evsel.c 56.9 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/tracing_path.h>
13 14 15
#include <traceevent/event-parse.h>
#include <linux/hw_breakpoint.h>
#include <linux/perf_event.h>
16
#include <linux/err.h>
17
#include <sys/resource.h>
18
#include "asm/bug.h"
19
#include "callchain.h"
20
#include "cgroup.h"
21
#include "evsel.h"
22
#include "evlist.h"
23
#include "util.h"
24
#include "cpumap.h"
25
#include "thread_map.h"
26
#include "target.h"
27
#include "perf_regs.h"
A
Adrian Hunter 已提交
28
#include "debug.h"
29
#include "trace-event.h"
30
#include "stat.h"
31

32 33 34
static struct {
	bool sample_id_all;
	bool exclude_guest;
35
	bool mmap2;
36
	bool cloexec;
37 38
	bool clockid;
	bool clockid_wrong;
39
	bool lbr_flags;
40 41
} perf_missing_features;

42 43
static clockid_t clockid;

A
Arnaldo Carvalho de Melo 已提交
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 79 80 81 82 83 84 85
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;
}

86 87
#define FD(e, x, y) (*(int *)xyarray__entry(e->fd, x, y))

88
int __perf_evsel__sample_size(u64 sample_type)
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103
{
	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;
}

104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169
/**
 * __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);
}

170 171 172 173 174 175
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);
176
		perf_evsel__calc_id_pos(evsel);
177 178 179 180 181 182 183 184 185
	}
}

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);
186
		perf_evsel__calc_id_pos(evsel);
187 188 189
	}
}

190 191
void perf_evsel__set_sample_id(struct perf_evsel *evsel,
			       bool can_sample_identifier)
192
{
193 194 195 196 197 198
	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);
	}
199 200 201
	evsel->attr.read_format |= PERF_FORMAT_ID;
}

202 203 204 205
void perf_evsel__init(struct perf_evsel *evsel,
		      struct perf_event_attr *attr, int idx)
{
	evsel->idx	   = idx;
206
	evsel->tracking	   = !idx;
207
	evsel->attr	   = *attr;
208
	evsel->leader	   = evsel;
209 210
	evsel->unit	   = "";
	evsel->scale	   = 1.0;
211
	evsel->evlist	   = NULL;
212
	evsel->bpf_fd	   = -1;
213
	INIT_LIST_HEAD(&evsel->node);
214
	INIT_LIST_HEAD(&evsel->config_terms);
A
Arnaldo Carvalho de Melo 已提交
215
	perf_evsel__object.init(evsel);
216
	evsel->sample_size = __perf_evsel__sample_size(attr->sample_type);
217
	perf_evsel__calc_id_pos(evsel);
218
	evsel->cmdline_group_boundary = false;
219 220
}

221
struct perf_evsel *perf_evsel__new_idx(struct perf_event_attr *attr, int idx)
222
{
A
Arnaldo Carvalho de Melo 已提交
223
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
224

225 226
	if (evsel != NULL)
		perf_evsel__init(evsel, attr, idx);
227

228
	if (perf_evsel__is_bpf_output(evsel)) {
229 230
		evsel->attr.sample_type |= (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					    PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
231 232 233
		evsel->attr.sample_period = 1;
	}

234 235 236
	return evsel;
}

237 238 239
/*
 * Returns pointer with encoded error via <linux/err.h> interface.
 */
240
struct perf_evsel *perf_evsel__newtp_idx(const char *sys, const char *name, int idx)
241
{
A
Arnaldo Carvalho de Melo 已提交
242
	struct perf_evsel *evsel = zalloc(perf_evsel__object.size);
243
	int err = -ENOMEM;
244

245 246 247
	if (evsel == NULL) {
		goto out_err;
	} else {
248
		struct perf_event_attr attr = {
249 250 251
			.type	       = PERF_TYPE_TRACEPOINT,
			.sample_type   = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME |
					  PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD),
252 253
		};

254 255 256
		if (asprintf(&evsel->name, "%s:%s", sys, name) < 0)
			goto out_free;

257
		evsel->tp_format = trace_event__tp_format(sys, name);
258 259
		if (IS_ERR(evsel->tp_format)) {
			err = PTR_ERR(evsel->tp_format);
260
			goto out_free;
261
		}
262

263
		event_attr_init(&attr);
264
		attr.config = evsel->tp_format->id;
265
		attr.sample_period = 1;
266 267 268 269 270 271
		perf_evsel__init(evsel, &attr, idx);
	}

	return evsel;

out_free:
272
	zfree(&evsel->name);
273
	free(evsel);
274 275
out_err:
	return ERR_PTR(err);
276 277
}

278
const char *perf_evsel__hw_names[PERF_COUNT_HW_MAX] = {
279 280 281 282 283 284 285 286 287 288 289 290
	"cycles",
	"instructions",
	"cache-references",
	"cache-misses",
	"branches",
	"branch-misses",
	"bus-cycles",
	"stalled-cycles-frontend",
	"stalled-cycles-backend",
	"ref-cycles",
};

291
static const char *__perf_evsel__hw_name(u64 config)
292 293 294 295 296 297 298
{
	if (config < PERF_COUNT_HW_MAX && perf_evsel__hw_names[config])
		return perf_evsel__hw_names[config];

	return "unknown-hardware";
}

299
static int perf_evsel__add_modifiers(struct perf_evsel *evsel, char *bf, size_t size)
300
{
301
	int colon = 0, r = 0;
302 303 304 305 306
	struct perf_event_attr *attr = &evsel->attr;
	bool exclude_guest_default = false;

#define MOD_PRINT(context, mod)	do {					\
		if (!attr->exclude_##context) {				\
307
			if (!colon) colon = ++r;			\
308 309 310 311 312 313 314 315 316 317 318 319
			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)
320
			colon = ++r;
321 322 323 324 325 326 327 328 329 330
		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)
331
		bf[colon - 1] = ':';
332 333 334
	return r;
}

335 336 337 338 339 340
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);
}

341
const char *perf_evsel__sw_names[PERF_COUNT_SW_MAX] = {
342 343 344 345
	"cpu-clock",
	"task-clock",
	"page-faults",
	"context-switches",
346
	"cpu-migrations",
347 348 349 350
	"minor-faults",
	"major-faults",
	"alignment-faults",
	"emulation-faults",
351
	"dummy",
352 353
};

354
static const char *__perf_evsel__sw_name(u64 config)
355 356 357 358 359 360 361 362 363 364 365 366
{
	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);
}

