
+};
+
+struct perf_record_schedstat_cpu {
+	struct perf_event_header header;
+	__u64			 timestamp;
+	__u32			 cpu;
+	__u16			 version;
+	/* Padding */
+	char			 __pad[2];
+	union {
+		struct perf_record_schedstat_cpu_v15 v15;
+		struct perf_record_schedstat_cpu_v16 v16;
+		struct perf_record_schedstat_cpu_v17 v17;
+	};
+};
+
+struct perf_record_schedstat_domain_v15 {
+#define DOMAIN_FIELD(_type, _name, _desc, _format, _is_jiffies, _ver)		_type _name
+#include "schedstat-v15.h"
+#undef DOMAIN_FIELD
+};
+
+struct perf_record_schedstat_domain_v16 {
+#define DOMAIN_FIELD(_type, _name, _desc, _format, _is_jiffies, _ver)		_type _name
+#include "schedstat-v16.h"
+#undef DOMAIN_FIELD
+};
+
+struct perf_record_schedstat_domain_v17 {
+#define DOMAIN_FIELD(_type, _name, _desc, _format, _is_jiffies, _ver)		_type _name
+#include "schedstat-v17.h"
+#undef DOMAIN_FIELD
+};
+
+#define DOMAIN_NAME_LEN		16
+
+struct perf_record_schedstat_domain {
+	struct perf_event_header header;
+	__u64			 timestamp;
+	__u32			 cpu;
+	__u16			 version;
+	__u16			 domain;
+	union {
+		struct perf_record_schedstat_domain_v15 v15;
+		struct perf_record_schedstat_domain_v16 v16;
+		struct perf_record_schedstat_domain_v17 v17;
+	};
+};
+
 enum perf_user_event_type { /* above any possible kernel type */
 	PERF_RECORD_USER_TYPE_START		= 64,
 	PERF_RECORD_HEADER_ATTR			= 64,
@@ -478,6 +582,10 @@ enum perf_user_event_type { /* above any possible kernel type */
 	PERF_RECORD_HEADER_FEATURE		= 80,
 	PERF_RECORD_COMPRESSED			= 81,
 	PERF_RECORD_FINISHED_INIT		= 82,
+	PERF_RECORD_COMPRESSED2			= 83,
+	PERF_RECORD_BPF_METADATA		= 84,
+	PERF_RECORD_SCHEDSTAT_CPU		= 85,
+	PERF_RECORD_SCHEDSTAT_DOMAIN		= 86,
 	PERF_RECORD_HEADER_MAX
 };
 
@@ -494,6 +602,7 @@ union perf_event {
 	struct perf_record_read			read;
 	struct perf_record_throttle		throttle;
 	struct perf_record_sample		sample;
+	struct perf_record_callchain_deferred	callchain_deferred;
 	struct perf_record_bpf_event		bpf;
 	struct perf_record_ksymbol		ksymbol;
 	struct perf_record_text_poke_event	text_poke;
@@ -518,6 +627,10 @@ union perf_event {
 	struct perf_record_time_conv		time_conv;
 	struct perf_record_header_feature	feat;
 	struct perf_record_compressed		pack;
+	struct perf_record_compressed2		pack2;
+	struct perf_record_bpf_metadata		bpf_metadata;
+	struct perf_record_schedstat_cpu	schedstat_cpu;
+	struct perf_record_schedstat_domain	schedstat_domain;
 };
 
 #endif /* __LIBPERF_EVENT_H */
diff --git a/tools/lib/perf/include/perf/schedstat-v15.h b/tools/lib/perf/include/perf/schedstat-v15.h
new file mode 100644
index 000000000000..639458df05f8
--- /dev/null
+++ b/tools/lib/perf/include/perf/schedstat-v15.h
@@ -0,0 +1,146 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifdef CPU_FIELD
+CPU_FIELD(__u32, yld_count, "sched_yield() count",
+	  "%11u", false, yld_count, v15);
+CPU_FIELD(__u32, array_exp, "Legacy counter can be ignored",
+	  "%11u", false, array_exp, v15);
+CPU_FIELD(__u32, sched_count, "schedule() called",
+	  "%11u", false, sched_count, v15);
+CPU_FIELD(__u32, sched_goidle, "schedule() left the processor idle",
+	  "%11u", true, sched_count, v15);
+CPU_FIELD(__u32, ttwu_count, "try_to_wake_up() was called",
+	  "%11u", false, ttwu_count, v15);
+CPU_FIELD(__u32, ttwu_local, "try_to_wake_up() was called to wake up the local cpu",
+	  "%11u", true, ttwu_count, v15);
+CPU_FIELD(__u64, rq_cpu_time, "total runtime by tasks on this processor (in jiffies)",
+	  "%11llu", false, rq_cpu_time, v15);
+CPU_FIELD(__u64, run_delay, "total waittime by tasks on this processor (in jiffies)",
+	  "%11llu", true, rq_cpu_time, v15);
+CPU_FIELD(__u64, pcount, "total timeslices run on this cpu",
+	  "%11llu", false, pcount, v15);
+#endif
+
+#ifdef DOMAIN_FIELD
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category idle> ");
+#endif
+DOMAIN_FIELD(__u32, idle_lb_count,
+	     "load_balance() count on cpu idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, idle_lb_balanced,
+	     "load_balance() found balanced on cpu idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, idle_lb_failed,
+	     "load_balance() move task failed on cpu idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, idle_lb_imbalance,
