[linux-block.git] / tools / perf / tests / switch-tracking.c

// SPDX-License-Identifier: GPL-2.0
#include <sys/time.h>
#include <sys/prctl.h>
#include <errno.h>
#include <time.h>
#include <stdlib.h>

#include "parse-events.h"
#include "evlist.h"
#include "evsel.h"
#include "thread_map.h"
#include "cpumap.h"
#include "tests.h"

static int spin_sleep(void)
{
	struct timeval start, now, diff, maxtime;
	struct timespec ts;
	int err, i;

	maxtime.tv_sec = 0;
	maxtime.tv_usec = 50000;

	err = gettimeofday(&start, NULL);
	if (err)
		return err;

	/* Spin for 50ms */
	while (1) {
		for (i = 0; i < 1000; i++)
			barrier();

		err = gettimeofday(&now, NULL);
		if (err)
			return err;

		timersub(&now, &start, &diff);
		if (timercmp(&diff, &maxtime, > /* For checkpatch */))
			break;
	}

	ts.tv_nsec = 50 * 1000 * 1000;
	ts.tv_sec = 0;

	/* Sleep for 50ms */
	err = nanosleep(&ts, NULL);
	if (err == EINTR)
		err = 0;

	return err;
}

struct switch_tracking {
	struct perf_evsel *switch_evsel;
	struct perf_evsel *cycles_evsel;
	pid_t *tids;
	int nr_tids;
	int comm_seen[4];
	int cycles_before_comm_1;
	int cycles_between_comm_2_and_comm_3;
	int cycles_after_comm_4;
};

static int check_comm(struct switch_tracking *switch_tracking,
		      union perf_event *event, const char *comm, int nr)
{
	if (event->header.type == PERF_RECORD_COMM &&
	    (pid_t)event->comm.pid == getpid() &&
	    (pid_t)event->comm.tid == getpid() &&
	    strcmp(event->comm.comm, comm) == 0) {
		if (switch_tracking->comm_seen[nr]) {
			pr_debug("Duplicate comm event\n");
			return -1;
		}
		switch_tracking->comm_seen[nr] = 1;
		pr_debug3("comm event: %s nr: %d\n", event->comm.comm, nr);
		return 1;
	}
	return 0;
}

static int check_cpu(struct switch_tracking *switch_tracking, int cpu)
{
	int i, nr = cpu + 1;

	if (cpu < 0)
		return -1;

	if (!switch_tracking->tids) {
		switch_tracking->tids = calloc(nr, sizeof(pid_t));
		if (!switch_tracking->tids)
			return -1;
		for (i = 0; i < nr; i++)
			switch_tracking->tids[i] = -1;
		switch_tracking->nr_tids = nr;
		return 0;
	}

	if (cpu >= switch_tracking->nr_tids) {
		void *addr;

		addr = realloc(switch_tracking->tids, nr * sizeof(pid_t));
		if (!addr)
			return -1;
		switch_tracking->tids = addr;
		for (i = switch_tracking->nr_tids; i < nr; i++)
			switch_tracking->tids[i] = -1;
		switch_tracking->nr_tids = nr;
		return 0;
	}

	return 0;
}

static int process_sample_event(struct perf_evlist *evlist,
				union perf_event *event,
				struct switch_tracking *switch_tracking)
{
	struct perf_sample sample;
	struct perf_evsel *evsel;
	pid_t next_tid, prev_tid;
	int cpu, err;

	if (perf_evlist__parse_sample(evlist, event, &sample)) {
		pr_debug("perf_evlist__parse_sample failed\n");
		return -1;
	}

	evsel = perf_evlist__id2evsel(evlist, sample.id);
	if (evsel == switch_tracking->switch_evsel) {
		next_tid = perf_evsel__intval(evsel, &sample, "next_pid");
		prev_tid = perf_evsel__intval(evsel, &sample, "prev_pid");
		cpu = sample.cpu;
		pr_debug3("sched_switch: cpu: %d prev_tid %d next_tid %d\n",
			  cpu, prev_tid, next_tid);
		err = check_cpu(switch_tracking, cpu);
		if (err)
			return err;
		/*
		 * Check for no missing sched_switch events i.e. that the
		 * evsel->system_wide flag has worked.
		 */
		if (switch_tracking->tids[cpu] != -1 &&
		    switch_tracking->tids[cpu] != prev_tid) {
			pr_debug("Missing sched_switch events\n");
			return -1;
		}
		switch_tracking->tids[cpu] = next_tid;
	}

