evsel.c 49.7 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 <api/fs/debugfs.h>
13 14 15
#include <traceevent/event-parse.h>
#include <linux/hw_breakpoint.h>
#include <linux/perf_event.h>
16
#include <sys/resource.h>
17
#include "asm/bug.h"
18
#include "callchain.h"
19
#include "evsel.h"
20
#include "evlist.h"
21
#include "util.h"
22
#include "cpumap.h"
23
#include "thread_map.h"
24
#include "target.h"
25
#include "perf_regs.h"
A
Adrian Hunter 已提交
26
#include "debug.h"
27
#include "trace-event.h"
28

29 30 31
static struct {
	bool sample_id_all;
	bool exclude_guest;
32
	bool mmap2;
33
	bool cloexec;
34 35
} perf_missing_features;

A
Arnaldo Carvalho de Melo 已提交
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
static int perf_evsel__no_extra_init(struct perf_evsel *evsel __maybe_unused)
{
	return 0;
}

static void perf_evsel__no_extra_fini(struct perf_evsel *evsel __maybe_unused)
{
}

static struct {
	size_t	size;
	int	(*init)(struct perf_evsel *evsel);
	void	(*fini)(struct perf_evsel *evsel);
} perf_evsel__object = {
	.size = sizeof(struct perf_evsel),
	.init = perf_evsel__no_extra_init,
	.fini = perf_evsel__no_extra_fini,
};

int perf_evsel__object_config(size_t object_size,
			      int (*init)(struct perf_evsel *evsel),
			      void (*fini)(struct perf_evsel *evsel))
{

	if (object_size == 0)
		goto set_methods;

	if (perf_evsel__object.size > object_size)
		return -EINVAL;

	perf_evsel__object.size = object_size;

set_methods:
	if (init != NULL)
		perf_evsel__object.init = init;

	if (fini != NULL)
		perf_evsel__object.fini = fini;

	return 0;
}

78 79
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))

80
int __perf_evsel__sample_size(u64 sample_type)
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
{
	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;
}

96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
/**
 * __perf_evsel__calc_id_pos - calculate id_pos.
 * @sample_type: sample type
 *
 * This function returns the position of the event id (PERF_SAMPLE_ID or
 * PERF_SAMPLE_IDENTIFIER) in a sample event i.e. in the array of struct
 * sample_event.
 */
static int __perf_evsel__calc_id_pos(u64 sample_type)
{
	int idx = 0;

	if (sample_type & PERF_SAMPLE_IDENTIFIER)
		return 0;

	if (!(sample_type & PERF_SAMPLE_ID))
		return -1;

	if (sample_type & PERF_SAMPLE_IP)
		idx += 1;

	if (sample_type & PERF_SAMPLE_TID)
		idx += 1;

	if (sample_type & PERF_SAMPLE_TIME)
		idx += 1;

	if (sample_type & PERF_SAMPLE_ADDR)
		idx += 1;

	return idx;
}

/**
 * __perf_evsel__calc_is_pos - calculate is_pos.
 * @sample_type: sample type
 *
 * This function returns the position (counting backwards) of the event id
 * (PERF_SAMPLE_ID or PERF_SAMPLE_IDENTIFIER) in a non-sample event i.e. if
 * sample_id_all is used there is an id sample appended to non-sample events.
 */
static int __perf_evsel__calc_is_pos(u64 sample_type)
{
	int idx = 1;

	if (sample_type & PERF_SAMPLE_IDENTIFIER)
		return 1;

	if (!(sample_type & PERF_SAMPLE_ID))
		return -1;

	if (sample_type & PERF_SAMPLE_CPU)
		idx += 1;

	if (sample_type & PERF_SAMPLE_STREAM_ID)
		idx += 1;

	return idx;
}

void perf_evsel__calc_id_pos(struct perf_evsel *evsel)
{
	evsel->id_pos = __perf_evsel__calc_id_pos(evsel->attr.sample_type);
	evsel->is_pos = __perf_evsel__calc_is_pos(evsel->attr.sample_type);
}

162 163 164 165 166 167
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);
168
		perf_evsel__calc_id_pos(evsel);
169 170 171 172 173 174 175 176 177
	}
}

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);
178
		perf_evsel__calc_id_pos(evsel);
179 180 181
	}
}

182 183
void perf_evsel__set_sample_id(struct perf_evsel *evsel,
			       bool can_sample_identifier)
184
{
185 186 187 188 189 190
	if (can_sample_identifier) {
		perf_evsel__reset_sample_bit(evsel, ID);
		perf_evsel__set_sample_bit(evsel, IDENTIFIER);
	} else {
		perf_evsel__set_sample_bit(evsel, ID);
	}
191 192 193
	evsel->attr.read_format |= PERF_FORMAT_ID;
}

194 195 196 197
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
198
	evsel->tracking	   = !idx;
199
	evsel->attr	   = *attr;
200
	evsel->leader	   = evsel;
201 202
	evsel->unit	   = "";
	evsel->scale	   = 1.0;
203
	INIT_LIST_HEAD(&evsel->node);
A
Arnaldo Carvalho de Melo 已提交
204
	perf_evsel__object.init(evsel);
205
	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
206
	perf_evsel__calc_id_pos(evsel);
207 208
}

209
struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx)
210
{
A
Arnaldo Carvalho de Melo 已提交
211
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
212

213 214
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
215 216 217 218

	return evsel;
}

219
struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx)
220
{
A
Arnaldo Carvalho de Melo 已提交
221
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
222 223 224

	if (evsel != NULL) {
		struct perf_event_attr attr = {
225 226 227
			.type	       = PERF_TYPE_TRACEPOINT,
			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
228 229
		};

230 231 232
		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
			goto out_free;

233
		evsel->tp_format = trace_event__tp_format(sys, name);
234 235 236
		if (evsel->tp_format == NULL)
			goto out_free;

237
		event_attr_init(&attr);
238
		attr.config = evsel->tp_format->id;
239
		attr.sample_period = 1;
240 241 242 243 244 245
		perf_evsel__init(evsel, &attr, idx);
	}

	return evsel;

out_free:
246
	zfree(&evsel->name);
247 248 249 250
	free(evsel);
	return NULL;
}

