Merge tag 'perf-tools-for-v6.9-2024-03-13' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools

Pull perf tools updates from Namhyung Kim:
 "perf stat:

   - Support new 'cluster' aggregation mode for shared resources
     depending on the hardware configuration:

        $ sudo perf stat -a --per-cluster -e cycles,instructions sleep 1

         Performance counter stats for 'system wide':

        S0-D0-CLS0    2         85,051,822      cycles
        S0-D0-CLS0    2         73,909,908      instructions      #    0.87  insn per cycle
        S0-D0-CLS2    2         93,365,918      cycles
        S0-D0-CLS2    2         83,006,158      instructions      #    0.89  insn per cycle
        S0-D0-CLS4    2        104,157,523      cycles
        S0-D0-CLS4    2         53,234,396      instructions      #    0.51  insn per cycle
        S0-D0-CLS6    2         65,891,079      cycles
        S0-D0-CLS6    2         41,478,273      instructions      #    0.63  insn per cycle

               1.002407989 seconds time elapsed

   - Various fixes and cleanups for event metrics including NaN handling

  perf script:

   - Use libcapstone if available to disassemble the instructions. This
     enables 'perf script -F disasm' and 'perf script --insn-trace=disasm'
     (for Intel-PT):

        $ perf script -F event,ip,disasm
        cycles:P:  ffffffffa988d428             wrmsr
        cycles:P:  ffffffffa9839d25             movq %rax, %r14
        cycles:P:  ffffffffa9cdcaf0             endbr64
        cycles:P:  ffffffffa988d428             wrmsr
        cycles:P:  ffffffffa988d428             wrmsr
        cycles:P:  ffffffffaa401f86             iretq
        cycles:P:  ffffffffa99c4de5             movq 0x30(%rcx), %r8
        cycles:P:  ffffffffa988d428             wrmsr
        cycles:P:  ffffffffaa401f86             iretq
        cycles:P:  ffffffffa9907983             movl 0x68(%rbx), %eax
        cycles:P:  ffffffffa988d428             wrmsr

   - Expose sample ID / stream ID to python scripts

  perf test:

   - Add more perf test cases from Redhat internal test suites. This
     time it adds the base infra and a few perf probe tests. More to
     come. :)

   - Add 'perf test -p' for parallel execution and fix some issues found
     by the parallel test

   - Support symbol test to print symbols in given (active) module:

        $ perf test -F -v Symbols --dso /lib/modules/$(uname -r)/kernel/fs/ext4/ext4.ko
        --- start ---
        Testing /lib/modules/6.5.13-1rodete2-amd64/kernel/fs/ext4/ext4.ko
        Overlapping symbols:
         7a990-7a9a0 l __pfx_ext4_exit_fs
         7a990-7a9a0 g __pfx_cleanup_module
        Overlapping symbols:
         7a9a0-7aa1c l ext4_exit_fs
         7a9a0-7aa1c g cleanup_module
        ...

  JSON metric updates:

   - A new round of Intel metric updates

   - Support Power11 PVR (compatible to Power10)

   - Fix cache latency events on Zen 4 to set SliceId properly

  Internal:

   - Fix reference counting for 'map' data structure, tireless work from
     Ian!

   - More memory optimization for struct thread and annotate histogram.
     Now, 'perf report' (TUI) and 'perf annotate' should be much
     lighter-weight in terms of memory footprint

   - Support cross-arch perf register access. Clean up the build
     configuration so that it can detect arch-register support at
     runtime. This can allow to parse register data in sample which was
     recorded in a different arch

  Others:

   - Sync task state in 'perf sched' to kernel using trace event fields.
     The task states have been changed so tools cannot assume a fixed
     encoding

   - Clean up 'perf mem' to generalize the arch-specific events

   - Add support for local and global variables to data type profiling.
     This would increase the success rate of type resolution with DWARF

   - Add short option -H for --hierarchy in 'perf report' and 'perf top'"

