evsel.c 20.5 KB
Newer Older
1 2 3 4 5 6 7 8 9
/*
 * Copyright (C) 2011, Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
 *
 * Parts came from builtin-{top,stat,record}.c, see those files for further
 * copyright notes.
 *
 * Released under the GPL v2. (and only v2, not any later version)
 */

10 11
#include <byteswap.h>
#include "asm/bug.h"
12
#include "evsel.h"
13
#include "evlist.h"
14
#include "util.h"
15
#include "cpumap.h"
16
#include "thread_map.h"
17
#include "target.h"
18

19
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
20
#define GROUP_FD(group_fd, cpu) (*(int *)xyarray__entry(group_fd, cpu, 0))
21

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
int __perf_evsel__sample_size(u64 sample_type)
{
	u64 mask = sample_type & PERF_SAMPLE_MASK;
	int size = 0;
	int i;

	for (i = 0; i < 64; i++) {
		if (mask & (1ULL << i))
			size++;
	}

	size *= sizeof(u64);

	return size;
}

38
void hists__init(struct hists *hists)
39 40 41 42 43 44 45 46 47
{
	memset(hists, 0, sizeof(*hists));
	hists->entries_in_array[0] = hists->entries_in_array[1] = RB_ROOT;
	hists->entries_in = &hists->entries_in_array[0];
	hists->entries_collapsed = RB_ROOT;
	hists->entries = RB_ROOT;
	pthread_mutex_init(&hists->lock, NULL);
}

48 49 50 51 52 53
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
	evsel->attr	   = *attr;
	INIT_LIST_HEAD(&evsel->node);
54
	hists__init(&evsel->hists);
55 56
}

57
struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
58 59 60
{
	struct perf_evsel *evsel = zalloc(sizeof(*evsel));

61 62
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
63 64 65 66

	return evsel;
}

67 68 69 70 71 72 73 74 75 76 77 78 79
static const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

80
static const char *__perf_evsel__hw_name(u64 config)
81 82 83 84 85 86 87
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

88
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
89
{
90
	int colon = 0, r = 0;
91 92 93 94 95
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
96
			if (!colon) colon = ++r;			\
97 98 99 100 101 102 103 104 105 106 107 108
			r += scnprintf(bf + r, size - r, "%c", mod);	\
		} } while(0)

	if (attr->exclude_kernel || attr->exclude_user || attr->exclude_hv) {
		MOD_PRINT(kernel, 'k');
		MOD_PRINT(user, 'u');
		MOD_PRINT(hv, 'h');
		exclude_guest_default = true;
	}

	if (attr->precise_ip) {
		if (!colon)
109
			colon = ++r;
110 111 112 113 114 115 116 117 118 119
		r += scnprintf(bf + r, size - r, "%.*s", attr->precise_ip, "ppp");
		exclude_guest_default = true;
	}

	if (attr->exclude_host || attr->exclude_guest == exclude_guest_default) {
		MOD_PRINT(host, 'H');
		MOD_PRINT(guest, 'G');
	}
#undef MOD_PRINT
	if (colon)
120
		bf[colon - 1] = ':';
121 122 123
	return r;
}

124 125 126 127 128 129
static int perf_evsel__hw_name(struct perf_evsel *evsel, char *bf, size_t size)
{
	int r = scnprintf(bf, size, "%s", __perf_evsel__hw_name(evsel->attr.config));
	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
}

130 131 132 133 134 135 136 137 138 139 140 141
static const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
	"CPU-migrations",
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
};

142
static const char *__perf_evsel__sw_name(u64 config)
143 144 145 146 147 148 149 150 151 152 153 154
{
	if (config < PERF_COUNT_SW_MAX && perf_evsel__sw_names[config])
		return perf_evsel__sw_names[config];
	return "unknown-software";
}

static int perf_evsel__sw_name(struct perf_evsel *evsel, char *bf, size_t size)
{
	int r = scnprintf(bf, size, "%s", __perf_evsel__sw_name(evsel->attr.config));
	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
}