251
const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
252 253 254 255 256 257 258 259 260 261 262 263
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

264
static const char *__perf_evsel__hw_name(u64 config)
265 266 267 268 269 270 271
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

272
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
273
{
274
	int colon = 0, r = 0;
275 276 277 278 279
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
280
			if (!colon) colon = ++r;			\
281 282 283 284 285 286 287 288 289 290 291 292
			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)
293
			colon = ++r;
294 295 296 297 298 299 300 301 302 303
		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)
304
		bf[colon - 1] = ':';
305 306 307
	return r;
}

308 309 310 311 312 313
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);
}

314
const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
315 316 317 318
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
319
	"cpu-migrations",
320 321 322 323
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
324
	"dummy",
325 326
};

327
static const char *__perf_evsel__sw_name(u64 config)
328 329 330 331 332 333 334 335 336 337 338 339
{
	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);
}

340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364
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);
}

365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
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]);
}

431
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463
{
	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);
}

464 465 466 467 468 469
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);
}

470
const char *perf_evsel__name(struct perf_evsel *evsel)
471
{
472
	char bf[128];
473

474 475
	if (evsel->name)
		return evsel->name;
476 477 478

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
479
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
480 481 482
		break;

	case PERF_TYPE_HARDWARE:
483
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
484
		break;
485 486

	case PERF_TYPE_HW_CACHE:
487
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
488 489
		break;

490
	case PERF_TYPE_SOFTWARE:
491
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
492 493
		break;

494
	case PERF_TYPE_TRACEPOINT:
495
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
496 497
		break;

498 499 500 501
	case PERF_TYPE_BREAKPOINT:
		perf_evsel__bp_name(evsel, bf, sizeof(bf));
		break;

502
	default:
503 504
		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
			  evsel->attr.type);
505
		break;
506 507
	}

508 509 510
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
511 512
}

513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537
const char *perf_evsel__group_name(struct perf_evsel *evsel)
{
	return evsel->group_name ?: "anon group";
}

int perf_evsel__group_desc(struct perf_evsel *evsel, char *buf, size_t size)
{
	int ret;
	struct perf_evsel *pos;
	const char *group_name = perf_evsel__group_name(evsel);

	ret = scnprintf(buf, size, "%s", group_name);

	ret += scnprintf(buf + ret, size - ret, " { %s",
			 perf_evsel__name(evsel));

	for_each_group_member(pos, evsel)
		ret += scnprintf(buf + ret, size - ret, ", %s",
				 perf_evsel__name(pos));

	ret += scnprintf(buf + ret, size - ret, " }");

	return ret;
}

538
static void
539
perf_evsel__config_callgraph(struct perf_evsel *evsel)
540 541 542 543 544 545
{
	bool function = perf_evsel__is_function_event(evsel);
	struct perf_event_attr *attr = &evsel->attr;

	perf_evsel__set_sample_bit(evsel, CALLCHAIN);

546
	if (callchain_param.record_mode == CALLCHAIN_DWARF) {
547 548 549 550
		if (!function) {
			perf_evsel__set_sample_bit(evsel, REGS_USER);
			perf_evsel__set_sample_bit(evsel, STACK_USER);
			attr->sample_regs_user = PERF_REGS_MASK;
551
			attr->sample_stack_user = callchain_param.dump_size;
552 553 554 555 556 557 558 559 560 561 562 563 564
			attr->exclude_callchain_user = 1;
		} else {
			pr_info("Cannot use DWARF unwind for function trace event,"
				" falling back to framepointers.\n");
		}
	}

	if (function) {
		pr_info("Disabling user space callchains for function trace event.\n");
		attr->exclude_callchain_user = 1;
	}
}

565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592
/*
 * 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.
 */
593
void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
594
{
595
	struct perf_evsel *leader = evsel->leader;
596
	struct perf_event_attr *attr = &evsel->attr;
597
	int track = evsel->tracking;
598
	bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread;
599

600
	attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1;
601 602
	attr->inherit	    = !opts->no_inherit;

603 604
	perf_evsel__set_sample_bit(evsel, IP);
	perf_evsel__set_sample_bit(evsel, TID);
605

606 607 608 609 610 611 612
	if (evsel->sample_read) {
		perf_evsel__set_sample_bit(evsel, READ);

		/*
		 * We need ID even in case of single event, because
		 * PERF_SAMPLE_READ process ID specific data.
		 */
613
		perf_evsel__set_sample_id(evsel, false);
614 615 616 617 618 619 620 621 622 623 624

		/*
		 * Apply group format only if we belong to group
		 * with more than one members.
		 */
		if (leader->nr_members > 1) {
			attr->read_format |= PERF_FORMAT_GROUP;
			attr->inherit = 0;
		}
	}

625
	/*
626
	 * We default some events to have a default interval. But keep
627 628
	 * it a weak assumption overridable by the user.
	 */
629
	if (!attr->sample_period || (opts->user_freq != UINT_MAX ||
630 631
				     opts->user_interval != ULLONG_MAX)) {
		if (opts->freq) {
632
			perf_evsel__set_sample_bit(evsel, PERIOD);
633 634 635 636 637 638 639
			attr->freq		= 1;
			attr->sample_freq	= opts->freq;
		} else {
			attr->sample_period = opts->default_interval;
		}
	}

640 641 642 643 644 645 646 647 648
	/*
	 * Disable sampling for all group members other
	 * than leader in case leader 'leads' the sampling.
	 */
	if ((leader != evsel) && leader->sample_read) {
		attr->sample_freq   = 0;
		attr->sample_period = 0;
	}

649 650 651 652 653 654 655
	if (opts->no_samples)
		attr->sample_freq = 0;

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

	if (opts->sample_address) {
656
		perf_evsel__set_sample_bit(evsel, ADDR);
657 658 659
		attr->mmap_data = track;
	}

660 661
	if (callchain_param.enabled && !evsel->no_aux_samples)
		perf_evsel__config_callgraph(evsel);
662

663
	if (target__has_cpu(&opts->target))
664
		perf_evsel__set_sample_bit(evsel, CPU);
665

