1 files changed, 366 insertions, 243 deletions

diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 5152fd5a6ead..8282ddf68b98 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -7,6 +7,7 @@
 #include <errno.h>
 #include <sys/ioctl.h>
 #include <sys/param.h>
+#include "cpumap.h"
 #include "term.h"
 #include "env.h"
 #include "evlist.h"
@@ -16,20 +17,22 @@
 #include "string2.h"
 #include "strbuf.h"
 #include "debug.h"
-#include <api/fs/tracing_path.h>
-#include <api/io_dir.h>
 #include <perf/cpumap.h>
 #include <util/parse-events-bison.h>
 #include <util/parse-events-flex.h>
 #include "pmu.h"
 #include "pmus.h"
+#include "tp_pmu.h"
 #include "asm/bug.h"
+#include "ui/ui.h"
 #include "util/parse-branch-options.h"
 #include "util/evsel_config.h"
 #include "util/event.h"
 #include "util/bpf-filter.h"
+#include "util/stat.h"
 #include "util/util.h"
 #include "tracepoint.h"
+#include <api/fs/tracing_path.h>
 
 #define MAX_NAME_LEN 100
 
@@ -81,77 +84,21 @@ const struct event_symbol event_symbols_hw[PERF_COUNT_HW_MAX] = {
 	},
 };
 
-const struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = {
-	[PERF_COUNT_SW_CPU_CLOCK] = {
-		.symbol = "cpu-clock",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_TASK_CLOCK] = {
-		.symbol = "task-clock",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_PAGE_FAULTS] = {
-		.symbol = "page-faults",
-		.alias  = "faults",
-	},
-	[PERF_COUNT_SW_CONTEXT_SWITCHES] = {
-		.symbol = "context-switches",
-		.alias  = "cs",
-	},
-	[PERF_COUNT_SW_CPU_MIGRATIONS] = {
-		.symbol = "cpu-migrations",
-		.alias  = "migrations",
-	},
-	[PERF_COUNT_SW_PAGE_FAULTS_MIN] = {
-		.symbol = "minor-faults",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_PAGE_FAULTS_MAJ] = {
-		.symbol = "major-faults",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_ALIGNMENT_FAULTS] = {
-		.symbol = "alignment-faults",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_EMULATION_FAULTS] = {
-		.symbol = "emulation-faults",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_DUMMY] = {
-		.symbol = "dummy",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_BPF_OUTPUT] = {
-		.symbol = "bpf-output",
-		.alias  = "",
-	},
-	[PERF_COUNT_SW_CGROUP_SWITCHES] = {
-		.symbol = "cgroup-switches",
-		.alias  = "",
-	},
+static const char *const event_types[] = {
+	[PERF_TYPE_HARDWARE]	= "hardware",
+	[PERF_TYPE_SOFTWARE]	= "software",
+	[PERF_TYPE_TRACEPOINT]	= "tracepoint",
+	[PERF_TYPE_HW_CACHE]	= "hardware-cache",
+	[PERF_TYPE_RAW]		= "raw",
+	[PERF_TYPE_BREAKPOINT]	= "breakpoint",
 };
 
-const char *event_type(int type)
+const char *event_type(size_t type)
 {
-	switch (type) {
-	case PERF_TYPE_HARDWARE:
-		return "hardware";
+	if (type >= PERF_TYPE_MAX)
+		return "unknown";
 
-	case PERF_TYPE_SOFTWARE:
-		return "software";
-
-	case PERF_TYPE_TRACEPOINT:
-		return "tracepoint";
-
-	case PERF_TYPE_HW_CACHE:
-		return "hardware-cache";
-
-	default:
-		break;
-	}
-
-	return "unknown";
+	return event_types[type];
 }
 
 static char *get_config_str(const struct parse_events_terms *head_terms,
@@ -179,6 +126,38 @@ static char *get_config_name(const struct parse_events_terms *head_terms)
 	return get_config_str(head_terms, PARSE_EVENTS__TERM_TYPE_NAME);
 }
 
+static struct perf_cpu_map *get_config_cpu(const struct parse_events_terms *head_terms)
+{
+	struct parse_events_term *term;
+	struct perf_cpu_map *cpus = NULL;
+
+	if (!head_terms)
+		return NULL;
+
+	list_for_each_entry(term, &head_terms->terms, list) {
+		if (term->type_term == PARSE_EVENTS__TERM_TYPE_CPU) {
+			struct perf_cpu_map *term_cpus;
+
+			if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) {
+				term_cpus = perf_cpu_map__new_int(term->val.num);
+			} else {
+				struct perf_pmu *pmu = perf_pmus__find(term->val.str);
+
+				if (pmu && perf_cpu_map__is_empty(pmu->cpus))
+					term_cpus = pmu->is_core ? cpu_map__online() : NULL;
+				else if (pmu)
+					term_cpus = perf_cpu_map__get(pmu->cpus);
+				else
+					term_cpus = perf_cpu_map__new(term->val.str);
+			}
+			perf_cpu_map__merge(&cpus, term_cpus);
+			perf_cpu_map__put(term_cpus);
+		}
+	}
+
+	return cpus;
+}
+
 /**
  * fix_raw - For each raw term see if there is an event (aka alias) in pmu that
  *           matches the raw's string value. If the string value matches an
@@ -228,49 +207,98 @@ __add_event(struct list_head *list, int *idx,
 	    struct perf_event_attr *attr,
 	    bool init_attr,
 	    const char *name, const char *metric_id, struct perf_pmu *pmu,
-	    struct list_head *config_terms, bool auto_merge_stats,
-	    struct perf_cpu_map *cpu_list, u64 alternate_hw_config)
+	    struct list_head *config_terms, struct evsel *first_wildcard_match,
+	    struct perf_cpu_map *user_cpus, u64 alternate_hw_config)
 {
 	struct evsel *evsel;
-	struct perf_cpu_map *cpus = perf_cpu_map__is_empty(cpu_list) && pmu ? pmu->cpus : cpu_list;
+	bool is_pmu_core;
+	struct perf_cpu_map *cpus, *pmu_cpus;
+	bool has_user_cpus = !perf_cpu_map__is_empty(user_cpus);
 