	if (evsel == switch_tracking->cycles_evsel) {
		pr_debug3("cycles event\n");
		if (!switch_tracking->comm_seen[0])
			switch_tracking->cycles_before_comm_1 = 1;
		if (switch_tracking->comm_seen[1] &&
		    !switch_tracking->comm_seen[2])
			switch_tracking->cycles_between_comm_2_and_comm_3 = 1;
		if (switch_tracking->comm_seen[3])
			switch_tracking->cycles_after_comm_4 = 1;
	}

	return 0;
}

static int process_event(struct perf_evlist *evlist, union perf_event *event,
			 struct switch_tracking *switch_tracking)
{
	if (event->header.type == PERF_RECORD_SAMPLE)
		return process_sample_event(evlist, event, switch_tracking);

	if (event->header.type == PERF_RECORD_COMM) {
		int err, done = 0;

		err = check_comm(switch_tracking, event, "Test COMM 1", 0);
		if (err < 0)
			return -1;
		done += err;
		err = check_comm(switch_tracking, event, "Test COMM 2", 1);
		if (err < 0)
			return -1;
		done += err;
		err = check_comm(switch_tracking, event, "Test COMM 3", 2);
		if (err < 0)
			return -1;
		done += err;
		err = check_comm(switch_tracking, event, "Test COMM 4", 3);
		if (err < 0)
			return -1;
		done += err;
		if (done != 1) {
			pr_debug("Unexpected comm event\n");
			return -1;
		}
	}

	return 0;
}

struct event_node {
	struct list_head list;
	union perf_event *event;
	u64 event_time;
};

static int add_event(struct perf_evlist *evlist, struct list_head *events,
		     union perf_event *event)
{
	struct perf_sample sample;
	struct event_node *node;

	node = malloc(sizeof(struct event_node));
	if (!node) {
		pr_debug("malloc failed\n");
		return -1;
	}
	node->event = event;
	list_add(&node->list, events);

	if (perf_evlist__parse_sample(evlist, event, &sample)) {
		pr_debug("perf_evlist__parse_sample failed\n");
		return -1;
	}

	if (!sample.time) {
		pr_debug("event with no time\n");
		return -1;
	}

	node->event_time = sample.time;

	return 0;
}

static void free_event_nodes(struct list_head *events)
{
	struct event_node *node;

	while (!list_empty(events)) {
		node = list_entry(events->next, struct event_node, list);
		list_del(&node->list);
		free(node);
	}
}

static int compar(const void *a, const void *b)
{
	const struct event_node *nodea = a;
	const struct event_node *nodeb = b;
	s64 cmp = nodea->event_time - nodeb->event_time;

	return cmp;
}

static int process_events(struct perf_evlist *evlist,
			  struct switch_tracking *switch_tracking)
{
	union perf_event *event;
	unsigned pos, cnt = 0;
	LIST_HEAD(events);
	struct event_node *events_array, *node;
	int i, ret;

	for (i = 0; i < evlist->nr_mmaps; i++) {
		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
			cnt += 1;
			ret = add_event(evlist, &events, event);
			perf_evlist__mmap_consume(evlist, i);
			if (ret < 0)
				goto out_free_nodes;
		}
	}

	events_array = calloc(cnt, sizeof(struct event_node));
	if (!events_array) {
		pr_debug("calloc failed\n");
		ret = -1;
		goto out_free_nodes;
	}

	pos = 0;
	list_for_each_entry(node, &events, list)
		events_array[pos++] = *node;

	qsort(events_array, cnt, sizeof(struct event_node), compar);

	for (pos = 0; pos < cnt; pos++) {
		ret = process_event(evlist, events_array[pos].event,
				    switch_tracking);
		if (ret < 0)
			goto out_free;
	}

	ret = 0;
out_free:
	pr_debug("%u events recorded\n", cnt);
	free(events_array);
out_free_nodes:
	free_event_nodes(&events);
	return ret;
}

