evsel.c 50.7 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 "cgroup.h"
20
#include "evsel.h"
21
#include "evlist.h"
22
#include "util.h"
23
#include "cpumap.h"
24
#include "thread_map.h"
25
#include "target.h"
26
#include "perf_regs.h"
A
Adrian Hunter 已提交
27
#include "debug.h"
28
#include "trace-event.h"
29

30 31 32
static struct {
	bool sample_id_all;
	bool exclude_guest;
33
	bool mmap2;
34
	bool cloexec;
35 36
} perf_missing_features;

A
Arnaldo Carvalho de Melo 已提交
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 78
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;
}

79 80
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))

81
int __perf_evsel__sample_size(u64 sample_type)
82 83 84 85 86 87 88 89 90 91 92 93 94 95 96
{
	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;
}

97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162
/**
 * __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);
}

163 164 165 166 167 168
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);
169
		perf_evsel__calc_id_pos(evsel);
170 171 172 173 174 175 176 177 178
	}
}

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);
179
		perf_evsel__calc_id_pos(evsel);
180 181 182
	}
}

183 184
void perf_evsel__set_sample_id(struct perf_evsel *evsel,
			       bool can_sample_identifier)
185
{
186 187 188 189 190 191
	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);
	}
192 193 194
	evsel->attr.read_format |= PERF_FORMAT_ID;
}

195 196 197 198
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
199
	evsel->tracking	   = !idx;
200
	evsel->attr	   = *attr;
201
	evsel->leader	   = evsel;
202 203
	evsel->unit	   = "";
	evsel->scale	   = 1.0;
204
	INIT_LIST_HEAD(&evsel->node);
A
Arnaldo Carvalho de Melo 已提交
205
	perf_evsel__object.init(evsel);
206
	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
207
	perf_evsel__calc_id_pos(evsel);
208 209
}

210
struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx)
211
{
A
Arnaldo Carvalho de Melo 已提交
212
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
213

214 215
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
216 217 218 219

	return evsel;
}

220
struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx)
221
{
A
Arnaldo Carvalho de Melo 已提交
222
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
223 224 225

	if (evsel != NULL) {
		struct perf_event_attr attr = {
226 227 228
			.type	       = PERF_TYPE_TRACEPOINT,
			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
229 230
		};

231 232 233
		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
			goto out_free;

234
		evsel->tp_format = trace_event__tp_format(sys, name);
235 236 237
		if (evsel->tp_format == NULL)
			goto out_free;

238
		event_attr_init(&attr);
239
		attr.config = evsel->tp_format->id;
240
		attr.sample_period = 1;
241 242 243 244 245 246
		perf_evsel__init(evsel, &attr, idx);
	}

	return evsel;

out_free:
247
	zfree(&evsel->name);
248 249 250 251
	free(evsel);
	return NULL;
}

252
const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
253 254 255 256 257 258 259 260 261 262 263 264
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

265
static const char *__perf_evsel__hw_name(u64 config)
266 267 268 269 270 271 272
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

273
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
274
{
275
	int colon = 0, r = 0;
276 277 278 279 280
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
281
			if (!colon) colon = ++r;			\
282 283 284 285 286 287 288 289 290 291 292 293
			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)
294
			colon = ++r;
295 296 297 298 299 300 301 302 303 304
		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)
305
		bf[colon - 1] = ':';
306 307 308
	return r;
}

309 310 311 312 313 314
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);
}

315
const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
316 317 318 319
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
320
	"cpu-migrations",
321 322 323 324
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
325
	"dummy",
326 327
};

328
static const char *__perf_evsel__sw_name(u64 config)
329 330 331 332 333 334 335 336 337 338 339 340
{
	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);
}

341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365
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);
}

366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431
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]);
}

432
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464
{
	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);
}

465 466 467 468 469 470
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);
}

471
const char *perf_evsel__name(struct perf_evsel *evsel)
472
{
473
	char bf[128];
474

475 476
	if (evsel->name)
		return evsel->name;
477 478 479

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
480
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
481 482 483
		break;

	case PERF_TYPE_HARDWARE:
484
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
485
		break;
486 487

	case PERF_TYPE_HW_CACHE:
488
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
489 490
		break;

491
	case PERF_TYPE_SOFTWARE:
492
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
493 494
		break;

495
	case PERF_TYPE_TRACEPOINT:
496
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
497 498
		break;