-	cpus = perf_cpu_map__get(cpus);
-	if (pmu)
-		perf_pmu__warn_invalid_formats(pmu);
+	/*
+	 * Ensure the first_wildcard_match's PMU matches that of the new event
+	 * being added. Otherwise try to match with another event further down
+	 * the evlist.
+	 */
+	if (first_wildcard_match) {
+		struct evsel *pos = list_prev_entry(first_wildcard_match, core.node);
+
+		first_wildcard_match = NULL;
+		list_for_each_entry_continue(pos, list, core.node) {
+			if (perf_pmu__name_no_suffix_match(pos->pmu, pmu->name)) {
+				first_wildcard_match = pos;
+				break;
+			}
+			if (pos->pmu->is_core && (!pmu || pmu->is_core)) {
+				first_wildcard_match = pos;
+				break;
+			}
+		}
+	}
 
-	if (pmu && (attr->type == PERF_TYPE_RAW || attr->type >= PERF_TYPE_MAX)) {
-		perf_pmu__warn_invalid_config(pmu, attr->config, name,
-					      PERF_PMU_FORMAT_VALUE_CONFIG, "config");
-		perf_pmu__warn_invalid_config(pmu, attr->config1, name,
-					      PERF_PMU_FORMAT_VALUE_CONFIG1, "config1");
-		perf_pmu__warn_invalid_config(pmu, attr->config2, name,
-					      PERF_PMU_FORMAT_VALUE_CONFIG2, "config2");
-		perf_pmu__warn_invalid_config(pmu, attr->config3, name,
-					      PERF_PMU_FORMAT_VALUE_CONFIG3, "config3");
+	if (pmu) {
+		perf_pmu__warn_invalid_formats(pmu);
+		if (attr->type == PERF_TYPE_RAW || attr->type >= PERF_TYPE_MAX) {
+			perf_pmu__warn_invalid_config(pmu, attr->config, name,
+						PERF_PMU_FORMAT_VALUE_CONFIG, "config");
+			perf_pmu__warn_invalid_config(pmu, attr->config1, name,
+						PERF_PMU_FORMAT_VALUE_CONFIG1, "config1");
+			perf_pmu__warn_invalid_config(pmu, attr->config2, name,
+						PERF_PMU_FORMAT_VALUE_CONFIG2, "config2");
+			perf_pmu__warn_invalid_config(pmu, attr->config3, name,
+						PERF_PMU_FORMAT_VALUE_CONFIG3, "config3");
+		}
+	}
+	/*
+	 * If a PMU wasn't given, such as for legacy events, find now that
+	 * warnings won't be generated.
+	 */
+	if (!pmu)
+		pmu = perf_pmus__find_by_attr(attr);
+
+	if (pmu) {
+		is_pmu_core = pmu->is_core;
+		pmu_cpus = perf_cpu_map__get(pmu->cpus);
+		if (perf_cpu_map__is_empty(pmu_cpus))
+			pmu_cpus = cpu_map__online();
+	} else {
+		is_pmu_core = (attr->type == PERF_TYPE_HARDWARE ||
+			       attr->type == PERF_TYPE_HW_CACHE);
+		pmu_cpus = is_pmu_core ? cpu_map__online() : NULL;
 	}
+
+	if (has_user_cpus)
+		cpus = perf_cpu_map__get(user_cpus);
+	else
+		cpus = perf_cpu_map__get(pmu_cpus);
+
 	if (init_attr)
 		event_attr_init(attr);
 
 	evsel = evsel__new_idx(attr, *idx);
-	if (!evsel) {
-		perf_cpu_map__put(cpus);
-		return NULL;
+	if (!evsel)
+		goto out_err;
+
+	if (name) {
+		evsel->name = strdup(name);
+		if (!evsel->name)
+			goto out_err;
+	}
+
+	if (metric_id) {
+		evsel->metric_id = strdup(metric_id);
+		if (!evsel->metric_id)
+			goto out_err;
 	}
 