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
const char *perf_evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX]
				[PERF_EVSEL__MAX_ALIASES] = {
 { "L1-dcache",	"l1-d",		"l1d",		"L1-data",		},
 { "L1-icache",	"l1-i",		"l1i",		"L1-instruction",	},
 { "LLC",	"L2",							},
 { "dTLB",	"d-tlb",	"Data-TLB",				},
 { "iTLB",	"i-tlb",	"Instruction-TLB",			},
 { "branch",	"branches",	"bpu",		"btb",		"bpc",	},
 { "node",								},
};

const char *perf_evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX]
				   [PERF_EVSEL__MAX_ALIASES] = {
 { "load",	"loads",	"read",					},
 { "store",	"stores",	"write",				},
 { "prefetch",	"prefetches",	"speculative-read", "speculative-load",	},
};

const char *perf_evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX]
				       [PERF_EVSEL__MAX_ALIASES] = {
 { "refs",	"Reference",	"ops",		"access",		},
 { "misses",	"miss",							},
};

#define C(x)		PERF_COUNT_HW_CACHE_##x
#define CACHE_READ	(1 << C(OP_READ))
#define CACHE_WRITE	(1 << C(OP_WRITE))
#define CACHE_PREFETCH	(1 << C(OP_PREFETCH))
#define COP(x)		(1 << x)

/*
 * cache operartion stat
 * L1I : Read and prefetch only
 * ITLB and BPU : Read-only
 */
static unsigned long perf_evsel__hw_cache_stat[C(MAX)] = {
 [C(L1D)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
 [C(L1I)]	= (CACHE_READ | CACHE_PREFETCH),
 [C(LL)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
 [C(DTLB)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
 [C(ITLB)]	= (CACHE_READ),
 [C(BPU)]	= (CACHE_READ),
 [C(NODE)]	= (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH),
};

bool perf_evsel__is_cache_op_valid(u8 type, u8 op)
{
	if (perf_evsel__hw_cache_stat[type] & COP(op))
		return true;	/* valid */
	else
		return false;	/* invalid */
}

int __perf_evsel__hw_cache_type_op_res_name(u8 type, u8 op, u8 result,
					    char *bf, size_t size)
{
	if (result) {
		return scnprintf(bf, size, "%s-%s-%s", perf_evsel__hw_cache[type][0],
				 perf_evsel__hw_cache_op[op][0],
				 perf_evsel__hw_cache_result[result][0]);
	}

	return scnprintf(bf, size, "%s-%s", perf_evsel__hw_cache[type][0],
			 perf_evsel__hw_cache_op[op][1]);
}

221
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
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
{
	u8 op, result, type = (config >>  0) & 0xff;
	const char *err = "unknown-ext-hardware-cache-type";

	if (type > PERF_COUNT_HW_CACHE_MAX)
		goto out_err;

	op = (config >>  8) & 0xff;
	err = "unknown-ext-hardware-cache-op";
	if (op > PERF_COUNT_HW_CACHE_OP_MAX)
		goto out_err;

	result = (config >> 16) & 0xff;
	err = "unknown-ext-hardware-cache-result";
	if (result > PERF_COUNT_HW_CACHE_RESULT_MAX)
		goto out_err;

	err = "invalid-cache";
	if (!perf_evsel__is_cache_op_valid(type, op))
		goto out_err;

	return __perf_evsel__hw_cache_type_op_res_name(type, op, result, bf, size);
out_err:
	return scnprintf(bf, size, "%s", err);
}

static int perf_evsel__hw_cache_name(struct perf_evsel *evsel, char *bf, size_t size)
{
	int ret = __perf_evsel__hw_cache_name(evsel->attr.config, bf, size);
	return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret);
}

254 255 256 257 258 259
static int perf_evsel__raw_name(struct perf_evsel *evsel, char *bf, size_t size)
{
	int ret = scnprintf(bf, size, "raw 0x%" PRIx64, evsel->attr.config);
	return ret + perf_evsel__add_modifiers(evsel, bf + ret, size - ret);
}

260
const char *perf_evsel__name(struct perf_evsel *evsel)
261
{
262
	char bf[128];
263

264 265
	if (evsel->name)
		return evsel->name;
266 267 268

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
269
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
270 271 272
		break;

	case PERF_TYPE_HARDWARE:
273
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
274
		break;
275 276

	case PERF_TYPE_HW_CACHE:
277
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
278 279
		break;

280
	case PERF_TYPE_SOFTWARE:
281
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
282 283
		break;

284
	case PERF_TYPE_TRACEPOINT:
285
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
286 287
		break;

