evsel.c 26.7 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
#include <byteswap.h>
11
#include <linux/bitops.h>
12
#include "asm/bug.h"
13
#include "debugfs.h"
14
#include "event-parse.h"
15
#include "evsel.h"
16
#include "evlist.h"
17
#include "util.h"
18
#include "cpumap.h"
19
#include "thread_map.h"
20
#include "target.h"
21 22
#include <linux/hw_breakpoint.h>
#include <linux/perf_event.h>
23
#include "perf_regs.h"
24

25 26
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))

27
static int __perf_evsel__sample_size(u64 sample_type)
28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
{
	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;
}

43
void hists__init(struct hists *hists)
44 45 46 47 48 49 50 51 52
{
	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);
}

53 54 55 56 57
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
	evsel->attr	   = *attr;
58
	evsel->leader	   = evsel;
59
	INIT_LIST_HEAD(&evsel->node);
60
	hists__init(&evsel->hists);
61
	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
62 63
}

64
struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
65 66 67
{
	struct perf_evsel *evsel = zalloc(sizeof(*evsel));

68 69
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
70 71 72 73

	return evsel;
}

74
struct event_format *event_format__new(const char *sys, const char *name)
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
{
	int fd, n;
	char *filename;
	void *bf = NULL, *nbf;
	size_t size = 0, alloc_size = 0;
	struct event_format *format = NULL;

	if (asprintf(&filename, "%s/%s/%s/format", tracing_events_path, sys, name) < 0)
		goto out;

	fd = open(filename, O_RDONLY);
	if (fd < 0)
		goto out_free_filename;

	do {
		if (size == alloc_size) {
			alloc_size += BUFSIZ;
			nbf = realloc(bf, alloc_size);
			if (nbf == NULL)
				goto out_free_bf;
			bf = nbf;
		}

		n = read(fd, bf + size, BUFSIZ);
		if (n < 0)
			goto out_free_bf;
		size += n;
	} while (n > 0);

	pevent_parse_format(&format, bf, size, sys);

out_free_bf:
	free(bf);
	close(fd);
out_free_filename:
	free(filename);
out:
	return format;
}

struct perf_evsel *perf_evsel__newtp(const char *sys, const char *name, int idx)
{
	struct perf_evsel *evsel = zalloc(sizeof(*evsel));

	if (evsel != NULL) {
		struct perf_event_attr attr = {
121 122 123
			.type	       = PERF_TYPE_TRACEPOINT,
			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
124 125
		};

126 127 128
		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
			goto out_free;

129 130 131 132
		evsel->tp_format = event_format__new(sys, name);
		if (evsel->tp_format == NULL)
			goto out_free;

133
		event_attr_init(&attr);
134
		attr.config = evsel->tp_format->id;
135
		attr.sample_period = 1;
136 137 138 139 140 141
		perf_evsel__init(evsel, &attr, idx);
	}

	return evsel;

out_free:
142
	free(evsel->name);
143 144 145 146
	free(evsel);
	return NULL;
}

147
const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
148 149 150 151 152 153 154 155 156 157 158 159
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

160
static const char *__perf_evsel__hw_name(u64 config)
161 162 163 164 165 166 167
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

168
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
169
{
170
	int colon = 0, r = 0;
171 172 173 174 175
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
176
			if (!colon) colon = ++r;			\
177 178 179 180 181 182 183 184 185 186 187 188
			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)
189
			colon = ++r;
190 191 192 193 194 195 196 197 198 199
		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)
200
		bf[colon - 1] = ':';
201 202 203
	return r;
}

204 205 206 207 208 209
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);
}

210
const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
211 212 213 214
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
215
	"cpu-migrations",
216 217 218 219 220 221
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
};

222
static const char *__perf_evsel__sw_name(u64 config)
223 224 225 226 227 228 229 230 231 232 233 234
{
	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);
}

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
static int __perf_evsel__bp_name(char *bf, size_t size, u64 addr, u64 type)
{
	int r;

	r = scnprintf(bf, size, "mem:0x%" PRIx64 ":", addr);

	if (type & HW_BREAKPOINT_R)
		r += scnprintf(bf + r, size - r, "r");

	if (type & HW_BREAKPOINT_W)
		r += scnprintf(bf + r, size - r, "w");

	if (type & HW_BREAKPOINT_X)
		r += scnprintf(bf + r, size - r, "x");

	return r;
}

static int perf_evsel__bp_name(struct perf_evsel *evsel, char *bf, size_t size)
{
	struct perf_event_attr *attr = &evsel->attr;
	int r = __perf_evsel__bp_name(bf, size, attr->bp_addr, attr->bp_type);
	return r + perf_evsel__add_modifiers(evsel, bf + r, size - r);
}

