header.c 65.3 KB
Newer Older
1
#include "util.h"
2
#include <sys/types.h>
3
#include <byteswap.h>
4 5 6
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
7
#include <linux/list.h>
8
#include <linux/kernel.h>
9
#include <linux/bitops.h>
10
#include <sys/utsname.h>
11

12
#include "evlist.h"
13
#include "evsel.h"
14
#include "header.h"
15 16
#include "../perf.h"
#include "trace-event.h"
17
#include "session.h"
18
#include "symbol.h"
19
#include "debug.h"
20
#include "cpumap.h"
21
#include "pmu.h"
22
#include "vdso.h"
23
#include "strbuf.h"
24
#include "build-id.h"
25

26 27
static bool no_buildid_cache = false;

28 29 30
static u32 header_argc;
static const char **header_argv;

31 32 33 34 35 36 37 38 39 40 41 42
/*
 * magic2 = "PERFILE2"
 * must be a numerical value to let the endianness
 * determine the memory layout. That way we are able
 * to detect endianness when reading the perf.data file
 * back.
 *
 * we check for legacy (PERFFILE) format.
 */
static const char *__perf_magic1 = "PERFFILE";
static const u64 __perf_magic2    = 0x32454c4946524550ULL;
static const u64 __perf_magic2_sw = 0x50455246494c4532ULL;
43

44
#define PERF_MAGIC	__perf_magic2
45 46

struct perf_file_attr {
47
	struct perf_event_attr	attr;
48 49 50
	struct perf_file_section	ids;
};

51
void perf_header__set_feat(struct perf_header *header, int feat)
52
{
53
	set_bit(feat, header->adds_features);
54 55
}

56
void perf_header__clear_feat(struct perf_header *header, int feat)
57
{
58
	clear_bit(feat, header->adds_features);
59 60
}

61
bool perf_header__has_feat(const struct perf_header *header, int feat)
62
{
63
	return test_bit(feat, header->adds_features);
64 65
}

66
static int do_write(int fd, const void *buf, size_t size)
67 68 69 70 71
{
	while (size) {
		int ret = write(fd, buf, size);

		if (ret < 0)
72
			return -errno;
73 74 75 76

		size -= ret;
		buf += ret;
	}
77 78

	return 0;
79 80
}

81 82 83 84 85 86 87 88 89 90 91 92 93 94
#define NAME_ALIGN 64

static int write_padded(int fd, const void *bf, size_t count,
			size_t count_aligned)
{
	static const char zero_buf[NAME_ALIGN];
	int err = do_write(fd, bf, count);

	if (!err)
		err = do_write(fd, zero_buf, count_aligned - count);

	return err;
}

95 96 97 98 99 100
static int do_write_string(int fd, const char *str)
{
	u32 len, olen;
	int ret;

	olen = strlen(str) + 1;
101
	len = PERF_ALIGN(olen, NAME_ALIGN);
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116

	/* write len, incl. \0 */
	ret = do_write(fd, &len, sizeof(len));
	if (ret < 0)
		return ret;

	return write_padded(fd, str, olen, len);
}

static char *do_read_string(int fd, struct perf_header *ph)
{
	ssize_t sz, ret;
	u32 len;
	char *buf;

117
	sz = readn(fd, &len, sizeof(len));
118 119 120 121 122 123 124 125 126 127
	if (sz < (ssize_t)sizeof(len))
		return NULL;

	if (ph->needs_swap)
		len = bswap_32(len);

	buf = malloc(len);
	if (!buf)
		return NULL;

128
	ret = readn(fd, buf, len);
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146
	if (ret == (ssize_t)len) {
		/*
		 * strings are padded by zeroes
		 * thus the actual strlen of buf
		 * may be less than len
		 */
		return buf;
	}

	free(buf);
	return NULL;
}

int
perf_header__set_cmdline(int argc, const char **argv)
{
	int i;

147 148 149 150 151 152 153 154 155
	/*
	 * If header_argv has already been set, do not override it.
	 * This allows a command to set the cmdline, parse args and
	 * then call another builtin function that implements a
	 * command -- e.g, cmd_kvm calling cmd_record.
	 */
	if (header_argv)
		return 0;