666
	if (opts->period)
667
		perf_evsel__set_sample_bit(evsel, PERIOD);
668

669 670 671 672 673 674
	/*
	 * When the user explicitely disabled time don't force it here.
	 */
	if (opts->sample_time &&
	    (!perf_missing_features.sample_id_all &&
	    (!opts->no_inherit || target__has_cpu(&opts->target) || per_cpu)))
675
		perf_evsel__set_sample_bit(evsel, TIME);
676

677
	if (opts->raw_samples && !evsel->no_aux_samples) {
678 679 680
		perf_evsel__set_sample_bit(evsel, TIME);
		perf_evsel__set_sample_bit(evsel, RAW);
		perf_evsel__set_sample_bit(evsel, CPU);
681 682
	}

683
	if (opts->sample_address)
684
		perf_evsel__set_sample_bit(evsel, DATA_SRC);
685

686
	if (opts->no_buffering) {
687 688 689
		attr->watermark = 0;
		attr->wakeup_events = 1;
	}
690
	if (opts->branch_stack && !evsel->no_aux_samples) {
691
		perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
692 693
		attr->branch_sample_type = opts->branch_stack;
	}
694

695
	if (opts->sample_weight)
696
		perf_evsel__set_sample_bit(evsel, WEIGHT);
697

698
	attr->mmap  = track;
699
	attr->mmap2 = track && !perf_missing_features.mmap2;
700
	attr->comm  = track;
701

702
	if (opts->sample_transaction)
703
		perf_evsel__set_sample_bit(evsel, TRANSACTION);
704

705 706 707 708 709 710
	/*
	 * XXX see the function comment above
	 *
	 * Disabling only independent events or group leaders,
	 * keeping group members enabled.
	 */
711
	if (perf_evsel__is_group_leader(evsel))
712 713 714 715 716 717
		attr->disabled = 1;

	/*
	 * Setting enable_on_exec for independent events and
	 * group leaders for traced executed by perf.
	 */
718 719
	if (target__none(&opts->target) && perf_evsel__is_group_leader(evsel) &&
		!opts->initial_delay)
720
		attr->enable_on_exec = 1;
721 722 723 724 725

	if (evsel->immediate) {
		attr->disabled = 0;
		attr->enable_on_exec = 0;
	}
726 727
}

728 729
int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
730
	int cpu, thread;
731 732 733 734

	if (evsel->system_wide)
		nthreads = 1;

735
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
736 737 738 739 740 741 742 743 744

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

745 746 747
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

748 749
static int perf_evsel__run_ioctl(struct perf_evsel *evsel, int ncpus, int nthreads,
			  int ioc,  void *arg)
750 751 752
{
	int cpu, thread;

753 754 755
	if (evsel->system_wide)
		nthreads = 1;

756 757 758
	for (cpu = 0; cpu < ncpus; cpu++) {
		for (thread = 0; thread < nthreads; thread++) {
			int fd = FD(evsel, cpu, thread),
759
			    err = ioctl(fd, ioc, arg);
760 761 762 763 764 765 766 767 768

			if (err)
				return err;
		}
	}

	return 0;
}

769 770 771 772 773 774 775 776 777 778 779 780 781 782 783
int perf_evsel__set_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
			   const char *filter)
{
	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
				     PERF_EVENT_IOC_SET_FILTER,
				     (void *)filter);
}

int perf_evsel__enable(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
				     PERF_EVENT_IOC_ENABLE,
				     0);
}

784 785
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
786 787 788
	if (evsel->system_wide)
		nthreads = 1;

789 790 791 792 793 794 795 796 797 798 799 800
	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;
801 802
}

803 804 805 806 807 808
void perf_evsel__reset_counts(struct perf_evsel *evsel, int ncpus)
{
	memset(evsel->counts, 0, (sizeof(*evsel->counts) +
				 (ncpus * sizeof(struct perf_counts_values))));
}

809 810 811 812 813 814 815
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;
}

816 817 818 819 820 821
void perf_evsel__free_fd(struct perf_evsel *evsel)
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

822 823
void perf_evsel__free_id(struct perf_evsel *evsel)
{
824 825
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
826
	zfree(&evsel->id);
827 828
}

829 830 831 832
void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	int cpu, thread;

833 834 835
	if (evsel->system_wide)
		nthreads = 1;

836 837 838 839 840 841 842
	for (cpu = 0; cpu < ncpus; cpu++)
		for (thread = 0; thread < nthreads; ++thread) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
}

843 844
void perf_evsel__free_counts(struct perf_evsel *evsel)
{
845
	zfree(&evsel->counts);
846 847
}

848
void perf_evsel__exit(struct perf_evsel *evsel)
849 850
{
	assert(list_empty(&evsel->node));
851 852
	perf_evsel__free_fd(evsel);
	perf_evsel__free_id(evsel);
A
Arnaldo Carvalho de Melo 已提交
853
	perf_evsel__object.fini(evsel);
854 855 856 857 858
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
S
Stephane Eranian 已提交
859
	close_cgroup(evsel->cgrp);
860
	zfree(&evsel->group_name);
861
	if (evsel->tp_format)
862
		pevent_free_format(evsel->tp_format);
863
	zfree(&evsel->name);
864 865
	free(evsel);
}
866

867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888
static inline void compute_deltas(struct perf_evsel *evsel,
				  int cpu,
				  struct perf_counts_values *count)
{
	struct perf_counts_values tmp;

	if (!evsel->prev_raw_counts)
		return;

	if (cpu == -1) {
		tmp = evsel->prev_raw_counts->aggr;
		evsel->prev_raw_counts->aggr = *count;
	} else {
		tmp = evsel->prev_raw_counts->cpu[cpu];
		evsel->prev_raw_counts->cpu[cpu] = *count;
	}

	count->val = count->val - tmp.val;
	count->ena = count->ena - tmp.ena;
	count->run = count->run - tmp.run;
}

889 890 891 892 893 894 895 896 897
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;

898 899 900
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

901 902 903
	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
		return -errno;

904 905
	compute_deltas(evsel, cpu, &count);

906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924
	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;

925 926 927
	if (evsel->system_wide)
		nthreads = 1;

928
	aggr->val = aggr->ena = aggr->run = 0;
929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946

