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
#include "../../include/linux/perf_event.h"
19

20
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))
21
#define GROUP_FD(group_fd, cpu) (*(int *)xyarray__entry(group_fd, cpu, 0))
22

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
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;
}

39
void hists__init(struct hists *hists)
40 41 42 43 44 45 46 47 48
{
	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);
}

49 50 51 52 53 54
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);
55
	hists__init(&evsel->hists);
56 57
}

58
struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
59 60 61
{
	struct perf_evsel *evsel = zalloc(sizeof(*evsel));

62 63
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
64 65 66 67

	return evsel;
}

68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
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",
};

const char *__perf_evsel__hw_name(u64 config)
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

89
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
90
{
91
	int colon = 0, r = 0;
92 93 94 95 96
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
97
			if (!colon) colon = ++r;			\
98 99 100 101 102 103 104 105 106 107 108 109
			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)
110
			colon = ++r;
111 112 113 114 115 116 117 118 119 120
		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)
121
		bf[colon - 1] = ':';
122 123 124
	return r;
}

125 126 127 128 129 130
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);
}

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
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",
};

const char *__perf_evsel__sw_name(u64 config)
{
	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);
}

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
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]);
}

int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
{
	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);
}

255 256 257 258 259 260 261 262 263 264 265 266
int perf_evsel__name(struct perf_evsel *evsel, char *bf, size_t size)
{
	int ret;

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
		ret = scnprintf(bf, size, "raw 0x%" PRIx64, evsel->attr.config);
		break;

	case PERF_TYPE_HARDWARE:
		ret = perf_evsel__hw_name(evsel, bf, size);
		break;
267 268 269 270 271

	case PERF_TYPE_HW_CACHE:
		ret = perf_evsel__hw_cache_name(evsel, bf, size);
		break;

272 273 274 275
	case PERF_TYPE_SOFTWARE:
		ret = perf_evsel__sw_name(evsel, bf, size);
		break;

276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
	default:
		/*
		 * FIXME
 		 *
		 * This is the minimal perf_evsel__name so that we can
		 * reconstruct event names taking into account event modifiers.
		 *
		 * The old event_name uses it now for raw anr hw events, so that
		 * we don't drag all the parsing stuff into the python binding.
		 *
		 * On the next devel cycle the rest of the event naming will be
		 * brought here.
 		 */
		return 0;
	}

	return ret;
}

295 296
void perf_evsel__config(struct perf_evsel *evsel, struct perf_record_opts *opts,
			struct perf_evsel *first)
297 298 299 300
{
	struct perf_event_attr *attr = &evsel->attr;
	int track = !evsel->idx; /* only the first counter needs these */

301
	attr->disabled = 1;
302
	attr->sample_id_all = opts->sample_id_all_missing ? 0 : 1;
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
	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;

339
	if (perf_target__has_cpu(&opts->target))
340 341
		attr->sample_type	|= PERF_SAMPLE_CPU;

342 343 344
	if (opts->period)
		attr->sample_type	|= PERF_SAMPLE_PERIOD;

345
	if (!opts->sample_id_all_missing &&
346
	    (opts->sample_time || !opts->no_inherit ||
347
	     perf_target__has_cpu(&opts->target)))
348 349 350 351 352 353 354 355 356 357 358 359
		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;
	}
360 361 362 363
	if (opts->branch_stack) {
		attr->sample_type	|= PERF_SAMPLE_BRANCH_STACK;
		attr->branch_sample_type = opts->branch_stack;
	}
364 365 366 367

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

368 369
	if (perf_target__none(&opts->target) &&
	    (!opts->group || evsel == first)) {
370 371 372 373
		attr->enable_on_exec = 1;
	}
}

374 375
int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
376
	int cpu, thread;
377
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
378 379 380 381 382 383 384 385 386

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

387 388 389
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

390 391
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
392 393 394 395 396 397 398 399 400 401 402 403
	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;
404 405
}

406 407 408 409 410 411 412
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;
}

413 414 415 416 417 418
void perf_evsel__free_fd(struct perf_evsel *evsel)
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

419 420
void perf_evsel__free_id(struct perf_evsel *evsel)
{
421 422 423
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
	free(evsel->id);
424 425 426
	evsel->id = NULL;
}

427 428 429 430 431 432 433 434 435 436 437
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;
		}
}

438
void perf_evsel__exit(struct perf_evsel *evsel)
439 440 441
{
	assert(list_empty(&evsel->node));
	xyarray__delete(evsel->fd);
442 443
	xyarray__delete(evsel->sample_id);
	free(evsel->id);
444 445 446 447 448
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
S
Stephane Eranian 已提交
449
	close_cgroup(evsel->cgrp);
450
	free(evsel->name);
451 452
	free(evsel);
}
453 454 455 456 457 458 459 460 461 462

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;

463 464 465
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
	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;

488
	aggr->val = aggr->ena = aggr->run = 0;
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