288
	default:
289
		scnprintf(bf, sizeof(bf), "%s", "unknown attr type");
290
		break;
291 292
	}

293 294 295
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
296 297
}

298 299
void perf_evsel__config(struct perf_evsel *evsel, struct perf_record_opts *opts,
			struct perf_evsel *first)
300 301 302 303
{
	struct perf_event_attr *attr = &evsel->attr;
	int track = !evsel->idx; /* only the first counter needs these */

304
	attr->disabled = 1;
305
	attr->sample_id_all = opts->sample_id_all_missing ? 0 : 1;
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
	attr->inherit	    = !opts->no_inherit;
	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;

	/*
	 * We default some events to a 1 default interval. But keep
	 * it a weak assumption overridable by the user.
	 */
	if (!attr->sample_period || (opts->user_freq != UINT_MAX &&
				     opts->user_interval != ULLONG_MAX)) {
		if (opts->freq) {
			attr->sample_type	|= PERF_SAMPLE_PERIOD;
			attr->freq		= 1;
			attr->sample_freq	= opts->freq;
		} else {
			attr->sample_period = opts->default_interval;
		}
	}

	if (opts->no_samples)
		attr->sample_freq = 0;

	if (opts->inherit_stat)
		attr->inherit_stat = 1;

	if (opts->sample_address) {
		attr->sample_type	|= PERF_SAMPLE_ADDR;
		attr->mmap_data = track;
	}

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

342
	if (perf_target__has_cpu(&opts->target))
343 344
		attr->sample_type	|= PERF_SAMPLE_CPU;

345 346 347
	if (opts->period)
		attr->sample_type	|= PERF_SAMPLE_PERIOD;

348
	if (!opts->sample_id_all_missing &&
349
	    (opts->sample_time || !opts->no_inherit ||
350
	     perf_target__has_cpu(&opts->target)))
351 352 353 354 355 356 357 358 359 360 361 362
		attr->sample_type	|= PERF_SAMPLE_TIME;

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

	if (opts->no_delay) {
		attr->watermark = 0;
		attr->wakeup_events = 1;
	}
363 364 365 366
	if (opts->branch_stack) {
		attr->sample_type	|= PERF_SAMPLE_BRANCH_STACK;
		attr->branch_sample_type = opts->branch_stack;
	}
367 368 369 370

	attr->mmap = track;
	attr->comm = track;

371 372
	if (perf_target__none(&opts->target) &&
	    (!opts->group || evsel == first)) {
373 374 375 376
		attr->enable_on_exec = 1;
	}
}

377 378
int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
379
	int cpu, thread;
380
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
381 382 383 384 385 386 387 388 389

	if (evsel->fd) {
		for (cpu = 0; cpu < ncpus; cpu++) {
			for (thread = 0; thread < nthreads; thread++) {
				FD(evsel, cpu, thread) = -1;
			}
		}
	}

390 391 392
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

393 394
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
395 396 397 398 399 400 401 402 403 404 405 406
	evsel->sample_id = xyarray__new(ncpus, nthreads, sizeof(struct perf_sample_id));
	if (evsel->sample_id == NULL)
		return -ENOMEM;

	evsel->id = zalloc(ncpus * nthreads * sizeof(u64));
	if (evsel->id == NULL) {
		xyarray__delete(evsel->sample_id);
		evsel->sample_id = NULL;
		return -ENOMEM;
	}

	return 0;
407 408
}

409 410 411 412 413 414 415
int perf_evsel__alloc_counts(struct perf_evsel *evsel, int ncpus)
{
	evsel->counts = zalloc((sizeof(*evsel->counts) +
				(ncpus * sizeof(struct perf_counts_values))));
	return evsel->counts != NULL ? 0 : -ENOMEM;
}

416 417 418 419 420 421
void perf_evsel__free_fd(struct perf_evsel *evsel)
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

422 423
void perf_evsel__free_id(struct perf_evsel *evsel)
{
424 425 426
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
	free(evsel->id);
427 428 429
	evsel->id = NULL;
}

430 431 432 433 434 435 436 437 438 439 440
void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	int cpu, thread;

	for (cpu = 0; cpu < ncpus; cpu++)
		for (thread = 0; thread < nthreads; ++thread) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
}