+	     "imbalance sum on cpu idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, idle_lb_gained,
+	     "pull_task() count on cpu idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, idle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, idle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, idle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu idle", "%11u", true, v15);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(idle_lb_success_count, "load_balance() success count on cpu idle", "%11u",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, v15);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(idle_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu idle)", "%11.2Lf",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, idle_lb_gained, v15);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category busy> ");
+#endif
+DOMAIN_FIELD(__u32, busy_lb_count,
+	     "load_balance() count on cpu busy", "%11u", true, v15);
+DOMAIN_FIELD(__u32, busy_lb_balanced,
+	     "load_balance() found balanced on cpu busy", "%11u", true, v15);
+DOMAIN_FIELD(__u32, busy_lb_failed,
+	     "load_balance() move task failed on cpu busy", "%11u", true, v15);
+DOMAIN_FIELD(__u32, busy_lb_imbalance,
+	     "imbalance sum on cpu busy", "%11u", false, v15);
+DOMAIN_FIELD(__u32, busy_lb_gained,
+	     "pull_task() count on cpu busy", "%11u", false, v15);
+DOMAIN_FIELD(__u32, busy_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu busy", "%11u", false, v15);
+DOMAIN_FIELD(__u32, busy_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu busy", "%11u", true, v15);
+DOMAIN_FIELD(__u32, busy_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu busy", "%11u", true, v15);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(busy_lb_success_count, "load_balance() success count on cpu busy", "%11u",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, v15);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(busy_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu busy)", "%11.2Lf",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, busy_lb_gained, v15);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category newidle> ");
+#endif
+DOMAIN_FIELD(__u32, newidle_lb_count,
+	     "load_balance() count on cpu newly idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, newidle_lb_balanced,
+	     "load_balance() found balanced on cpu newly idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, newidle_lb_failed,
+	     "load_balance() move task failed on cpu newly idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance,
+	     "imbalance sum on cpu newly idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, newidle_lb_gained,
+	     "pull_task() count on cpu newly idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, newidle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu newly idle", "%11u", false, v15);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu newly idle", "%11u", true, v15);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu newly idle", "%11u", true, v15);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(newidle_lb_success_count,
+		  "load_balance() success count on cpu newly idle", "%11u",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, v15);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(newidle_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu newly idle)", "%11.2Lf",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, newidle_lb_gained, v15);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category active_load_balance()> ");
+#endif
+DOMAIN_FIELD(__u32, alb_count,
+	     "active_load_balance() count", "%11u", false, v15);
+DOMAIN_FIELD(__u32, alb_failed,
+	     "active_load_balance() move task failed", "%11u", false, v15);
+DOMAIN_FIELD(__u32, alb_pushed,
+	     "active_load_balance() successfully moved a task", "%11u", false, v15);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_exec()> ");
+#endif
+DOMAIN_FIELD(__u32, sbe_count,
+	     "sbe_count is not used", "%11u", false, v15);
+DOMAIN_FIELD(__u32, sbe_balanced,
+	     "sbe_balanced is not used", "%11u", false, v15);
+DOMAIN_FIELD(__u32, sbe_pushed,