156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
	header_argc = (u32)argc;

	/* do not include NULL termination */
	header_argv = calloc(argc, sizeof(char *));
	if (!header_argv)
		return -ENOMEM;

	/*
	 * must copy argv contents because it gets moved
	 * around during option parsing
	 */
	for (i = 0; i < argc ; i++)
		header_argv[i] = argv[i];

	return 0;
}

173 174 175 176 177 178
#define dsos__for_each_with_build_id(pos, head)	\
	list_for_each_entry(pos, head, node)	\
		if (!pos->has_build_id)		\
			continue;		\
		else

179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
static int write_buildid(char *name, size_t name_len, u8 *build_id,
			 pid_t pid, u16 misc, int fd)
{
	int err;
	struct build_id_event b;
	size_t len;

	len = name_len + 1;
	len = PERF_ALIGN(len, NAME_ALIGN);

	memset(&b, 0, sizeof(b));
	memcpy(&b.build_id, build_id, BUILD_ID_SIZE);
	b.pid = pid;
	b.header.misc = misc;
	b.header.size = sizeof(b) + len;

	err = do_write(fd, &b, sizeof(b));
	if (err < 0)
		return err;

	return write_padded(fd, name, name_len + 1, len);
}

202 203 204 205 206 207 208
static int __dsos__write_buildid_table(struct list_head *head, pid_t pid,
				u16 misc, int fd)
{
	struct dso *pos;

	dsos__for_each_with_build_id(pos, head) {
		int err;
209 210
		char  *name;
		size_t name_len;
211 212 213

		if (!pos->hit)
			continue;
214 215 216 217 218 219 220 221 222 223 224 225

		if (is_vdso_map(pos->short_name)) {
			name = (char *) VDSO__MAP_NAME;
			name_len = sizeof(VDSO__MAP_NAME) + 1;
		} else {
			name = pos->long_name;
			name_len = pos->long_name_len + 1;
		}

		err = write_buildid(name, name_len, pos->build_id,
				    pid, misc, fd);
		if (err)
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
			return err;
	}

	return 0;
}

static int machine__write_buildid_table(struct machine *machine, int fd)
{
	int err;
	u16 kmisc = PERF_RECORD_MISC_KERNEL,
	    umisc = PERF_RECORD_MISC_USER;

	if (!machine__is_host(machine)) {
		kmisc = PERF_RECORD_MISC_GUEST_KERNEL;
		umisc = PERF_RECORD_MISC_GUEST_USER;
	}

	err = __dsos__write_buildid_table(&machine->kernel_dsos, machine->pid,
					  kmisc, fd);
	if (err == 0)
		err = __dsos__write_buildid_table(&machine->user_dsos,
						  machine->pid, umisc, fd);
	return err;
}

static int dsos__write_buildid_table(struct perf_header *header, int fd)
{
	struct perf_session *session = container_of(header,
			struct perf_session, header);
	struct rb_node *nd;
256
	int err = machine__write_buildid_table(&session->machines.host, fd);
257 258 259 260

	if (err)
		return err;

261
	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
262 263 264 265 266 267 268 269 270
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		err = machine__write_buildid_table(pos, fd);
		if (err)
			break;
	}
	return err;
}

int build_id_cache__add_s(const char *sbuild_id, const char *debugdir,
271
			  const char *name, bool is_kallsyms, bool is_vdso)
272 273 274 275 276
{
	const size_t size = PATH_MAX;
	char *realname, *filename = zalloc(size),
	     *linkname = zalloc(size), *targetname;
	int len, err = -1;
277
	bool slash = is_kallsyms || is_vdso;
278 279 280 281

	if (is_kallsyms) {
		if (symbol_conf.kptr_restrict) {
			pr_debug("Not caching a kptr_restrict'ed /proc/kallsyms\n");
282 283
			err = 0;
			goto out_free;
284
		}
285
		realname = (char *) name;
286 287 288 289 290 291
	} else
		realname = realpath(name, NULL);

	if (realname == NULL || filename == NULL || linkname == NULL)
		goto out_free;