499 500 501 502
	case PERF_TYPE_BREAKPOINT:
		perf_evsel__bp_name(evsel, bf, sizeof(bf));
		break;

503
	default:
504 505
		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
			  evsel->attr.type);
506
		break;
507 508
	}

509 510 511
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
512 513
}

514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538
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;
}

539
static void
540
perf_evsel__config_callgraph(struct perf_evsel *evsel)
541 542 543 544 545 546
{
	bool function = perf_evsel__is_function_event(evsel);
	struct perf_event_attr *attr = &evsel->attr;

	perf_evsel__set_sample_bit(evsel, CALLCHAIN);

547
	if (callchain_param.record_mode == CALLCHAIN_DWARF) {
548 549 550 551
		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;
552
			attr->sample_stack_user = callchain_param.dump_size;
553 554 555 556 557 558 559 560 561 562 563 564 565
			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;
	}
}

566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593
/*
 * 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.
 */
594
void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)
595
{
596
	struct perf_evsel *leader = evsel->leader;
597
	struct perf_event_attr *attr = &evsel->attr;
598
	int track = evsel->tracking;
599
	bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread;
600

601
	attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1;
602 603
	attr->inherit	    = !opts->no_inherit;

604 605
	perf_evsel__set_sample_bit(evsel, IP);
	perf_evsel__set_sample_bit(evsel, TID);
606

607 608 609 610 611 612 613
	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.
		 */
614
		perf_evsel__set_sample_id(evsel, false);
615 616 617 618 619 620 621 622 623 624 625

		/*
		 * 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;
		}
	}

626
	/*
627
	 * We default some events to have a default interval. But keep
628 629
	 * it a weak assumption overridable by the user.
	 */
630
	if (!attr->sample_period || (opts->user_freq != UINT_MAX ||
631 632
				     opts->user_interval != ULLONG_MAX)) {
		if (opts->freq) {
633
			perf_evsel__set_sample_bit(evsel, PERIOD);
634 635 636 637 638 639 640
			attr->freq		= 1;
			attr->sample_freq	= opts->freq;
		} else {
			attr->sample_period = opts->default_interval;
		}
	}

641 642 643 644 645 646 647 648 649
	/*
	 * 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;
	}

650 651 652 653 654 655 656
	if (opts->no_samples)
		attr->sample_freq = 0;

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

	if (opts->sample_address) {
657
		perf_evsel__set_sample_bit(evsel, ADDR);
658 659 660
		attr->mmap_data = track;
	}

661 662 663 664 665 666 667 668
	/*
	 * We don't allow user space callchains for  function trace
	 * event, due to issues with page faults while tracing page
	 * fault handler and its overall trickiness nature.
	 */
	if (perf_evsel__is_function_event(evsel))
		evsel->attr.exclude_callchain_user = 1;

669 670
	if (callchain_param.enabled && !evsel->no_aux_samples)
		perf_evsel__config_callgraph(evsel);
671

672 673 674 675 676
	if (opts->sample_intr_regs) {
		attr->sample_regs_intr = PERF_REGS_MASK;
		perf_evsel__set_sample_bit(evsel, REGS_INTR);
	}

677
	if (target__has_cpu(&opts->target))
678
		perf_evsel__set_sample_bit(evsel, CPU);
679

680
	if (opts->period)
681
		perf_evsel__set_sample_bit(evsel, PERIOD);
682

683 684 685 686 687 688
	/*
	 * 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)))
689
		perf_evsel__set_sample_bit(evsel, TIME);
690

691
	if (opts->raw_samples && !evsel->no_aux_samples) {
692 693 694
		perf_evsel__set_sample_bit(evsel, TIME);
		perf_evsel__set_sample_bit(evsel, RAW);
		perf_evsel__set_sample_bit(evsel, CPU);
695 696
	}

697
	if (opts->sample_address)
698
		perf_evsel__set_sample_bit(evsel, DATA_SRC);
699

700
	if (opts->no_buffering) {
701 702 703
		attr->watermark = 0;
		attr->wakeup_events = 1;
	}
704
	if (opts->branch_stack && !evsel->no_aux_samples) {
705
		perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
706 707
		attr->branch_sample_type = opts->branch_stack;
	}
708

709
	if (opts->sample_weight)
710
		perf_evsel__set_sample_bit(evsel, WEIGHT);
711