+	     "sbe_pushed is not used", "%11u", false, v15);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_fork()> ");
+#endif
+DOMAIN_FIELD(__u32, sbf_count,
+	     "sbf_count is not used", "%11u", false, v15);
+DOMAIN_FIELD(__u32, sbf_balanced,
+	     "sbf_balanced is not used", "%11u", false, v15);
+DOMAIN_FIELD(__u32, sbf_pushed,
+	     "sbf_pushed is not used", "%11u", false, v15);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Wakeup Info> ");
+#endif
+DOMAIN_FIELD(__u32, ttwu_wake_remote,
+	     "try_to_wake_up() awoke a task that last ran on a diff cpu", "%11u", false, v15);
+DOMAIN_FIELD(__u32, ttwu_move_affine,
+	     "try_to_wake_up() moved task because cache-cold on own cpu", "%11u", false, v15);
+DOMAIN_FIELD(__u32, ttwu_move_balance,
+	     "try_to_wake_up() started passive balancing", "%11u", false, v15);
+#endif /* DOMAIN_FIELD */
diff --git a/tools/lib/perf/include/perf/schedstat-v16.h b/tools/lib/perf/include/perf/schedstat-v16.h
new file mode 100644
index 000000000000..3462b79c29af
--- /dev/null
+++ b/tools/lib/perf/include/perf/schedstat-v16.h
@@ -0,0 +1,146 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifdef CPU_FIELD
+CPU_FIELD(__u32, yld_count, "sched_yield() count",
+	  "%11u", false, yld_count, v16);
+CPU_FIELD(__u32, array_exp, "Legacy counter can be ignored",
+	  "%11u", false, array_exp, v16);
+CPU_FIELD(__u32, sched_count, "schedule() called",
+	  "%11u", false, sched_count, v16);
+CPU_FIELD(__u32, sched_goidle, "schedule() left the processor idle",
+	  "%11u", true, sched_count, v16);
+CPU_FIELD(__u32, ttwu_count, "try_to_wake_up() was called",
+	  "%11u", false, ttwu_count, v16);
+CPU_FIELD(__u32, ttwu_local, "try_to_wake_up() was called to wake up the local cpu",
+	  "%11u", true, ttwu_count, v16);
+CPU_FIELD(__u64, rq_cpu_time, "total runtime by tasks on this processor (in jiffies)",
+	  "%11llu", false, rq_cpu_time, v16);
+CPU_FIELD(__u64, run_delay, "total waittime by tasks on this processor (in jiffies)",
+	  "%11llu", true, rq_cpu_time, v16);
+CPU_FIELD(__u64, pcount, "total timeslices run on this cpu",
+	  "%11llu", false, pcount, v16);
+#endif /* CPU_FIELD */
+
+#ifdef DOMAIN_FIELD
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category busy> ");
+#endif
+DOMAIN_FIELD(__u32, busy_lb_count,
+	     "load_balance() count on cpu busy", "%11u", true, v16);
+DOMAIN_FIELD(__u32, busy_lb_balanced,
+	     "load_balance() found balanced on cpu busy", "%11u", true, v16);
+DOMAIN_FIELD(__u32, busy_lb_failed,
+	     "load_balance() move task failed on cpu busy", "%11u", true, v16);
+DOMAIN_FIELD(__u32, busy_lb_imbalance,
+	     "imbalance sum on cpu busy", "%11u", false, v16);
+DOMAIN_FIELD(__u32, busy_lb_gained,
+	     "pull_task() count on cpu busy", "%11u", false, v16);
+DOMAIN_FIELD(__u32, busy_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu busy", "%11u", false, v16);
+DOMAIN_FIELD(__u32, busy_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu busy", "%11u", true, v16);
+DOMAIN_FIELD(__u32, busy_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu busy", "%11u", true, v16);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(busy_lb_success_count, "load_balance() success count on cpu busy", "%11u",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, v16);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(busy_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu busy)", "%11.2Lf",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, busy_lb_gained, v16);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category idle> ");
+#endif
+DOMAIN_FIELD(__u32, idle_lb_count,
+	     "load_balance() count on cpu idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, idle_lb_balanced,
+	     "load_balance() found balanced on cpu idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, idle_lb_failed,
+	     "load_balance() move task failed on cpu idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, idle_lb_imbalance,
+	     "imbalance sum on cpu idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, idle_lb_gained,