292
	len = scnprintf(filename, size, "%s%s%s",
293 294
		       debugdir, slash ? "/" : "",
		       is_vdso ? VDSO__MAP_NAME : realname);
295 296 297
	if (mkdir_p(filename, 0755))
		goto out_free;

298
	snprintf(filename + len, size - len, "/%s", sbuild_id);
299 300 301 302 303 304 305 306 307

	if (access(filename, F_OK)) {
		if (is_kallsyms) {
			 if (copyfile("/proc/kallsyms", filename))
				goto out_free;
		} else if (link(realname, filename) && copyfile(name, filename))
			goto out_free;
	}

308
	len = scnprintf(linkname, size, "%s/.build-id/%.2s",
309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
		       debugdir, sbuild_id);

	if (access(linkname, X_OK) && mkdir_p(linkname, 0755))
		goto out_free;

	snprintf(linkname + len, size - len, "/%s", sbuild_id + 2);
	targetname = filename + strlen(debugdir) - 5;
	memcpy(targetname, "../..", 5);

	if (symlink(targetname, linkname) == 0)
		err = 0;
out_free:
	if (!is_kallsyms)
		free(realname);
	free(filename);
	free(linkname);
	return err;
}

static int build_id_cache__add_b(const u8 *build_id, size_t build_id_size,
				 const char *name, const char *debugdir,
330
				 bool is_kallsyms, bool is_vdso)
331 332 333 334 335
{
	char sbuild_id[BUILD_ID_SIZE * 2 + 1];

	build_id__sprintf(build_id, build_id_size, sbuild_id);

336 337
	return build_id_cache__add_s(sbuild_id, debugdir, name,
				     is_kallsyms, is_vdso);
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
}

int build_id_cache__remove_s(const char *sbuild_id, const char *debugdir)
{
	const size_t size = PATH_MAX;
	char *filename = zalloc(size),
	     *linkname = zalloc(size);
	int err = -1;

	if (filename == NULL || linkname == NULL)
		goto out_free;

	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
		 debugdir, sbuild_id, sbuild_id + 2);

	if (access(linkname, F_OK))
		goto out_free;

	if (readlink(linkname, filename, size - 1) < 0)
		goto out_free;

	if (unlink(linkname))
		goto out_free;

	/*
	 * Since the link is relative, we must make it absolute:
	 */
	snprintf(linkname, size, "%s/.build-id/%.2s/%s",
		 debugdir, sbuild_id, filename);

	if (unlink(linkname))
		goto out_free;

	err = 0;
out_free:
	free(filename);
	free(linkname);
	return err;
}

static int dso__cache_build_id(struct dso *dso, const char *debugdir)
{
	bool is_kallsyms = dso->kernel && dso->long_name[0] != '/';
381
	bool is_vdso = is_vdso_map(dso->short_name);
382 383

	return build_id_cache__add_b(dso->build_id, sizeof(dso->build_id),
384 385
				     dso->long_name, debugdir,
				     is_kallsyms, is_vdso);
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
}

static int __dsos__cache_build_ids(struct list_head *head, const char *debugdir)
{
	struct dso *pos;
	int err = 0;

	dsos__for_each_with_build_id(pos, head)
		if (dso__cache_build_id(pos, debugdir))
			err = -1;

	return err;
}

static int machine__cache_build_ids(struct machine *machine, const char *debugdir)
{
	int ret = __dsos__cache_build_ids(&machine->kernel_dsos, debugdir);
	ret |= __dsos__cache_build_ids(&machine->user_dsos, debugdir);
	return ret;
}

static int perf_session__cache_build_ids(struct perf_session *session)
{
	struct rb_node *nd;
	int ret;
	char debugdir[PATH_MAX];

	snprintf(debugdir, sizeof(debugdir), "%s", buildid_dir);

	if (mkdir(debugdir, 0755) != 0 && errno != EEXIST)
		return -1;