712
	attr->task  = track;
713
	attr->mmap  = track;
714
	attr->mmap2 = track && !perf_missing_features.mmap2;
715
	attr->comm  = track;
716

717
	if (opts->sample_transaction)
718
		perf_evsel__set_sample_bit(evsel, TRANSACTION);
719

720 721 722 723 724 725
	/*
	 * XXX see the function comment above
	 *
	 * Disabling only independent events or group leaders,
	 * keeping group members enabled.
	 */
726
	if (perf_evsel__is_group_leader(evsel))
727 728 729 730 731 732
		attr->disabled = 1;

	/*
	 * Setting enable_on_exec for independent events and
	 * group leaders for traced executed by perf.
	 */
733 734
	if (target__none(&opts->target) && perf_evsel__is_group_leader(evsel) &&
		!opts->initial_delay)
735
		attr->enable_on_exec = 1;
736 737 738 739 740

	if (evsel->immediate) {
		attr->disabled = 0;
		attr->enable_on_exec = 0;
	}
741 742
}

743
static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
744
{
745
	int cpu, thread;
746 747 748 749

	if (evsel->system_wide)
		nthreads = 1;

750
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
751 752 753 754 755 756 757 758 759

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

760 761 762
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

763 764
static int perf_evsel__run_ioctl(struct perf_evsel *evsel, int ncpus, int nthreads,
			  int ioc,  void *arg)
765 766 767
{
	int cpu, thread;

768 769 770
	if (evsel->system_wide)
		nthreads = 1;

771 772 773
	for (cpu = 0; cpu < ncpus; cpu++) {
		for (thread = 0; thread < nthreads; thread++) {
			int fd = FD(evsel, cpu, thread),
774
			    err = ioctl(fd, ioc, arg);
775 776 777 778 779 780 781 782 783

			if (err)
				return err;
		}
	}

	return 0;
}

784 785 786 787 788 789 790 791 792 793 794 795 796 797 798
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);
}

799 800
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
801 802 803
	if (ncpus == 0 || nthreads == 0)
		return 0;

804 805 806
	if (evsel->system_wide)
		nthreads = 1;

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

821 822 823 824 825 826
void perf_evsel__reset_counts(struct perf_evsel *evsel, int ncpus)
{
	memset(evsel->counts, 0, (sizeof(*evsel->counts) +
				 (ncpus * sizeof(struct perf_counts_values))));
}

827 828 829 830 831 832 833
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;
}

834
static void perf_evsel__free_fd(struct perf_evsel *evsel)
835 836 837 838 839
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

840
static void perf_evsel__free_id(struct perf_evsel *evsel)
841
{
842 843
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
844
	zfree(&evsel->id);
845 846
}

847 848 849 850
void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	int cpu, thread;

851 852 853
	if (evsel->system_wide)
		nthreads = 1;

854 855 856 857 858 859 860
	for (cpu = 0; cpu < ncpus; cpu++)
		for (thread = 0; thread < nthreads; ++thread) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
}

861 862
void perf_evsel__free_counts(struct perf_evsel *evsel)
{
863
	zfree(&evsel->counts);
864 865
}

866
void perf_evsel__exit(struct perf_evsel *evsel)
867 868
{
	assert(list_empty(&evsel->node));
869 870
	perf_evsel__free_fd(evsel);
	perf_evsel__free_id(evsel);
871 872 873
	close_cgroup(evsel->cgrp);
	zfree(&evsel->group_name);
	zfree(&evsel->name);
A
Arnaldo Carvalho de Melo 已提交
874
	perf_evsel__object.fini(evsel);
875 876 877 878 879
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
880 881
	free(evsel);
}
882

883 884
void perf_evsel__compute_deltas(struct perf_evsel *evsel, int cpu,
				struct perf_counts_values *count)
885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903
{
	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;
}

904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923
void perf_counts_values__scale(struct perf_counts_values *count,
			       bool scale, s8 *pscaled)
{
	s8 scaled = 0;

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

	if (pscaled)
		*pscaled = scaled;
}

924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939
int perf_evsel__read_cb(struct perf_evsel *evsel, int cpu, int thread,
			perf_evsel__read_cb_t cb)
{
	struct perf_counts_values count;

	memset(&count, 0, sizeof(count));

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

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

	return cb(evsel, cpu, thread, &count);
}

940 941 942 943 944 945 946 947 948
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;

949 950 951
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1) < 0)
		return -ENOMEM;