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
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]);
}

326
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
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
{
	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);
}

359 360 361 362 363 364
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);
}

365
const char *perf_evsel__name(struct perf_evsel *evsel)
366
{
367
	char bf[128];
368

369 370
	if (evsel->name)
		return evsel->name;
371 372 373

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
374
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
375 376 377
		break;

	case PERF_TYPE_HARDWARE:
378
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
379
		break;
380 381

	case PERF_TYPE_HW_CACHE:
382
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
383 384
		break;

385
	case PERF_TYPE_SOFTWARE:
386
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
387 388
		break;

389
	case PERF_TYPE_TRACEPOINT:
390
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
391 392
		break;

393 394 395 396
	case PERF_TYPE_BREAKPOINT:
		perf_evsel__bp_name(evsel, bf, sizeof(bf));
		break;

397
	default:
398 399
		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
			  evsel->attr.type);
400
		break;
401 402
	}

403 404 405
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
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
/*
 * The enable_on_exec/disabled value strategy:
 *
 *  1) For any type of traced program:
 *    - all independent events and group leaders are disabled
 *    - all group members are enabled
 *
 *     Group members are ruled by group leaders. They need to
 *     be enabled, because the group scheduling relies on that.
 *
 *  2) For traced programs executed by perf:
 *     - all independent events and group leaders have
 *       enable_on_exec set
 *     - we don't specifically enable or disable any event during
 *       the record command
 *
 *     Independent events and group leaders are initially disabled
 *     and get enabled by exec. Group members are ruled by group
 *     leaders as stated in 1).
 *
 *  3) For traced programs attached by perf (pid/tid):
 *     - we specifically enable or disable all events during
 *       the record command
 *
 *     When attaching events to already running traced we
 *     enable/disable events specifically, as there's no
 *     initial traced exec call.
 */
436 437
void perf_evsel__config(struct perf_evsel *evsel,
			struct perf_record_opts *opts)
438 439 440 441
{
	struct perf_event_attr *attr = &evsel->attr;
	int track = !evsel->idx; /* only the first counter needs these */

442
	attr->sample_id_all = opts->sample_id_all_missing ? 0 : 1;
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
	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;
	}

476
	if (opts->call_graph) {
477 478
		attr->sample_type	|= PERF_SAMPLE_CALLCHAIN;

479 480 481 482 483 484 485 486 487
		if (opts->call_graph == CALLCHAIN_DWARF) {
			attr->sample_type |= PERF_SAMPLE_REGS_USER |
					     PERF_SAMPLE_STACK_USER;
			attr->sample_regs_user = PERF_REGS_MASK;
			attr->sample_stack_user = opts->stack_dump_size;
			attr->exclude_callchain_user = 1;
		}
	}

488
	if (perf_target__has_cpu(&opts->target))
489 490
		attr->sample_type	|= PERF_SAMPLE_CPU;

491 492 493
	if (opts->period)
		attr->sample_type	|= PERF_SAMPLE_PERIOD;

494
	if (!opts->sample_id_all_missing &&
495
	    (opts->sample_time || !opts->no_inherit ||
496
	     perf_target__has_cpu(&opts->target)))
497 498 499 500 501 502 503 504 505 506 507 508
		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;
	}
509 510 511 512
	if (opts->branch_stack) {
		attr->sample_type	|= PERF_SAMPLE_BRANCH_STACK;
		attr->branch_sample_type = opts->branch_stack;
	}
513 514 515 516

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

517 518 519 520 521 522
	/*
	 * XXX see the function comment above
	 *
	 * Disabling only independent events or group leaders,
	 * keeping group members enabled.
	 */
523
	if (perf_evsel__is_group_leader(evsel))
524 525 526 527 528 529
		attr->disabled = 1;

	/*
	 * Setting enable_on_exec for independent events and
	 * group leaders for traced executed by perf.
	 */
530
	if (perf_target__none(&opts->target) && perf_evsel__is_group_leader(evsel))
531 532 533
		attr->enable_on_exec = 1;
}

534 535
int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
536
	int cpu, thread;
537
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
538 539 540 541 542 543 544 545 546

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

547 548 549
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567
int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
			   const char *filter)
{
	int cpu, thread;

	for (cpu = 0; cpu < ncpus; cpu++) {
		for (thread = 0; thread < nthreads; thread++) {
			int fd = FD(evsel, cpu, thread),
			    err = ioctl(fd, PERF_EVENT_IOC_SET_FILTER, filter);

			if (err)
				return err;
		}
	}

	return 0;
}

568 569
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
570 571 572 573 574 575 576 577 578 579 580 581
	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;
582 583
}