 	(*idx)++;
 	evsel->core.cpus = cpus;
-	evsel->core.own_cpus = perf_cpu_map__get(cpus);
+	evsel->core.pmu_cpus = pmu_cpus;
 	evsel->core.requires_cpu = pmu ? pmu->is_uncore : false;
-	evsel->core.is_pmu_core = pmu ? pmu->is_core : false;
-	evsel->auto_merge_stats = auto_merge_stats;
+	evsel->core.is_pmu_core = is_pmu_core;
 	evsel->pmu = pmu;
 	evsel->alternate_hw_config = alternate_hw_config;
-
-	if (name)
-		evsel->name = strdup(name);
-
-	if (metric_id)
-		evsel->metric_id = strdup(metric_id);
+	evsel->first_wildcard_match = first_wildcard_match;
 
 	if (config_terms)
 		list_splice_init(config_terms, &evsel->config_terms);
@@ -278,7 +306,17 @@ __add_event(struct list_head *list, int *idx,
 	if (list)
 		list_add_tail(&evsel->core.node, list);
 
+	if (has_user_cpus)
+		evsel__warn_user_requested_cpus(evsel, user_cpus);
+
 	return evsel;
+out_err:
+	perf_cpu_map__put(cpus);
+	perf_cpu_map__put(pmu_cpus);
+	zfree(&evsel->name);
+	zfree(&evsel->metric_id);
+	free(evsel);
+	return NULL;
 }
 
 struct evsel *parse_events__add_event(int idx, struct perf_event_attr *attr,
@@ -287,7 +325,7 @@ struct evsel *parse_events__add_event(int idx, struct perf_event_attr *attr,
 {
 	return __add_event(/*list=*/NULL, &idx, attr, /*init_attr=*/false, name,
 			   metric_id, pmu, /*config_terms=*/NULL,
-			   /*auto_merge_stats=*/false, /*cpu_list=*/NULL,
+			   /*first_wildcard_match=*/NULL, /*cpu_list=*/NULL,
 			   /*alternate_hw_config=*/PERF_COUNT_HW_MAX);
 }
 
@@ -298,7 +336,7 @@ static int add_event(struct list_head *list, int *idx,
 {
 	return __add_event(list, idx, attr, /*init_attr*/true, name, metric_id,
 			   /*pmu=*/NULL, config_terms,
-			   /*auto_merge_stats=*/false, /*cpu_list=*/NULL,
+			   /*first_wildcard_match=*/NULL, /*cpu_list=*/NULL,
 			   alternate_hw_config) ? 0 : -ENOMEM;
 }
 
@@ -423,7 +461,7 @@ bool parse_events__filter_pmu(const struct parse_events_state *parse_state,
 static int parse_events_add_pmu(struct parse_events_state *parse_state,
 				struct list_head *list, struct perf_pmu *pmu,
 				const struct parse_events_terms *const_parsed_terms,
-				bool auto_merge_stats, u64 alternate_hw_config);
+				struct evsel *first_wildcard_match, u64 alternate_hw_config);
 
 int parse_events_add_cache(struct list_head *list, int *idx, const char *name,
 			   struct parse_events_state *parse_state,
@@ -433,11 +471,13 @@ int parse_events_add_cache(struct list_head *list, int *idx, const char *name,
 	bool found_supported = false;
 	const char *config_name = get_config_name(parsed_terms);
 	const char *metric_id = get_config_metric_id(parsed_terms);
+	struct perf_cpu_map *cpus = get_config_cpu(parsed_terms);
+	int ret = 0;
+	struct evsel *first_wildcard_match = NULL;
 
-	while ((pmu = perf_pmus__scan(pmu)) != NULL) {
+	while ((pmu = perf_pmus__scan_for_event(pmu, name)) != NULL) {
 		LIST_HEAD(config_terms);
 		struct perf_event_attr attr;
-		int ret;
 
 		if (parse_events__filter_pmu(parse_state, pmu))
 			continue;
@@ -449,10 +489,13 @@ int parse_events_add_cache(struct list_head *list, int *idx, const char *name,
 			 */
 			ret = parse_events_add_pmu(parse_state, list, pmu,
 						   parsed_terms,
-						   perf_pmu__auto_merge_stats(pmu),
+						   first_wildcard_match,
 						   /*alternate_hw_config=*/PERF_COUNT_HW_MAX);
 			if (ret)
-				return ret;
+				goto out_err;
+			if (first_wildcard_match == NULL)
+				first_wildcard_match =
+					container_of(list->prev, struct evsel, core.node);
 			continue;
 		}
 
@@ -472,21 +515,29 @@ int parse_events_add_cache(struct list_head *list, int *idx, const char *name,
 
 		if (parsed_terms) {
 			if (config_attr(&attr, parsed_terms, parse_state->error,
-					config_term_common))
-				return -EINVAL;
-
-			if (get_config_terms(parsed_terms, &config_terms))
-				return -ENOMEM;
+					config_term_common)) {
+				ret = -EINVAL;
+				goto out_err;
+			}
+			if (get_config_terms(parsed_terms, &config_terms)) {
+				ret = -ENOMEM;
+				goto out_err;
+			}
 		}
 
 		if (__add_event(list, idx, &attr, /*init_attr*/true, config_name ?: name,
-				metric_id, pmu, &config_terms, /*auto_merge_stats=*/false,
-				/*cpu_list=*/NULL,
-				/*alternate_hw_config=*/PERF_COUNT_HW_MAX) == NULL)
-			return -ENOMEM;
+				metric_id, pmu, &config_terms, first_wildcard_match,
+				cpus, /*alternate_hw_config=*/PERF_COUNT_HW_MAX) == NULL)
+			ret = -ENOMEM;
 