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
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);
}

392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457
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]);
}

458
static int __perf_evsel__hw_cache_name(u64 config, char *bf, size_t size)
459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
{
	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);
}

491 492 493 494 495 496
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);
}

497
const char *perf_evsel__name(struct perf_evsel *evsel)
498
{
499
	char bf[128];
500

501 502
	if (evsel->name)
		return evsel->name;
503 504 505

	switch (evsel->attr.type) {
	case PERF_TYPE_RAW:
506
		perf_evsel__raw_name(evsel, bf, sizeof(bf));
507 508 509
		break;

	case PERF_TYPE_HARDWARE:
510
		perf_evsel__hw_name(evsel, bf, sizeof(bf));
511
		break;
512 513

	case PERF_TYPE_HW_CACHE:
514
		perf_evsel__hw_cache_name(evsel, bf, sizeof(bf));
515 516
		break;

517
	case PERF_TYPE_SOFTWARE:
518
		perf_evsel__sw_name(evsel, bf, sizeof(bf));
519 520
		break;

521
	case PERF_TYPE_TRACEPOINT:
522
		scnprintf(bf, sizeof(bf), "%s", "unknown tracepoint");
523 524
		break;

525 526 527 528
	case PERF_TYPE_BREAKPOINT:
		perf_evsel__bp_name(evsel, bf, sizeof(bf));
		break;

529
	default:
530 531
		scnprintf(bf, sizeof(bf), "unknown attr type: %d",
			  evsel->attr.type);
532
		break;
533 534
	}

535 536 537
	evsel->name = strdup(bf);

	return evsel->name ?: "unknown";
538 539
}

540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
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;
}

565 566 567
void perf_evsel__config_callchain(struct perf_evsel *evsel,
				  struct record_opts *opts,
				  struct callchain_param *param)
568 569 570 571 572 573
{
	bool function = perf_evsel__is_function_event(evsel);
	struct perf_event_attr *attr = &evsel->attr;

	perf_evsel__set_sample_bit(evsel, CALLCHAIN);

574
	if (param->record_mode == CALLCHAIN_LBR) {
575 576 577 578 579 580 581 582
		if (!opts->branch_stack) {
			if (attr->exclude_user) {
				pr_warning("LBR callstack option is only available "
					   "to get user callchain information. "
					   "Falling back to framepointers.\n");
			} else {
				perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
				attr->branch_sample_type = PERF_SAMPLE_BRANCH_USER |
583 584 585
							PERF_SAMPLE_BRANCH_CALL_STACK |
							PERF_SAMPLE_BRANCH_NO_CYCLES |
							PERF_SAMPLE_BRANCH_NO_FLAGS;
586 587 588 589 590 591
			}
		} else
			 pr_warning("Cannot use LBR callstack with branch stack. "
				    "Falling back to framepointers.\n");
	}

592
	if (param->record_mode == CALLCHAIN_DWARF) {
593 594 595 596
		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;
597
			attr->sample_stack_user = param->dump_size;
598 599 600 601 602 603 604 605 606 607 608 609 610
			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;
	}
}

611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630
static void
perf_evsel__reset_callgraph(struct perf_evsel *evsel,
			    struct callchain_param *param)
{
	struct perf_event_attr *attr = &evsel->attr;

	perf_evsel__reset_sample_bit(evsel, CALLCHAIN);
	if (param->record_mode == CALLCHAIN_LBR) {
		perf_evsel__reset_sample_bit(evsel, BRANCH_STACK);
		attr->branch_sample_type &= ~(PERF_SAMPLE_BRANCH_USER |
					      PERF_SAMPLE_BRANCH_CALL_STACK);
	}
	if (param->record_mode == CALLCHAIN_DWARF) {
		perf_evsel__reset_sample_bit(evsel, REGS_USER);
		perf_evsel__reset_sample_bit(evsel, STACK_USER);
	}
}

static void apply_config_terms(struct perf_evsel *evsel,
			       struct record_opts *opts)
631 632
{
	struct perf_evsel_config_term *term;
K
Kan Liang 已提交
633 634
	struct list_head *config_terms = &evsel->config_terms;
	struct perf_event_attr *attr = &evsel->attr;
635 636 637 638 639 640
	struct callchain_param param;
	u32 dump_size = 0;
	char *callgraph_buf = NULL;

	/* callgraph default */
	param.record_mode = callchain_param.record_mode;
641 642 643

	list_for_each_entry(term, config_terms, list) {
		switch (term->type) {
644 645
		case PERF_EVSEL__CONFIG_TERM_PERIOD:
			attr->sample_period = term->val.period;
646
			attr->freq = 0;
K
Kan Liang 已提交
647
			break;
648 649 650 651
		case PERF_EVSEL__CONFIG_TERM_FREQ:
			attr->sample_freq = term->val.freq;
			attr->freq = 1;
			break;
K
Kan Liang 已提交
652 653 654 655 656 657
		case PERF_EVSEL__CONFIG_TERM_TIME:
			if (term->val.time)
				perf_evsel__set_sample_bit(evsel, TIME);
			else
				perf_evsel__reset_sample_bit(evsel, TIME);
			break;
658 659 660 661 662 663
		case PERF_EVSEL__CONFIG_TERM_CALLGRAPH:
			callgraph_buf = term->val.callgraph;
			break;
		case PERF_EVSEL__CONFIG_TERM_STACK_USER:
			dump_size = term->val.stack_user;
			break;
664 665 666 667 668 669 670 671 672
		case PERF_EVSEL__CONFIG_TERM_INHERIT:
			/*
			 * attr->inherit should has already been set by
			 * perf_evsel__config. If user explicitly set
			 * inherit using config terms, override global
			 * opt->no_inherit setting.
			 */
			attr->inherit = term->val.inherit ? 1 : 0;
			break;
673 674 675 676
		default:
			break;
		}
	}
677 678 679 680 681 682

	/* User explicitly set per-event callgraph, clear the old setting and reset. */
	if ((callgraph_buf != NULL) || (dump_size > 0)) {

		/* parse callgraph parameters */
		if (callgraph_buf != NULL) {
683 684 685 686 687 688 689 690 691 692 693
			if (!strcmp(callgraph_buf, "no")) {
				param.enabled = false;
				param.record_mode = CALLCHAIN_NONE;
			} else {
				param.enabled = true;
				if (parse_callchain_record(callgraph_buf, &param)) {
					pr_err("per-event callgraph setting for %s failed. "
					       "Apply callgraph global setting for it\n",
					       evsel->name);
					return;
				}
694 695 696 697 698 699 700 701 702 703 704 705 706
			}
		}
		if (dump_size > 0) {
			dump_size = round_up(dump_size, sizeof(u64));
			param.dump_size = dump_size;
		}