952 953 954
	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
		return -errno;

955
	perf_evsel__compute_deltas(evsel, cpu, &count);
956
	perf_counts_values__scale(&count, scale, NULL);
957 958 959 960
	evsel->counts->cpu[cpu] = count;
	return 0;
}

961 962 963 964 965
static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
{
	struct perf_evsel *leader = evsel->leader;
	int fd;

966
	if (perf_evsel__is_group_leader(evsel))
967 968 969 970 971 972 973 974 975 976 977 978 979 980
		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
981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015
#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);
1016
	ret += PRINT_ATTR2(mmap2, comm_exec);
A
Adrian Hunter 已提交
1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034
	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);
1035
	ret += PRINT_ATTR_X64(sample_regs_intr);
A
Adrian Hunter 已提交
1036 1037 1038 1039 1040 1041

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

	return ret;
}

1042
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1043
			      struct thread_map *threads)
1044
{
1045
	int cpu, thread, nthreads;
1046
	unsigned long flags = PERF_FLAG_FD_CLOEXEC;
1047
	int pid = -1, err;
1048
	enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit = NO_CHANGE;
1049

1050 1051 1052 1053 1054
	if (evsel->system_wide)
		nthreads = 1;
	else
		nthreads = threads->nr;

1055
	if (evsel->fd == NULL &&
1056
	    perf_evsel__alloc_fd(evsel, cpus->nr, nthreads) < 0)
1057
		return -ENOMEM;
1058

S
Stephane Eranian 已提交
1059
	if (evsel->cgrp) {
1060
		flags |= PERF_FLAG_PID_CGROUP;
S
Stephane Eranian 已提交
1061 1062 1063
		pid = evsel->cgrp->fd;
	}

1064
fallback_missing_features:
1065 1066
	if (perf_missing_features.cloexec)
		flags &= ~(unsigned long)PERF_FLAG_FD_CLOEXEC;
1067 1068
	if (perf_missing_features.mmap2)
		evsel->attr.mmap2 = 0;
1069 1070 1071 1072 1073 1074
	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
1075 1076 1077
	if (verbose >= 2)
		perf_event_attr__fprintf(&evsel->attr, stderr);

1078
	for (cpu = 0; cpu < cpus->nr; cpu++) {
1079

1080
		for (thread = 0; thread < nthreads; thread++) {
1081
			int group_fd;
S
Stephane Eranian 已提交
1082

1083
			if (!evsel->cgrp && !evsel->system_wide)
S
Stephane Eranian 已提交
1084 1085
				pid = threads->map[thread];

1086
			group_fd = get_group_fd(evsel, cpu, thread);
1087
retry_open:
1088
			pr_debug2("sys_perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx\n",
A
Adrian Hunter 已提交
1089 1090
				  pid, cpus->map[cpu], group_fd, flags);

1091
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
1092
								     pid,
1093
								     cpus->map[cpu],
S
Stephane Eranian 已提交
1094
								     group_fd, flags);
1095 1096
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
1097
				pr_debug2("sys_perf_event_open failed, error %d\n",
1098
					  err);
1099
				goto try_fallback;
1100
			}
1101
			set_rlimit = NO_CHANGE;
1102
		}
1103 1104 1105 1106
	}

	return 0;

1107
try_fallback:
1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
	/*
	 * 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;
	}

1132 1133 1134
	if (err != -EINVAL || cpu > 0 || thread > 0)
		goto out_close;

1135 1136 1137 1138
	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) {
1139 1140 1141 1142
		perf_missing_features.mmap2 = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.exclude_guest &&
		   (evsel->attr.exclude_guest || evsel->attr.exclude_host)) {
1143 1144 1145 1146 1147 1148 1149
		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;
	}

1150
out_close:
1151 1152 1153 1154 1155
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
1156
		thread = nthreads;
1157
	} while (--cpu >= 0);
1158 1159 1160 1161 1162 1163 1164 1165 1166 1167
	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);
1168 1169
}

1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185
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, },
};

1186
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1187
		     struct thread_map *threads)
1188
{
1189 1190 1191
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
1192 1193
	}

1194 1195
	if (threads == NULL)
		threads = &empty_thread_map.map;
1196

1197
	return __perf_evsel__open(evsel, cpus, threads);
1198 1199
}

1200
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
1201
			     struct cpu_map *cpus)
1202
{
1203
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
1204
}
1205