+		if (first_wildcard_match == NULL)
+			first_wildcard_match = container_of(list->prev, struct evsel, core.node);
 		free_config_terms(&config_terms);
+		if (ret)
+			goto out_err;
 	}
+out_err:
+	perf_cpu_map__put(cpus);
 	return found_supported ? 0 : -EINVAL;
 }
 
@@ -548,105 +599,82 @@ static int add_tracepoint(struct parse_events_state *parse_state,
 	return 0;
 }
 
-static int add_tracepoint_multi_event(struct parse_events_state *parse_state,
-				      struct list_head *list,
-				      const char *sys_name, const char *evt_name,
-				      struct parse_events_error *err,
-				      struct parse_events_terms *head_config, YYLTYPE *loc)
-{
-	char *evt_path;
-	struct io_dirent64 *evt_ent;
-	struct io_dir evt_dir;
-	int ret = 0, found = 0;
-
-	evt_path = get_events_file(sys_name);
-	if (!evt_path) {
-		tracepoint_error(err, errno, sys_name, evt_name, loc->first_column);
-		return -1;
-	}
-	io_dir__init(&evt_dir, open(evt_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY));
-	if (evt_dir.dirfd < 0) {
-		put_events_file(evt_path);
-		tracepoint_error(err, errno, sys_name, evt_name, loc->first_column);
-		return -1;
-	}
+struct add_tracepoint_multi_args {
+	struct parse_events_state *parse_state;
+	struct list_head *list;
+	const char *sys_glob;
+	const char *evt_glob;
+	struct parse_events_error *err;
+	struct parse_events_terms *head_config;
+	YYLTYPE *loc;
+	int found;
+};
 
-	while (!ret && (evt_ent = io_dir__readdir(&evt_dir))) {
-		if (!strcmp(evt_ent->d_name, ".")
-		    || !strcmp(evt_ent->d_name, "..")
-		    || !strcmp(evt_ent->d_name, "enable")
-		    || !strcmp(evt_ent->d_name, "filter"))
-			continue;
+static int add_tracepoint_multi_event_cb(void *state, const char *sys_name, const char *evt_name)
+{
+	struct add_tracepoint_multi_args *args = state;
+	int ret;
 
-		if (!strglobmatch(evt_ent->d_name, evt_name))
-			continue;
+	if (!strglobmatch(evt_name, args->evt_glob))
+		return 0;
 
-		found++;
+	args->found++;
+	ret = add_tracepoint(args->parse_state, args->list, sys_name, evt_name,
+			     args->err, args->head_config, args->loc);
 
-		ret = add_tracepoint(parse_state, list, sys_name, evt_ent->d_name,
-				     err, head_config, loc);
-	}
+	return ret;
+}
 
-	if (!found) {
-		tracepoint_error(err, ENOENT, sys_name, evt_name, loc->first_column);
-		ret = -1;
+static int add_tracepoint_multi_event(struct add_tracepoint_multi_args *args, const char *sys_name)
+{
+	if (strpbrk(args->evt_glob, "*?") == NULL) {
+		/* Not a glob. */
+		args->found++;
+		return add_tracepoint(args->parse_state, args->list, sys_name, args->evt_glob,
+				      args->err, args->head_config, args->loc);
 	}
 
-	put_events_file(evt_path);
-	close(evt_dir.dirfd);
-	return ret;
+	return tp_pmu__for_each_tp_event(sys_name, args, add_tracepoint_multi_event_cb);
 }
 
-static int add_tracepoint_event(struct parse_events_state *parse_state,
-				struct list_head *list,
-				const char *sys_name, const char *evt_name,
-				struct parse_events_error *err,
-				struct parse_events_terms *head_config, YYLTYPE *loc)
+static int add_tracepoint_multi_sys_cb(void *state, const char *sys_name)
 {
-	return strpbrk(evt_name, "*?") ?
-		add_tracepoint_multi_event(parse_state, list, sys_name, evt_name,
-					   err, head_config, loc) :
-		add_tracepoint(parse_state, list, sys_name, evt_name,
-			       err, head_config, loc);
+	struct add_tracepoint_multi_args *args = state;
+
+	if (!strglobmatch(sys_name, args->sys_glob))
+		return 0;
+
+	return add_tracepoint_multi_event(args, sys_name);
 }
 
 static int add_tracepoint_multi_sys(struct parse_events_state *parse_state,
 				    struct list_head *list,
-				    const char *sys_name, const char *evt_name,
+				    const char *sys_glob, const char *evt_glob,
 				    struct parse_events_error *err,
 				    struct parse_events_terms *head_config, YYLTYPE *loc)
 {
-	struct io_dirent64 *events_ent;
-	struct io_dir events_dir;
-	int ret = 0;
-	char *events_dir_path = get_tracing_file("events");
+	struct add_tracepoint_multi_args args = {
+		.parse_state = parse_state,
+		.list = list,
+		.sys_glob = sys_glob,
+		.evt_glob = evt_glob,
+		.err = err,
+		.head_config = head_config,
+		.loc = loc,
+		.found = 0,
+	};
+	int ret;
 