* tag 'perf-tools-for-v6.9-2024-03-13' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (154 commits)
  perf annotate: Add comments in the data structures
  perf annotate: Remove sym_hist.addr[] array
  perf annotate: Calculate instruction overhead using hashmap
  perf annotate: Add a hashmap for symbol histogram
  perf threads: Reduce table size from 256 to 8
  perf threads: Switch from rbtree to hashmap
  perf threads: Move threads to its own files
  perf machine: Move machine's threads into its own abstraction
  perf machine: Move fprintf to for_each loop and a callback
  perf trace: Ignore thread hashing in summary
  perf report: Sort child tasks by tid
  perf vendor events amd: Fix Zen 4 cache latency events
  perf version: Display availability of OpenCSD support
  perf vendor events intel: Add umasks/occ_sel to PCU events.
  perf map: Fix map reference count issues
  libperf evlist: Avoid out-of-bounds access
  perf lock contention: Account contending locks too
  perf metrics: Fix segv for metrics with no events
  perf metrics: Fix metric matching
  perf pmu: Fix a potential memory leak in perf_pmu__lookup()
  ...

1 files changed, 114 insertions, 106 deletions

diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c
index dd6065afbbaf..b248c433529a 100644
--- a/tools/perf/builtin-sched.c
+++ b/tools/perf/builtin-sched.c
@@ -92,24 +92,6 @@ struct sched_atom {
 	struct task_desc	*wakee;
 };
 
-#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKWP"
-
-/* task state bitmask, copied from include/linux/sched.h */
-#define TASK_RUNNING		0
-#define TASK_INTERRUPTIBLE	1
-#define TASK_UNINTERRUPTIBLE	2
-#define __TASK_STOPPED		4
-#define __TASK_TRACED		8
-/* in tsk->exit_state */
-#define EXIT_DEAD		16
-#define EXIT_ZOMBIE		32
-#define EXIT_TRACE		(EXIT_ZOMBIE | EXIT_DEAD)
-/* in tsk->state again */
-#define TASK_DEAD		64
-#define TASK_WAKEKILL		128
-#define TASK_WAKING		256
-#define TASK_PARKED		512
-
 enum thread_state {
 	THREAD_SLEEPING = 0,
 	THREAD_WAIT_CPU,
@@ -266,7 +248,7 @@ struct thread_runtime {
 	u64 total_preempt_time;
 	u64 total_delay_time;
 
-	int last_state;
+	char last_state;
 
 	char shortname[3];
 	bool comm_changed;
@@ -436,7 +418,7 @@ static void add_sched_event_wakeup(struct perf_sched *sched, struct task_desc *t
 }
 
 static void add_sched_event_sleep(struct perf_sched *sched, struct task_desc *task,
-				  u64 timestamp, u64 task_state __maybe_unused)
+				  u64 timestamp, const char task_state __maybe_unused)
 {
 	struct sched_atom *event = get_new_event(task, timestamp);
 
@@ -860,7 +842,7 @@ static int replay_switch_event(struct perf_sched *sched,
 		   *next_comm  = evsel__strval(evsel, sample, "next_comm");
 	const u32 prev_pid = evsel__intval(evsel, sample, "prev_pid"),
 		  next_pid = evsel__intval(evsel, sample, "next_pid");
-	const u64 prev_state = evsel__intval(evsel, sample, "prev_state");
+	const char prev_state = evsel__taskstate(evsel, sample, "prev_state");
 	struct task_desc *prev, __maybe_unused *next;
 	u64 timestamp0, timestamp = sample->time;
 	int cpu = sample->cpu;
@@ -1050,13 +1032,6 @@ static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread)
 	return 0;
 }
 
-static char sched_out_state(u64 prev_state)
-{
-	const char *str = TASK_STATE_TO_CHAR_STR;
-
-	return str[prev_state];
-}
-
 static int
 add_sched_out_event(struct work_atoms *atoms,
 		    char run_state,
@@ -1132,7 +1107,7 @@ static int latency_switch_event(struct perf_sched *sched,
 {
 	const u32 prev_pid = evsel__intval(evsel, sample, "prev_pid"),
 		  next_pid = evsel__intval(evsel, sample, "next_pid");
-	const u64 prev_state = evsel__intval(evsel, sample, "prev_state");
+	const char prev_state = evsel__taskstate(evsel, sample, "prev_state");
 	struct work_atoms *out_events, *in_events;
 	struct thread *sched_out, *sched_in;
 	u64 timestamp0, timestamp = sample->time;
@@ -1168,7 +1143,7 @@ static int latency_switch_event(struct perf_sched *sched,
 			goto out_put;
 		}
 	}
-	if (add_sched_out_event(out_events, sched_out_state(prev_state), timestamp))
+	if (add_sched_out_event(out_events, prev_state, timestamp))
 		return -1;
 
 	in_events = thread_atoms_search(&sched->atom_root, sched_in, &sched->cmp_pid);
@@ -2033,24 +2008,12 @@ static void timehist_header(struct perf_sched *sched)
 	printf("\n");
 }
 