/**
 * test__switch_tracking - test using sched_switch and tracking events.
 *
 * This function implements a test that checks that sched_switch events and
 * tracking events can be recorded for a workload (current process) using the
 * evsel->system_wide and evsel->tracking flags (respectively) with other events
 * sometimes enabled or disabled.
 */
int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_unused)
{
	const char *sched_switch = "sched:sched_switch";
	struct switch_tracking switch_tracking = { .tids = NULL, };
	struct record_opts opts = {
		.mmap_pages	     = UINT_MAX,
		.user_freq	     = UINT_MAX,
		.user_interval	     = ULLONG_MAX,
		.freq		     = 4000,
		.target		     = {
			.uses_mmap   = true,
		},
	};
	struct thread_map *threads = NULL;
	struct cpu_map *cpus = NULL;
	struct perf_evlist *evlist = NULL;
	struct perf_evsel *evsel, *cpu_clocks_evsel, *cycles_evsel;
	struct perf_evsel *switch_evsel, *tracking_evsel;
	const char *comm;
	int err = -1;

	threads = thread_map__new(-1, getpid(), UINT_MAX);
	if (!threads) {
		pr_debug("thread_map__new failed!\n");
		goto out_err;
	}

	cpus = cpu_map__new(NULL);
	if (!cpus) {
		pr_debug("cpu_map__new failed!\n");
		goto out_err;
	}

	evlist = perf_evlist__new();
	if (!evlist) {
		pr_debug("perf_evlist__new failed!\n");
		goto out_err;
	}

	perf_evlist__set_maps(evlist, cpus, threads);

	/* First event */
	err = parse_events(evlist, "cpu-clock:u", NULL);
	if (err) {
		pr_debug("Failed to parse event dummy:u\n");
		goto out_err;
	}

	cpu_clocks_evsel = perf_evlist__last(evlist);

	/* Second event */
	err = parse_events(evlist, "cycles:u", NULL);
	if (err) {
		pr_debug("Failed to parse event cycles:u\n");
		goto out_err;
	}

	cycles_evsel = perf_evlist__last(evlist);

	/* Third event */
	if (!perf_evlist__can_select_event(evlist, sched_switch)) {
		pr_debug("No sched_switch\n");
		err = 0;
		goto out;
	}

	err = parse_events(evlist, sched_switch, NULL);
	if (err) {
		pr_debug("Failed to parse event %s\n", sched_switch);
		goto out_err;
	}

	switch_evsel = perf_evlist__last(evlist);

	perf_evsel__set_sample_bit(switch_evsel, CPU);
	perf_evsel__set_sample_bit(switch_evsel, TIME);

	switch_evsel->system_wide = true;
	switch_evsel->no_aux_samples = true;
	switch_evsel->immediate = true;

	/* Test moving an event to the front */
	if (cycles_evsel == perf_evlist__first(evlist)) {
		pr_debug("cycles event already at front");
		goto out_err;
	}
	perf_evlist__to_front(evlist, cycles_evsel);
	if (cycles_evsel != perf_evlist__first(evlist)) {
		pr_debug("Failed to move cycles event to front");
		goto out_err;
	}

	perf_evsel__set_sample_bit(cycles_evsel, CPU);
	perf_evsel__set_sample_bit(cycles_evsel, TIME);

	/* Fourth event */
	err = parse_events(evlist, "dummy:u", NULL);
	if (err) {
		pr_debug("Failed to parse event dummy:u\n");
		goto out_err;
	}

	tracking_evsel = perf_evlist__last(evlist);

	perf_evlist__set_tracking_event(evlist, tracking_evsel);

	tracking_evsel->attr.freq = 0;
	tracking_evsel->attr.sample_period = 1;

	perf_evsel__set_sample_bit(tracking_evsel, TIME);

	/* Config events */
	perf_evlist__config(evlist, &opts, NULL);

	/* Check moved event is still at the front */
	if (cycles_evsel != perf_evlist__first(evlist)) {
		pr_debug("Front event no longer at front");
		goto out_err;
	}

	/* Check tracking event is tracking */
	if (!tracking_evsel->attr.mmap || !tracking_evsel->attr.comm) {
		pr_debug("Tracking event not tracking\n");
		goto out_err;
	}

	/* Check non-tracking events are not tracking */
	evlist__for_each_entry(evlist, evsel) {
		if (evsel != tracking_evsel) {
			if (evsel->attr.mmap || evsel->attr.comm) {
				pr_debug("Non-tracking event is tracking\n");
				goto out_err;
			}
		}
	}

	if (perf_evlist__open(evlist) < 0) {
		pr_debug("Not supported\n");
		err = 0;
		goto out;
	}

	err = perf_evlist__mmap(evlist, UINT_MAX, false);
	if (err) {
		pr_debug("perf_evlist__mmap failed!\n");
		goto out_err;
	}

	perf_evlist__enable(evlist);