-	if (!events_dir_path) {
-		tracepoint_error(err, errno, sys_name, evt_name, loc->first_column);
-		return -1;
+	if (strpbrk(sys_glob, "*?") == NULL) {
+		/* Not a glob. */
+		ret = add_tracepoint_multi_event(&args, sys_glob);
+	} else {
+		ret = tp_pmu__for_each_tp_sys(&args, add_tracepoint_multi_sys_cb);
 	}
-	io_dir__init(&events_dir, open(events_dir_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY));
-	put_events_file(events_dir_path);
-	if (events_dir.dirfd < 0) {
-		tracepoint_error(err, errno, sys_name, evt_name, loc->first_column);
-		return -1;
+	if (args.found == 0) {
+		tracepoint_error(err, ENOENT, sys_glob, evt_glob, loc->first_column);
+		return -ENOENT;
 	}
-
-	while (!ret && (events_ent = io_dir__readdir(&events_dir))) {
-		if (!strcmp(events_ent->d_name, ".")
-		    || !strcmp(events_ent->d_name, "..")
-		    || !strcmp(events_ent->d_name, "enable")
-		    || !strcmp(events_ent->d_name, "header_event")
-		    || !strcmp(events_ent->d_name, "header_page"))
-			continue;
-
-		if (!strglobmatch(events_ent->d_name, sys_name))
-			continue;
-
-		ret = add_tracepoint_event(parse_state, list, events_ent->d_name,
-					   evt_name, err, head_config, loc);
-	}
-	close(events_dir.dirfd);
 	return ret;
 }
 
@@ -805,6 +833,7 @@ const char *parse_events__term_type_str(enum parse_events__term_type term_type)
 		[PARSE_EVENTS__TERM_TYPE_RAW]                   = "raw",
 		[PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE]          = "legacy-cache",
 		[PARSE_EVENTS__TERM_TYPE_HARDWARE]              = "hardware",
+		[PARSE_EVENTS__TERM_TYPE_CPU]			= "cpu",
 	};
 	if ((unsigned int)term_type >= __PARSE_EVENTS__TERM_TYPE_NR)
 		return "unknown term";
@@ -834,6 +863,7 @@ config_term_avail(enum parse_events__term_type term_type, struct parse_events_er
 	case PARSE_EVENTS__TERM_TYPE_METRIC_ID:
 	case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD:
 	case PARSE_EVENTS__TERM_TYPE_PERCORE:
+	case PARSE_EVENTS__TERM_TYPE_CPU:
 		return true;
 	case PARSE_EVENTS__TERM_TYPE_USER:
 	case PARSE_EVENTS__TERM_TYPE_SAMPLE_FREQ:
@@ -981,6 +1011,32 @@ do {									   \
 			return -EINVAL;
 		}
 		break;
+	case PARSE_EVENTS__TERM_TYPE_CPU: {
+		struct perf_cpu_map *map;
+
+		if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) {
+			if (term->val.num >= (u64)cpu__max_present_cpu().cpu) {
+				parse_events_error__handle(err, term->err_val,
+							strdup("too big"),
+							/*help=*/NULL);
+				return -EINVAL;
+			}
+			break;
+		}
+		assert(term->type_val == PARSE_EVENTS__TERM_TYPE_STR);
+		if (perf_pmus__find(term->val.str) != NULL)
+			break;
+
+		map = perf_cpu_map__new(term->val.str);
+		if (!map) {
+			parse_events_error__handle(err, term->err_val,
+						   strdup("not a valid PMU or CPU number"),
+						   /*help=*/NULL);
+			return -EINVAL;
+		}
+		perf_cpu_map__put(map);
+		break;
+	}
 	case PARSE_EVENTS__TERM_TYPE_DRV_CFG:
 	case PARSE_EVENTS__TERM_TYPE_USER:
 	case PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE:
@@ -1108,6 +1164,7 @@ static int config_term_tracepoint(struct perf_event_attr *attr,
 	case PARSE_EVENTS__TERM_TYPE_RAW:
 	case PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE:
 	case PARSE_EVENTS__TERM_TYPE_HARDWARE:
+	case PARSE_EVENTS__TERM_TYPE_CPU:
 	default:
 		if (err) {
 			parse_events_error__handle(err, term->err_term,
@@ -1242,6 +1299,7 @@ do {								\
 		case PARSE_EVENTS__TERM_TYPE_RAW:
 		case PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE:
 		case PARSE_EVENTS__TERM_TYPE_HARDWARE:
+		case PARSE_EVENTS__TERM_TYPE_CPU:
 		default:
 			break;
 		}
@@ -1296,6 +1354,7 @@ static int get_config_chgs(struct perf_pmu *pmu, struct parse_events_terms *head
 		case PARSE_EVENTS__TERM_TYPE_RAW:
 		case PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE:
 		case PARSE_EVENTS__TERM_TYPE_HARDWARE:
+		case PARSE_EVENTS__TERM_TYPE_CPU:
 		default:
 			break;
 		}
@@ -1324,22 +1383,20 @@ int parse_events_add_tracepoint(struct parse_events_state *parse_state,
 			return -EINVAL;
 	}
 