1206
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
1207
				struct thread_map *threads)
1208
{
1209
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
1210
}
1211

1212 1213 1214
static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
				       const union perf_event *event,
				       struct perf_sample *sample)
1215
{
1216
	u64 type = evsel->attr.sample_type;
1217
	const u64 *array = event->sample.array;
1218
	bool swapped = evsel->needs_swap;
1219
	union u64_swap u;
1220 1221 1222 1223

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

1224 1225 1226 1227 1228
	if (type & PERF_SAMPLE_IDENTIFIER) {
		sample->id = *array;
		array--;
	}

1229
	if (type & PERF_SAMPLE_CPU) {
1230 1231 1232 1233 1234 1235 1236 1237
		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];
1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256
		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) {
1257 1258 1259 1260 1261 1262 1263 1264 1265 1266
		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];
1267
		array--;
1268 1269 1270 1271 1272
	}

	return 0;
}

1273 1274
static inline bool overflow(const void *endp, u16 max_size, const void *offset,
			    u64 size)
1275
{
1276 1277
	return size > max_size || offset + size > endp;
}
1278

1279 1280 1281 1282 1283
#define OVERFLOW_CHECK(offset, size, max_size)				\
	do {								\
		if (overflow(endp, (max_size), (offset), (size)))	\
			return -EFAULT;					\
	} while (0)
1284

1285 1286
#define OVERFLOW_CHECK_u64(offset) \
	OVERFLOW_CHECK(offset, sizeof(u64), sizeof(u64))
1287

1288
int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
1289
			     struct perf_sample *data)
1290
{
1291
	u64 type = evsel->attr.sample_type;
1292
	bool swapped = evsel->needs_swap;
1293
	const u64 *array;
1294 1295 1296
	u16 max_size = event->header.size;
	const void *endp = (void *)event + max_size;
	u64 sz;
1297

1298 1299 1300 1301
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
1302
	union u64_swap u;
1303

1304
	memset(data, 0, sizeof(*data));
1305 1306
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
1307
	data->period = evsel->attr.sample_period;
1308
	data->weight = 0;
1309 1310

	if (event->header.type != PERF_RECORD_SAMPLE) {
1311
		if (!evsel->attr.sample_id_all)
1312
			return 0;
1313
		return perf_evsel__parse_id_sample(evsel, event, data);
1314 1315 1316 1317
	}

	array = event->sample.array;

1318 1319 1320 1321 1322
	/*
	 * 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.
	 */
1323
	if (evsel->sample_size + sizeof(event->header) > event->header.size)
1324 1325
		return -EFAULT;

1326 1327 1328 1329 1330 1331
	data->id = -1ULL;
	if (type & PERF_SAMPLE_IDENTIFIER) {
		data->id = *array;
		array++;
	}

1332
	if (type & PERF_SAMPLE_IP) {
1333
		data->ip = *array;
1334 1335 1336 1337
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
1338 1339 1340 1341 1342 1343 1344 1345 1346 1347
		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];
1348 1349 1350 1351 1352 1353 1354 1355
		array++;
	}

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

1356
	data->addr = 0;
1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372
	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) {
1373 1374 1375 1376 1377 1378 1379 1380 1381

		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];
1382 1383 1384 1385 1386 1387 1388 1389 1390
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
1391 1392
		u64 read_format = evsel->attr.read_format;

1393
		OVERFLOW_CHECK_u64(array);
1394 1395 1396 1397 1398 1399 1400 1401
		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) {
1402
			OVERFLOW_CHECK_u64(array);
1403 1404 1405 1406 1407
			data->read.time_enabled = *array;
			array++;
		}

		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
1408
			OVERFLOW_CHECK_u64(array);
1409 1410 1411 1412 1413 1414
			data->read.time_running = *array;
			array++;
		}

		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
		if (read_format & PERF_FORMAT_GROUP) {
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425
			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;
1426
		} else {
1427
			OVERFLOW_CHECK_u64(array);
1428 1429 1430
			data->read.one.id = *array;
			array++;
		}
1431 1432 1433
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
1434
		const u64 max_callchain_nr = UINT64_MAX / sizeof(u64);
1435