	err = perf_evsel__disable(cpu_clocks_evsel);
	if (err) {
		pr_debug("perf_evlist__disable_event failed!\n");
		goto out_err;
	}

	err = spin_sleep();
	if (err) {
		pr_debug("spin_sleep failed!\n");
		goto out_err;
	}

	comm = "Test COMM 1";
	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	if (err) {
		pr_debug("PR_SET_NAME failed!\n");
		goto out_err;
	}

	err = perf_evsel__disable(cycles_evsel);
	if (err) {
		pr_debug("perf_evlist__disable_event failed!\n");
		goto out_err;
	}

	comm = "Test COMM 2";
	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	if (err) {
		pr_debug("PR_SET_NAME failed!\n");
		goto out_err;
	}

	err = spin_sleep();
	if (err) {
		pr_debug("spin_sleep failed!\n");
		goto out_err;
	}

	comm = "Test COMM 3";
	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	if (err) {
		pr_debug("PR_SET_NAME failed!\n");
		goto out_err;
	}

	err = perf_evsel__enable(cycles_evsel);
	if (err) {
		pr_debug("perf_evlist__disable_event failed!\n");
		goto out_err;
	}

	comm = "Test COMM 4";
	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	if (err) {
		pr_debug("PR_SET_NAME failed!\n");
		goto out_err;
	}

	err = spin_sleep();
	if (err) {
		pr_debug("spin_sleep failed!\n");
		goto out_err;
	}

	perf_evlist__disable(evlist);

	switch_tracking.switch_evsel = switch_evsel;
	switch_tracking.cycles_evsel = cycles_evsel;

	err = process_events(evlist, &switch_tracking);

	zfree(&switch_tracking.tids);

	if (err)
		goto out_err;

	/* Check all 4 comm events were seen i.e. that evsel->tracking works */
	if (!switch_tracking.comm_seen[0] || !switch_tracking.comm_seen[1] ||
	    !switch_tracking.comm_seen[2] || !switch_tracking.comm_seen[3]) {
		pr_debug("Missing comm events\n");
		goto out_err;
	}

	/* Check cycles event got enabled */
	if (!switch_tracking.cycles_before_comm_1) {
		pr_debug("Missing cycles events\n");
		goto out_err;
	}

	/* Check cycles event got disabled */
	if (switch_tracking.cycles_between_comm_2_and_comm_3) {
		pr_debug("cycles events even though event was disabled\n");
		goto out_err;
	}

	/* Check cycles event got enabled again */
	if (!switch_tracking.cycles_after_comm_4) {
		pr_debug("Missing cycles events\n");
		goto out_err;
	}
out:
	if (evlist) {
		perf_evlist__disable(evlist);
		perf_evlist__delete(evlist);
	} else {
		cpu_map__put(cpus);
		thread_map__put(threads);
	}

	return err;