	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;
			}
		}
	}

947 948
	compute_deltas(evsel, -1, aggr);

949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965
	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;
}
966

967 968 969 970 971
static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
{
	struct perf_evsel *leader = evsel->leader;
	int fd;

972
	if (perf_evsel__is_group_leader(evsel))
973 974 975 976 977 978 979 980 981 982 983 984 985 986
		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;
}

A
Adrian Hunter 已提交
987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021
#define __PRINT_ATTR(fmt, cast, field)  \
	fprintf(fp, "  %-19s "fmt"\n", #field, cast attr->field)

#define PRINT_ATTR_U32(field)  __PRINT_ATTR("%u" , , field)
#define PRINT_ATTR_X32(field)  __PRINT_ATTR("%#x", , field)
#define PRINT_ATTR_U64(field)  __PRINT_ATTR("%" PRIu64, (uint64_t), field)
#define PRINT_ATTR_X64(field)  __PRINT_ATTR("%#"PRIx64, (uint64_t), field)

#define PRINT_ATTR2N(name1, field1, name2, field2)	\
	fprintf(fp, "  %-19s %u    %-19s %u\n",		\
	name1, attr->field1, name2, attr->field2)

#define PRINT_ATTR2(field1, field2) \
	PRINT_ATTR2N(#field1, field1, #field2, field2)

static size_t perf_event_attr__fprintf(struct perf_event_attr *attr, FILE *fp)
{
	size_t ret = 0;

	ret += fprintf(fp, "%.60s\n", graph_dotted_line);
	ret += fprintf(fp, "perf_event_attr:\n");

	ret += PRINT_ATTR_U32(type);
	ret += PRINT_ATTR_U32(size);
	ret += PRINT_ATTR_X64(config);
	ret += PRINT_ATTR_U64(sample_period);
	ret += PRINT_ATTR_U64(sample_freq);
	ret += PRINT_ATTR_X64(sample_type);
	ret += PRINT_ATTR_X64(read_format);

	ret += PRINT_ATTR2(disabled, inherit);
	ret += PRINT_ATTR2(pinned, exclusive);
	ret += PRINT_ATTR2(exclude_user, exclude_kernel);
	ret += PRINT_ATTR2(exclude_hv, exclude_idle);
	ret += PRINT_ATTR2(mmap, comm);
1022
	ret += PRINT_ATTR2(mmap2, comm_exec);
A
Adrian Hunter 已提交
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046
	ret += PRINT_ATTR2(freq, inherit_stat);
	ret += PRINT_ATTR2(enable_on_exec, task);
	ret += PRINT_ATTR2(watermark, precise_ip);
	ret += PRINT_ATTR2(mmap_data, sample_id_all);
	ret += PRINT_ATTR2(exclude_host, exclude_guest);
	ret += PRINT_ATTR2N("excl.callchain_kern", exclude_callchain_kernel,
			    "excl.callchain_user", exclude_callchain_user);

	ret += PRINT_ATTR_U32(wakeup_events);
	ret += PRINT_ATTR_U32(wakeup_watermark);
	ret += PRINT_ATTR_X32(bp_type);
	ret += PRINT_ATTR_X64(bp_addr);
	ret += PRINT_ATTR_X64(config1);
	ret += PRINT_ATTR_U64(bp_len);
	ret += PRINT_ATTR_X64(config2);
	ret += PRINT_ATTR_X64(branch_sample_type);
	ret += PRINT_ATTR_X64(sample_regs_user);
	ret += PRINT_ATTR_U32(sample_stack_user);

	ret += fprintf(fp, "%.60s\n", graph_dotted_line);

	return ret;
}

1047
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1048
			      struct thread_map *threads)
1049
{
1050
	int cpu, thread, nthreads;
1051
	unsigned long flags = PERF_FLAG_FD_CLOEXEC;
1052
	int pid = -1, err;
1053
	enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit = NO_CHANGE;
1054

1055 1056 1057 1058 1059
	if (evsel->system_wide)
		nthreads = 1;
	else
		nthreads = threads->nr;

1060
	if (evsel->fd == NULL &&
1061
	    perf_evsel__alloc_fd(evsel, cpus->nr, nthreads) < 0)
1062
		return -ENOMEM;
1063

S
Stephane Eranian 已提交
1064
	if (evsel->cgrp) {
1065
		flags |= PERF_FLAG_PID_CGROUP;
S
Stephane Eranian 已提交
1066 1067 1068
		pid = evsel->cgrp->fd;
	}

1069
fallback_missing_features:
1070 1071
	if (perf_missing_features.cloexec)
		flags &= ~(unsigned long)PERF_FLAG_FD_CLOEXEC;
1072 1073
	if (perf_missing_features.mmap2)
		evsel->attr.mmap2 = 0;
1074 1075 1076 1077 1078 1079
	if (perf_missing_features.exclude_guest)
		evsel->attr.exclude_guest = evsel->attr.exclude_host = 0;
retry_sample_id:
	if (perf_missing_features.sample_id_all)
		evsel->attr.sample_id_all = 0;

A
Adrian Hunter 已提交
1080 1081 1082
	if (verbose >= 2)
		perf_event_attr__fprintf(&evsel->attr, stderr);

1083
	for (cpu = 0; cpu < cpus->nr; cpu++) {
1084

1085
		for (thread = 0; thread < nthreads; thread++) {
1086
			int group_fd;
S
Stephane Eranian 已提交
1087

1088
			if (!evsel->cgrp && !evsel->system_wide)
S
Stephane Eranian 已提交
1089 1090
				pid = threads->map[thread];

1091
			group_fd = get_group_fd(evsel, cpu, thread);
1092
retry_open:
1093
			pr_debug2("sys_perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx\n",
A
Adrian Hunter 已提交
1094 1095
				  pid, cpus->map[cpu], group_fd, flags);

1096
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
1097
								     pid,
1098
								     cpus->map[cpu],
S
Stephane Eranian 已提交
1099
								     group_fd, flags);
1100 1101
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
1102
				pr_debug2("sys_perf_event_open failed, error %d\n",
1103
					  err);
1104
				goto try_fallback;
1105
			}
1106
			set_rlimit = NO_CHANGE;
1107
		}
1108 1109 1110 1111
	}

	return 0;