		/* If global callgraph set, clear it */
		if (callchain_param.enabled)
			perf_evsel__reset_callgraph(evsel, &callchain_param);

		/* set perf-event callgraph */
		if (param.enabled)
707
			perf_evsel__config_callchain(evsel, opts, &param);
708
	}
709 710
}

711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738
/*
 * 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.
 */
739 740
void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts,
			struct callchain_param *callchain)
741
{
742
	struct perf_evsel *leader = evsel->leader;
743
	struct perf_event_attr *attr = &evsel->attr;
744
	int track = evsel->tracking;
745
	bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread;
746

747
	attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1;
748 749
	attr->inherit	    = !opts->no_inherit;

750 751
	perf_evsel__set_sample_bit(evsel, IP);
	perf_evsel__set_sample_bit(evsel, TID);
752

753 754 755 756 757 758 759
	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.
		 */
760
		perf_evsel__set_sample_id(evsel, false);
761 762 763 764 765 766 767 768 769 770 771

		/*
		 * 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;
		}
	}

772
	/*
773
	 * We default some events to have a default interval. But keep
774 775
	 * it a weak assumption overridable by the user.
	 */
776
	if (!attr->sample_period || (opts->user_freq != UINT_MAX ||
777 778
				     opts->user_interval != ULLONG_MAX)) {
		if (opts->freq) {
779
			perf_evsel__set_sample_bit(evsel, PERIOD);
780 781 782 783 784 785 786
			attr->freq		= 1;
			attr->sample_freq	= opts->freq;
		} else {
			attr->sample_period = opts->default_interval;
		}
	}

787 788 789 790 791 792 793 794 795
	/*
	 * 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;
	}

796 797 798 799 800 801 802
	if (opts->no_samples)
		attr->sample_freq = 0;

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

	if (opts->sample_address) {
803
		perf_evsel__set_sample_bit(evsel, ADDR);
804 805 806
		attr->mmap_data = track;
	}

807 808 809 810 811 812 813 814
	/*
	 * 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;

815
	if (callchain && callchain->enabled && !evsel->no_aux_samples)
816
		perf_evsel__config_callchain(evsel, opts, callchain);
817

818
	if (opts->sample_intr_regs) {
819
		attr->sample_regs_intr = opts->sample_intr_regs;
820 821 822
		perf_evsel__set_sample_bit(evsel, REGS_INTR);
	}

823
	if (target__has_cpu(&opts->target))
824
		perf_evsel__set_sample_bit(evsel, CPU);
825

826
	if (opts->period)
827
		perf_evsel__set_sample_bit(evsel, PERIOD);
828

829 830 831 832 833
	/*
	 * When the user explicitely disabled time don't force it here.
	 */
	if (opts->sample_time &&
	    (!perf_missing_features.sample_id_all &&
834 835
	    (!opts->no_inherit || target__has_cpu(&opts->target) || per_cpu ||
	     opts->sample_time_set)))
836
		perf_evsel__set_sample_bit(evsel, TIME);
837

838
	if (opts->raw_samples && !evsel->no_aux_samples) {
839 840 841
		perf_evsel__set_sample_bit(evsel, TIME);
		perf_evsel__set_sample_bit(evsel, RAW);
		perf_evsel__set_sample_bit(evsel, CPU);
842 843
	}

844
	if (opts->sample_address)
845
		perf_evsel__set_sample_bit(evsel, DATA_SRC);
846

847
	if (opts->no_buffering) {
848 849 850
		attr->watermark = 0;
		attr->wakeup_events = 1;
	}
851
	if (opts->branch_stack && !evsel->no_aux_samples) {
852
		perf_evsel__set_sample_bit(evsel, BRANCH_STACK);
853 854
		attr->branch_sample_type = opts->branch_stack;
	}
855

856
	if (opts->sample_weight)
857
		perf_evsel__set_sample_bit(evsel, WEIGHT);
858

859
	attr->task  = track;
860
	attr->mmap  = track;
861
	attr->mmap2 = track && !perf_missing_features.mmap2;
862
	attr->comm  = track;
863

864 865 866
	if (opts->record_switch_events)
		attr->context_switch = track;

867
	if (opts->sample_transaction)
868
		perf_evsel__set_sample_bit(evsel, TRANSACTION);
869

870 871 872 873 874 875
	if (opts->running_time) {
		evsel->attr.read_format |=
			PERF_FORMAT_TOTAL_TIME_ENABLED |
			PERF_FORMAT_TOTAL_TIME_RUNNING;
	}

876 877 878 879 880 881
	/*
	 * XXX see the function comment above
	 *
	 * Disabling only independent events or group leaders,
	 * keeping group members enabled.
	 */
882
	if (perf_evsel__is_group_leader(evsel))
883 884 885 886 887 888
		attr->disabled = 1;

	/*
	 * Setting enable_on_exec for independent events and
	 * group leaders for traced executed by perf.
	 */
889 890
	if (target__none(&opts->target) && perf_evsel__is_group_leader(evsel) &&
		!opts->initial_delay)
891
		attr->enable_on_exec = 1;
892 893 894 895 896

	if (evsel->immediate) {
		attr->disabled = 0;
		attr->enable_on_exec = 0;
	}
897 898 899 900 901 902

	clockid = opts->clockid;
	if (opts->use_clockid) {
		attr->use_clockid = 1;
		attr->clockid = opts->clockid;
	}
903

904 905 906
	if (evsel->precise_max)
		perf_event_attr__set_max_precise_ip(attr);

907 908 909 910 911 912 913 914 915 916
	if (opts->all_user) {
		attr->exclude_kernel = 1;
		attr->exclude_user   = 0;
	}

	if (opts->all_kernel) {
		attr->exclude_kernel = 0;
		attr->exclude_user   = 1;
	}

917 918 919 920
	/*
	 * Apply event specific term settings,
	 * it overloads any global configuration.
	 */
921
	apply_config_terms(evsel, opts);
922 923
}

924
static int perf_evsel__alloc_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
925
{
926
	int cpu, thread;
927 928 929 930

	if (evsel->system_wide)
		nthreads = 1;

931
	evsel->fd = xyarray__new(ncpus, nthreads, sizeof(int));
932 933 934 935 936 937 938 939 940

