summaryrefslogtreecommitdiff
path: root/tools/perf/builtin-record.c
blob: 4decbd14eaed080ac046b8a10c793b721280c1df (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
/*
 * builtin-record.c
 *
 * Builtin record command: Record the profile of a workload
 * (or a CPU, or a PID) into the perf.data output file - for
 * later analysis via perf report.
 */
#include "builtin.h"

#include "perf.h"

#include "util/util.h"
#include "util/parse-options.h"
#include "util/parse-events.h"
#include "util/string.h"

#include "util/header.h"
#include "util/event.h"
#include "util/debug.h"
#include "util/session.h"
#include "util/symbol.h"

#include <unistd.h>
#include <sched.h>

static int			fd[MAX_NR_CPUS][MAX_COUNTERS];

static long			default_interval		=      0;

static int			nr_cpus				=      0;
static unsigned int		page_size;
static unsigned int		mmap_pages			=    128;
static int			freq				=   1000;
static int			output;
static const char		*output_name			= "perf.data";
static int			group				=      0;
static unsigned int		realtime_prio			=      0;
static int			raw_samples			=      0;
static int			system_wide			=      0;
static int			profile_cpu			=     -1;
static pid_t			target_pid			=     -1;
static pid_t			child_pid			=     -1;
static int			inherit				=      1;
static int			force				=      0;
static int			append_file			=      0;
static int			call_graph			=      0;
static int			inherit_stat			=      0;
static int			no_samples			=      0;
static int			sample_address			=      0;
static int			multiplex			=      0;
static int			multiplex_fd			=     -1;

static long			samples				=      0;
static struct timeval		last_read;
static struct timeval		this_read;

static u64			bytes_written			=      0;

static struct pollfd		event_array[MAX_NR_CPUS * MAX_COUNTERS];

static int			nr_poll				=      0;
static int			nr_cpu				=      0;

static int			file_new			=      1;

static struct perf_session	*session;

struct mmap_data {
	int			counter;
	void			*base;
	unsigned int		mask;
	unsigned int		prev;
};

static struct mmap_data		mmap_array[MAX_NR_CPUS][MAX_COUNTERS];

static unsigned long mmap_read_head(struct mmap_data *md)
{
	struct perf_event_mmap_page *pc = md->base;
	long head;

	head = pc->data_head;
	rmb();

	return head;
}

static void mmap_write_tail(struct mmap_data *md, unsigned long tail)
{
	struct perf_event_mmap_page *pc = md->base;

	/*
	 * ensure all reads are done before we write the tail out.
	 */
	/* mb(); */
	pc->data_tail = tail;
}

static void write_output(void *buf, size_t size)
{
	while (size) {
		int ret = write(output, buf, size);

		if (ret < 0)
			die("failed to write");

		size -= ret;
		buf += ret;

		bytes_written += ret;
	}
}

static void write_event(event_t *buf, size_t size)
{
	/*
	* Add it to the list of DSOs, so that when we finish this
	 * record session we can pick the available build-ids.
	 */
	if (buf->header.type == PERF_RECORD_MMAP)
		dsos__findnew(buf->mmap.filename);

	write_output(buf, size);
}

static int process_synthesized_event(event_t *event)
{
	write_event(event, event->header.size);
	return 0;
}

static void mmap_read(struct mmap_data *md)
{
	unsigned int head = mmap_read_head(md);
	unsigned int old = md->prev;
	unsigned char *data = md->base + page_size;
	unsigned long size;
	void *buf;
	int diff;

	gettimeofday(&this_read, NULL);

	/*
	 * If we're further behind than half the buffer, there's a chance
	 * the writer will bite our tail and mess up the samples under us.
	 *
	 * If we somehow ended up ahead of the head, we got messed up.
	 *
	 * In either case, truncate and restart at head.
	 */
	diff = head - old;
	if (diff < 0) {
		struct timeval iv;
		unsigned long msecs;

		timersub(&this_read, &last_read, &iv);
		msecs = iv.tv_sec*1000 + iv.tv_usec/1000;

		fprintf(stderr, "WARNING: failed to keep up with mmap data."
				"  Last read %lu msecs ago.\n", msecs);