418
	ret = machine__cache_build_ids(&session->machines.host, debugdir);
419

420
	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		ret |= machine__cache_build_ids(pos, debugdir);
	}
	return ret ? -1 : 0;
}

static bool machine__read_build_ids(struct machine *machine, bool with_hits)
{
	bool ret = __dsos__read_build_ids(&machine->kernel_dsos, with_hits);
	ret |= __dsos__read_build_ids(&machine->user_dsos, with_hits);
	return ret;
}

static bool perf_session__read_build_ids(struct perf_session *session, bool with_hits)
{
	struct rb_node *nd;
437
	bool ret = machine__read_build_ids(&session->machines.host, with_hits);
438

439
	for (nd = rb_first(&session->machines.guests); nd; nd = rb_next(nd)) {
440 441 442 443 444 445 446
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		ret |= machine__read_build_ids(pos, with_hits);
	}

	return ret;
}

447
static int write_tracing_data(int fd, struct perf_header *h __maybe_unused,
448 449 450 451 452 453 454
			    struct perf_evlist *evlist)
{
	return read_tracing_data(fd, &evlist->entries);
}


static int write_build_id(int fd, struct perf_header *h,
455
			  struct perf_evlist *evlist __maybe_unused)
456 457 458 459 460 461
{
	struct perf_session *session;
	int err;

	session = container_of(h, struct perf_session, header);

462 463 464
	if (!perf_session__read_build_ids(session, true))
		return -1;

465 466 467 468 469 470 471 472 473 474 475
	err = dsos__write_buildid_table(h, fd);
	if (err < 0) {
		pr_debug("failed to write buildid table\n");
		return err;
	}
	if (!no_buildid_cache)
		perf_session__cache_build_ids(session);

	return 0;
}

476 477
static int write_hostname(int fd, struct perf_header *h __maybe_unused,
			  struct perf_evlist *evlist __maybe_unused)
478 479 480 481 482 483 484 485 486 487 488
{
	struct utsname uts;
	int ret;

	ret = uname(&uts);
	if (ret < 0)
		return -1;

	return do_write_string(fd, uts.nodename);
}

489 490
static int write_osrelease(int fd, struct perf_header *h __maybe_unused,
			   struct perf_evlist *evlist __maybe_unused)
491 492 493 494 495 496 497 498 499 500 501
{
	struct utsname uts;
	int ret;

	ret = uname(&uts);
	if (ret < 0)
		return -1;

	return do_write_string(fd, uts.release);
}

502 503
static int write_arch(int fd, struct perf_header *h __maybe_unused,
		      struct perf_evlist *evlist __maybe_unused)
504 505 506 507 508 509 510 511 512 513 514
{
	struct utsname uts;
	int ret;

	ret = uname(&uts);
	if (ret < 0)
		return -1;

	return do_write_string(fd, uts.machine);
}

515 516
static int write_version(int fd, struct perf_header *h __maybe_unused,
			 struct perf_evlist *evlist __maybe_unused)
517 518 519 520
{
	return do_write_string(fd, perf_version_string);
}

521 522
static int write_cpudesc(int fd, struct perf_header *h __maybe_unused,
		       struct perf_evlist *evlist __maybe_unused)
523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
{
#ifndef CPUINFO_PROC
#define CPUINFO_PROC NULL
#endif
	FILE *file;
	char *buf = NULL;
	char *s, *p;
	const char *search = CPUINFO_PROC;
	size_t len = 0;
	int ret = -1;

	if (!search)
		return -1;

	file = fopen("/proc/cpuinfo", "r");
	if (!file)
		return -1;

	while (getline(&buf, &len, file) > 0) {
		ret = strncmp(buf, search, strlen(search));
		if (!ret)
			break;
	}

	if (ret)
		goto done;

	s = buf;

	p = strchr(buf, ':');
	if (p && *(p+1) == ' ' && *(p+2))
		s = p + 2;
	p = strchr(s, '\n');
	if (p)
		*p = '\0';

	/* squash extra space characters (branding string) */
	p = s;
	while (*p) {
		if (isspace(*p)) {
			char *r = p + 1;
			char *q = r;
			*p = ' ';
			while (*q && isspace(*q))
				q++;
			if (q != (p+1))
				while ((*r++ = *q++));
		}
		p++;
	}
	ret = do_write_string(fd, s);
done:
	free(buf);
	fclose(file);
	return ret;
}