-	if (strpbrk(sys, "*?"))
-		return add_tracepoint_multi_sys(parse_state, list, sys, event,
-						err, head_config, loc);
-	else
-		return add_tracepoint_event(parse_state, list, sys, event,
-					    err, head_config, loc);
+	return add_tracepoint_multi_sys(parse_state, list, sys, event,
+					err, head_config, loc);
 }
 
 static int __parse_events_add_numeric(struct parse_events_state *parse_state,
 				struct list_head *list,
 				struct perf_pmu *pmu, u32 type, u32 extended_type,
-				u64 config, const struct parse_events_terms *head_config)
+				u64 config, const struct parse_events_terms *head_config,
+				struct evsel *first_wildcard_match)
 {
 	struct perf_event_attr attr;
 	LIST_HEAD(config_terms);
 	const char *name, *metric_id;
+	struct perf_cpu_map *cpus;
 	int ret;
 
 	memset(&attr, 0, sizeof(attr));
@@ -1361,10 +1418,11 @@ static int __parse_events_add_numeric(struct parse_events_state *parse_state,
 
 	name = get_config_name(head_config);
 	metric_id = get_config_metric_id(head_config);
+	cpus = get_config_cpu(head_config);
 	ret = __add_event(list, &parse_state->idx, &attr, /*init_attr*/true, name,
-			  metric_id, pmu, &config_terms, /*auto_merge_stats=*/false,
-			  /*cpu_list=*/NULL, /*alternate_hw_config=*/PERF_COUNT_HW_MAX
-		) == NULL ? -ENOMEM : 0;
+			metric_id, pmu, &config_terms, first_wildcard_match,
+			cpus, /*alternate_hw_config=*/PERF_COUNT_HW_MAX) ? 0 : -ENOMEM;
+	perf_cpu_map__put(cpus);
 	free_config_terms(&config_terms);
 	return ret;
 }
@@ -1380,6 +1438,7 @@ int parse_events_add_numeric(struct parse_events_state *parse_state,
 
 	/* Wildcards on numeric values are only supported by core PMUs. */
 	if (wildcard && perf_pmus__supports_extended_type()) {
+		struct evsel *first_wildcard_match = NULL;
 		while ((pmu = perf_pmus__scan_core(pmu)) != NULL) {
 			int ret;
 
@@ -1389,15 +1448,20 @@ int parse_events_add_numeric(struct parse_events_state *parse_state,
 
 			ret = __parse_events_add_numeric(parse_state, list, pmu,
 							 type, pmu->type,
-							 config, head_config);
+							 config, head_config,
+							 first_wildcard_match);
 			if (ret)
 				return ret;
+			if (first_wildcard_match == NULL)
+				first_wildcard_match =
+					container_of(list->prev, struct evsel, core.node);
 		}
 		if (found_supported)
 			return 0;
 	}
 	return __parse_events_add_numeric(parse_state, list, perf_pmus__find_by_type(type),
-					type, /*extended_type=*/0, config, head_config);
+					type, /*extended_type=*/0, config, head_config,
+					/*first_wildcard_match=*/NULL);
 }
 
 static bool config_term_percore(struct list_head *config_terms)
@@ -1415,7 +1479,7 @@ static bool config_term_percore(struct list_head *config_terms)
 static int parse_events_add_pmu(struct parse_events_state *parse_state,
 				struct list_head *list, struct perf_pmu *pmu,
 				const struct parse_events_terms *const_parsed_terms,
-				bool auto_merge_stats, u64 alternate_hw_config)
+				struct evsel *first_wildcard_match, u64 alternate_hw_config)
 {
 	struct perf_event_attr attr;
 	struct perf_pmu_info info;
@@ -1424,6 +1488,7 @@ static int parse_events_add_pmu(struct parse_events_state *parse_state,
 	LIST_HEAD(config_terms);
 	struct parse_events_terms parsed_terms;
 	bool alias_rewrote_terms = false;
+	struct perf_cpu_map *term_cpu = NULL;
 
 	if (verbose > 1) {
 		struct strbuf sb;
@@ -1451,7 +1516,7 @@ static int parse_events_add_pmu(struct parse_events_state *parse_state,
 		evsel = __add_event(list, &parse_state->idx, &attr,
 				    /*init_attr=*/true, /*name=*/NULL,
 				    /*metric_id=*/NULL, pmu,
-				    /*config_terms=*/NULL, auto_merge_stats,
+				    /*config_terms=*/NULL, first_wildcard_match,
 				    /*cpu_list=*/NULL, alternate_hw_config);
 		return evsel ? 0 : -ENOMEM;
 	}
@@ -1518,11 +1583,12 @@ static int parse_events_add_pmu(struct parse_events_state *parse_state,
 		return -EINVAL;
 	}
 
+	term_cpu = get_config_cpu(&parsed_terms);
 	evsel = __add_event(list, &parse_state->idx, &attr, /*init_attr=*/true,
 			    get_config_name(&parsed_terms),
 			    get_config_metric_id(&parsed_terms), pmu,
-			    &config_terms, auto_merge_stats, /*cpu_list=*/NULL,
-			    alternate_hw_config);
+			    &config_terms, first_wildcard_match, term_cpu, alternate_hw_config);
+	perf_cpu_map__put(term_cpu);
 	if (!evsel) {
 		parse_events_terms__exit(&parsed_terms);
 		return -ENOMEM;
@@ -1539,6 +1605,10 @@ static int parse_events_add_pmu(struct parse_events_state *parse_state,
 	evsel->scale = info.scale;
 	evsel->per_pkg = info.per_pkg;
 	evsel->snapshot = info.snapshot;
+	evsel->retirement_latency.mean = info.retirement_latency_mean;
+	evsel->retirement_latency.min = info.retirement_latency_min;
+	evsel->retirement_latency.max = info.retirement_latency_max;
+
 	return 0;
 }
 