1112
try_fallback:
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136
	/*
	 * perf stat needs between 5 and 22 fds per CPU. When we run out
	 * of them try to increase the limits.
	 */
	if (err == -EMFILE && set_rlimit < INCREASED_MAX) {
		struct rlimit l;
		int old_errno = errno;

		if (getrlimit(RLIMIT_NOFILE, &l) == 0) {
			if (set_rlimit == NO_CHANGE)
				l.rlim_cur = l.rlim_max;
			else {
				l.rlim_cur = l.rlim_max + 1000;
				l.rlim_max = l.rlim_cur;
			}
			if (setrlimit(RLIMIT_NOFILE, &l) == 0) {
				set_rlimit++;
				errno = old_errno;
				goto retry_open;
			}
		}
		errno = old_errno;
	}

1137 1138 1139
	if (err != -EINVAL || cpu > 0 || thread > 0)
		goto out_close;

1140 1141 1142 1143
	if (!perf_missing_features.cloexec && (flags & PERF_FLAG_FD_CLOEXEC)) {
		perf_missing_features.cloexec = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.mmap2 && evsel->attr.mmap2) {
1144 1145 1146 1147
		perf_missing_features.mmap2 = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.exclude_guest &&
		   (evsel->attr.exclude_guest || evsel->attr.exclude_host)) {
1148 1149 1150 1151 1152 1153 1154
		perf_missing_features.exclude_guest = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.sample_id_all) {
		perf_missing_features.sample_id_all = true;
		goto retry_sample_id;
	}

1155
out_close:
1156 1157 1158 1159 1160
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
1161
		thread = nthreads;
1162
	} while (--cpu >= 0);
1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
	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);
1173 1174
}

1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190
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, },
};

1191
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1192
		     struct thread_map *threads)
1193
{
1194 1195 1196
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
1197 1198
	}

1199 1200
	if (threads == NULL)
		threads = &empty_thread_map.map;
1201

1202
	return __perf_evsel__open(evsel, cpus, threads);
1203 1204
}

1205
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
1206
			     struct cpu_map *cpus)
1207
{
1208
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
1209
}
1210

1211
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
1212
				struct thread_map *threads)
1213
{
1214
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
1215
}
1216

1217 1218 1219
static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
				       const union perf_event *event,
				       struct perf_sample *sample)
1220
{
1221
	u64 type = evsel->attr.sample_type;
1222
	const u64 *array = event->sample.array;
1223
	bool swapped = evsel->needs_swap;
1224
	union u64_swap u;
1225 1226 1227 1228

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

1229 1230 1231 1232 1233
	if (type & PERF_SAMPLE_IDENTIFIER) {
		sample->id = *array;
		array--;
	}

1234
	if (type & PERF_SAMPLE_CPU) {
1235 1236 1237 1238 1239 1240 1241 1242
		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];
1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261
		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) {
1262 1263 1264 1265 1266 1267 1268 1269 1270 1271
		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];
1272
		array--;
1273 1274 1275 1276 1277
	}

	return 0;
}

1278 1279
static inline bool overflow(const void *endp, u16 max_size, const void *offset,
			    u64 size)
1280
{
1281 1282
	return size > max_size || offset + size > endp;
}
1283

1284 1285 1286 1287 1288
#define OVERFLOW_CHECK(offset, size, max_size)				\
	do {								\
		if (overflow(endp, (max_size), (offset), (size)))	\
			return -EFAULT;					\
	} while (0)
1289

1290 1291
#define OVERFLOW_CHECK_u64(offset) \
	OVERFLOW_CHECK(offset, sizeof(u64), sizeof(u64))
1292

1293
int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
1294
			     struct perf_sample *data)
1295
{
1296
	u64 type = evsel->attr.sample_type;
1297
	bool swapped = evsel->needs_swap;
1298
	const u64 *array;
1299 1300 1301
	u16 max_size = event->header.size;
	const void *endp = (void *)event + max_size;
	u64 sz;
1302

1303 1304 1305 1306
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
1307
	union u64_swap u;
1308

1309
	memset(data, 0, sizeof(*data));
1310 1311
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
1312
	data->period = evsel->attr.sample_period;
1313
	data->weight = 0;
1314 1315

	if (event->header.type != PERF_RECORD_SAMPLE) {
1316
		if (!evsel->attr.sample_id_all)
1317
			return 0;
1318
		return perf_evsel__parse_id_sample(evsel, event, data);
1319 1320 1321 1322
	}

	array = event->sample.array;

1323 1324 1325 1326 1327
	/*
	 * The evsel's sample_size is based on PERF_SAMPLE_MASK which includes
	 * up to PERF_SAMPLE_PERIOD.  After that overflow() must be used to
	 * check the format does not go past the end of the event.
	 */
1328
	if (evsel->sample_size + sizeof(event->header) > event->header.size)
1329 1330
		return -EFAULT;

1331 1332 1333 1334 1335 1336
	data->id = -1ULL;
	if (type & PERF_SAMPLE_IDENTIFIER) {
		data->id = *array;
		array++;
	}

1337
	if (type & PERF_SAMPLE_IP) {
1338
		data->ip = *array;
1339 1340 1341 1342
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
1343 1344 1345 1346 1347 1348 1349 1350 1351 1352
		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];
1353 1354 1355 1356 1357 1358 1359 1360
		array++;
	}

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

1361
	data->addr = 0;
1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377
	if (type & PERF_SAMPLE_ADDR) {
		data->addr = *array;
		array++;
	}

	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) {
1378 1379 1380 1381 1382 1383 1384 1385 1386

		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];
1387 1388 1389 1390 1391 1392 1393 1394 1395
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
1396 1397
		u64 read_format = evsel->attr.read_format;

1398
		OVERFLOW_CHECK_u64(array);
1399 1400 1401 1402 1403 1404 1405 1406
		if (read_format & PERF_FORMAT_GROUP)
			data->read.group.nr = *array;
		else
			data->read.one.value = *array;

		array++;

		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
1407
			OVERFLOW_CHECK_u64(array);
1408 1409 1410 1411 1412
			data->read.time_enabled = *array;
			array++;
		}