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

941 942 943
	return evsel->fd != NULL ? 0 : -ENOMEM;
}

944 945
static int perf_evsel__run_ioctl(struct perf_evsel *evsel, int ncpus, int nthreads,
			  int ioc,  void *arg)
946 947 948
{
	int cpu, thread;

949 950 951
	if (evsel->system_wide)
		nthreads = 1;

952 953 954
	for (cpu = 0; cpu < ncpus; cpu++) {
		for (thread = 0; thread < nthreads; thread++) {
			int fd = FD(evsel, cpu, thread),
955
			    err = ioctl(fd, ioc, arg);
956 957 958 959 960 961 962 963 964

			if (err)
				return err;
		}
	}

	return 0;
}

965 966
int perf_evsel__apply_filter(struct perf_evsel *evsel, int ncpus, int nthreads,
			     const char *filter)
967 968 969 970 971 972
{
	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
				     PERF_EVENT_IOC_SET_FILTER,
				     (void *)filter);
}

973 974 975 976 977 978 979 980 981 982 983 984 985
int perf_evsel__set_filter(struct perf_evsel *evsel, const char *filter)
{
	char *new_filter = strdup(filter);

	if (new_filter != NULL) {
		free(evsel->filter);
		evsel->filter = new_filter;
		return 0;
	}

	return -1;
}

986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002
int perf_evsel__append_filter(struct perf_evsel *evsel,
			      const char *op, const char *filter)
{
	char *new_filter;

	if (evsel->filter == NULL)
		return perf_evsel__set_filter(evsel, filter);

	if (asprintf(&new_filter,"(%s) %s (%s)", evsel->filter, op, filter) > 0) {
		free(evsel->filter);
		evsel->filter = new_filter;
		return 0;
	}

	return -1;
}

1003
int perf_evsel__enable(struct perf_evsel *evsel)
1004
{
1005 1006 1007
	int nthreads = thread_map__nr(evsel->threads);
	int ncpus = cpu_map__nr(evsel->cpus);

1008 1009 1010 1011 1012
	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
				     PERF_EVENT_IOC_ENABLE,
				     0);
}

J
Jiri Olsa 已提交
1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
int perf_evsel__disable(struct perf_evsel *evsel)
{
	int nthreads = thread_map__nr(evsel->threads);
	int ncpus = cpu_map__nr(evsel->cpus);

	return perf_evsel__run_ioctl(evsel, ncpus, nthreads,
				     PERF_EVENT_IOC_DISABLE,
				     0);
}

1023 1024
int perf_evsel__alloc_id(struct perf_evsel *evsel, int ncpus, int nthreads)
{
1025 1026 1027
	if (ncpus == 0 || nthreads == 0)
		return 0;

1028 1029 1030
	if (evsel->system_wide)
		nthreads = 1;

1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042
	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;
1043 1044
}

1045
static void perf_evsel__free_fd(struct perf_evsel *evsel)
1046 1047 1048 1049 1050
{
	xyarray__delete(evsel->fd);
	evsel->fd = NULL;
}

1051
static void perf_evsel__free_id(struct perf_evsel *evsel)
1052
{
1053 1054
	xyarray__delete(evsel->sample_id);
	evsel->sample_id = NULL;
1055
	zfree(&evsel->id);
1056 1057
}

1058 1059 1060 1061 1062 1063 1064 1065 1066 1067
static void perf_evsel__free_config_terms(struct perf_evsel *evsel)
{
	struct perf_evsel_config_term *term, *h;

	list_for_each_entry_safe(term, h, &evsel->config_terms, list) {
		list_del(&term->list);
		free(term);
	}
}

1068 1069 1070 1071
void perf_evsel__close_fd(struct perf_evsel *evsel, int ncpus, int nthreads)
{
	int cpu, thread;

1072 1073 1074
	if (evsel->system_wide)
		nthreads = 1;

1075 1076 1077 1078 1079 1080 1081
	for (cpu = 0; cpu < ncpus; cpu++)
		for (thread = 0; thread < nthreads; ++thread) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
}

1082
void perf_evsel__exit(struct perf_evsel *evsel)
1083 1084
{
	assert(list_empty(&evsel->node));
1085
	assert(evsel->evlist == NULL);
1086 1087
	perf_evsel__free_fd(evsel);
	perf_evsel__free_id(evsel);
1088
	perf_evsel__free_config_terms(evsel);
1089
	close_cgroup(evsel->cgrp);
1090
	cpu_map__put(evsel->cpus);
A
Adrian Hunter 已提交
1091
	cpu_map__put(evsel->own_cpus);
1092
	thread_map__put(evsel->threads);
1093 1094
	zfree(&evsel->group_name);
	zfree(&evsel->name);
A
Arnaldo Carvalho de Melo 已提交
1095
	perf_evsel__object.fini(evsel);
1096 1097 1098 1099 1100
}

void perf_evsel__delete(struct perf_evsel *evsel)
{
	perf_evsel__exit(evsel);
1101 1102
	free(evsel);
}
1103

1104
void perf_evsel__compute_deltas(struct perf_evsel *evsel, int cpu, int thread,
1105
				struct perf_counts_values *count)
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115
{
	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 {
1116 1117
		tmp = *perf_counts(evsel->prev_raw_counts, cpu, thread);
		*perf_counts(evsel->prev_raw_counts, cpu, thread) = *count;
1118 1119 1120 1121 1122 1123 1124
	}

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

1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144
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;
}

1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158
int perf_evsel__read(struct perf_evsel *evsel, int cpu, int thread,
		     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 0;
}

1159 1160 1161 1162 1163 1164 1165 1166 1167
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;

1168
	if (evsel->counts == NULL && perf_evsel__alloc_counts(evsel, cpu + 1, thread + 1) < 0)
1169 1170
		return -ENOMEM;

1171 1172 1173
	if (readn(FD(evsel, cpu, thread), &count, nv * sizeof(u64)) < 0)
		return -errno;

1174
	perf_evsel__compute_deltas(evsel, cpu, thread, &count);
1175
	perf_counts_values__scale(&count, scale, NULL);
1176
	*perf_counts(evsel->counts, cpu, thread) = count;
1177 1178 1179
	return 0;
}

1180 1181 1182 1183 1184
static int get_group_fd(struct perf_evsel *evsel, int cpu, int thread)
{
	struct perf_evsel *leader = evsel->leader;
	int fd;

1185
	if (perf_evsel__is_group_leader(evsel))
1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199
		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;
}