@@ -1554,6 +1624,7 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state,
 	int ok = 0;
 	const char *config;
 	struct parse_events_terms parsed_terms;
+	struct evsel *first_wildcard_match = NULL;
 
 	*listp = NULL;
 
@@ -1585,8 +1656,7 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state,
 
 	INIT_LIST_HEAD(list);
 
-	while ((pmu = perf_pmus__scan(pmu)) != NULL) {
-		bool auto_merge_stats;
+	while ((pmu = perf_pmus__scan_for_event(pmu, event_name)) != NULL) {
 
 		if (parse_events__filter_pmu(parse_state, pmu))
 			continue;
@@ -1594,9 +1664,8 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state,
 		if (!perf_pmu__have_event(pmu, event_name))
 			continue;
 
-		auto_merge_stats = perf_pmu__auto_merge_stats(pmu);
 		if (!parse_events_add_pmu(parse_state, list, pmu,
-					  &parsed_terms, auto_merge_stats, hw_config)) {
+					  &parsed_terms, first_wildcard_match, hw_config)) {
 			struct strbuf sb;
 
 			strbuf_init(&sb, /*hint=*/ 0);
@@ -1605,11 +1674,13 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state,
 			strbuf_release(&sb);
 			ok++;
 		}
+		if (first_wildcard_match == NULL)
+			first_wildcard_match = container_of(list->prev, struct evsel, core.node);
 	}
 
 	if (parse_state->fake_pmu) {
 		if (!parse_events_add_pmu(parse_state, list, perf_pmus__fake_pmu(), &parsed_terms,
-					  /*auto_merge_stats=*/true, hw_config)) {
+					 first_wildcard_match, hw_config)) {
 			struct strbuf sb;
 
 			strbuf_init(&sb, /*hint=*/ 0);
@@ -1640,6 +1711,7 @@ int parse_events_multi_pmu_add_or_add_pmu(struct parse_events_state *parse_state
 	struct perf_pmu *pmu;
 	int ok = 0;
 	char *help;
+	struct evsel *first_wildcard_match = NULL;
 
 	*listp = malloc(sizeof(**listp));
 	if (!*listp)
@@ -1650,32 +1722,35 @@ int parse_events_multi_pmu_add_or_add_pmu(struct parse_events_state *parse_state
 	/* Attempt to add to list assuming event_or_pmu is a PMU name. */
 	pmu = perf_pmus__find(event_or_pmu);
 	if (pmu && !parse_events_add_pmu(parse_state, *listp, pmu, const_parsed_terms,
-					 /*auto_merge_stats=*/false,
+					 first_wildcard_match,
 					 /*alternate_hw_config=*/PERF_COUNT_HW_MAX))
 		return 0;
 
 	if (parse_state->fake_pmu) {
 		if (!parse_events_add_pmu(parse_state, *listp, perf_pmus__fake_pmu(),
 					  const_parsed_terms,
-					  /*auto_merge_stats=*/false,
+					  first_wildcard_match,
 					  /*alternate_hw_config=*/PERF_COUNT_HW_MAX))
 			return 0;
 	}
 
 	pmu = NULL;
 	/* Failed to add, try wildcard expansion of event_or_pmu as a PMU name. */
-	while ((pmu = perf_pmus__scan(pmu)) != NULL) {
-		if (!parse_events__filter_pmu(parse_state, pmu) &&
-		    perf_pmu__wildcard_match(pmu, event_or_pmu)) {
-			bool auto_merge_stats = perf_pmu__auto_merge_stats(pmu);
-
-			if (!parse_events_add_pmu(parse_state, *listp, pmu,
-						  const_parsed_terms,
-						  auto_merge_stats,
-						  /*alternate_hw_config=*/PERF_COUNT_HW_MAX)) {
-				ok++;
-				parse_state->wild_card_pmus = true;
-			}
+	while ((pmu = perf_pmus__scan_matching_wildcard(pmu, event_or_pmu)) != NULL) {
+
+		if (parse_events__filter_pmu(parse_state, pmu))
+			continue;
+
+		if (!parse_events_add_pmu(parse_state, *listp, pmu,
+					  const_parsed_terms,
+					  first_wildcard_match,
+					  /*alternate_hw_config=*/PERF_COUNT_HW_MAX)) {
+			ok++;
+			parse_state->wild_card_pmus = true;
+		}
+		if (first_wildcard_match == NULL) {
+			first_wildcard_match =
+				container_of((*listp)->prev, struct evsel, core.node);
 		}
 	}
 	if (ok)
@@ -1733,13 +1808,11 @@ static int parse_events__modifier_list(struct parse_events_state *parse_state,
 		int eH = group ? evsel->core.attr.exclude_host : 0;
 		int eG = group ? evsel->core.attr.exclude_guest : 0;
 		int exclude = eu | ek | eh;
-		int exclude_GH = group ? evsel->exclude_GH : 0;
+		int exclude_GH = eG | eH;
 
 		if (mod.user) {
 			if (!exclude)
 				exclude = eu = ek = eh = 1;
-			if (!exclude_GH && !perf_guest && exclude_GH_default)
-				eG = 1;
 			eu = 0;
 		}
 		if (mod.kernel) {
@@ -1762,6 +1835,13 @@ static int parse_events__modifier_list(struct parse_events_state *parse_state,
 				exclude_GH = eG = eH = 1;
 			eH = 0;
 		}
+		if (!exclude_GH && exclude_GH_default) {
+			if (perf_host)
+				eG = 1;
+			else if (perf_guest)
+				eH = 1;
+		}
+
 		evsel->core.attr.exclude_user   = eu;
 		evsel->core.attr.exclude_kernel = ek;
 		evsel->core.attr.exclude_hv     = eh;
@@ -2032,6 +2112,11 @@ static int evlist__cmp(void *_fg_idx, const struct list_head *l, const struct li
 	return arch_evlist__cmp(lhs, rhs);
 }
 