580 581
static int write_nrcpus(int fd, struct perf_header *h __maybe_unused,
			struct perf_evlist *evlist __maybe_unused)
582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605
{
	long nr;
	u32 nrc, nra;
	int ret;

	nr = sysconf(_SC_NPROCESSORS_CONF);
	if (nr < 0)
		return -1;

	nrc = (u32)(nr & UINT_MAX);

	nr = sysconf(_SC_NPROCESSORS_ONLN);
	if (nr < 0)
		return -1;

	nra = (u32)(nr & UINT_MAX);

	ret = do_write(fd, &nrc, sizeof(nrc));
	if (ret < 0)
		return ret;

	return do_write(fd, &nra, sizeof(nra));
}

606
static int write_event_desc(int fd, struct perf_header *h __maybe_unused,
607 608
			    struct perf_evlist *evlist)
{
609
	struct perf_evsel *evsel;
610
	u32 nre, nri, sz;
611 612
	int ret;

613
	nre = evlist->nr_entries;
614 615 616 617 618 619 620 621 622 623 624

	/*
	 * write number of events
	 */
	ret = do_write(fd, &nre, sizeof(nre));
	if (ret < 0)
		return ret;

	/*
	 * size of perf_event_attr struct
	 */
625
	sz = (u32)sizeof(evsel->attr);
626 627 628 629
	ret = do_write(fd, &sz, sizeof(sz));
	if (ret < 0)
		return ret;

630
	list_for_each_entry(evsel, &evlist->entries, node) {
631

632
		ret = do_write(fd, &evsel->attr, sz);
633 634 635 636 637 638 639 640 641
		if (ret < 0)
			return ret;
		/*
		 * write number of unique id per event
		 * there is one id per instance of an event
		 *
		 * copy into an nri to be independent of the
		 * type of ids,
		 */
642
		nri = evsel->ids;
643 644 645 646 647 648 649
		ret = do_write(fd, &nri, sizeof(nri));
		if (ret < 0)
			return ret;

		/*
		 * write event string as passed on cmdline
		 */
650
		ret = do_write_string(fd, perf_evsel__name(evsel));
651 652 653 654 655
		if (ret < 0)
			return ret;
		/*
		 * write unique ids for this event
		 */
656
		ret = do_write(fd, evsel->id, evsel->ids * sizeof(u64));
657 658 659 660 661 662
		if (ret < 0)
			return ret;
	}
	return 0;
}

663 664
static int write_cmdline(int fd, struct perf_header *h __maybe_unused,
			 struct perf_evlist *evlist __maybe_unused)
665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718
{
	char buf[MAXPATHLEN];
	char proc[32];
	u32 i, n;
	int ret;

	/*
	 * actual atual path to perf binary
	 */
	sprintf(proc, "/proc/%d/exe", getpid());
	ret = readlink(proc, buf, sizeof(buf));
	if (ret <= 0)
		return -1;

	/* readlink() does not add null termination */
	buf[ret] = '\0';

	/* account for binary path */
	n = header_argc + 1;

	ret = do_write(fd, &n, sizeof(n));
	if (ret < 0)
		return ret;

	ret = do_write_string(fd, buf);
	if (ret < 0)
		return ret;

	for (i = 0 ; i < header_argc; i++) {
		ret = do_write_string(fd, header_argv[i]);
		if (ret < 0)
			return ret;
	}
	return 0;
}

#define CORE_SIB_FMT \
	"/sys/devices/system/cpu/cpu%d/topology/core_siblings_list"
#define THRD_SIB_FMT \
	"/sys/devices/system/cpu/cpu%d/topology/thread_siblings_list"

struct cpu_topo {
	u32 core_sib;
	u32 thread_sib;
	char **core_siblings;
	char **thread_siblings;
};

static int build_cpu_topo(struct cpu_topo *tp, int cpu)
{
	FILE *fp;
	char filename[MAXPATHLEN];
	char *buf = NULL, *p;
	size_t len = 0;
719
	ssize_t sret;
720 721 722 723 724 725
	u32 i = 0;
	int ret = -1;

	sprintf(filename, CORE_SIB_FMT, cpu);
	fp = fopen(filename, "r");
	if (!fp)
726
		goto try_threads;
727