1436 1437 1438
		OVERFLOW_CHECK_u64(array);
		data->callchain = (struct ip_callchain *)array++;
		if (data->callchain->nr > max_callchain_nr)
1439
			return -EFAULT;
1440 1441 1442
		sz = data->callchain->nr * sizeof(u64);
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1443 1444 1445
	}

	if (type & PERF_SAMPLE_RAW) {
1446
		OVERFLOW_CHECK_u64(array);
1447 1448 1449 1450 1451 1452 1453 1454 1455
		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];
1456
		array = (void *)array + sizeof(u32);
1457

1458 1459 1460
		OVERFLOW_CHECK(array, data->raw_size, max_size);
		data->raw_data = (void *)array;
		array = (void *)array + data->raw_size;
1461 1462
	}

1463
	if (type & PERF_SAMPLE_BRANCH_STACK) {
1464 1465
		const u64 max_branch_nr = UINT64_MAX /
					  sizeof(struct branch_entry);
1466

1467 1468
		OVERFLOW_CHECK_u64(array);
		data->branch_stack = (struct branch_stack *)array++;
1469

1470 1471
		if (data->branch_stack->nr > max_branch_nr)
			return -EFAULT;
1472
		sz = data->branch_stack->nr * sizeof(struct branch_entry);
1473 1474
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1475
	}
1476 1477

	if (type & PERF_SAMPLE_REGS_USER) {
1478
		OVERFLOW_CHECK_u64(array);
1479 1480
		data->user_regs.abi = *array;
		array++;
1481

1482
		if (data->user_regs.abi) {
1483
			u64 mask = evsel->attr.sample_regs_user;
1484

1485
			sz = hweight_long(mask) * sizeof(u64);
1486
			OVERFLOW_CHECK(array, sz, max_size);
1487
			data->user_regs.mask = mask;
1488
			data->user_regs.regs = (u64 *)array;
1489
			array = (void *)array + sz;
1490 1491 1492 1493
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
1494 1495
		OVERFLOW_CHECK_u64(array);
		sz = *array++;
1496 1497 1498 1499

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

1500
		if (!sz) {
1501 1502
			data->user_stack.size = 0;
		} else {
1503
			OVERFLOW_CHECK(array, sz, max_size);
1504
			data->user_stack.data = (char *)array;
1505 1506
			array = (void *)array + sz;
			OVERFLOW_CHECK_u64(array);
1507
			data->user_stack.size = *array++;
1508 1509 1510
			if (WARN_ONCE(data->user_stack.size > sz,
				      "user stack dump failure\n"))
				return -EFAULT;
1511 1512 1513
		}
	}

1514 1515
	data->weight = 0;
	if (type & PERF_SAMPLE_WEIGHT) {
1516
		OVERFLOW_CHECK_u64(array);
1517 1518 1519 1520
		data->weight = *array;
		array++;
	}

1521 1522
	data->data_src = PERF_MEM_DATA_SRC_NONE;
	if (type & PERF_SAMPLE_DATA_SRC) {
1523
		OVERFLOW_CHECK_u64(array);
1524 1525 1526 1527
		data->data_src = *array;
		array++;
	}

1528 1529
	data->transaction = 0;
	if (type & PERF_SAMPLE_TRANSACTION) {
1530
		OVERFLOW_CHECK_u64(array);
1531 1532 1533 1534
		data->transaction = *array;
		array++;
	}

1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551
	data->intr_regs.abi = PERF_SAMPLE_REGS_ABI_NONE;
	if (type & PERF_SAMPLE_REGS_INTR) {
		OVERFLOW_CHECK_u64(array);
		data->intr_regs.abi = *array;
		array++;

		if (data->intr_regs.abi != PERF_SAMPLE_REGS_ABI_NONE) {
			u64 mask = evsel->attr.sample_regs_intr;

			sz = hweight_long(mask) * sizeof(u64);
			OVERFLOW_CHECK(array, sz, max_size);
			data->intr_regs.mask = mask;
			data->intr_regs.regs = (u64 *)array;
			array = (void *)array + sz;
		}
	}

1552 1553
	return 0;
}
1554

1555
size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type,
1556
				     u64 read_format)
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 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621
{
	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);
1622
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643
			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);

1644 1645 1646
	if (type & PERF_SAMPLE_TRANSACTION)
		result += sizeof(u64);

1647 1648 1649 1650 1651 1652 1653 1654 1655 1656
	if (type & PERF_SAMPLE_REGS_INTR) {
		if (sample->intr_regs.abi) {
			result += sizeof(u64);
			sz = hweight_long(sample->intr_regs.mask) * sizeof(u64);
			result += sz;
		} else {
			result += sizeof(u64);
		}
	}