1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
struct bit_names {
	int bit;
	const char *name;
};

static void __p_bits(char *buf, size_t size, u64 value, struct bit_names *bits)
{
	bool first_bit = true;
	int i = 0;

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

static void __p_sample_type(char *buf, size_t size, 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),
1226
		bit_name(IDENTIFIER), bit_name(REGS_INTR), bit_name(DATA_SRC),
1227
		bit_name(WEIGHT),
1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247
		{ .name = NULL, }
	};
#undef bit_name
	__p_bits(buf, size, value, bits);
}

static void __p_read_format(char *buf, size_t size, 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
	__p_bits(buf, size, value, bits);
}

#define BUF_SIZE		1024

1248
#define p_hex(val)		snprintf(buf, BUF_SIZE, "%#"PRIx64, (uint64_t)(val))
1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301
#define p_unsigned(val)		snprintf(buf, BUF_SIZE, "%"PRIu64, (uint64_t)(val))
#define p_signed(val)		snprintf(buf, BUF_SIZE, "%"PRId64, (int64_t)(val))
#define p_sample_type(val)	__p_sample_type(buf, BUF_SIZE, val)
#define p_read_format(val)	__p_read_format(buf, BUF_SIZE, val)

#define PRINT_ATTRn(_n, _f, _p)				\
do {							\
	if (attr->_f) {					\
		_p(attr->_f);				\
		ret += attr__fprintf(fp, _n, buf, priv);\
	}						\
} while (0)

#define PRINT_ATTRf(_f, _p)	PRINT_ATTRn(#_f, _f, _p)

int perf_event_attr__fprintf(FILE *fp, struct perf_event_attr *attr,
			     attr__fprintf_f attr__fprintf, void *priv)
{
	char buf[BUF_SIZE];
	int ret = 0;

	PRINT_ATTRf(type, p_unsigned);
	PRINT_ATTRf(size, p_unsigned);
	PRINT_ATTRf(config, p_hex);
	PRINT_ATTRn("{ sample_period, sample_freq }", sample_period, p_unsigned);
	PRINT_ATTRf(sample_type, p_sample_type);
	PRINT_ATTRf(read_format, p_read_format);

	PRINT_ATTRf(disabled, p_unsigned);
	PRINT_ATTRf(inherit, p_unsigned);
	PRINT_ATTRf(pinned, p_unsigned);
	PRINT_ATTRf(exclusive, p_unsigned);
	PRINT_ATTRf(exclude_user, p_unsigned);
	PRINT_ATTRf(exclude_kernel, p_unsigned);
	PRINT_ATTRf(exclude_hv, p_unsigned);
	PRINT_ATTRf(exclude_idle, p_unsigned);
	PRINT_ATTRf(mmap, p_unsigned);
	PRINT_ATTRf(comm, p_unsigned);
	PRINT_ATTRf(freq, p_unsigned);
	PRINT_ATTRf(inherit_stat, p_unsigned);
	PRINT_ATTRf(enable_on_exec, p_unsigned);
	PRINT_ATTRf(task, p_unsigned);
	PRINT_ATTRf(watermark, p_unsigned);
	PRINT_ATTRf(precise_ip, p_unsigned);
	PRINT_ATTRf(mmap_data, p_unsigned);
	PRINT_ATTRf(sample_id_all, p_unsigned);
	PRINT_ATTRf(exclude_host, p_unsigned);
	PRINT_ATTRf(exclude_guest, p_unsigned);
	PRINT_ATTRf(exclude_callchain_kernel, p_unsigned);
	PRINT_ATTRf(exclude_callchain_user, p_unsigned);
	PRINT_ATTRf(mmap2, p_unsigned);
	PRINT_ATTRf(comm_exec, p_unsigned);
	PRINT_ATTRf(use_clockid, p_unsigned);
1302
	PRINT_ATTRf(context_switch, p_unsigned);
1303 1304 1305 1306 1307

	PRINT_ATTRn("{ wakeup_events, wakeup_watermark }", wakeup_events, p_unsigned);
	PRINT_ATTRf(bp_type, p_unsigned);
	PRINT_ATTRn("{ bp_addr, config1 }", bp_addr, p_hex);
	PRINT_ATTRn("{ bp_len, config2 }", bp_len, p_hex);
1308
	PRINT_ATTRf(branch_sample_type, p_unsigned);
1309 1310 1311 1312
	PRINT_ATTRf(sample_regs_user, p_hex);
	PRINT_ATTRf(sample_stack_user, p_unsigned);
	PRINT_ATTRf(clockid, p_signed);
	PRINT_ATTRf(sample_regs_intr, p_hex);
1313
	PRINT_ATTRf(aux_watermark, p_unsigned);
A
Adrian Hunter 已提交
1314 1315 1316 1317

	return ret;
}

1318 1319 1320 1321 1322 1323
static int __open_attr__fprintf(FILE *fp, const char *name, const char *val,
				void *priv __attribute__((unused)))
{
	return fprintf(fp, "  %-32s %s\n", name, val);
}

1324
static int __perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1325
			      struct thread_map *threads)
1326
{
1327
	int cpu, thread, nthreads;
1328
	unsigned long flags = PERF_FLAG_FD_CLOEXEC;
1329
	int pid = -1, err;
1330
	enum { NO_CHANGE, SET_TO_MAX, INCREASED_MAX } set_rlimit = NO_CHANGE;
1331

1332 1333 1334 1335 1336
	if (evsel->system_wide)
		nthreads = 1;
	else
		nthreads = threads->nr;

1337
	if (evsel->fd == NULL &&
1338
	    perf_evsel__alloc_fd(evsel, cpus->nr, nthreads) < 0)
1339
		return -ENOMEM;
1340

S
Stephane Eranian 已提交
1341
	if (evsel->cgrp) {
1342
		flags |= PERF_FLAG_PID_CGROUP;
S
Stephane Eranian 已提交
1343 1344 1345
		pid = evsel->cgrp->fd;
	}