out_err:
	err = -1;
	goto out;
}
Commit	Line	Data
b2441318	1	// SPDX-License-Identifier: GPL-2.0
d44bc558 AH	2	#include <sys/time.h>
d44bc558 AH	3	#include <sys/prctl.h>
a43783ae	4	#include <errno.h>
d44bc558 AH	5	#include <time.h>
	6	#include <stdlib.h>
	7
	8	#include "parse-events.h"
	9	#include "evlist.h"
	10	#include "evsel.h"
	11	#include "thread_map.h"
	12	#include "cpumap.h"
	13	#include "tests.h"
	14
	15	static int spin_sleep(void)
	16	{
	17	struct timeval start, now, diff, maxtime;
	18	struct timespec ts;
	19	int err, i;
	20
	21	maxtime.tv_sec = 0;
	22	maxtime.tv_usec = 50000;
	23
	24	err = gettimeofday(&start, NULL);
	25	if (err)
	26	return err;
	27
	28	/* Spin for 50ms */
	29	while (1) {
	30	for (i = 0; i < 1000; i++)
	31	barrier();
	32
	33	err = gettimeofday(&now, NULL);
	34	if (err)
	35	return err;
	36
	37	timersub(&now, &start, &diff);
	38	if (timercmp(&diff, &maxtime, > /* For checkpatch */))
	39	break;
	40	}
	41
	42	ts.tv_nsec = 50 * 1000 * 1000;
	43	ts.tv_sec = 0;
	44
	45	/* Sleep for 50ms */
	46	err = nanosleep(&ts, NULL);
	47	if (err == EINTR)
	48	err = 0;
	49
	50	return err;
	51	}
	52
	53	struct switch_tracking {
	54	struct perf_evsel *switch_evsel;
	55	struct perf_evsel *cycles_evsel;
	56	pid_t *tids;
	57	int nr_tids;
	58	int comm_seen[4];
	59	int cycles_before_comm_1;
	60	int cycles_between_comm_2_and_comm_3;
	61	int cycles_after_comm_4;
	62	};
	63
	64	static int check_comm(struct switch_tracking *switch_tracking,
	65	union perf_event event, const char comm, int nr)
	66	{
	67	if (event->header.type == PERF_RECORD_COMM &&
	68	(pid_t)event->comm.pid == getpid() &&
69	(pid_t)event->comm.tid == getpid() &&
70	strcmp(event->comm.comm, comm) == 0) {
71	if (switch_tracking->comm_seen[nr]) {
72	pr_debug("Duplicate comm event\n");
73	return -1;
74	}
75	switch_tracking->comm_seen[nr] = 1;
76	pr_debug3("comm event: %s nr: %d\n", event->comm.comm, nr);
77	return 1;
78	}
79	return 0;
80	}
81
82	static int check_cpu(struct switch_tracking *switch_tracking, int cpu)
83	{
84	int i, nr = cpu + 1;
85
86	if (cpu < 0)
87	return -1;
88
89	if (!switch_tracking->tids) {
90	switch_tracking->tids = calloc(nr, sizeof(pid_t));
91	if (!switch_tracking->tids)
92	return -1;
93	for (i = 0; i < nr; i++)
94	switch_tracking->tids[i] = -1;
95	switch_tracking->nr_tids = nr;
96	return 0;
97	}
98
99	if (cpu >= switch_tracking->nr_tids) {
100	void *addr;
101
102	addr = realloc(switch_tracking->tids, nr * sizeof(pid_t));
103	if (!addr)
104	return -1;
105	switch_tracking->tids = addr;
106	for (i = switch_tracking->nr_tids; i < nr; i++)
107	switch_tracking->tids[i] = -1;
108	switch_tracking->nr_tids = nr;
109	return 0;
110	}
111
112	return 0;
113	}
114
115	static int process_sample_event(struct perf_evlist *evlist,
116	union perf_event *event,
117	struct switch_tracking *switch_tracking)
118	{
119	struct perf_sample sample;
120	struct perf_evsel *evsel;
121	pid_t next_tid, prev_tid;
122	int cpu, err;
123
124	if (perf_evlist__parse_sample(evlist, event, &sample)) {
125	pr_debug("perf_evlist__parse_sample failed\n");
126	return -1;
127	}
128
129	evsel = perf_evlist__id2evsel(evlist, sample.id);
130	if (evsel == switch_tracking->switch_evsel) {
131	next_tid = perf_evsel__intval(evsel, &sample, "next_pid");
132	prev_tid = perf_evsel__intval(evsel, &sample, "prev_pid");
133	cpu = sample.cpu;
134	pr_debug3("sched_switch: cpu: %d prev_tid %d next_tid %d\n",
135	cpu, prev_tid, next_tid);
136	err = check_cpu(switch_tracking, cpu);
137	if (err)
138	return err;
139	/*
140	* Check for no missing sched_switch events i.e. that the
141	* evsel->system_wide flag has worked.
142	*/
143	if (switch_tracking->tids[cpu] != -1 &&
144	switch_tracking->tids[cpu] != prev_tid) {
145	pr_debug("Missing sched_switch events\n");
146	return -1;
147	}
148	switch_tracking->tids[cpu] = next_tid;