+	     "pull_task() count on cpu idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, idle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, idle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, idle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu idle", "%11u", true, v16);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(idle_lb_success_count, "load_balance() success count on cpu idle", "%11u",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, v16);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(idle_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu idle)", "%11.2Lf",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, idle_lb_gained, v16);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category newidle> ");
+#endif
+DOMAIN_FIELD(__u32, newidle_lb_count,
+	     "load_balance() count on cpu newly idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, newidle_lb_balanced,
+	     "load_balance() found balanced on cpu newly idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, newidle_lb_failed,
+	     "load_balance() move task failed on cpu newly idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance,
+	     "imbalance sum on cpu newly idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, newidle_lb_gained,
+	     "pull_task() count on cpu newly idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, newidle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu newly idle", "%11u", false, v16);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu newly idle", "%11u", true, v16);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu newly idle", "%11u", true, v16);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(newidle_lb_success_count,
+		  "load_balance() success count on cpu newly idle", "%11u",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, v16);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(newidle_lb_avg_count,
+		  "avg task pulled per successful lb attempt (cpu newly idle)", "%11.2Lf",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, newidle_lb_gained, v16);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category active_load_balance()> ");
+#endif
+DOMAIN_FIELD(__u32, alb_count,
+	     "active_load_balance() count", "%11u", false, v16);
+DOMAIN_FIELD(__u32, alb_failed,
+	     "active_load_balance() move task failed", "%11u", false, v16);
+DOMAIN_FIELD(__u32, alb_pushed,
+	     "active_load_balance() successfully moved a task", "%11u", false, v16);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_exec()> ");
+#endif
+DOMAIN_FIELD(__u32, sbe_count,
+	     "sbe_count is not used", "%11u", false, v16);
+DOMAIN_FIELD(__u32, sbe_balanced,
+	     "sbe_balanced is not used", "%11u", false, v16);
+DOMAIN_FIELD(__u32, sbe_pushed,
+	     "sbe_pushed is not used", "%11u", false, v16);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_fork()> ");
+#endif
+DOMAIN_FIELD(__u32, sbf_count,
+	     "sbf_count is not used", "%11u", false, v16);
+DOMAIN_FIELD(__u32, sbf_balanced,
+	     "sbf_balanced is not used", "%11u", false, v16);
+DOMAIN_FIELD(__u32, sbf_pushed,
+	     "sbf_pushed is not used", "%11u", false, v16);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Wakeup Info> ");
+#endif
+DOMAIN_FIELD(__u32, ttwu_wake_remote,
+	     "try_to_wake_up() awoke a task that last ran on a diff cpu", "%11u", false, v16);
+DOMAIN_FIELD(__u32, ttwu_move_affine,
+	     "try_to_wake_up() moved task because cache-cold on own cpu", "%11u", false, v16);
+DOMAIN_FIELD(__u32, ttwu_move_balance,
+	     "try_to_wake_up() started passive balancing", "%11u", false, v16);
+#endif /* DOMAIN_FIELD */
diff --git a/tools/lib/perf/include/perf/schedstat-v17.h b/tools/lib/perf/include/perf/schedstat-v17.h
new file mode 100644
index 000000000000..865dc7c1039c
--- /dev/null
+++ b/tools/lib/perf/include/perf/schedstat-v17.h
@@ -0,0 +1,164 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#ifdef CPU_FIELD
+CPU_FIELD(__u32, yld_count, "sched_yield() count",
+	  "%11u", false, yld_count, v17);
+CPU_FIELD(__u32, array_exp, "Legacy counter can be ignored",
+	  "%11u", false, array_exp, v17);
+CPU_FIELD(__u32, sched_count, "schedule() called",
+	  "%11u", false, sched_count, v17);
+CPU_FIELD(__u32, sched_goidle, "schedule() left the processor idle",
+	  "%11u", true, sched_count, v17);
+CPU_FIELD(__u32, ttwu_count, "try_to_wake_up() was called",
+	  "%11u", false, ttwu_count, v17);