1346
fallback_missing_features:
1347 1348 1349 1350 1351 1352
	if (perf_missing_features.clockid_wrong)
		evsel->attr.clockid = CLOCK_MONOTONIC; /* should always work */
	if (perf_missing_features.clockid) {
		evsel->attr.use_clockid = 0;
		evsel->attr.clockid = 0;
	}
1353 1354
	if (perf_missing_features.cloexec)
		flags &= ~(unsigned long)PERF_FLAG_FD_CLOEXEC;
1355 1356
	if (perf_missing_features.mmap2)
		evsel->attr.mmap2 = 0;
1357 1358
	if (perf_missing_features.exclude_guest)
		evsel->attr.exclude_guest = evsel->attr.exclude_host = 0;
1359 1360 1361
	if (perf_missing_features.lbr_flags)
		evsel->attr.branch_sample_type &= ~(PERF_SAMPLE_BRANCH_NO_FLAGS |
				     PERF_SAMPLE_BRANCH_NO_CYCLES);
1362 1363 1364 1365
retry_sample_id:
	if (perf_missing_features.sample_id_all)
		evsel->attr.sample_id_all = 0;

1366 1367 1368 1369 1370 1371
	if (verbose >= 2) {
		fprintf(stderr, "%.60s\n", graph_dotted_line);
		fprintf(stderr, "perf_event_attr:\n");
		perf_event_attr__fprintf(stderr, &evsel->attr, __open_attr__fprintf, NULL);
		fprintf(stderr, "%.60s\n", graph_dotted_line);
	}
A
Adrian Hunter 已提交
1372

1373
	for (cpu = 0; cpu < cpus->nr; cpu++) {
1374

1375
		for (thread = 0; thread < nthreads; thread++) {
1376
			int group_fd;
S
Stephane Eranian 已提交
1377

1378
			if (!evsel->cgrp && !evsel->system_wide)
1379
				pid = thread_map__pid(threads, thread);
S
Stephane Eranian 已提交
1380

1381
			group_fd = get_group_fd(evsel, cpu, thread);
1382
retry_open:
1383
			pr_debug2("sys_perf_event_open: pid %d  cpu %d  group_fd %d  flags %#lx\n",
A
Adrian Hunter 已提交
1384 1385
				  pid, cpus->map[cpu], group_fd, flags);

1386
			FD(evsel, cpu, thread) = sys_perf_event_open(&evsel->attr,
S
Stephane Eranian 已提交
1387
								     pid,
1388
								     cpus->map[cpu],
S
Stephane Eranian 已提交
1389
								     group_fd, flags);
1390 1391
			if (FD(evsel, cpu, thread) < 0) {
				err = -errno;
1392
				pr_debug2("sys_perf_event_open failed, error %d\n",
1393
					  err);
1394
				goto try_fallback;
1395
			}
1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411

			if (evsel->bpf_fd >= 0) {
				int evt_fd = FD(evsel, cpu, thread);
				int bpf_fd = evsel->bpf_fd;

				err = ioctl(evt_fd,
					    PERF_EVENT_IOC_SET_BPF,
					    bpf_fd);
				if (err && errno != EEXIST) {
					pr_err("failed to attach bpf fd %d: %s\n",
					       bpf_fd, strerror(errno));
					err = -EINVAL;
					goto out_close;
				}
			}

1412
			set_rlimit = NO_CHANGE;
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423

			/*
			 * If we succeeded but had to kill clockid, fail and
			 * have perf_evsel__open_strerror() print us a nice
			 * error.
			 */
			if (perf_missing_features.clockid ||
			    perf_missing_features.clockid_wrong) {
				err = -EINVAL;
				goto out_close;
			}
1424
		}
1425 1426 1427 1428
	}

	return 0;

1429
try_fallback:
1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453
	/*
	 * 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;
	}

1454 1455 1456
	if (err != -EINVAL || cpu > 0 || thread > 0)
		goto out_close;

1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
	/*
	 * Must probe features in the order they were added to the
	 * perf_event_attr interface.
	 */
	if (!perf_missing_features.clockid_wrong && evsel->attr.use_clockid) {
		perf_missing_features.clockid_wrong = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.clockid && evsel->attr.use_clockid) {
		perf_missing_features.clockid = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.cloexec && (flags & PERF_FLAG_FD_CLOEXEC)) {
1468 1469 1470
		perf_missing_features.cloexec = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.mmap2 && evsel->attr.mmap2) {
1471 1472 1473 1474
		perf_missing_features.mmap2 = true;
		goto fallback_missing_features;
	} else if (!perf_missing_features.exclude_guest &&
		   (evsel->attr.exclude_guest || evsel->attr.exclude_host)) {
1475 1476 1477 1478 1479
		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;
1480 1481 1482 1483 1484 1485
	} else if (!perf_missing_features.lbr_flags &&
			(evsel->attr.branch_sample_type &
			 (PERF_SAMPLE_BRANCH_NO_CYCLES |
			  PERF_SAMPLE_BRANCH_NO_FLAGS))) {
		perf_missing_features.lbr_flags = true;
		goto fallback_missing_features;
1486 1487
	}

1488
out_close:
1489 1490 1491 1492 1493
	do {
		while (--thread >= 0) {
			close(FD(evsel, cpu, thread));
			FD(evsel, cpu, thread) = -1;
		}
1494
		thread = nthreads;
1495
	} while (--cpu >= 0);
1496 1497 1498 1499 1500 1501 1502 1503 1504 1505
	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);
1506 1507
}

1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523
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, },
};

1524
int perf_evsel__open(struct perf_evsel *evsel, struct cpu_map *cpus,
1525
		     struct thread_map *threads)
1526
{
1527 1528 1529
	if (cpus == NULL) {
		/* Work around old compiler warnings about strict aliasing */
		cpus = &empty_cpu_map.map;
1530 1531
	}

1532 1533
	if (threads == NULL)
		threads = &empty_thread_map.map;
1534

1535
	return __perf_evsel__open(evsel, cpus, threads);
1536 1537
}

1538
int perf_evsel__open_per_cpu(struct perf_evsel *evsel,
1539
			     struct cpu_map *cpus)
1540
{
1541
	return __perf_evsel__open(evsel, cpus, &empty_thread_map.map);
1542
}
1543

1544
int perf_evsel__open_per_thread(struct perf_evsel *evsel,
1545
				struct thread_map *threads)
1546
{
1547
	return __perf_evsel__open(evsel, &empty_cpu_map.map, threads);
1548
}
1549

1550 1551 1552
static int perf_evsel__parse_id_sample(const struct perf_evsel *evsel,
				       const union perf_event *event,
				       struct perf_sample *sample)
1553
{
1554
	u64 type = evsel->attr.sample_type;
1555
	const u64 *array = event->sample.array;
1556
	bool swapped = evsel->needs_swap;
1557
	union u64_swap u;
1558 1559 1560 1561