		/*
		 * head points to a known good entry, start there.
		 */
		old = head;
	}

	last_read = this_read;

	if (old != head)
		samples++;

	size = head - old;

	if ((old & md->mask) + size != (head & md->mask)) {
		buf = &data[old & md->mask];
		size = md->mask + 1 - (old & md->mask);
		old += size;

		write_event(buf, size);
	}

	buf = &data[old & md->mask];
	size = head - old;
	old += size;

	write_event(buf, size);

	md->prev = old;
	mmap_write_tail(md, old);
}

static volatile int done = 0;
static volatile int signr = -1;

static void sig_handler(int sig)
{
	done = 1;
	signr = sig;
}

static void sig_atexit(void)
{
	if (child_pid != -1)
		kill(child_pid, SIGTERM);

	if (signr == -1)
		return;

	signal(signr, SIG_DFL);
	kill(getpid(), signr);
}

static int group_fd;

static struct perf_header_attr *get_header_attr(struct perf_event_attr *a, int nr)
{
	struct perf_header_attr *h_attr;

	if (nr < session->header.attrs) {
		h_attr = session->header.attr[nr];
	} else {
		h_attr = perf_header_attr__new(a);
		if (h_attr != NULL)
			if (perf_header__add_attr(&session->header, h_attr) < 0) {
				perf_header_attr__delete(h_attr);
				h_attr = NULL;
			}
	}

	return h_attr;
}

static void create_counter(int counter, int cpu, pid_t pid)
{
	char *filter = filters[counter];
	struct perf_event_attr *attr = attrs + counter;
	struct perf_header_attr *h_attr;
	int track = !counter; /* only the first counter needs these */
	int ret;
	struct {
		u64 count;
		u64 time_enabled;
		u64 time_running;
		u64 id;
	} read_data;

	attr->read_format	= PERF_FORMAT_TOTAL_TIME_ENABLED |
				  PERF_FORMAT_TOTAL_TIME_RUNNING |
				  PERF_FORMAT_ID;

	attr->sample_type	|= PERF_SAMPLE_IP | PERF_SAMPLE_TID;

	if (freq) {
		attr->sample_type	|= PERF_SAMPLE_PERIOD;
		attr->freq		= 1;
		attr->sample_freq	= freq;
	}

	if (no_samples)
		attr->sample_freq = 0;

	if (inherit_stat)
		attr->inherit_stat = 1;

	if (sample_address)
		attr->sample_type	|= PERF_SAMPLE_ADDR;

	if (call_graph)
		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;

	if (raw_samples) {
		attr->sample_type	|= PERF_SAMPLE_TIME;
		attr->sample_type	|= PERF_SAMPLE_RAW;
		attr->sample_type	|= PERF_SAMPLE_CPU;
	}

	attr->mmap		= track;
	attr->comm		= track;
	attr->inherit		= (cpu < 0) && inherit;
	attr->disabled		= 1;

try_again:
	fd[nr_cpu][counter] = sys_perf_event_open(attr, pid, cpu, group_fd, 0);

	if (fd[nr_cpu][counter] < 0) {
		int err = errno;

		if (err == EPERM || err == EACCES)
			die("Permission error - are you root?\n");
		else if (err ==  ENODEV && profile_cpu != -1)
			die("No such device - did you specify an out-of-range profile CPU?\n");

		/*
		 * If it's cycles then fall back to hrtimer
		 * based cpu-clock-tick sw counter, which
		 * is always available even if no PMU support:
		 */
		if (attr->type == PERF_TYPE_HARDWARE
			&& attr->config == PERF_COUNT_HW_CPU_CYCLES) {

			if (verbose)
				warning(" ... trying to fall back to cpu-clock-ticks\n");
			attr->type = PERF_TYPE_SOFTWARE;
			attr->config = PERF_COUNT_SW_CPU_CLOCK;
			goto try_again;
		}
		printf("\n");
		error("perfcounter syscall returned with %d (%s)\n",
			fd[nr_cpu][counter], strerror(err));

#if defined(__i386__) || defined(__x86_64__)
		if (attr->type == PERF_TYPE_HARDWARE && err == EOPNOTSUPP)
			die("No hardware sampling interrupt available. No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it.\n");
#endif

		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
		exit(-1);
	}

	h_attr = get_header_attr(attr, counter);
	if (h_attr == NULL)
		die("nomem\n");

	if (!file_new) {
		if (memcmp(&h_attr->attr, attr, sizeof(*attr))) {
			fprintf(stderr, "incompatible append\n");
			exit(-1);
		}
	}

	if (read(fd[nr_cpu][counter], &read_data, sizeof(read_data)) == -1) {
		perror("Unable to read perf file descriptor\n");
		exit(-1);
	}

	if (perf_header_attr__add_id(h_attr, read_data.id) < 0) {
		pr_warning("Not enough memory to add id\n");
		exit(-1);
	}

	assert(fd[nr_cpu][counter] >= 0);
	fcntl(fd[nr_cpu][counter], F_SETFL, O_NONBLOCK);