441
void perf_evsel__exit(struct perf_evsel *evsel)
442 443 444
{
	assert(list_empty(&evsel->node));
	xyarray__delete(evsel->fd);
445 446
	xyarray__delete(evsel->sample_id);
	free(evsel->id);
447 448 449 450 451
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
S
Stephane Eranian 已提交
452
	close_cgroup(evsel->cgrp);
453
	free(evsel->name);
454 455
	free(evsel);
}
456 457 458 459 460 461 462 463 464 465

int __perf_evsel__read_on_cpu(struct perf_evsel *evsel,
			      int cpu, int thread, bool scale)
{
	struct perf_counts_values count;
	size_t nv = scale ? 3 : 1;

	if (FD(evsel, cpu, thread) < 0)
		return -EINVAL;

466 467 468
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
		return -errno;

	if (scale) {
		if (count.run == 0)
			count.val = 0;
		else if (count.run < count.ena)
			count.val = (u64)((double)count.val * count.ena / count.run + 0.5);
	} else
		count.ena = count.run = 0;

	evsel->counts->cpu[cpu] = count;
	return 0;
}

int __perf_evsel__read(struct perf_evsel *evsel,
		       int ncpus, int nthreads, bool scale)
{
	size_t nv = scale ? 3 : 1;
	int cpu, thread;
	struct perf_counts_values *aggr = &evsel->counts->aggr, count;

491
	aggr->val = aggr->ena = aggr->run = 0;
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

	for (cpu = 0; cpu < ncpus; cpu++) {
		for (thread = 0; thread < nthreads; thread++) {
			if (FD(evsel, cpu, thread) < 0)
				continue;

			if (readn(FD(evsel, cpu, thread),
				  &count, nv * sizeof(u64)) < 0)
				return -errno;

			aggr->val += count.val;
			if (scale) {
				aggr->ena += count.ena;
				aggr->run += count.run;
			}
		}
	}

	evsel->counts->scaled = 0;
	if (scale) {
		if (aggr->run == 0) {
			evsel->counts->scaled = -1;
			aggr->val = 0;
			return 0;
		}

		if (aggr->run < aggr->ena) {
			evsel->counts->scaled = 1;
			aggr->val = (u64)((double)aggr->val * aggr->ena / aggr->run + 0.5);
		}
	} else
		aggr->ena = aggr->run = 0;

	return 0;
}
527

528
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
529 530
			      struct thread_map *threads, bool group,
			      struct xyarray *group_fds)
531
{
532
	int cpu, thread;
S
Stephane Eranian 已提交
533
	unsigned long flags = 0;
534
	int pid = -1, err;
535

536 537
	if (evsel->fd == NULL &&
	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
538
		return -ENOMEM;
539

S
Stephane Eranian 已提交
540 541 542 543 544
	if (evsel->cgrp) {
		flags = PERF_FLAG_PID_CGROUP;
		pid = evsel->cgrp->fd;
	}

545
	for (cpu = 0; cpu < cpus->nr; cpu++) {
546
		int group_fd = group_fds ? GROUP_FD(group_fds, cpu) : -1;
547

548
		for (thread = 0; thread < threads->nr; thread++) {
S
Stephane Eranian 已提交
549 550 551 552

			if (!evsel->cgrp)
				pid = threads->map[thread];

553
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
554
								     pid,
555
								     cpus->map[cpu],
S
Stephane Eranian 已提交
556
								     group_fd, flags);
557 558
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
559
				goto out_close;
560
			}
561 562 563

			if (group && group_fd == -1)
				group_fd = FD(evsel, cpu, thread);
564
		}
565 566 567 568 569
	}

	return 0;

out_close:
570 571 572 573 574 575 576
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
		thread = threads->nr;
	} while (--cpu >= 0);
577 578 579 580 581 582 583 584 585 586 587
	return err;
}

void perf_evsel__close(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	if (evsel->fd == NULL)
		return;

	perf_evsel__close_fd(evsel, ncpus, nthreads);
	perf_evsel__free_fd(evsel);
	evsel->fd = NULL;
588 589
}

590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605
static struct {
	struct cpu_map map;
	int cpus[1];
} empty_cpu_map = {
	.map.nr	= 1,
	.cpus	= { -1, },
};

static struct {
	struct thread_map map;
	int threads[1];
} empty_thread_map = {
	.map.nr	 = 1,
	.threads = { -1, },
};