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

1562 1563 1564 1565 1566
	if (type & PERF_SAMPLE_IDENTIFIER) {
		sample->id = *array;
		array--;
	}

1567
	if (type & PERF_SAMPLE_CPU) {
1568 1569 1570 1571 1572 1573 1574 1575
		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];
1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594
		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) {
1595 1596 1597 1598 1599 1600 1601 1602 1603 1604
		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];
1605
		array--;
1606 1607 1608 1609 1610
	}

	return 0;
}

1611 1612
static inline bool overflow(const void *endp, u16 max_size, const void *offset,
			    u64 size)
1613
{
1614 1615
	return size > max_size || offset + size > endp;
}
1616

1617 1618 1619 1620 1621
#define OVERFLOW_CHECK(offset, size, max_size)				\
	do {								\
		if (overflow(endp, (max_size), (offset), (size)))	\
			return -EFAULT;					\
	} while (0)
1622

1623 1624
#define OVERFLOW_CHECK_u64(offset) \
	OVERFLOW_CHECK(offset, sizeof(u64), sizeof(u64))
1625

1626
int perf_evsel__parse_sample(struct perf_evsel *evsel, union perf_event *event,
1627
			     struct perf_sample *data)
1628
{
1629
	u64 type = evsel->attr.sample_type;
1630
	bool swapped = evsel->needs_swap;
1631
	const u64 *array;
1632 1633 1634
	u16 max_size = event->header.size;
	const void *endp = (void *)event + max_size;
	u64 sz;
1635

1636 1637 1638 1639
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
1640
	union u64_swap u;
1641

1642
	memset(data, 0, sizeof(*data));
1643 1644
	data->cpu = data->pid = data->tid = -1;
	data->stream_id = data->id = data->time = -1ULL;
1645
	data->period = evsel->attr.sample_period;
1646
	data->weight = 0;
1647
	data->cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
1648 1649

	if (event->header.type != PERF_RECORD_SAMPLE) {
1650
		if (!evsel->attr.sample_id_all)
1651
			return 0;
1652
		return perf_evsel__parse_id_sample(evsel, event, data);
1653 1654 1655 1656
	}

	array = event->sample.array;

1657 1658 1659 1660 1661
	/*
	 * 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.
	 */
1662
	if (evsel->sample_size + sizeof(event->header) > event->header.size)
1663 1664
		return -EFAULT;

1665 1666 1667 1668 1669 1670
	data->id = -1ULL;
	if (type & PERF_SAMPLE_IDENTIFIER) {
		data->id = *array;
		array++;
	}

1671
	if (type & PERF_SAMPLE_IP) {
1672
		data->ip = *array;
1673 1674 1675 1676
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
1677 1678 1679 1680 1681 1682 1683 1684 1685 1686
		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];
1687 1688 1689 1690 1691 1692 1693 1694
		array++;
	}

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

1695
	data->addr = 0;
1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711
	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) {
1712 1713 1714 1715 1716 1717 1718 1719 1720

		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];
1721 1722 1723 1724 1725 1726 1727 1728 1729
		array++;
	}

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

	if (type & PERF_SAMPLE_READ) {
1730 1731
		u64 read_format = evsel->attr.read_format;

1732
		OVERFLOW_CHECK_u64(array);
1733 1734 1735 1736 1737 1738 1739 1740
		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) {
1741
			OVERFLOW_CHECK_u64(array);
1742 1743 1744 1745 1746
			data->read.time_enabled = *array;
			array++;
		}

		if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
1747
			OVERFLOW_CHECK_u64(array);
1748 1749 1750 1751 1752 1753
			data->read.time_running = *array;
			array++;
		}

		/* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */
		if (read_format & PERF_FORMAT_GROUP) {
1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764
			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;
1765
		} else {
1766
			OVERFLOW_CHECK_u64(array);
1767 1768 1769
			data->read.one.id = *array;
			array++;
		}
1770 1771 1772
	}

	if (type & PERF_SAMPLE_CALLCHAIN) {
1773
		const u64 max_callchain_nr = UINT64_MAX / sizeof(u64);
1774

1775 1776 1777
		OVERFLOW_CHECK_u64(array);
		data->callchain = (struct ip_callchain *)array++;
		if (data->callchain->nr > max_callchain_nr)
1778
			return -EFAULT;
1779 1780 1781
		sz = data->callchain->nr * sizeof(u64);
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1782 1783 1784
	}

	if (type & PERF_SAMPLE_RAW) {
1785
		OVERFLOW_CHECK_u64(array);
1786 1787 1788 1789 1790 1791 1792 1793 1794
		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];
1795
		array = (void *)array + sizeof(u32);
1796

1797 1798 1799
		OVERFLOW_CHECK(array, data->raw_size, max_size);
		data->raw_data = (void *)array;
		array = (void *)array + data->raw_size;
1800 1801
	}

1802
	if (type & PERF_SAMPLE_BRANCH_STACK) {
1803 1804
		const u64 max_branch_nr = UINT64_MAX /
					  sizeof(struct branch_entry);
1805

1806 1807
		OVERFLOW_CHECK_u64(array);
		data->branch_stack = (struct branch_stack *)array++;
1808

1809 1810
		if (data->branch_stack->nr > max_branch_nr)
			return -EFAULT;
1811
		sz = data->branch_stack->nr * sizeof(struct branch_entry);
1812 1813
		OVERFLOW_CHECK(array, sz, max_size);
		array = (void *)array + sz;
1814
	}
1815 1816

	if (type & PERF_SAMPLE_REGS_USER) {
1817
		OVERFLOW_CHECK_u64(array);
1818 1819
		data->user_regs.abi = *array;
		array++;
1820

1821
		if (data->user_regs.abi) {
1822
			u64 mask = evsel->attr.sample_regs_user;
1823

1824
			sz = hweight_long(mask) * sizeof(u64);
1825
			OVERFLOW_CHECK(array, sz, max_size);
1826
			data->user_regs.mask = mask;
1827
			data->user_regs.regs = (u64 *)array;
1828
			array = (void *)array + sz;
1829 1830 1831 1832
		}
	}

	if (type & PERF_SAMPLE_STACK_USER) {
1833 1834
		OVERFLOW_CHECK_u64(array);
		sz = *array++;
1835 1836 1837 1838

