evsel.c 27.2 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 58 59 60 61 62 63 64 65 66 67 68 69 70
void __perf_evsel__set_sample_bit(struct perf_evsel *evsel,
				  enum perf_event_sample_format bit)
{
	if (!(evsel->attr.sample_type & bit)) {
		evsel->attr.sample_type |= bit;
		evsel->sample_size += sizeof(u64);
	}
}

void __perf_evsel__reset_sample_bit(struct perf_evsel *evsel,
				    enum perf_event_sample_format bit)
{
	if (evsel->attr.sample_type & bit) {
		evsel->attr.sample_type &= ~bit;
		evsel->sample_size -= sizeof(u64);
	}
}

71 72 73 74 75
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
	evsel->attr	   = *attr;
76
	evsel->leader	   = evsel;
77
	INIT_LIST_HEAD(&evsel->node);
78
	hists__init(&evsel->hists);
79
	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
80 81
}

82
struct perf_evsel *perf_evsel__new(struct perf_event_attr *attr, int idx)
83 84 85
{
	struct perf_evsel *evsel = zalloc(sizeof(*evsel));

86 87
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
88 89 90 91

	return evsel;
}

92
struct event_format *event_format__new(const char *sys, const char *name)
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
{
	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 = {
139 140 141
			.type	       = PERF_TYPE_TRACEPOINT,
			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
142 143
		};

144 145 146
		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
			goto out_free;

147 148 149 150
		evsel->tp_format = event_format__new(sys, name);
		if (evsel->tp_format == NULL)
			goto out_free;

151
		event_attr_init(&attr);
152
		attr.config = evsel->tp_format->id;
153
		attr.sample_period = 1;
154 155 156 157 158 159
		perf_evsel__init(evsel, &attr, idx);
	}

	return evsel;

out_free:
160
	free(evsel->name);
161 162 163 164
	free(evsel);
	return NULL;
}

165
const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
166 167 168 169 170 171 172 173 174 175 176 177
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

178
static const char *__perf_evsel__hw_name(u64 config)
179 180 181 182 183 184 185
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

186
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
187
{
188
	int colon = 0, r = 0;
189 190 191 192 193
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
194
			if (!colon) colon = ++r;			\
195 196 197 198 199 200 201 202 203 204 205 206
			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)
207
			colon = ++r;
208 209 210 211 212 213 214 215 216 217
		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)
218
		bf[colon - 1] = ':';
219 220 221
	return r;
}

222 223 224 225 226 227
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);
}

228
const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
229 230 231 232
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
233
	"cpu-migrations",
234 235 236 237 238 239
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
};

240
static const char *__perf_evsel__sw_name(u64 config)
241 242 243 244 245 246 247 248 249 250 251 252
{
	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);
}

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
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);
}

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
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]);
}

344
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
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
{
	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);
}

377 378 379 380 381 382
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);
}

383
const char *perf_evsel__name(struct perf_evsel *evsel)
384
{
385
	char bf[128];
386

387 388
	if (evsel->name)
		return evsel->name;
389 390 391

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
392
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
393 394 395
		break;

	case PERF_TYPE_HARDWARE:
396
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
397
		break;
398 399

	case PERF_TYPE_HW_CACHE:
400
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
401 402
		break;

403
	case PERF_TYPE_SOFTWARE:
404
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
405 406
		break;

407
	case PERF_TYPE_TRACEPOINT:
408
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
409 410
		break;

411 412 413 414
	case PERF_TYPE_BREAKPOINT:
		perf_evsel__bp_name(evsel, bf, sizeof(bf));
		break;

415
	default:
416 417
		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
			  evsel->attr.type);
418
		break;
419 420
	}

421 422 423
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
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
/*
 * 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.
 */
454 455
void perf_evsel__config(struct perf_evsel *evsel,
			struct perf_record_opts *opts)
456 457 458 459
{
	struct perf_event_attr *attr = &evsel->attr;
	int track = !evsel->idx; /* only the first counter needs these */

460
	attr->sample_id_all = opts->sample_id_all_missing ? 0 : 1;
461 462 463 464 465
	attr->inherit	    = !opts->no_inherit;
	attr->read_format   = PERF_FORMAT_TOTAL_TIME_ENABLED |
			      PERF_FORMAT_TOTAL_TIME_RUNNING |
			      PERF_FORMAT_ID;

466 467
	perf_evsel__set_sample_bit(evsel, IP);
	perf_evsel__set_sample_bit(evsel, TID);
468 469 470 471 472 473 474 475

	/*
	 * 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) {
476
			perf_evsel__set_sample_bit(evsel, PERIOD);
477 478 479 480 481 482 483 484 485 486 487 488 489 490
			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) {
491
		perf_evsel__set_sample_bit(evsel, ADDR);
492 493 494
		attr->mmap_data = track;
	}