584 585 586 587 588 589 590
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;
}

591 592 593 594 595 596
void perf_evsel__free_fd(struct perf_evsel *evsel)
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

597 598
void perf_evsel__free_id(struct perf_evsel *evsel)
{
599 600 601
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
	free(evsel->id);
602 603 604
	evsel->id = NULL;
}

605 606 607 608 609 610 611 612 613 614 615
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;
		}
}

616
void perf_evsel__exit(struct perf_evsel *evsel)
617 618 619
{
	assert(list_empty(&evsel->node));
	xyarray__delete(evsel->fd);
620 621
	xyarray__delete(evsel->sample_id);
	free(evsel->id);
622 623 624 625 626
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
S
Stephane Eranian 已提交
627
	close_cgroup(evsel->cgrp);
628
	free(evsel->group_name);
629
	if (evsel->tp_format)
630
		pevent_free_format(evsel->tp_format);
631
	free(evsel->name);
632 633
	free(evsel);
}
634 635 636 637 638 639 640 641 642 643

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;

644 645 646
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668
	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;

669
	aggr->val = aggr->ena = aggr->run = 0;
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704

	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;
}
705

706 707 708 709 710
static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
{
	struct perf_evsel *leader = evsel->leader;
	int fd;

711
	if (perf_evsel__is_group_leader(evsel))
712 713 714 715 716 717 718 719 720 721 722 723 724 725
		return -1;

	/*
	 * Leader must be already processed/open,
	 * if not it's a bug.
	 */
	BUG_ON(!leader->fd);

	fd = FD(leader, cpu, thread);
	BUG_ON(fd == -1);

	return fd;
}

726
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
727
			      struct thread_map *threads)
728
{
729
	int cpu, thread;
S
Stephane Eranian 已提交
730
	unsigned long flags = 0;
731
	int pid = -1, err;
732

733 734
	if (evsel->fd == NULL &&
	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
735
		return -ENOMEM;
736

S
Stephane Eranian 已提交
737 738 739 740 741
	if (evsel->cgrp) {
		flags = PERF_FLAG_PID_CGROUP;
		pid = evsel->cgrp->fd;
	}

742
	for (cpu = 0; cpu < cpus->nr; cpu++) {
743

744
		for (thread = 0; thread < threads->nr; thread++) {
745
			int group_fd;
S
Stephane Eranian 已提交
746 747 748 749

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

750 751
			group_fd = get_group_fd(evsel, cpu, thread);

752
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
753
								     pid,
754
								     cpus->map[cpu],
S
Stephane Eranian 已提交
755
								     group_fd, flags);
756 757
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
758
				goto out_close;
759
			}
760
		}
761 762 763 764 765
	}

	return 0;

out_close:
766 767 768 769 770 771 772
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
		thread = threads->nr;
	} while (--cpu >= 0);
773 774 775 776 777 778 779 780 781 782 783
	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;
784 785
}

786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
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, },
};

802
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
803
		     struct thread_map *threads)
804
{
805 806 807
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
808 809
	}

810 811
	if (threads == NULL)
		threads = &empty_thread_map.map;
812

813
	return __perf_evsel__open(evsel, cpus, threads);
814 815
}

816
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
817
			     struct cpu_map *cpus)
818
{
819
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
820
}
821

822
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
823
				struct thread_map *threads)
824
{
825
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
826
}
827

828 829 830
static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
				       const union perf_event *event,
				       struct perf_sample *sample)
831
{
832
	u64 type = evsel->attr.sample_type;
833
	const u64 *array = event->sample.array;
834
	bool swapped = evsel->needs_swap;
835
	union u64_swap u;
836 837 838 839 840

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

	if (type & PERF_SAMPLE_CPU) {
841 842 843 844 845 846 847 848
		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];
849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867
		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) {
868 869 870 871 872 873 874 875 876 877
		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];
878 879 880 881 882
	}

	return 0;
}

883 884 885 886 887 888 889 890 891 892 893
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;
}

894
int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
895
			     struct perf_sample *data)
896
{
897
	u64 type = evsel->attr.sample_type;
898
	u64 regs_user = evsel->attr.sample_regs_user;
899
	bool swapped = evsel->needs_swap;
900 901
	const u64 *array;

902 903 904 905
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
906
	union u64_swap u;
907