+CPU_FIELD(__u32, ttwu_local, "try_to_wake_up() was called to wake up the local cpu",
+	  "%11u", true, ttwu_count, v17);
+CPU_FIELD(__u64, rq_cpu_time, "total runtime by tasks on this processor (in jiffies)",
+	  "%11llu", false, rq_cpu_time, v17);
+CPU_FIELD(__u64, run_delay, "total waittime by tasks on this processor (in jiffies)",
+	  "%11llu", true, rq_cpu_time, v17);
+CPU_FIELD(__u64, pcount, "total timeslices run on this cpu",
+	  "%11llu", false, pcount, v17);
+#endif /* CPU_FIELD */
+
+#ifdef DOMAIN_FIELD
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category busy> ");
+#endif
+DOMAIN_FIELD(__u32, busy_lb_count,
+	     "load_balance() count on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_balanced,
+	     "load_balance() found balanced on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_failed,
+	     "load_balance() move task failed on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_load,
+	     "imbalance in load on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_util,
+	     "imbalance in utilization on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_task,
+	     "imbalance in number of tasks on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_imbalance_misfit,
+	     "imbalance in misfit tasks on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_gained,
+	     "pull_task() count on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu busy", "%11u", false, v17);
+DOMAIN_FIELD(__u32, busy_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu busy", "%11u", true, v17);
+DOMAIN_FIELD(__u32, busy_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu busy", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(busy_lb_success_count, "load_balance() success count on cpu busy", "%11u",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(busy_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu busy)", "%11.2Lf",
+		  busy_lb_count, busy_lb_balanced, busy_lb_failed, busy_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category idle> ");
+#endif
+DOMAIN_FIELD(__u32, idle_lb_count,
+	     "load_balance() count on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_balanced,
+	     "load_balance() found balanced on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_failed,
+	     "load_balance() move task failed on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_load,
+	     "imbalance in load on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_util,
+	     "imbalance in utilization on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_task,
+	     "imbalance in number of tasks on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_imbalance_misfit,
+	     "imbalance in misfit tasks on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_gained,
+	     "pull_task() count on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, idle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, idle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu idle", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(idle_lb_success_count, "load_balance() success count on cpu idle", "%11u",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(idle_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu idle)", "%11.2Lf",
+		  idle_lb_count, idle_lb_balanced, idle_lb_failed, idle_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category newidle> ");
+#endif
+DOMAIN_FIELD(__u32, newidle_lb_count,
+	     "load_balance() count on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_balanced,
+	     "load_balance() found balanced on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_failed,
+	     "load_balance() move task failed on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_load,
+	     "imbalance in load on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_util,
+	     "imbalance in utilization on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_task,
+	     "imbalance in number of tasks on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_imbalance_misfit,
+	     "imbalance in misfit tasks on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_gained,
+	     "pull_task() count on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_hot_gained,