495
	if (opts->call_graph) {
496
		perf_evsel__set_sample_bit(evsel, CALLCHAIN);
497

498
		if (opts->call_graph == CALLCHAIN_DWARF) {
499 500
			perf_evsel__set_sample_bit(evsel, REGS_USER);
			perf_evsel__set_sample_bit(evsel, STACK_USER);
501 502 503 504 505 506
			attr->sample_regs_user = PERF_REGS_MASK;
			attr->sample_stack_user = opts->stack_dump_size;
			attr->exclude_callchain_user = 1;
		}
	}

507
	if (perf_target__has_cpu(&opts->target))
508
		perf_evsel__set_sample_bit(evsel, CPU);
509

510
	if (opts->period)
511
		perf_evsel__set_sample_bit(evsel, PERIOD);
512

513
	if (!opts->sample_id_all_missing &&
514
	    (opts->sample_time || !opts->no_inherit ||
515
	     perf_target__has_cpu(&opts->target)))
516
		perf_evsel__set_sample_bit(evsel, TIME);
517 518

	if (opts->raw_samples) {
519 520 521
		perf_evsel__set_sample_bit(evsel, TIME);
		perf_evsel__set_sample_bit(evsel, RAW);
		perf_evsel__set_sample_bit(evsel, CPU);
522 523 524 525 526 527
	}

	if (opts->no_delay) {
		attr->watermark = 0;
		attr->wakeup_events = 1;
	}
528
	if (opts->branch_stack) {
529
		perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
530 531
		attr->branch_sample_type = opts->branch_stack;
	}
532 533 534 535

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

536 537 538 539 540 541
	/*
	 * XXX see the function comment above
	 *
	 * Disabling only independent events or group leaders,
	 * keeping group members enabled.
	 */
542
	if (perf_evsel__is_group_leader(evsel))
543 544 545 546 547 548
		attr->disabled = 1;

	/*
	 * Setting enable_on_exec for independent events and
	 * group leaders for traced executed by perf.
	 */
549
	if (perf_target__none(&opts->target) && perf_evsel__is_group_leader(evsel))
550 551 552
		attr->enable_on_exec = 1;
}

553 554
int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
555
	int cpu, thread;
556
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
557 558 559 560 561 562 563 564 565

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

566 567 568
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586
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;
}

587 588
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
589 590 591 592 593 594 595 596 597 598 599 600
	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;
601 602
}

603 604 605 606 607 608 609
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;
}

610 611 612 613 614 615
void perf_evsel__free_fd(struct perf_evsel *evsel)
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

616 617
void perf_evsel__free_id(struct perf_evsel *evsel)
{
618 619 620
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
	free(evsel->id);
621 622 623
	evsel->id = NULL;
}

624 625 626 627 628 629 630 631 632 633 634
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;
		}
}

635
void perf_evsel__exit(struct perf_evsel *evsel)
636 637 638
{
	assert(list_empty(&evsel->node));
	xyarray__delete(evsel->fd);
639 640
	xyarray__delete(evsel->sample_id);
	free(evsel->id);
641 642 643 644 645
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
S
Stephane Eranian 已提交
646
	close_cgroup(evsel->cgrp);
647
	free(evsel->group_name);
648
	if (evsel->tp_format)
649
		pevent_free_format(evsel->tp_format);
650
	free(evsel->name);
651 652
	free(evsel);
}
653 654 655 656 657 658 659 660 661 662

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;

663 664 665
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687
	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;

688
	aggr->val = aggr->ena = aggr->run = 0;
689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723

	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;
}
724

725 726 727 728 729
static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
{
	struct perf_evsel *leader = evsel->leader;
	int fd;

730
	if (perf_evsel__is_group_leader(evsel))
731 732 733 734 735 736 737 738 739 740 741 742 743 744
		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;
}

745
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
746
			      struct thread_map *threads)
747
{
748
	int cpu, thread;
S
Stephane Eranian 已提交
749
	unsigned long flags = 0;
750
	int pid = -1, err;
751

752 753
	if (evsel->fd == NULL &&
	    perf_evsel__alloc_fd(evsel, cpus->nr, threads->nr) < 0)
754
		return -ENOMEM;
755

S
Stephane Eranian 已提交
756 757 758 759 760
	if (evsel->cgrp) {
		flags = PERF_FLAG_PID_CGROUP;
		pid = evsel->cgrp->fd;
	}

761
	for (cpu = 0; cpu < cpus->nr; cpu++) {
762

763
		for (thread = 0; thread < threads->nr; thread++) {
764
			int group_fd;
S
Stephane Eranian 已提交
765 766 767 768

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

769 770
			group_fd = get_group_fd(evsel, cpu, thread);