908
	memset(data, 0, sizeof(*data));
909 910
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
911
	data->period = 1;
912 913

	if (event->header.type != PERF_RECORD_SAMPLE) {
914
		if (!evsel->attr.sample_id_all)
915
			return 0;
916
		return perf_evsel__parse_id_sample(evsel, event, data);
917 918 919 920
	}

	array = event->sample.array;

921
	if (evsel->sample_size + sizeof(event->header) > event->header.size)
922 923
		return -EFAULT;

924 925 926 927 928 929
	if (type & PERF_SAMPLE_IP) {
		data->ip = event->ip.ip;
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
930 931 932 933 934 935 936 937 938 939
		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];
940 941 942 943 944 945 946 947
		array++;
	}

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

948
	data->addr = 0;
949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965
	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) {
966 967 968 969 970 971 972 973 974

		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];
975 976 977 978 979 980 981 982 983
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
M
Masanari Iida 已提交
984
		fprintf(stderr, "PERF_SAMPLE_READ is unsupported for now\n");
985 986 987 988
		return -1;
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
989 990 991
		if (sample_overlap(event, array, sizeof(data->callchain->nr)))
			return -EFAULT;

992
		data->callchain = (struct ip_callchain *)array;
993 994 995 996

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

997 998 999 1000
		array += 1 + data->callchain->nr;
	}

	if (type & PERF_SAMPLE_RAW) {
J
Jiri Olsa 已提交
1001 1002
		const u64 *pdata;

1003 1004 1005 1006 1007 1008 1009 1010
		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]);
		}
1011 1012 1013 1014

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

1015
		data->raw_size = u.val32[0];
J
Jiri Olsa 已提交
1016
		pdata = (void *) array + sizeof(u32);
1017

J
Jiri Olsa 已提交
1018
		if (sample_overlap(event, pdata, data->raw_size))
1019 1020
			return -EFAULT;

J
Jiri Olsa 已提交
1021
		data->raw_data = (void *) pdata;
1022 1023

		array = (void *)array + data->raw_size + sizeof(u32);
1024 1025
	}

1026 1027 1028 1029 1030 1031 1032 1033 1034 1035
	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;
	}
1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061

	if (type & PERF_SAMPLE_REGS_USER) {
		/* First u64 tells us if we have any regs in sample. */
		u64 avail = *array++;

		if (avail) {
			data->user_regs.regs = (u64 *)array;
			array += hweight_long(regs_user);
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
		u64 size = *array++;

		data->user_stack.offset = ((char *)(array - 1)
					  - (char *) event);

		if (!size) {
			data->user_stack.size = 0;
		} else {
			data->user_stack.data = (char *)array;
			array += size / sizeof(*array);
			data->user_stack.size = *array;
		}
	}

1062 1063
	return 0;
}
1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074

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.
	 */
1075
	union u64_swap u;
1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088

	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) {
			/*
1089
			 * Inverse of what is done in perf_evsel__parse_sample
1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123
			 */
			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) {
			/*
1124
			 * Inverse of what is done in perf_evsel__parse_sample
1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139
			 */
			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;
}
1140

1141 1142 1143 1144 1145
struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
{
	return pevent_find_field(evsel->tp_format, name);
}

1146
void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
1147 1148
			 const char *name)
{
1149
	struct format_field *field = perf_evsel__field(evsel, name);
1150 1151
	int offset;

1152 1153
	if (!field)
		return NULL;
1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167

	offset = field->offset;

	if (field->flags & FIELD_IS_DYNAMIC) {
		offset = *(int *)(sample->raw_data + field->offset);
		offset &= 0xffff;
	}

	return sample->raw_data + offset;
}

u64 perf_evsel__intval(struct perf_evsel *evsel, struct perf_sample *sample,
		       const char *name)
{
1168
	struct format_field *field = perf_evsel__field(evsel, name);
1169 1170
	void *ptr;
	u64 value;
1171

1172 1173
	if (!field)
		return 0;
1174

1175
	ptr = sample->raw_data + field->offset;
1176

1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207
	switch (field->size) {
	case 1:
		return *(u8 *)ptr;
	case 2:
		value = *(u16 *)ptr;
		break;
	case 4:
		value = *(u32 *)ptr;
		break;
	case 8:
		value = *(u64 *)ptr;
		break;
	default:
		return 0;
	}

	if (!evsel->needs_swap)
		return value;

	switch (field->size) {
	case 2:
		return bswap_16(value);
	case 4:
		return bswap_32(value);
	case 8:
		return bswap_64(value);
	default:
		return 0;
	}

	return 0;
1208
}