		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
1413
			OVERFLOW_CHECK_u64(array);
1414 1415 1416 1417 1418 1419
			data->read.time_running = *array;
			array++;
		}

		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
		if (read_format & PERF_FORMAT_GROUP) {
1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430
			const u64 max_group_nr = UINT64_MAX /
					sizeof(struct sample_read_value);

			if (data->read.group.nr > max_group_nr)
				return -EFAULT;
			sz = data->read.group.nr *
			     sizeof(struct sample_read_value);
			OVERFLOW_CHECK(array, sz, max_size);
			data->read.group.values =
					(struct sample_read_value *)array;
			array = (void *)array + sz;
1431
		} else {
1432
			OVERFLOW_CHECK_u64(array);
1433 1434 1435
			data->read.one.id = *array;
			array++;
		}
1436 1437 1438
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
1439
		const u64 max_callchain_nr = UINT64_MAX / sizeof(u64);
1440

1441 1442 1443
		OVERFLOW_CHECK_u64(array);
		data->callchain = (struct ip_callchain *)array++;
		if (data->callchain->nr > max_callchain_nr)
1444
			return -EFAULT;
1445 1446 1447
		sz = data->callchain->nr * sizeof(u64);
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1448 1449 1450
	}

	if (type & PERF_SAMPLE_RAW) {
1451
		OVERFLOW_CHECK_u64(array);
1452 1453 1454 1455 1456 1457 1458 1459 1460
		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]);
		}
		data->raw_size = u.val32[0];
1461
		array = (void *)array + sizeof(u32);
1462

1463 1464 1465
		OVERFLOW_CHECK(array, data->raw_size, max_size);
		data->raw_data = (void *)array;
		array = (void *)array + data->raw_size;
1466 1467
	}

1468
	if (type & PERF_SAMPLE_BRANCH_STACK) {
1469 1470
		const u64 max_branch_nr = UINT64_MAX /
					  sizeof(struct branch_entry);
1471

1472 1473
		OVERFLOW_CHECK_u64(array);
		data->branch_stack = (struct branch_stack *)array++;
1474

1475 1476
		if (data->branch_stack->nr > max_branch_nr)
			return -EFAULT;
1477
		sz = data->branch_stack->nr * sizeof(struct branch_entry);
1478 1479
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1480
	}
1481 1482

	if (type & PERF_SAMPLE_REGS_USER) {
1483
		OVERFLOW_CHECK_u64(array);
1484 1485
		data->user_regs.abi = *array;
		array++;
1486

1487
		if (data->user_regs.abi) {
1488
			u64 mask = evsel->attr.sample_regs_user;
1489

1490
			sz = hweight_long(mask) * sizeof(u64);
1491
			OVERFLOW_CHECK(array, sz, max_size);
1492
			data->user_regs.mask = mask;
1493
			data->user_regs.regs = (u64 *)array;
1494
			array = (void *)array + sz;
1495 1496 1497 1498
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
1499 1500
		OVERFLOW_CHECK_u64(array);
		sz = *array++;
1501 1502 1503 1504

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

1505
		if (!sz) {
1506 1507
			data->user_stack.size = 0;
		} else {
1508
			OVERFLOW_CHECK(array, sz, max_size);
1509
			data->user_stack.data = (char *)array;
1510 1511
			array = (void *)array + sz;
			OVERFLOW_CHECK_u64(array);
1512
			data->user_stack.size = *array++;
1513 1514 1515
			if (WARN_ONCE(data->user_stack.size > sz,
				      "user stack dump failure\n"))
				return -EFAULT;
1516 1517 1518
		}
	}

1519 1520
	data->weight = 0;
	if (type & PERF_SAMPLE_WEIGHT) {
1521
		OVERFLOW_CHECK_u64(array);
1522 1523 1524 1525
		data->weight = *array;
		array++;
	}

1526 1527
	data->data_src = PERF_MEM_DATA_SRC_NONE;
	if (type & PERF_SAMPLE_DATA_SRC) {
1528
		OVERFLOW_CHECK_u64(array);
1529 1530 1531 1532
		data->data_src = *array;
		array++;
	}

1533 1534
	data->transaction = 0;
	if (type & PERF_SAMPLE_TRANSACTION) {
1535
		OVERFLOW_CHECK_u64(array);
1536 1537 1538 1539
		data->transaction = *array;
		array++;
	}

1540 1541
	return 0;
}
1542

1543
size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type,
1544
				     u64 read_format)
1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609
{
	size_t sz, result = sizeof(struct sample_event);

	if (type & PERF_SAMPLE_IDENTIFIER)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_IP)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_TID)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_TIME)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_ADDR)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_ID)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_STREAM_ID)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_CPU)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_PERIOD)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_READ) {
		result += sizeof(u64);
		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
			result += sizeof(u64);
		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
			result += sizeof(u64);
		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
		if (read_format & PERF_FORMAT_GROUP) {
			sz = sample->read.group.nr *
			     sizeof(struct sample_read_value);
			result += sz;
		} else {
			result += sizeof(u64);
		}
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
		sz = (sample->callchain->nr + 1) * sizeof(u64);
		result += sz;
	}

	if (type & PERF_SAMPLE_RAW) {
		result += sizeof(u32);
		result += sample->raw_size;
	}

	if (type & PERF_SAMPLE_BRANCH_STACK) {
		sz = sample->branch_stack->nr * sizeof(struct branch_entry);
		sz += sizeof(u64);
		result += sz;
	}

	if (type & PERF_SAMPLE_REGS_USER) {
		if (sample->user_regs.abi) {
			result += sizeof(u64);
1610
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631
			result += sz;
		} else {
			result += sizeof(u64);
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
		sz = sample->user_stack.size;
		result += sizeof(u64);
		if (sz) {
			result += sz;
			result += sizeof(u64);
		}
	}

	if (type & PERF_SAMPLE_WEIGHT)
		result += sizeof(u64);

	if (type & PERF_SAMPLE_DATA_SRC)
		result += sizeof(u64);

1632 1633 1634
	if (type & PERF_SAMPLE_TRANSACTION)
		result += sizeof(u64);