-static char task_state_char(struct thread *thread, int state)
-{
-	static const char state_to_char[] = TASK_STATE_TO_CHAR_STR;
-	unsigned bit = state ? ffs(state) : 0;
-
-	/* 'I' for idle */
-	if (thread__tid(thread) == 0)
-		return 'I';
-
-	return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?';
-}
-
 static void timehist_print_sample(struct perf_sched *sched,
 				  struct evsel *evsel,
 				  struct perf_sample *sample,
 				  struct addr_location *al,
 				  struct thread *thread,
-				  u64 t, int state)
+				  u64 t, const char state)
 {
 	struct thread_runtime *tr = thread__priv(thread);
 	const char *next_comm = evsel__strval(evsel, sample, "next_comm");
@@ -2091,7 +2054,7 @@ static void timehist_print_sample(struct perf_sched *sched,
 	print_sched_time(tr->dt_run, 6);
 
 	if (sched->show_state)
-		printf(" %5c ", task_state_char(thread, state));
+		printf(" %5c ", thread__tid(thread) == 0 ? 'I' : state);
 
 	if (sched->show_next) {
 		snprintf(nstr, sizeof(nstr), "next: %s[%d]", next_comm, next_pid);
@@ -2163,9 +2126,9 @@ static void timehist_update_runtime_stats(struct thread_runtime *r,
 		else if (r->last_time) {
 			u64 dt_wait = tprev - r->last_time;
 
-			if (r->last_state == TASK_RUNNING)
+			if (r->last_state == 'R')
 				r->dt_preempt = dt_wait;
-			else if (r->last_state == TASK_UNINTERRUPTIBLE)
+			else if (r->last_state == 'D')
 				r->dt_iowait = dt_wait;
 			else
 				r->dt_sleep = dt_wait;
@@ -2590,7 +2553,7 @@ static int timehist_sched_change_event(struct perf_tool *tool,
 	struct thread_runtime *tr = NULL;
 	u64 tprev, t = sample->time;
 	int rc = 0;
-	int state = evsel__intval(evsel, sample, "prev_state");
+	const char state = evsel__taskstate(evsel, sample, "prev_state");
 
 	addr_location__init(&al);
 	if (machine__resolve(machine, &al, sample) < 0) {
@@ -3204,14 +3167,44 @@ static void perf_sched__merge_lat(struct perf_sched *sched)
 	}
 }
 
+static int setup_cpus_switch_event(struct perf_sched *sched)
+{
+	unsigned int i;
+
+	sched->cpu_last_switched = calloc(MAX_CPUS, sizeof(*(sched->cpu_last_switched)));
+	if (!sched->cpu_last_switched)
+		return -1;
+
+	sched->curr_pid = malloc(MAX_CPUS * sizeof(*(sched->curr_pid)));
+	if (!sched->curr_pid) {
+		zfree(&sched->cpu_last_switched);
+		return -1;
+	}
+
+	for (i = 0; i < MAX_CPUS; i++)
+		sched->curr_pid[i] = -1;
+
+	return 0;
+}
+
+static void free_cpus_switch_event(struct perf_sched *sched)
+{
+	zfree(&sched->curr_pid);
+	zfree(&sched->cpu_last_switched);
+}
+
 static int perf_sched__lat(struct perf_sched *sched)
 {
+	int rc = -1;
 	struct rb_node *next;
 
 	setup_pager();
 
+	if (setup_cpus_switch_event(sched))
+		return rc;
+
 	if (perf_sched__read_events(sched))
-		return -1;
+		goto out_free_cpus_switch_event;
 
 	perf_sched__merge_lat(sched);
 	perf_sched__sort_lat(sched);
@@ -3240,13 +3233,15 @@ static int perf_sched__lat(struct perf_sched *sched)
 	print_bad_events(sched);
 	printf("\n");
 