149	}
150
151	if (evsel == switch_tracking->cycles_evsel) {
152	pr_debug3("cycles event\n");
153	if (!switch_tracking->comm_seen[0])
154	switch_tracking->cycles_before_comm_1 = 1;
155	if (switch_tracking->comm_seen[1] &&
156	!switch_tracking->comm_seen[2])
157	switch_tracking->cycles_between_comm_2_and_comm_3 = 1;
158	if (switch_tracking->comm_seen[3])
159	switch_tracking->cycles_after_comm_4 = 1;
160	}
161
162	return 0;
163	}
164
165	static int process_event(struct perf_evlist evlist, union perf_event event,
166	struct switch_tracking *switch_tracking)
167	{
168	if (event->header.type == PERF_RECORD_SAMPLE)
169	return process_sample_event(evlist, event, switch_tracking);
170
171	if (event->header.type == PERF_RECORD_COMM) {
172	int err, done = 0;
173
174	err = check_comm(switch_tracking, event, "Test COMM 1", 0);
175	if (err < 0)
176	return -1;
177	done += err;
178	err = check_comm(switch_tracking, event, "Test COMM 2", 1);
179	if (err < 0)
180	return -1;
181	done += err;
182	err = check_comm(switch_tracking, event, "Test COMM 3", 2);
183	if (err < 0)
184	return -1;
185	done += err;
186	err = check_comm(switch_tracking, event, "Test COMM 4", 3);
187	if (err < 0)
188	return -1;
189	done += err;
190	if (done != 1) {
191	pr_debug("Unexpected comm event\n");
192	return -1;
193	}
194	}
195
196	return 0;
197	}
198
199	struct event_node {
200	struct list_head list;
201	union perf_event *event;
202	u64 event_time;
203	};
204
205	static int add_event(struct perf_evlist evlist, struct list_head events,
206	union perf_event *event)
207	{
208	struct perf_sample sample;
209	struct event_node *node;
210
211	node = malloc(sizeof(struct event_node));
212	if (!node) {
213	pr_debug("malloc failed\n");
214	return -1;
215	}
216	node->event = event;
217	list_add(&node->list, events);
218
219	if (perf_evlist__parse_sample(evlist, event, &sample)) {
220	pr_debug("perf_evlist__parse_sample failed\n");
221	return -1;
222	}
223
224	if (!sample.time) {
225	pr_debug("event with no time\n");
226	return -1;
227	}
228
229	node->event_time = sample.time;
230
231	return 0;
232	}
233
234	static void free_event_nodes(struct list_head *events)
235	{
236	struct event_node *node;
237
238	while (!list_empty(events)) {
239	node = list_entry(events->next, struct event_node, list);
240	list_del(&node->list);
241	free(node);
242	}
243	}
244
245	static int compar(const void a, const void b)
246	{
247	const struct event_node *nodea = a;
248	const struct event_node *nodeb = b;
249	s64 cmp = nodea->event_time - nodeb->event_time;
250
251	return cmp;
252	}
253
254	static int process_events(struct perf_evlist *evlist,
255	struct switch_tracking *switch_tracking)
256	{
257	union perf_event *event;
258	unsigned pos, cnt = 0;
259	LIST_HEAD(events);
260	struct event_node events_array, node;
261	int i, ret;
262
263	for (i = 0; i < evlist->nr_mmaps; i++) {
264	while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
265	cnt += 1;
266	ret = add_event(evlist, &events, event);
267	perf_evlist__mmap_consume(evlist, i);
268	if (ret < 0)
269	goto out_free_nodes;
270	}
271	}
272
273	events_array = calloc(cnt, sizeof(struct event_node));
274	if (!events_array) {
275	pr_debug("calloc failed\n");
276	ret = -1;
277	goto out_free_nodes;
278	}
279
280	pos = 0;
281	list_for_each_entry(node, &events, list)
282	events_array[pos++] = *node;
283
284	qsort(events_array, cnt, sizeof(struct event_node), compar);
285
286	for (pos = 0; pos < cnt; pos++) {
287	ret = process_event(evlist, events_array[pos].event,
288	switch_tracking);
289	if (ret < 0)
290	goto out_free;
291	}
292
293	ret = 0;
294	out_free:
295	pr_debug("%u events recorded\n", cnt);
296	free(events_array);
297	out_free_nodes:
298	free_event_nodes(&events);
299	return ret;
300	}
301
302	/**