	/*
	 * First counter acts as the group leader:
	 */
	if (group && group_fd == -1)
		group_fd = fd[nr_cpu][counter];
	if (multiplex && multiplex_fd == -1)
		multiplex_fd = fd[nr_cpu][counter];

	if (multiplex && fd[nr_cpu][counter] != multiplex_fd) {

		ret = ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_SET_OUTPUT, multiplex_fd);
		assert(ret != -1);
	} else {
		event_array[nr_poll].fd = fd[nr_cpu][counter];
		event_array[nr_poll].events = POLLIN;
		nr_poll++;

		mmap_array[nr_cpu][counter].counter = counter;
		mmap_array[nr_cpu][counter].prev = 0;
		mmap_array[nr_cpu][counter].mask = mmap_pages*page_size - 1;
		mmap_array[nr_cpu][counter].base = mmap(NULL, (mmap_pages+1)*page_size,
				PROT_READ|PROT_WRITE, MAP_SHARED, fd[nr_cpu][counter], 0);
		if (mmap_array[nr_cpu][counter].base == MAP_FAILED) {
			error("failed to mmap with %d (%s)\n", errno, strerror(errno));
			exit(-1);
		}
	}

	if (filter != NULL) {
		ret = ioctl(fd[nr_cpu][counter],
			    PERF_EVENT_IOC_SET_FILTER, filter);
		if (ret) {
			error("failed to set filter with %d (%s)\n", errno,
			      strerror(errno));
			exit(-1);
		}
	}

	ioctl(fd[nr_cpu][counter], PERF_EVENT_IOC_ENABLE);
}

static void open_counters(int cpu, pid_t pid)
{
	int counter;

	group_fd = -1;
	for (counter = 0; counter < nr_counters; counter++)
		create_counter(counter, cpu, pid);

	nr_cpu++;
}

static void atexit_header(void)
{
	session->header.data_size += bytes_written;

	perf_header__write(&session->header, output, true);
}

static int __cmd_record(int argc, const char **argv)
{
	int i, counter;
	struct stat st;
	pid_t pid = 0;
	int flags;
	int err;
	unsigned long waking = 0;

	page_size = sysconf(_SC_PAGE_SIZE);
	nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
	assert(nr_cpus <= MAX_NR_CPUS);
	assert(nr_cpus >= 0);

	atexit(sig_atexit);
	signal(SIGCHLD, sig_handler);
	signal(SIGINT, sig_handler);

	if (!stat(output_name, &st) && st.st_size) {
		if (!force && !append_file) {
			fprintf(stderr, "Error, output file %s exists, use -A to append or -f to overwrite.\n",
					output_name);
			exit(-1);
		}
	} else {
		append_file = 0;
	}

	flags = O_CREAT|O_RDWR;
	if (append_file)
		file_new = 0;
	else
		flags |= O_TRUNC;

	output = open(output_name, flags, S_IRUSR|S_IWUSR);
	if (output < 0) {
		perror("failed to create output file");
		exit(-1);
	}

	session = perf_session__new(output_name, O_WRONLY, force);
	if (session == NULL) {
		pr_err("Not enough memory for reading perf file header\n");
		return -1;
	}

	if (!file_new) {
		err = perf_header__read(&session->header, output);
		if (err < 0)
			return err;
	}

	if (raw_samples) {
		perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
	} else {
		for (i = 0; i < nr_counters; i++) {
			if (attrs[i].sample_type & PERF_SAMPLE_RAW) {
				perf_header__set_feat(&session->header, HEADER_TRACE_INFO);
				break;
			}
		}
	}

	atexit(atexit_header);

	if (!system_wide) {
		pid = target_pid;
		if (pid == -1)
			pid = getpid();

		open_counters(profile_cpu, pid);
	} else {
		if (profile_cpu != -1) {
			open_counters(profile_cpu, target_pid);
		} else {
			for (i = 0; i < nr_cpus; i++)
				open_counters(i, target_pid);
		}
	}

	if (file_new) {
		err = perf_header__write(&session->header, output, false);
		if (err < 0)
			return err;
	}

	if (!system_wide)
		event__synthesize_thread(pid, process_synthesized_event);
	else
		event__synthesize_threads(process_synthesized_event);

	if (target_pid == -1 && argc) {
		pid = fork();
		if (pid < 0)
			die("failed to fork");

		if (!pid) {
			if (execvp(argv[0], (char **)argv)) {
				perror(argv[0]);
				exit(-1);
			}
		} else {
			/*
			 * Wait a bit for the execv'ed child to appear
			 * and be updated in /proc
			 * FIXME: Do you know a less heuristical solution?
			 */
			usleep(1000);
			event__synthesize_thread(pid,
						 process_synthesized_event);
		}