1635 1636 1637
	return result;
}

1638
int perf_event__synthesize_sample(union perf_event *event, u64 type,
1639
				  u64 read_format,
1640 1641 1642 1643
				  const struct perf_sample *sample,
				  bool swapped)
{
	u64 *array;
1644
	size_t sz;
1645 1646 1647 1648
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
1649
	union u64_swap u;
1650 1651 1652

	array = event->sample.array;

1653 1654 1655 1656 1657
	if (type & PERF_SAMPLE_IDENTIFIER) {
		*array = sample->id;
		array++;
	}

1658
	if (type & PERF_SAMPLE_IP) {
1659
		*array = sample->ip;
1660 1661 1662 1663 1664 1665 1666 1667
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
		u.val32[0] = sample->pid;
		u.val32[1] = sample->tid;
		if (swapped) {
			/*
1668
			 * Inverse of what is done in perf_evsel__parse_sample
1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702
			 */
			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) {
			/*
1703
			 * Inverse of what is done in perf_evsel__parse_sample
1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716
			 */
			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++;
	}

1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779
	if (type & PERF_SAMPLE_READ) {
		if (read_format & PERF_FORMAT_GROUP)
			*array = sample->read.group.nr;
		else
			*array = sample->read.one.value;
		array++;

		if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
			*array = sample->read.time_enabled;
			array++;
		}

		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
			*array = sample->read.time_running;
			array++;
		}

		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
		if (read_format & PERF_FORMAT_GROUP) {
			sz = sample->read.group.nr *
			     sizeof(struct sample_read_value);
			memcpy(array, sample->read.group.values, sz);
			array = (void *)array + sz;
		} else {
			*array = sample->read.one.id;
			array++;
		}
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
		sz = (sample->callchain->nr + 1) * sizeof(u64);
		memcpy(array, sample->callchain, sz);
		array = (void *)array + sz;
	}

	if (type & PERF_SAMPLE_RAW) {
		u.val32[0] = sample->raw_size;
		if (WARN_ONCE(swapped,
			      "Endianness of raw data not corrected!\n")) {
			/*
			 * Inverse of what is done in perf_evsel__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 = (void *)array + sizeof(u32);

		memcpy(array, sample->raw_data, sample->raw_size);
		array = (void *)array + sample->raw_size;
	}

	if (type & PERF_SAMPLE_BRANCH_STACK) {
		sz = sample->branch_stack->nr * sizeof(struct branch_entry);
		sz += sizeof(u64);
		memcpy(array, sample->branch_stack, sz);
		array = (void *)array + sz;
	}

	if (type & PERF_SAMPLE_REGS_USER) {
		if (sample->user_regs.abi) {
			*array++ = sample->user_regs.abi;
1780
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807
			memcpy(array, sample->user_regs.regs, sz);
			array = (void *)array + sz;
		} else {
			*array++ = 0;
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
		sz = sample->user_stack.size;
		*array++ = sz;
		if (sz) {
			memcpy(array, sample->user_stack.data, sz);
			array = (void *)array + sz;
			*array++ = sz;
		}
	}

	if (type & PERF_SAMPLE_WEIGHT) {
		*array = sample->weight;
		array++;
	}

	if (type & PERF_SAMPLE_DATA_SRC) {
		*array = sample->data_src;
		array++;
	}

1808 1809 1810 1811 1812
	if (type & PERF_SAMPLE_TRANSACTION) {
		*array = sample->transaction;
		array++;
	}

1813 1814
	return 0;
}
1815

1816 1817 1818 1819 1820
struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
{
	return pevent_find_field(evsel->tp_format, name);
}

1821
void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
1822 1823
			 const char *name)
{
1824
	struct format_field *field = perf_evsel__field(evsel, name);
1825 1826
	int offset;

1827 1828
	if (!field)
		return NULL;
1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842

	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)
{
1843
	struct format_field *field = perf_evsel__field(evsel, name);
1844 1845
	void *ptr;
	u64 value;
1846

1847 1848
	if (!field)
		return 0;
1849

1850
	ptr = sample->raw_data + field->offset;
1851

1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882
	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;
1883
}
1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912

static int comma_fprintf(FILE *fp, bool *first, const char *fmt, ...)
{
	va_list args;
	int ret = 0;

	if (!*first) {
		ret += fprintf(fp, ",");
	} else {
		ret += fprintf(fp, ":");
		*first = false;
	}

	va_start(args, fmt);
	ret += vfprintf(fp, fmt, args);
	va_end(args);
	return ret;
}

static int __if_fprintf(FILE *fp, bool *first, const char *field, u64 value)
{
	if (value == 0)
		return 0;

	return comma_fprintf(fp, first, " %s: %" PRIu64, field, value);
}

#define if_print(field) printed += __if_fprintf(fp, &first, #field, evsel->attr.field)

1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941
struct bit_names {
	int bit;
	const char *name;
};

static int bits__fprintf(FILE *fp, const char *field, u64 value,
			 struct bit_names *bits, bool *first)
{
	int i = 0, printed = comma_fprintf(fp, first, " %s: ", field);
	bool first_bit = true;

	do {
		if (value & bits[i].bit) {
			printed += fprintf(fp, "%s%s", first_bit ? "" : "|", bits[i].name);
			first_bit = false;
		}
	} while (bits[++i].name != NULL);

	return printed;
}

static int sample_type__fprintf(FILE *fp, bool *first, u64 value)
{
#define bit_name(n) { PERF_SAMPLE_##n, #n }
	struct bit_names bits[] = {
		bit_name(IP), bit_name(TID), bit_name(TIME), bit_name(ADDR),
		bit_name(READ), bit_name(CALLCHAIN), bit_name(ID), bit_name(CPU),
		bit_name(PERIOD), bit_name(STREAM_ID), bit_name(RAW),
		bit_name(BRANCH_STACK), bit_name(REGS_USER), bit_name(STACK_USER),
1942
		bit_name(IDENTIFIER),
1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
		{ .name = NULL, }
	};
#undef bit_name
	return bits__fprintf(fp, "sample_type", value, bits, first);
}

static int read_format__fprintf(FILE *fp, bool *first, u64 value)
{
#define bit_name(n) { PERF_FORMAT_##n, #n }
	struct bit_names bits[] = {
		bit_name(TOTAL_TIME_ENABLED), bit_name(TOTAL_TIME_RUNNING),
		bit_name(ID), bit_name(GROUP),
		{ .name = NULL, }
	};
#undef bit_name
	return bits__fprintf(fp, "read_format", value, bits, first);
}