-	return 0;
+	rc = 0;
+
+out_free_cpus_switch_event:
+	free_cpus_switch_event(sched);
+	return rc;
 }
 
 static int setup_map_cpus(struct perf_sched *sched)
 {
-	struct perf_cpu_map *map;
-
 	sched->max_cpu.cpu  = sysconf(_SC_NPROCESSORS_CONF);
 
 	if (sched->map.comp) {
@@ -3255,16 +3250,15 @@ static int setup_map_cpus(struct perf_sched *sched)
 			return -1;
 	}
 
-	if (!sched->map.cpus_str)
-		return 0;
-
-	map = perf_cpu_map__new(sched->map.cpus_str);
-	if (!map) {
-		pr_err("failed to get cpus map from %s\n", sched->map.cpus_str);
-		return -1;
+	if (sched->map.cpus_str) {
+		sched->map.cpus = perf_cpu_map__new(sched->map.cpus_str);
+		if (!sched->map.cpus) {
+			pr_err("failed to get cpus map from %s\n", sched->map.cpus_str);
+			zfree(&sched->map.comp_cpus);
+			return -1;
+		}
 	}
 
-	sched->map.cpus = map;
 	return 0;
 }
 
@@ -3304,33 +3298,69 @@ static int setup_color_cpus(struct perf_sched *sched)
 
 static int perf_sched__map(struct perf_sched *sched)
 {
+	int rc = -1;
+
+	sched->curr_thread = calloc(MAX_CPUS, sizeof(*(sched->curr_thread)));
+	if (!sched->curr_thread)
+		return rc;
+
+	if (setup_cpus_switch_event(sched))
+		goto out_free_curr_thread;
+
 	if (setup_map_cpus(sched))
-		return -1;
+		goto out_free_cpus_switch_event;
 
 	if (setup_color_pids(sched))
-		return -1;
+		goto out_put_map_cpus;
 
 	if (setup_color_cpus(sched))
-		return -1;
+		goto out_put_color_pids;
 
 	setup_pager();
 	if (perf_sched__read_events(sched))
-		return -1;
+		goto out_put_color_cpus;
+
+	rc = 0;
 	print_bad_events(sched);
-	return 0;
+
+out_put_color_cpus:
+	perf_cpu_map__put(sched->map.color_cpus);
+
+out_put_color_pids:
+	perf_thread_map__put(sched->map.color_pids);
+
+out_put_map_cpus:
+	zfree(&sched->map.comp_cpus);
+	perf_cpu_map__put(sched->map.cpus);
+
+out_free_cpus_switch_event:
+	free_cpus_switch_event(sched);
+
+out_free_curr_thread:
+	zfree(&sched->curr_thread);
+	return rc;
 }
 
 static int perf_sched__replay(struct perf_sched *sched)
 {
+	int ret;
 	unsigned long i;
 
+	mutex_init(&sched->start_work_mutex);
+	mutex_init(&sched->work_done_wait_mutex);
+
+	ret = setup_cpus_switch_event(sched);
+	if (ret)
+		goto out_mutex_destroy;
+
 	calibrate_run_measurement_overhead(sched);
 	calibrate_sleep_measurement_overhead(sched);
 
 	test_calibrations(sched);
 
-	if (perf_sched__read_events(sched))
-		return -1;
+	ret = perf_sched__read_events(sched);
+	if (ret)
+		goto out_free_cpus_switch_event;
 
 	printf("nr_run_events:        %ld\n", sched->nr_run_events);
 	printf("nr_sleep_events:      %ld\n", sched->nr_sleep_events);
@@ -3355,7 +3385,14 @@ static int perf_sched__replay(struct perf_sched *sched)
 
 	sched->thread_funcs_exit = true;
 	destroy_tasks(sched);
-	return 0;
+
+out_free_cpus_switch_event:
+	free_cpus_switch_event(sched);
+
+out_mutex_destroy:
+	mutex_destroy(&sched->start_work_mutex);
+	mutex_destroy(&sched->work_done_wait_mutex);
+	return ret;
 }
 