303	* test__switch_tracking - test using sched_switch and tracking events.
304	*
305	* This function implements a test that checks that sched_switch events and
306	* tracking events can be recorded for a workload (current process) using the
307	* evsel->system_wide and evsel->tracking flags (respectively) with other events
308	* sometimes enabled or disabled.
309	*/
81f17c90	310	int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_unused)
d44bc558 AH	311	{
	312	const char *sched_switch = "sched:sched_switch";
	313	struct switch_tracking switch_tracking = { .tids = NULL, };
	314	struct record_opts opts = {
	315	.mmap_pages = UINT_MAX,
	316	.user_freq = UINT_MAX,
	317	.user_interval = ULLONG_MAX,
	318	.freq = 4000,
	319	.target = {
	320	.uses_mmap = true,
	321	},
	322	};
	323	struct thread_map *threads = NULL;
	324	struct cpu_map *cpus = NULL;
	325	struct perf_evlist *evlist = NULL;
	326	struct perf_evsel evsel, cpu_clocks_evsel, *cycles_evsel;
	327	struct perf_evsel switch_evsel, tracking_evsel;
	328	const char *comm;
	329	int err = -1;
	330
	331	threads = thread_map__new(-1, getpid(), UINT_MAX);
	332	if (!threads) {
	333	pr_debug("thread_map__new failed!\n");
	334	goto out_err;
	335	}
	336
	337	cpus = cpu_map__new(NULL);
	338	if (!cpus) {
	339	pr_debug("cpu_map__new failed!\n");
	340	goto out_err;
	341	}
	342
	343	evlist = perf_evlist__new();
	344	if (!evlist) {
	345	pr_debug("perf_evlist__new failed!\n");
	346	goto out_err;
	347	}
	348
	349	perf_evlist__set_maps(evlist, cpus, threads);
	350
	351	/* First event */
b39b8393	352	err = parse_events(evlist, "cpu-clock:u", NULL);
d44bc558 AH	353	if (err) {
	354	pr_debug("Failed to parse event dummy:u\n");
	355	goto out_err;
	356	}
	357
	358	cpu_clocks_evsel = perf_evlist__last(evlist);
	359
	360	/* Second event */
b39b8393	361	err = parse_events(evlist, "cycles:u", NULL);
d44bc558 AH	362	if (err) {
	363	pr_debug("Failed to parse event cycles:u\n");
	364	goto out_err;
	365	}
	366
	367	cycles_evsel = perf_evlist__last(evlist);
	368
	369	/* Third event */
	370	if (!perf_evlist__can_select_event(evlist, sched_switch)) {
597bdeb4	371	pr_debug("No sched_switch\n");
d44bc558 AH	372	err = 0;
	373	goto out;
	374	}
	375
b39b8393	376	err = parse_events(evlist, sched_switch, NULL);
d44bc558 AH	377	if (err) {
	378	pr_debug("Failed to parse event %s\n", sched_switch);
	379	goto out_err;
	380	}
	381
	382	switch_evsel = perf_evlist__last(evlist);
	383
	384	perf_evsel__set_sample_bit(switch_evsel, CPU);
	385	perf_evsel__set_sample_bit(switch_evsel, TIME);
	386
	387	switch_evsel->system_wide = true;
	388	switch_evsel->no_aux_samples = true;
	389	switch_evsel->immediate = true;
	390
	391	/* Test moving an event to the front */
	392	if (cycles_evsel == perf_evlist__first(evlist)) {
	393	pr_debug("cycles event already at front");
	394	goto out_err;
	395	}
	396	perf_evlist__to_front(evlist, cycles_evsel);
	397	if (cycles_evsel != perf_evlist__first(evlist)) {
	398	pr_debug("Failed to move cycles event to front");
	399	goto out_err;
	400	}
	401
	402	perf_evsel__set_sample_bit(cycles_evsel, CPU);
	403	perf_evsel__set_sample_bit(cycles_evsel, TIME);
	404
	405	/* Fourth event */
b39b8393	406	err = parse_events(evlist, "dummy:u", NULL);
d44bc558 AH	407	if (err) {
	408	pr_debug("Failed to parse event dummy:u\n");
	409	goto out_err;
	410	}
	411
	412	tracking_evsel = perf_evlist__last(evlist);
	413
	414	perf_evlist__set_tracking_event(evlist, tracking_evsel);
	415
	416	tracking_evsel->attr.freq = 0;
	417	tracking_evsel->attr.sample_period = 1;
	418
	419	perf_evsel__set_sample_bit(tracking_evsel, TIME);
	420
	421	/* Config events */
e68ae9cf	422	perf_evlist__config(evlist, &opts, NULL);
d44bc558 AH	423
	424	/* Check moved event is still at the front */
	425	if (cycles_evsel != perf_evlist__first(evlist)) {