1961 1962 1963 1964
int perf_evsel__fprintf(struct perf_evsel *evsel,
			struct perf_attr_details *details, FILE *fp)
{
	bool first = true;
N
Namhyung Kim 已提交
1965 1966
	int printed = 0;

1967
	if (details->event_group) {
N
Namhyung Kim 已提交
1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985
		struct perf_evsel *pos;

		if (!perf_evsel__is_group_leader(evsel))
			return 0;

		if (evsel->nr_members > 1)
			printed += fprintf(fp, "%s{", evsel->group_name ?: "");

		printed += fprintf(fp, "%s", perf_evsel__name(evsel));
		for_each_group_member(pos, evsel)
			printed += fprintf(fp, ",%s", perf_evsel__name(pos));

		if (evsel->nr_members > 1)
			printed += fprintf(fp, "}");
		goto out;
	}

	printed += fprintf(fp, "%s", perf_evsel__name(evsel));
1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

	if (details->verbose || details->freq) {
		printed += comma_fprintf(fp, &first, " sample_freq=%" PRIu64,
					 (u64)evsel->attr.sample_freq);
	}

	if (details->verbose) {
		if_print(type);
		if_print(config);
		if_print(config1);
		if_print(config2);
		if_print(size);
1998 1999 2000
		printed += sample_type__fprintf(fp, &first, evsel->attr.sample_type);
		if (evsel->attr.read_format)
			printed += read_format__fprintf(fp, &first, evsel->attr.read_format);
2001 2002 2003 2004 2005 2006 2007 2008 2009
		if_print(disabled);
		if_print(inherit);
		if_print(pinned);
		if_print(exclusive);
		if_print(exclude_user);
		if_print(exclude_kernel);
		if_print(exclude_hv);
		if_print(exclude_idle);
		if_print(mmap);
2010
		if_print(mmap2);
2011
		if_print(comm);
2012
		if_print(comm_exec);
2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027
		if_print(freq);
		if_print(inherit_stat);
		if_print(enable_on_exec);
		if_print(task);
		if_print(watermark);
		if_print(precise_ip);
		if_print(mmap_data);
		if_print(sample_id_all);
		if_print(exclude_host);
		if_print(exclude_guest);
		if_print(__reserved_1);
		if_print(wakeup_events);
		if_print(bp_type);
		if_print(branch_sample_type);
	}
N
Namhyung Kim 已提交
2028
out:
2029 2030 2031
	fputc('\n', fp);
	return ++printed;
}
2032 2033 2034 2035

bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
			  char *msg, size_t msgsize)
{
2036
	if ((err == ENOENT || err == ENXIO || err == ENODEV) &&
2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052
	    evsel->attr.type   == PERF_TYPE_HARDWARE &&
	    evsel->attr.config == PERF_COUNT_HW_CPU_CYCLES) {
		/*
		 * If it's cycles then fall back to hrtimer based
		 * cpu-clock-tick sw counter, which is always available even if
		 * no PMU support.
		 *
		 * PPC returns ENXIO until 2.6.37 (behavior changed with commit
		 * b0a873e).
		 */
		scnprintf(msg, msgsize, "%s",
"The cycles event is not supported, trying to fall back to cpu-clock-ticks");

		evsel->attr.type   = PERF_TYPE_SOFTWARE;
		evsel->attr.config = PERF_COUNT_SW_CPU_CLOCK;

2053
		zfree(&evsel->name);
2054 2055 2056 2057 2058
		return true;
	}

	return false;
}
2059

2060
int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
2061 2062
			      int err, char *msg, size_t size)
{
2063 2064
	char sbuf[STRERR_BUFSIZE];

2065 2066 2067
	switch (err) {
	case EPERM:
	case EACCES:
2068
		return scnprintf(msg, size,
2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098
		 "You may not have permission to collect %sstats.\n"
		 "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n"
		 " -1 - Not paranoid at all\n"
		 "  0 - Disallow raw tracepoint access for unpriv\n"
		 "  1 - Disallow cpu events for unpriv\n"
		 "  2 - Disallow kernel profiling for unpriv",
				 target->system_wide ? "system-wide " : "");
	case ENOENT:
		return scnprintf(msg, size, "The %s event is not supported.",
				 perf_evsel__name(evsel));
	case EMFILE:
		return scnprintf(msg, size, "%s",
			 "Too many events are opened.\n"
			 "Try again after reducing the number of events.");
	case ENODEV:
		if (target->cpu_list)
			return scnprintf(msg, size, "%s",
	 "No such device - did you specify an out-of-range profile CPU?\n");
		break;
	case EOPNOTSUPP:
		if (evsel->attr.precise_ip)
			return scnprintf(msg, size, "%s",
	"\'precise\' request may not be supported. Try removing 'p' modifier.");
#if defined(__i386__) || defined(__x86_64__)
		if (evsel->attr.type == PERF_TYPE_HARDWARE)
			return scnprintf(msg, size, "%s",
	"No hardware sampling interrupt available.\n"
	"No APIC? If so then you can boot the kernel with the \"lapic\" boot parameter to force-enable it.");
#endif
		break;
2099 2100 2101 2102 2103 2104
	case EBUSY:
		if (find_process("oprofiled"))
			return scnprintf(msg, size,
	"The PMU counters are busy/taken by another profiler.\n"
	"We found oprofile daemon running, please stop it and try again.");
		break;
2105 2106 2107 2108 2109
	default:
		break;
	}

	return scnprintf(msg, size,
2110
	"The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n"
2111 2112
	"/bin/dmesg may provide additional information.\n"
	"No CONFIG_PERF_EVENTS=y kernel support configured?\n",
2113 2114
			 err, strerror_r(err, sbuf, sizeof(sbuf)),
			 perf_evsel__name(evsel));
2115
}