728
	sret = getline(&buf, &len, fp);
729
	fclose(fp);
730 731
	if (sret <= 0)
		goto try_threads;
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746

	p = strchr(buf, '\n');
	if (p)
		*p = '\0';

	for (i = 0; i < tp->core_sib; i++) {
		if (!strcmp(buf, tp->core_siblings[i]))
			break;
	}
	if (i == tp->core_sib) {
		tp->core_siblings[i] = buf;
		tp->core_sib++;
		buf = NULL;
		len = 0;
	}
747
	ret = 0;
748

749
try_threads:
750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
	sprintf(filename, THRD_SIB_FMT, cpu);
	fp = fopen(filename, "r");
	if (!fp)
		goto done;

	if (getline(&buf, &len, fp) <= 0)
		goto done;

	p = strchr(buf, '\n');
	if (p)
		*p = '\0';

	for (i = 0; i < tp->thread_sib; i++) {
		if (!strcmp(buf, tp->thread_siblings[i]))
			break;
	}
	if (i == tp->thread_sib) {
		tp->thread_siblings[i] = buf;
		tp->thread_sib++;
		buf = NULL;
	}
	ret = 0;
done:
	if(fp)
		fclose(fp);
	free(buf);
	return ret;
}

static void free_cpu_topo(struct cpu_topo *tp)
{
	u32 i;

	if (!tp)
		return;

	for (i = 0 ; i < tp->core_sib; i++)
		free(tp->core_siblings[i]);

	for (i = 0 ; i < tp->thread_sib; i++)
		free(tp->thread_siblings[i]);

	free(tp);
}

static struct cpu_topo *build_cpu_topology(void)
{
	struct cpu_topo *tp;
	void *addr;
	u32 nr, i;
	size_t sz;
	long ncpus;
	int ret = -1;

	ncpus = sysconf(_SC_NPROCESSORS_CONF);
	if (ncpus < 0)
		return NULL;

	nr = (u32)(ncpus & UINT_MAX);

	sz = nr * sizeof(char *);

	addr = calloc(1, sizeof(*tp) + 2 * sz);
	if (!addr)
		return NULL;

	tp = addr;

	addr += sizeof(*tp);
	tp->core_siblings = addr;
	addr += sz;
	tp->thread_siblings = addr;

	for (i = 0; i < nr; i++) {
		ret = build_cpu_topo(tp, i);
		if (ret < 0)
			break;
	}
	if (ret) {
		free_cpu_topo(tp);
		tp = NULL;
	}
	return tp;
}

835 836
static int write_cpu_topology(int fd, struct perf_header *h __maybe_unused,
			  struct perf_evlist *evlist __maybe_unused)
837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870
{
	struct cpu_topo *tp;
	u32 i;
	int ret;

	tp = build_cpu_topology();
	if (!tp)
		return -1;

	ret = do_write(fd, &tp->core_sib, sizeof(tp->core_sib));
	if (ret < 0)
		goto done;

	for (i = 0; i < tp->core_sib; i++) {
		ret = do_write_string(fd, tp->core_siblings[i]);
		if (ret < 0)
			goto done;
	}
	ret = do_write(fd, &tp->thread_sib, sizeof(tp->thread_sib));
	if (ret < 0)
		goto done;

	for (i = 0; i < tp->thread_sib; i++) {
		ret = do_write_string(fd, tp->thread_siblings[i]);
		if (ret < 0)
			break;
	}
done:
	free_cpu_topo(tp);
	return ret;
}