 static void setup_sorting(struct perf_sched *sched, const struct option *options,
@@ -3590,28 +3627,7 @@ int cmd_sched(int argc, const char **argv)
 		.switch_event	    = replay_switch_event,
 		.fork_event	    = replay_fork_event,
 	};
-	unsigned int i;
-	int ret = 0;
-
-	mutex_init(&sched.start_work_mutex);
-	mutex_init(&sched.work_done_wait_mutex);
-	sched.curr_thread = calloc(MAX_CPUS, sizeof(*sched.curr_thread));
-	if (!sched.curr_thread) {
-		ret = -ENOMEM;
-		goto out;
-	}
-	sched.cpu_last_switched = calloc(MAX_CPUS, sizeof(*sched.cpu_last_switched));
-	if (!sched.cpu_last_switched) {
-		ret = -ENOMEM;
-		goto out;
-	}
-	sched.curr_pid = malloc(MAX_CPUS * sizeof(*sched.curr_pid));
-	if (!sched.curr_pid) {
-		ret = -ENOMEM;
-		goto out;
-	}
-	for (i = 0; i < MAX_CPUS; i++)
-		sched.curr_pid[i] = -1;
+	int ret;
 
 	argc = parse_options_subcommand(argc, argv, sched_options, sched_subcommands,
 					sched_usage, PARSE_OPT_STOP_AT_NON_OPTION);
@@ -3622,9 +3638,9 @@ int cmd_sched(int argc, const char **argv)
 	 * Aliased to 'perf script' for now:
 	 */
 	if (!strcmp(argv[0], "script")) {
-		ret = cmd_script(argc, argv);
+		return cmd_script(argc, argv);
 	} else if (strlen(argv[0]) > 2 && strstarts("record", argv[0])) {
-		ret = __cmd_record(argc, argv);
+		return __cmd_record(argc, argv);
 	} else if (strlen(argv[0]) > 2 && strstarts("latency", argv[0])) {
 		sched.tp_handler = &lat_ops;
 		if (argc > 1) {
@@ -3633,7 +3649,7 @@ int cmd_sched(int argc, const char **argv)
 				usage_with_options(latency_usage, latency_options);
 		}
 		setup_sorting(&sched, latency_options, latency_usage);
-		ret = perf_sched__lat(&sched);
+		return perf_sched__lat(&sched);
 	} else if (!strcmp(argv[0], "map")) {
 		if (argc) {
 			argc = parse_options(argc, argv, map_options, map_usage, 0);
@@ -3642,7 +3658,7 @@ int cmd_sched(int argc, const char **argv)
 		}
 		sched.tp_handler = &map_ops;
 		setup_sorting(&sched, latency_options, latency_usage);
-		ret = perf_sched__map(&sched);
+		return perf_sched__map(&sched);
 	} else if (strlen(argv[0]) > 2 && strstarts("replay", argv[0])) {
 		sched.tp_handler = &replay_ops;
 		if (argc) {
@@ -3650,7 +3666,7 @@ int cmd_sched(int argc, const char **argv)
 			if (argc)
 				usage_with_options(replay_usage, replay_options);
 		}
-		ret = perf_sched__replay(&sched);
+		return perf_sched__replay(&sched);
 	} else if (!strcmp(argv[0], "timehist")) {
 		if (argc) {
 			argc = parse_options(argc, argv, timehist_options,
@@ -3666,24 +3682,16 @@ int cmd_sched(int argc, const char **argv)
 				parse_options_usage(NULL, timehist_options, "w", true);
 			if (sched.show_next)
 				parse_options_usage(NULL, timehist_options, "n", true);
-			ret = -EINVAL;
-			goto out;
+			return -EINVAL;
 		}
 		ret = symbol__validate_sym_arguments();
 		if (ret)
-			goto out;
+			return ret;
 
-		ret = perf_sched__timehist(&sched);
+		return perf_sched__timehist(&sched);
 	} else {
 		usage_with_options(sched_usage, sched_options);
 	}
 
-out:
-	free(sched.curr_pid);
-	free(sched.cpu_last_switched);
-	free(sched.curr_thread);
-	mutex_destroy(&sched.start_work_mutex);
-	mutex_destroy(&sched.work_done_wait_mutex);
-
-	return ret;
+	return 0;
 }