606
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
607 608
		     struct thread_map *threads, bool group,
		     struct xyarray *group_fd)
609
{
610 611 612
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
613 614
	}

615 616
	if (threads == NULL)
		threads = &empty_thread_map.map;
617

618
	return __perf_evsel__open(evsel, cpus, threads, group, group_fd);
619 620
}

621
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
622 623
			     struct cpu_map *cpus, bool group,
			     struct xyarray *group_fd)
624
{
625 626
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map, group,
				  group_fd);
627
}
628

629
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
630 631
				struct thread_map *threads, bool group,
				struct xyarray *group_fd)
632
{
633 634
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads, group,
				  group_fd);
635
}
636

637
static int perf_event__parse_id_sample(const union perf_event *event, u64 type,
638 639
				       struct perf_sample *sample,
				       bool swapped)
640 641
{
	const u64 *array = event->sample.array;
642
	union u64_swap u;
643 644 645 646 647

	array += ((event->header.size -
		   sizeof(event->header)) / sizeof(u64)) - 1;

	if (type & PERF_SAMPLE_CPU) {
648 649 650 651 652 653 654 655
		u.val64 = *array;
		if (swapped) {
			/* undo swap of u64, then swap on individual u32s */
			u.val64 = bswap_64(u.val64);
			u.val32[0] = bswap_32(u.val32[0]);
		}

		sample->cpu = u.val32[0];
656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674
		array--;
	}

	if (type & PERF_SAMPLE_STREAM_ID) {
		sample->stream_id = *array;
		array--;
	}

	if (type & PERF_SAMPLE_ID) {
		sample->id = *array;
		array--;
	}

	if (type & PERF_SAMPLE_TIME) {
		sample->time = *array;
		array--;
	}

	if (type & PERF_SAMPLE_TID) {
675 676 677 678 679 680 681 682 683 684
		u.val64 = *array;
		if (swapped) {
			/* undo swap of u64, then swap on individual u32s */
			u.val64 = bswap_64(u.val64);
			u.val32[0] = bswap_32(u.val32[0]);
			u.val32[1] = bswap_32(u.val32[1]);
		}

		sample->pid = u.val32[0];
		sample->tid = u.val32[1];
685 686 687 688 689
	}

	return 0;
}

690 691 692 693 694 695 696 697 698 699 700
static bool sample_overlap(const union perf_event *event,
			   const void *offset, u64 size)
{
	const void *base = event;

	if (offset + size > base + event->header.size)
		return true;

	return false;
}

701
int perf_event__parse_sample(const union perf_event *event, u64 type,
702
			     int sample_size, bool sample_id_all,
703
			     struct perf_sample *data, bool swapped)
704 705 706
{
	const u64 *array;

707 708 709 710
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
711
	union u64_swap u;
712

713
	memset(data, 0, sizeof(*data));
714 715
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
716
	data->period = 1;
717 718 719 720

	if (event->header.type != PERF_RECORD_SAMPLE) {
		if (!sample_id_all)
			return 0;
721
		return perf_event__parse_id_sample(event, type, data, swapped);
722 723 724 725
	}

	array = event->sample.array;

726 727 728
	if (sample_size + sizeof(event->header) > event->header.size)
		return -EFAULT;

729 730 731 732 733 734
	if (type & PERF_SAMPLE_IP) {
		data->ip = event->ip.ip;
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
735 736 737 738 739 740 741 742 743 744
		u.val64 = *array;
		if (swapped) {
			/* undo swap of u64, then swap on individual u32s */
			u.val64 = bswap_64(u.val64);
			u.val32[0] = bswap_32(u.val32[0]);
			u.val32[1] = bswap_32(u.val32[1]);
		}

		data->pid = u.val32[0];
		data->tid = u.val32[1];
745 746 747 748 749 750 751 752
		array++;
	}

	if (type & PERF_SAMPLE_TIME) {
		data->time = *array;
		array++;
	}

753
	data->addr = 0;
754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770
	if (type & PERF_SAMPLE_ADDR) {
		data->addr = *array;
		array++;
	}

	data->id = -1ULL;
	if (type & PERF_SAMPLE_ID) {
		data->id = *array;
		array++;
	}