771
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
772
								     pid,
773
								     cpus->map[cpu],
S
Stephane Eranian 已提交
774
								     group_fd, flags);
775 776
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
777
				goto out_close;
778
			}
779
		}
780 781 782 783 784
	}

	return 0;

out_close:
785 786 787 788 789 790 791
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
		thread = threads->nr;
	} while (--cpu >= 0);
792 793 794 795 796 797 798 799 800 801 802
	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;
803 804
}

805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820
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, },
};

821
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
822
		     struct thread_map *threads)
823
{
824 825 826
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
827 828
	}

829 830
	if (threads == NULL)
		threads = &empty_thread_map.map;
831

832
	return __perf_evsel__open(evsel, cpus, threads);
833 834
}

835
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
836
			     struct cpu_map *cpus)
837
{
838
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
839
}
840

841
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
842
				struct thread_map *threads)
843
{
844
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
845
}
846

847 848 849
static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
				       const union perf_event *event,
				       struct perf_sample *sample)
850
{
851
	u64 type = evsel->attr.sample_type;
852
	const u64 *array = event->sample.array;
853
	bool swapped = evsel->needs_swap;
854
	union u64_swap u;
855 856 857 858 859

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

	if (type & PERF_SAMPLE_CPU) {
860 861 862 863 864 865 866 867
		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];
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886
		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) {
887 888 889 890 891 892 893 894 895 896
		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];
897 898 899 900 901
	}

	return 0;
}

902 903 904 905 906 907 908 909 910 911 912
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;
}

913
int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
914
			     struct perf_sample *data)
915
{
916
	u64 type = evsel->attr.sample_type;
917
	u64 regs_user = evsel->attr.sample_regs_user;
918
	bool swapped = evsel->needs_swap;
919 920
	const u64 *array;

921 922 923 924
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
925
	union u64_swap u;
926

927
	memset(data, 0, sizeof(*data));
928 929
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
930
	data->period = 1;
931 932

	if (event->header.type != PERF_RECORD_SAMPLE) {
933
		if (!evsel->attr.sample_id_all)
934
			return 0;
935
		return perf_evsel__parse_id_sample(evsel, event, data);
936 937 938 939
	}

	array = event->sample.array;

940
	if (evsel->sample_size + sizeof(event->header) > event->header.size)
941 942
		return -EFAULT;

943 944 945 946 947 948
	if (type & PERF_SAMPLE_IP) {
		data->ip = event->ip.ip;
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
949 950 951 952 953 954 955 956 957 958
		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];
959 960 961 962 963 964 965 966
		array++;
	}

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

967
	data->addr = 0;
968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984
	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) {
985 986 987 988 989 990 991 992 993

		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];
994 995 996 997 998 999 1000 1001 1002
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
M
Masanari Iida 已提交
1003
		fprintf(stderr, "PERF_SAMPLE_READ is unsupported for now\n");
1004 1005 1006 1007
		return -1;
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
1008 1009 1010
		if (sample_overlap(event, array, sizeof(data->callchain->nr)))
			return -EFAULT;

1011
		data->callchain = (struct ip_callchain *)array;
1012 1013 1014 1015

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

1016 1017 1018 1019
		array += 1 + data->callchain->nr;
	}

	if (type & PERF_SAMPLE_RAW) {
J
Jiri Olsa 已提交
1020 1021
		const u64 *pdata;

1022 1023 1024 1025 1026 1027 1028 1029
		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]);
		}
1030 1031 1032 1033

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

1034
		data->raw_size = u.val32[0];
J
Jiri Olsa 已提交
1035
		pdata = (void *) array + sizeof(u32);
1036

J
Jiri Olsa 已提交
1037
		if (sample_overlap(event, pdata, data->raw_size))
1038 1039
			return -EFAULT;

J
Jiri Olsa 已提交
1040
		data->raw_data = (void *) pdata;
1041 1042

		array = (void *)array + data->raw_size + sizeof(u32);
1043 1044
	}

1045 1046 1047 1048 1049 1050 1051 1052 1053 1054
	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;
	}
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080

	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;
		}
	}

1081 1082
	return 0;
}
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093

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.
	 */
1094
	union u64_swap u;
1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107

	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) {
			/*
1108
			 * Inverse of what is done in perf_evsel__parse_sample
1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142
			 */
			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) {
			/*
1143
			 * Inverse of what is done in perf_evsel__parse_sample
1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
			 */
			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;
}
1159

1160 1161 1162 1163 1164
struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
{
	return pevent_find_field(evsel->tp_format, name);
}

1165
void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
1166 1167
			 const char *name)
{
1168
	struct format_field *field = perf_evsel__field(evsel, name);
1169 1170
	int offset;

1171 1172
	if (!field)
		return NULL;
1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186

	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)
{
1187
	struct format_field *field = perf_evsel__field(evsel, name);
1188 1189
	void *ptr;
	u64 value;
1190

1191 1192
	if (!field)
		return 0;
1193

1194
	ptr = sample->raw_data + field->offset;
1195

1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
	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;
1227
}