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

1839
		if (!sz) {
1840 1841
			data->user_stack.size = 0;
		} else {
1842
			OVERFLOW_CHECK(array, sz, max_size);
1843
			data->user_stack.data = (char *)array;
1844 1845
			array = (void *)array + sz;
			OVERFLOW_CHECK_u64(array);
1846
			data->user_stack.size = *array++;
1847 1848 1849
			if (WARN_ONCE(data->user_stack.size > sz,
				      "user stack dump failure\n"))
				return -EFAULT;
1850 1851 1852
		}
	}

1853 1854
	data->weight = 0;
	if (type & PERF_SAMPLE_WEIGHT) {
1855
		OVERFLOW_CHECK_u64(array);
1856 1857 1858 1859
		data->weight = *array;
		array++;
	}

1860 1861
	data->data_src = PERF_MEM_DATA_SRC_NONE;
	if (type & PERF_SAMPLE_DATA_SRC) {
1862
		OVERFLOW_CHECK_u64(array);
1863 1864 1865 1866
		data->data_src = *array;
		array++;
	}

1867 1868
	data->transaction = 0;
	if (type & PERF_SAMPLE_TRANSACTION) {
1869
		OVERFLOW_CHECK_u64(array);
1870 1871 1872 1873
		data->transaction = *array;
		array++;
	}

1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890
	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;
		}
	}

1891 1892
	return 0;
}
1893

1894
size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type,
1895
				     u64 read_format)
1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 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 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960
{
	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);
1961
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982
			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);

1983 1984 1985
	if (type & PERF_SAMPLE_TRANSACTION)
		result += sizeof(u64);

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
	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);
		}
	}

1996 1997 1998
	return result;
}

1999
int perf_event__synthesize_sample(union perf_event *event, u64 type,
2000
				  u64 read_format,
2001 2002 2003 2004
				  const struct perf_sample *sample,
				  bool swapped)
{
	u64 *array;
2005
	size_t sz;
2006 2007 2008 2009
	/*
	 * used for cross-endian analysis. See git commit 65014ab3
	 * for why this goofiness is needed.
	 */
2010
	union u64_swap u;
2011 2012 2013

	array = event->sample.array;

2014 2015 2016 2017 2018
	if (type & PERF_SAMPLE_IDENTIFIER) {
		*array = sample->id;
		array++;
	}

2019
	if (type & PERF_SAMPLE_IP) {
2020
		*array = sample->ip;
2021 2022 2023 2024 2025 2026 2027 2028
		array++;
	}

	if (type & PERF_SAMPLE_TID) {
		u.val32[0] = sample->pid;
		u.val32[1] = sample->tid;
		if (swapped) {
			/*
2029
			 * Inverse of what is done in perf_evsel__parse_sample
2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063
			 */
			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) {
			/*
2064
			 * Inverse of what is done in perf_evsel__parse_sample
2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077
			 */
			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++;
	}

2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 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 2132 2133 2134 2135 2136 2137 2138 2139 2140
	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;
2141
			sz = hweight_long(sample->user_regs.mask) * sizeof(u64);
2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168
			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++;
	}

2169 2170 2171 2172 2173
	if (type & PERF_SAMPLE_TRANSACTION) {
		*array = sample->transaction;
		array++;
	}

2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184
	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;
		}
	}

2185 2186
	return 0;
}
2187

2188 2189 2190 2191 2192
struct format_field *perf_evsel__field(struct perf_evsel *evsel, const char *name)
{
	return pevent_find_field(evsel->tp_format, name);
}

2193
void *perf_evsel__rawptr(struct perf_evsel *evsel, struct perf_sample *sample,
2194 2195
			 const char *name)
{
2196
	struct format_field *field = perf_evsel__field(evsel, name);
2197 2198
	int offset;

2199 2200
	if (!field)
		return NULL;
2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214

	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)
{
2215
	struct format_field *field = perf_evsel__field(evsel, name);
2216 2217
	void *ptr;
	u64 value;
2218

2219 2220
	if (!field)
		return 0;
2221

2222
	ptr = sample->raw_data + field->offset;
2223

2224 2225 2226 2227 2228 2229 2230 2231 2232 2233
	switch (field->size) {
	case 1:
		return *(u8 *)ptr;
	case 2:
		value = *(u16 *)ptr;
		break;
	case 4:
		value = *(u32 *)ptr;
		break;
	case 8:
2234
		memcpy(&value, ptr, sizeof(u64));
2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254
		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;
2255
}
2256

2257 2258 2259
bool perf_evsel__fallback(struct perf_evsel *evsel, int err,
			  char *msg, size_t msgsize)
{
2260
	if ((err == ENOENT || err == ENXIO || err == ENODEV) &&
2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276
	    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;

2277
		zfree(&evsel->name);
2278 2279 2280 2281 2282
		return true;
	}

	return false;
}
2283

2284
int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
2285 2286
			      int err, char *msg, size_t size)
{
2287 2288
	char sbuf[STRERR_BUFSIZE];

2289 2290 2291
	switch (err) {
	case EPERM:
	case EACCES:
2292
		return scnprintf(msg, size,
2293 2294 2295 2296 2297 2298 2299 2300 2301
		 "You may not have permission to collect %sstats.\n\n"
		 "Consider tweaking /proc/sys/kernel/perf_event_paranoid,\n"
		 "which controls use of the performance events system by\n"
		 "unprivileged users (without CAP_SYS_ADMIN).\n\n"
		 "The default value is 1:\n\n"
		 "  -1: Allow use of (almost) all events by all users\n"
		 ">= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK\n"
		 ">= 1: Disallow CPU event access by users without CAP_SYS_ADMIN\n"
		 ">= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN",
2302 2303 2304 2305 2306 2307 2308
				 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"
2309 2310 2311
			 "Probably the maximum number of open file descriptors has been reached.\n"
			 "Hint: Try again after reducing the number of events.\n"
			 "Hint: Try increasing the limit with 'ulimit -n <limit>'");
2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327
	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;
2328 2329 2330 2331 2332 2333
	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;
2334 2335 2336 2337 2338 2339
	case EINVAL:
		if (perf_missing_features.clockid)
			return scnprintf(msg, size, "clockid feature not supported.");
		if (perf_missing_features.clockid_wrong)
			return scnprintf(msg, size, "wrong clockid (%d).", clockid);
		break;
2340 2341 2342 2343 2344
	default:
		break;
	}

	return scnprintf(msg, size,
2345
	"The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n"
2346 2347
	"/bin/dmesg may provide additional information.\n"
	"No CONFIG_PERF_EVENTS=y kernel support configured?\n",
2348 2349
			 err, strerror_r(err, sbuf, sizeof(sbuf)),
			 perf_evsel__name(evsel));
2350
}