	426	pr_debug("Front event no longer at front");
	427	goto out_err;
	428	}
	429
	430	/* Check tracking event is tracking */
	431	if (!tracking_evsel->attr.mmap \|\| !tracking_evsel->attr.comm) {
	432	pr_debug("Tracking event not tracking\n");
	433	goto out_err;
	434	}
	435
	436	/* Check non-tracking events are not tracking */
e5cadb93	437	evlist__for_each_entry(evlist, evsel) {
d44bc558 AH	438	if (evsel != tracking_evsel) {
	439	if (evsel->attr.mmap \|\| evsel->attr.comm) {
	440	pr_debug("Non-tracking event is tracking\n");
	441	goto out_err;
	442	}
	443	}
	444	}
	445
	446	if (perf_evlist__open(evlist) < 0) {
597bdeb4	447	pr_debug("Not supported\n");
d44bc558 AH	448	err = 0;
	449	goto out;
	450	}
	451
	452	err = perf_evlist__mmap(evlist, UINT_MAX, false);
	453	if (err) {
	454	pr_debug("perf_evlist__mmap failed!\n");
	455	goto out_err;
	456	}
	457
	458	perf_evlist__enable(evlist);
	459
d2190a80	460	err = perf_evsel__disable(cpu_clocks_evsel);
d44bc558 AH	461	if (err) {
	462	pr_debug("perf_evlist__disable_event failed!\n");
	463	goto out_err;
	464	}
	465
	466	err = spin_sleep();
	467	if (err) {
	468	pr_debug("spin_sleep failed!\n");
	469	goto out_err;
	470	}
	471
	472	comm = "Test COMM 1";
	473	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	474	if (err) {
	475	pr_debug("PR_SET_NAME failed!\n");
	476	goto out_err;
	477	}
	478
d2190a80	479	err = perf_evsel__disable(cycles_evsel);
d44bc558 AH	480	if (err) {
	481	pr_debug("perf_evlist__disable_event failed!\n");
	482	goto out_err;
	483	}
	484
	485	comm = "Test COMM 2";
	486	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	487	if (err) {
	488	pr_debug("PR_SET_NAME failed!\n");
	489	goto out_err;
	490	}
	491
	492	err = spin_sleep();
	493	if (err) {
	494	pr_debug("spin_sleep failed!\n");
	495	goto out_err;
	496	}
	497
	498	comm = "Test COMM 3";
	499	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	500	if (err) {
	501	pr_debug("PR_SET_NAME failed!\n");
	502	goto out_err;
	503	}
	504
d2190a80	505	err = perf_evsel__enable(cycles_evsel);
d44bc558 AH	506	if (err) {
	507	pr_debug("perf_evlist__disable_event failed!\n");
	508	goto out_err;
	509	}
	510
	511	comm = "Test COMM 4";
	512	err = prctl(PR_SET_NAME, (unsigned long)comm, 0, 0, 0);
	513	if (err) {
	514	pr_debug("PR_SET_NAME failed!\n");
	515	goto out_err;
	516	}
	517
	518	err = spin_sleep();
	519	if (err) {
	520	pr_debug("spin_sleep failed!\n");
	521	goto out_err;
	522	}
	523
	524	perf_evlist__disable(evlist);
	525
	526	switch_tracking.switch_evsel = switch_evsel;
	527	switch_tracking.cycles_evsel = cycles_evsel;
	528
	529	err = process_events(evlist, &switch_tracking);
	530
	531	zfree(&switch_tracking.tids);
	532
	533	if (err)
	534	goto out_err;
	535
	536	/* Check all 4 comm events were seen i.e. that evsel->tracking works */
	537	if (!switch_tracking.comm_seen[0] \|\| !switch_tracking.comm_seen[1] \|\|
	538	!switch_tracking.comm_seen[2] \|\| !switch_tracking.comm_seen[3]) {
	539	pr_debug("Missing comm events\n");
	540	goto out_err;
	541	}
	542
	543	/* Check cycles event got enabled */
	544	if (!switch_tracking.cycles_before_comm_1) {
	545	pr_debug("Missing cycles events\n");
	546	goto out_err;
	547	}
	548
	549	/* Check cycles event got disabled */
	550	if (switch_tracking.cycles_between_comm_2_and_comm_3) {
	551	pr_debug("cycles events even though event was disabled\n");
	552	goto out_err;
	553	}
	554
	555	/* Check cycles event got enabled again */
	556	if (!switch_tracking.cycles_after_comm_4) {
	557	pr_debug("Missing cycles events\n");
	558	goto out_err;
	559	}
	560	out:
	561	if (evlist) {
	562	perf_evlist__disable(evlist);
	563	perf_evlist__delete(evlist);
	564	} else {
f30a79b0	565	cpu_map__put(cpus);
186fbb74	566	thread_map__put(threads);
d44bc558 AH	567	}
	568
	569	return err;
	570
	571	out_err:
	572	err = -1;
	573	goto out;
	574	}