+int __weak arch_evlist__add_required_events(struct list_head *list __always_unused)
+{
+	return 0;
+}
+
 static int parse_events__sort_events_and_fix_groups(struct list_head *list)
 {
 	int idx = 0, force_grouped_idx = -1;
@@ -2043,6 +2128,11 @@ static int parse_events__sort_events_and_fix_groups(struct list_head *list)
 	struct evsel *force_grouped_leader = NULL;
 	bool last_event_was_forced_leader = false;
 
+	/* On x86 topdown metrics events require a slots event. */
+	ret = arch_evlist__add_required_events(list);
+	if (ret)
+		return ret;
+
 	/*
 	 * Compute index to insert ungrouped events at. Place them where the
 	 * first ungrouped event appears.
@@ -2196,14 +2286,23 @@ int __parse_events(struct evlist *evlist, const char *str, const char *pmu_filte
 	if (ret2 < 0)
 		return ret;
 
-	if (ret2 && warn_if_reordered && !parse_state.wild_card_pmus)
-		pr_warning("WARNING: events were regrouped to match PMUs\n");
-
 	/*
 	 * Add list to the evlist even with errors to allow callers to clean up.
 	 */
 	evlist__splice_list_tail(evlist, &parse_state.list);
 
+	if (ret2 && warn_if_reordered && !parse_state.wild_card_pmus) {
+		pr_warning("WARNING: events were regrouped to match PMUs\n");
+
+		if (verbose > 0) {
+			struct strbuf sb = STRBUF_INIT;
+
+			evlist__uniquify_evsel_names(evlist, &stat_config);
+			evlist__format_evsels(evlist, &sb, 2048);
+			pr_debug("evlist after sorting/fixing: '%s'\n", sb.buf);
+			strbuf_release(&sb);
+		}
+	}
 	if (!ret) {
 		struct evsel *last;
 
@@ -2456,12 +2555,17 @@ foreach_evsel_in_last_glob(struct evlist *evlist,
 	return 0;
 }
 
+/* Will a tracepoint filter work for str or should a BPF filter be used? */
+static bool is_possible_tp_filter(const char *str)
+{
+	return strstr(str, "uid") == NULL;
+}
+
 static int set_filter(struct evsel *evsel, const void *arg)
 {
 	const char *str = arg;
-	bool found = false;
 	int nr_addr_filters = 0;
-	struct perf_pmu *pmu = NULL;
+	struct perf_pmu *pmu;
 
 	if (evsel == NULL) {
 		fprintf(stderr,
@@ -2469,7 +2573,7 @@ static int set_filter(struct evsel *evsel, const void *arg)
 		return -1;
 	}
 
-	if (evsel->core.attr.type == PERF_TYPE_TRACEPOINT) {
+	if (evsel->core.attr.type == PERF_TYPE_TRACEPOINT && is_possible_tp_filter(str)) {
 		if (evsel__append_tp_filter(evsel, str) < 0) {
 			fprintf(stderr,
 				"not enough memory to hold filter string\n");
@@ -2479,16 +2583,11 @@ static int set_filter(struct evsel *evsel, const void *arg)
 		return 0;
 	}
 
-	while ((pmu = perf_pmus__scan(pmu)) != NULL)
-		if (pmu->type == evsel->core.attr.type) {
-			found = true;
-			break;
-		}
-
-	if (found)
+	pmu = evsel__find_pmu(evsel);
+	if (pmu) {
 		perf_pmu__scan_file(pmu, "nr_addr_filters",
 				    "%d", &nr_addr_filters);
-
+	}
 	if (!nr_addr_filters)
 		return perf_bpf_filter__parse(&evsel->bpf_filters, str);
 
@@ -2510,6 +2609,30 @@ int parse_filter(const struct option *opt, const char *str,
 					  (const void *)str);
 }
 
+int parse_uid_filter(struct evlist *evlist, uid_t uid)
+{
+	struct option opt = {
+		.value = &evlist,
+	};
+	char buf[128];
+	int ret;
+
+	snprintf(buf, sizeof(buf), "uid == %d", uid);
+	ret = parse_filter(&opt, buf, /*unset=*/0);
+	if (ret) {
+		if (use_browser >= 1) {
+			/*
+			 * Use ui__warning so a pop up appears above the
+			 * underlying BPF error message.
+			 */
+			ui__warning("Failed to add UID filtering that uses BPF filtering.\n");
+		} else {
+			fprintf(stderr, "Failed to add UID filtering that uses BPF filtering.\n");
+		}
+	}
+	return ret;
+}
+
 static int add_exclude_perf_filter(struct evsel *evsel,
 				   const void *arg __maybe_unused)
 {