	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;
}
524

525
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
526 527
			      struct thread_map *threads, bool group,
			      struct xyarray *group_fds)
528
{
529
	int cpu, thread;
S
Stephane Eranian 已提交
530
	unsigned long flags = 0;
531
	int pid = -1, err;
532

533 534
	if (evsel->fd == NULL &&
	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
535
		return -ENOMEM;
536

S
Stephane Eranian 已提交
537 538 539 540 541
	if (evsel->cgrp) {
		flags = PERF_FLAG_PID_CGROUP;
		pid = evsel->cgrp->fd;
	}

542
	for (cpu = 0; cpu < cpus->nr; cpu++) {
543
		int group_fd = group_fds ? GROUP_FD(group_fds, cpu) : -1;
544

545
		for (thread = 0; thread < threads->nr; thread++) {
S
Stephane Eranian 已提交
546 547 548 549

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

550
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
551
								     pid,
552
								     cpus->map[cpu],
S
Stephane Eranian 已提交
553
								     group_fd, flags);
554 555
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
556
				goto out_close;
557
			}
558 559 560

			if (group && group_fd == -1)
				group_fd = FD(evsel, cpu, thread);
561
		}
562 563 564 565 566
	}

	return 0;

out_close:
567 568 569 570 571 572 573
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
		thread = threads->nr;
	} while (--cpu >= 0);
574 575 576 577 578 579 580 581 582 583 584
	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;
585 586
}

587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602
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, },
};

603
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
604 605
		     struct thread_map *threads, bool group,
		     struct xyarray *group_fd)
606
{
607 608 609
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
610 611
	}

612 613
	if (threads == NULL)
		threads = &empty_thread_map.map;
614

615
	return __perf_evsel__open(evsel, cpus, threads, group, group_fd);
616 617
}

618
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
619 620
			     struct cpu_map *cpus, bool group,
			     struct xyarray *group_fd)
621
{
622 623
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map, group,
				  group_fd);
624
}
625

626
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
627 628
				struct thread_map *threads, bool group,
				struct xyarray *group_fd)
629
{
630 631
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads, group,
				  group_fd);
632
}
633

634
static int perf_event__parse_id_sample(const union perf_event *event, u64 type,
635 636
				       struct perf_sample *sample,
				       bool swapped)
637 638
{
	const u64 *array = event->sample.array;
639
	union u64_swap u;
640 641 642 643 644

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

	if (type & PERF_SAMPLE_CPU) {
645 646 647 648 649 650 651 652
		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];
653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671
		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) {
672 673 674 675 676 677 678 679 680 681
		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];
682 683 684 685 686
	}

	return 0;
}

687 688 689 690 691 692 693 694 695 696 697
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;
}

698
int perf_event__parse_sample(const union perf_event *event, u64 type,
699
			     int sample_size, bool sample_id_all,
700
			     struct perf_sample *data, bool swapped)
701 702 703
{
	const u64 *array;

704 705 706 707
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
708
	union u64_swap u;
709

710
	memset(data, 0, sizeof(*data));
711 712
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
713
	data->period = 1;
714 715 716 717

	if (event->header.type != PERF_RECORD_SAMPLE) {
		if (!sample_id_all)
			return 0;
718
		return perf_event__parse_id_sample(event, type, data, swapped);
719 720 721 722
	}

	array = event->sample.array;

723 724 725
	if (sample_size + sizeof(event->header) > event->header.size)
		return -EFAULT;

726 727 728 729 730 731
	if (type & PERF_SAMPLE_IP) {
		data->ip = event->ip.ip;
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
732 733 734 735 736 737 738 739 740 741
		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];
742 743 744 745 746 747 748 749
		array++;
	}

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

750
	data->addr = 0;
751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767
	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) {
768 769 770 771 772 773 774 775 776

		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];
777 778 779 780 781 782 783 784 785
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
M
Masanari Iida 已提交
786
		fprintf(stderr, "PERF_SAMPLE_READ is unsupported for now\n");
787 788 789 790
		return -1;
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
791 792 793
		if (sample_overlap(event, array, sizeof(data->callchain->nr)))
			return -EFAULT;

794
		data->callchain = (struct ip_callchain *)array;
795 796 797 798

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

799 800 801 802
		array += 1 + data->callchain->nr;
	}

	if (type & PERF_SAMPLE_RAW) {
J
Jiri Olsa 已提交
803 804
		const u64 *pdata;

805 806 807 808 809 810 811 812
		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]);
		}
813 814 815 816

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

817
		data->raw_size = u.val32[0];
J
Jiri Olsa 已提交
818
		pdata = (void *) array + sizeof(u32);
819

J
Jiri Olsa 已提交
820
		if (sample_overlap(event, pdata, data->raw_size))
821 822
			return -EFAULT;

J
Jiri Olsa 已提交
823
		data->raw_data = (void *) pdata;
824 825

		array = (void *)array + data->raw_size + sizeof(u32);
826 827
	}

828 829 830 831 832 833 834 835 836 837
	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;
	}
838 839
	return 0;
}
840 841 842 843 844 845 846 847 848 849 850

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.
	 */
851
	union u64_swap u;
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915

	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;
}