1657 1658 1659
	return result;
}

1660
int perf_event__synthesize_sample(union perf_event *event, u64 type,
1661
				  u64 read_format,
1662 1663 1664 1665
				  const struct perf_sample *sample,
				  bool swapped)
{
	u64 *array;
1666
	size_t sz;
1667 1668 1669 1670
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
1671
	union u64_swap u;
1672 1673 1674

	array = event->sample.array;

1675 1676 1677 1678 1679
	if (type & PERF_SAMPLE_IDENTIFIER) {
		*array = sample->id;
		array++;
	}

1680
	if (type & PERF_SAMPLE_IP) {
1681
		*array = sample->ip;
1682 1683 1684 1685 1686 1687 1688 1689
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
		u.val32[0] = sample->pid;
		u.val32[1] = sample->tid;
		if (swapped) {
			/*
1690
			 * Inverse of what is done in perf_evsel__parse_sample
1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724
			 */
			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) {
			/*
1725
			 * Inverse of what is done in perf_evsel__parse_sample
1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738
			 */
			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++;
	}

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 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801
	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;
1802
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829
			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++;
	}

1830 1831 1832 1833 1834
	if (type & PERF_SAMPLE_TRANSACTION) {
		*array = sample->transaction;
		array++;
	}

1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845
	if (type & PERF_SAMPLE_REGS_INTR) {
		if (sample->intr_regs.abi) {
			*array++ = sample->intr_regs.abi;
			sz = hweight_long(sample->intr_regs.mask) * sizeof(u64);
			memcpy(array, sample->intr_regs.regs, sz);
			array = (void *)array + sz;
		} else {
			*array++ = 0;
		}
	}

1846 1847
	return 0;
}
1848

1849 1850 1851 1852 1853
struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
{
	return pevent_find_field(evsel->tp_format, name);
}

1854
void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
1855 1856
			 const char *name)
{
1857
	struct format_field *field = perf_evsel__field(evsel, name);
1858 1859
	int offset;

1860 1861
	if (!field)
		return NULL;
1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875

	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)
{
1876
	struct format_field *field = perf_evsel__field(evsel, name);
1877 1878
	void *ptr;
	u64 value;
1879

1880 1881
	if (!field)
		return 0;
1882

1883
	ptr = sample->raw_data + field->offset;
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 1913 1914 1915
	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;
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 1942 1943 1944 1945

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)

1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974
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),
1975
		bit_name(IDENTIFIER), bit_name(REGS_INTR),
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
		{ .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);
}

1994 1995 1996 1997
int perf_evsel__fprintf(struct perf_evsel *evsel,
			struct perf_attr_details *details, FILE *fp)
{
	bool first = true;
N
Namhyung Kim 已提交
1998 1999
	int printed = 0;

2000
	if (details->event_group) {
N
Namhyung Kim 已提交
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
		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));
2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030

	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);
2031 2032 2033
		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);
2034 2035 2036 2037 2038 2039 2040 2041 2042
		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);
2043
		if_print(mmap2);
2044
		if_print(comm);
2045
		if_print(comm_exec);
2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060
		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
2061
out:
2062 2063 2064
	fputc('\n', fp);
	return ++printed;
}
2065 2066 2067 2068

bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
			  char *msg, size_t msgsize)
{
2069
	if ((err == ENOENT || err == ENXIO || err == ENODEV) &&
2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085
	    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;

2086
		zfree(&evsel->name);
2087 2088 2089 2090 2091
		return true;
	}

	return false;
}
2092

2093
int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
2094 2095
			      int err, char *msg, size_t size)
{
2096 2097
	char sbuf[STRERR_BUFSIZE];

2098 2099 2100
	switch (err) {
	case EPERM:
	case EACCES:
2101
		return scnprintf(msg, size,
2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131
		 "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;
2132 2133 2134 2135 2136 2137
	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;
2138 2139 2140 2141 2142
	default:
		break;
	}

	return scnprintf(msg, size,
2143
	"The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n"
2144 2145
	"/bin/dmesg may provide additional information.\n"
	"No CONFIG_PERF_EVENTS=y kernel support configured?\n",
2146 2147
			 err, strerror_r(err, sbuf, sizeof(sbuf)),
			 perf_evsel__name(evsel));
2148
}