		child_pid = pid;
	}

	if (realtime_prio) {
		struct sched_param param;

		param.sched_priority = realtime_prio;
		if (sched_setscheduler(0, SCHED_FIFO, &param)) {
			pr_err("Could not set realtime priority.\n");
			exit(-1);
		}
	}

	for (;;) {
		int hits = samples;

		for (i = 0; i < nr_cpu; i++) {
			for (counter = 0; counter < nr_counters; counter++) {
				if (mmap_array[i][counter].base)
					mmap_read(&mmap_array[i][counter]);
			}
		}

		if (hits == samples) {
			if (done)
				break;
			err = poll(event_array, nr_poll, -1);
			waking++;
		}

		if (done) {
			for (i = 0; i < nr_cpu; i++) {
				for (counter = 0; counter < nr_counters; counter++)
					ioctl(fd[i][counter], PERF_EVENT_IOC_DISABLE);
			}
		}
	}

	fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);

	/*
	 * Approximate RIP event size: 24 bytes.
	 */
	fprintf(stderr,
		"[ perf record: Captured and wrote %.3f MB %s (~%lld samples) ]\n",
		(double)bytes_written / 1024.0 / 1024.0,
		output_name,
		bytes_written / 24);

	return 0;
}

static const char * const record_usage[] = {
	"perf record [<options>] [<command>]",
	"perf record [<options>] -- <command> [<options>]",
	NULL
};

static const struct option options[] = {
	OPT_CALLBACK('e', "event", NULL, "event",
		     "event selector. use 'perf list' to list available events",
		     parse_events),
	OPT_CALLBACK(0, "filter", NULL, "filter",
		     "event filter", parse_filter),
	OPT_INTEGER('p', "pid", &target_pid,
		    "record events on existing pid"),
	OPT_INTEGER('r', "realtime", &realtime_prio,
		    "collect data with this RT SCHED_FIFO priority"),
	OPT_BOOLEAN('R', "raw-samples", &raw_samples,
		    "collect raw sample records from all opened counters"),
	OPT_BOOLEAN('a', "all-cpus", &system_wide,
			    "system-wide collection from all CPUs"),
	OPT_BOOLEAN('A', "append", &append_file,
			    "append to the output file to do incremental profiling"),
	OPT_INTEGER('C', "profile_cpu", &profile_cpu,
			    "CPU to profile on"),
	OPT_BOOLEAN('f', "force", &force,
			"overwrite existing data file"),
	OPT_LONG('c', "count", &default_interval,
		    "event period to sample"),
	OPT_STRING('o', "output", &output_name, "file",
		    "output file name"),
	OPT_BOOLEAN('i', "inherit", &inherit,
		    "child tasks inherit counters"),
	OPT_INTEGER('F', "freq", &freq,
		    "profile at this frequency"),
	OPT_INTEGER('m', "mmap-pages", &mmap_pages,
		    "number of mmap data pages"),
	OPT_BOOLEAN('g', "call-graph", &call_graph,
		    "do call-graph (stack chain/backtrace) recording"),
	OPT_BOOLEAN('v', "verbose", &verbose,
		    "be more verbose (show counter open errors, etc)"),
	OPT_BOOLEAN('s', "stat", &inherit_stat,
		    "per thread counts"),
	OPT_BOOLEAN('d', "data", &sample_address,
		    "Sample addresses"),
	OPT_BOOLEAN('n', "no-samples", &no_samples,
		    "don't sample"),
	OPT_BOOLEAN('M', "multiplex", &multiplex,
		    "multiplex counter output in a single channel"),
	OPT_END()
};

int cmd_record(int argc, const char **argv, const char *prefix __used)
{
	int counter;

	symbol__init(0);

	argc = parse_options(argc, argv, options, record_usage,
		PARSE_OPT_STOP_AT_NON_OPTION);
	if (!argc && target_pid == -1 && !system_wide)
		usage_with_options(record_usage, options);

	if (!nr_counters) {
		nr_counters	= 1;
		attrs[0].type	= PERF_TYPE_HARDWARE;
		attrs[0].config = PERF_COUNT_HW_CPU_CYCLES;
	}

	/*
	 * User specified count overrides default frequency.
	 */
	if (default_interval)
		freq = 0;
	else if (freq) {
		default_interval = freq;
	} else {
		fprintf(stderr, "frequency and count are zero, aborting\n");
		exit(EXIT_FAILURE);
	}

	for (counter = 0; counter < nr_counters; counter++) {
		if (attrs[counter].sample_period)
			continue;

		attrs[counter].sample_period = default_interval;
	}

	return __cmd_record(argc, argv);
}