+	     "pull_task() when target task was cache-hot on cpu newly idle", "%11u", false, v17);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyq,
+	     "load_balance() failed to find busier queue on cpu newly idle", "%11u", true, v17);
+DOMAIN_FIELD(__u32, newidle_lb_nobusyg,
+	     "load_balance() failed to find busier group on cpu newly idle", "%11u", true, v17);
+#ifdef DERIVED_CNT_FIELD
+DERIVED_CNT_FIELD(newidle_lb_success_count,
+		  "load_balance() success count on cpu newly idle", "%11u",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, v17);
+#endif
+#ifdef DERIVED_AVG_FIELD
+DERIVED_AVG_FIELD(newidle_lb_avg_pulled,
+		  "avg task pulled per successful lb attempt (cpu newly idle)", "%11.2Lf",
+		  newidle_lb_count, newidle_lb_balanced, newidle_lb_failed, newidle_lb_gained, v17);
+#endif
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category active_load_balance()> ");
+#endif
+DOMAIN_FIELD(__u32, alb_count,
+	     "active_load_balance() count", "%11u", false, v17);
+DOMAIN_FIELD(__u32, alb_failed,
+	     "active_load_balance() move task failed", "%11u", false, v17);
+DOMAIN_FIELD(__u32, alb_pushed,
+	     "active_load_balance() successfully moved a task", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_exec()> ");
+#endif
+DOMAIN_FIELD(__u32, sbe_count,
+	     "sbe_count is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbe_balanced,
+	     "sbe_balanced is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbe_pushed,
+	     "sbe_pushed is not used", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Category sched_balance_fork()> ");
+#endif
+DOMAIN_FIELD(__u32, sbf_count,
+	     "sbf_count is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbf_balanced,
+	     "sbf_balanced is not used", "%11u", false, v17);
+DOMAIN_FIELD(__u32, sbf_pushed,
+	     "sbf_pushed is not used", "%11u", false, v17);
+#ifdef DOMAIN_CATEGORY
+DOMAIN_CATEGORY(" <Wakeup Info> ");
+#endif
+DOMAIN_FIELD(__u32, ttwu_wake_remote,
+	     "try_to_wake_up() awoke a task that last ran on a diff cpu", "%11u", false, v17);
+DOMAIN_FIELD(__u32, ttwu_move_affine,
+	     "try_to_wake_up() moved task because cache-cold on own cpu", "%11u", false, v17);
+DOMAIN_FIELD(__u32, ttwu_move_balance,
+	     "try_to_wake_up() started passive balancing", "%11u", false, v17);
+#endif /* DOMAIN_FIELD */
diff --git a/tools/lib/perf/include/perf/threadmap.h b/tools/lib/perf/include/perf/threadmap.h
index 8b40e7777cea..44deb815b817 100644
--- a/tools/lib/perf/include/perf/threadmap.h
+++ b/tools/lib/perf/include/perf/threadmap.h
@@ -14,6 +14,7 @@ LIBPERF_API void perf_thread_map__set_pid(struct perf_thread_map *map, int idx,
 LIBPERF_API char *perf_thread_map__comm(struct perf_thread_map *map, int idx);
 LIBPERF_API int perf_thread_map__nr(struct perf_thread_map *threads);
 LIBPERF_API pid_t perf_thread_map__pid(struct perf_thread_map *map, int idx);
+LIBPERF_API int perf_thread_map__idx(struct perf_thread_map *map, pid_t pid);
 
 LIBPERF_API struct perf_thread_map *perf_thread_map__get(struct perf_thread_map *map);
 LIBPERF_API void perf_thread_map__put(struct perf_thread_map *map);
diff --git a/tools/lib/perf/mmap.c b/tools/lib/perf/mmap.c
index c1a51d925e0e..ec124eb0ec0a 100644
--- a/tools/lib/perf/mmap.c
+++ b/tools/lib/perf/mmap.c
@@ -508,7 +508,7 @@ int perf_mmap__read_self(struct perf_mmap *map, struct perf_counts_values *count
 		idx = READ_ONCE(pc->index);
 		cnt = READ_ONCE(pc->offset);
 		if (pc->cap_user_rdpmc && idx) {
-			s64 evcnt = read_perf_counter(idx - 1);
+			u64 evcnt = read_perf_counter(idx - 1);
 			u16 width = READ_ONCE(pc->pmc_width);
 
 			evcnt <<= 64 - width;
diff --git a/tools/lib/perf/threadmap.c b/tools/lib/perf/threadmap.c
index 07968f3ea093..db431b036f57 100644
--- a/tools/lib/perf/threadmap.c
+++ b/tools/lib/perf/threadmap.c
@@ -97,5 +97,22 @@ int perf_thread_map__nr(struct perf_thread_map *threads)
 
 pid_t perf_thread_map__pid(struct perf_thread_map *map, int idx)
 {
+	if (!map) {
+		assert(idx == 0);
+		return -1;
+	}
+
 	return map->map[idx].pid;
 }
+
+int perf_thread_map__idx(struct perf_thread_map *threads, pid_t pid)
+{
+	if (!threads)
+		return pid == -1 ? 0 : -1;
+
+	for (int i = 0; i < threads->nr; ++i) {
+		if (threads->map[i].pid == pid)
+			return i;
+	}
+	return -1;
+}