	if (type & PERF_SAMPLE_STREAM_ID) {
		data->stream_id = *array;
		array++;
	}

	if (type & PERF_SAMPLE_CPU) {
771 772 773 774 775 776 777 778 779

		u.val64 = *array;
		if (swapped) {
			/* undo swap of u64, then swap on individual u32s */
			u.val64 = bswap_64(u.val64);
			u.val32[0] = bswap_32(u.val32[0]);
		}

		data->cpu = u.val32[0];
780 781 782 783 784 785 786 787 788
		array++;
	}

	if (type & PERF_SAMPLE_PERIOD) {
		data->period = *array;
		array++;
	}

	if (type & PERF_SAMPLE_READ) {
M
Masanari Iida 已提交
789
		fprintf(stderr, "PERF_SAMPLE_READ is unsupported for now\n");
790 791 792 793
		return -1;
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
794 795 796
		if (sample_overlap(event, array, sizeof(data->callchain->nr)))
			return -EFAULT;

797
		data->callchain = (struct ip_callchain *)array;
798 799 800 801

		if (sample_overlap(event, array, data->callchain->nr))
			return -EFAULT;

802 803 804 805
		array += 1 + data->callchain->nr;
	}

	if (type & PERF_SAMPLE_RAW) {
J
Jiri Olsa 已提交
806 807
		const u64 *pdata;

808 809 810 811 812 813 814 815
		u.val64 = *array;
		if (WARN_ONCE(swapped,
			      "Endianness of raw data not corrected!\n")) {
			/* undo swap of u64, then swap on individual u32s */
			u.val64 = bswap_64(u.val64);
			u.val32[0] = bswap_32(u.val32[0]);
			u.val32[1] = bswap_32(u.val32[1]);
		}
816 817 818 819

		if (sample_overlap(event, array, sizeof(u32)))
			return -EFAULT;

820
		data->raw_size = u.val32[0];
J
Jiri Olsa 已提交
821
		pdata = (void *) array + sizeof(u32);
822

J
Jiri Olsa 已提交
823
		if (sample_overlap(event, pdata, data->raw_size))
824 825
			return -EFAULT;

J
Jiri Olsa 已提交
826
		data->raw_data = (void *) pdata;
827 828

		array = (void *)array + data->raw_size + sizeof(u32);
829 830
	}

831 832 833 834 835 836 837 838 839 840
	if (type & PERF_SAMPLE_BRANCH_STACK) {
		u64 sz;

		data->branch_stack = (struct branch_stack *)array;
		array++; /* nr */

		sz = data->branch_stack->nr * sizeof(struct branch_entry);
		sz /= sizeof(u64);
		array += sz;
	}
841 842
	return 0;
}
843 844 845 846 847 848 849 850 851 852 853

int perf_event__synthesize_sample(union perf_event *event, u64 type,
				  const struct perf_sample *sample,
				  bool swapped)
{
	u64 *array;

	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
854
	union u64_swap u;
855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918

	array = event->sample.array;

	if (type & PERF_SAMPLE_IP) {
		event->ip.ip = sample->ip;
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
		u.val32[0] = sample->pid;
		u.val32[1] = sample->tid;
		if (swapped) {
			/*
			 * Inverse of what is done in perf_event__parse_sample
			 */
			u.val32[0] = bswap_32(u.val32[0]);
			u.val32[1] = bswap_32(u.val32[1]);
			u.val64 = bswap_64(u.val64);
		}

		*array = u.val64;
		array++;
	}

	if (type & PERF_SAMPLE_TIME) {
		*array = sample->time;
		array++;
	}

	if (type & PERF_SAMPLE_ADDR) {
		*array = sample->addr;
		array++;
	}

	if (type & PERF_SAMPLE_ID) {
		*array = sample->id;
		array++;
	}

	if (type & PERF_SAMPLE_STREAM_ID) {
		*array = sample->stream_id;
		array++;
	}

	if (type & PERF_SAMPLE_CPU) {
		u.val32[0] = sample->cpu;
		if (swapped) {
			/*
			 * Inverse of what is done in perf_event__parse_sample
			 */
			u.val32[0] = bswap_32(u.val32[0]);
			u.val64 = bswap_64(u.val64);
		}
		*array = u.val64;
		array++;
	}

	if (type & PERF_SAMPLE_PERIOD) {
		*array = sample->period;
		array++;
	}

	return 0;
}