871 872
static int write_total_mem(int fd, struct perf_header *h __maybe_unused,
			  struct perf_evlist *evlist __maybe_unused)
873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926
{
	char *buf = NULL;
	FILE *fp;
	size_t len = 0;
	int ret = -1, n;
	uint64_t mem;

	fp = fopen("/proc/meminfo", "r");
	if (!fp)
		return -1;

	while (getline(&buf, &len, fp) > 0) {
		ret = strncmp(buf, "MemTotal:", 9);
		if (!ret)
			break;
	}
	if (!ret) {
		n = sscanf(buf, "%*s %"PRIu64, &mem);
		if (n == 1)
			ret = do_write(fd, &mem, sizeof(mem));
	}
	free(buf);
	fclose(fp);
	return ret;
}

static int write_topo_node(int fd, int node)
{
	char str[MAXPATHLEN];
	char field[32];
	char *buf = NULL, *p;
	size_t len = 0;
	FILE *fp;
	u64 mem_total, mem_free, mem;
	int ret = -1;

	sprintf(str, "/sys/devices/system/node/node%d/meminfo", node);
	fp = fopen(str, "r");
	if (!fp)
		return -1;

	while (getline(&buf, &len, fp) > 0) {
		/* skip over invalid lines */
		if (!strchr(buf, ':'))
			continue;
		if (sscanf(buf, "%*s %*d %s %"PRIu64, field, &mem) != 2)
			goto done;
		if (!strcmp(field, "MemTotal:"))
			mem_total = mem;
		if (!strcmp(field, "MemFree:"))
			mem_free = mem;
	}

	fclose(fp);
927
	fp = NULL;
928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953

	ret = do_write(fd, &mem_total, sizeof(u64));
	if (ret)
		goto done;

	ret = do_write(fd, &mem_free, sizeof(u64));
	if (ret)
		goto done;

	ret = -1;
	sprintf(str, "/sys/devices/system/node/node%d/cpulist", node);

	fp = fopen(str, "r");
	if (!fp)
		goto done;

	if (getline(&buf, &len, fp) <= 0)
		goto done;

	p = strchr(buf, '\n');
	if (p)
		*p = '\0';

	ret = do_write_string(fd, buf);
done:
	free(buf);
954 955
	if (fp)
		fclose(fp);
956 957 958
	return ret;
}

959 960
static int write_numa_topology(int fd, struct perf_header *h __maybe_unused,
			  struct perf_evlist *evlist __maybe_unused)
961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
{
	char *buf = NULL;
	size_t len = 0;
	FILE *fp;
	struct cpu_map *node_map = NULL;
	char *c;
	u32 nr, i, j;
	int ret = -1;

	fp = fopen("/sys/devices/system/node/online", "r");
	if (!fp)
		return -1;

	if (getline(&buf, &len, fp) <= 0)
		goto done;

	c = strchr(buf, '\n');
	if (c)
		*c = '\0';

	node_map = cpu_map__new(buf);
	if (!node_map)
		goto done;

	nr = (u32)node_map->nr;

	ret = do_write(fd, &nr, sizeof(nr));
	if (ret < 0)
		goto done;

	for (i = 0; i < nr; i++) {
		j = (u32)node_map->map[i];
		ret = do_write(fd, &j, sizeof(j));
		if (ret < 0)
			break;

		ret = write_topo_node(fd, i);
		if (ret < 0)
			break;
	}
done:
	free(buf);
	fclose(fp);
	free(node_map);
	return ret;
}

1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019
/*
 * File format:
 *
 * struct pmu_mappings {
 *	u32	pmu_num;
 *	struct pmu_map {
 *		u32	type;
 *		char	name[];
 *	}[pmu_num];
 * };
 */

1020 1021
static int write_pmu_mappings(int fd, struct perf_header *h __maybe_unused,
			      struct perf_evlist *evlist __maybe_unused)
1022 1023 1024 1025
{
	struct perf_pmu *pmu = NULL;
	off_t offset = lseek(fd, 0, SEEK_CUR);
	__u32 pmu_num = 0;
1026
	int ret;
1027 1028

	/* write real pmu_num later */
1029 1030 1031
	ret = do_write(fd, &pmu_num, sizeof(pmu_num));
	if (ret < 0)
		return ret;
1032 1033 1034 1035 1036

	while ((pmu = perf_pmu__scan(pmu))) {
		if (!pmu->name)
			continue;
		pmu_num++;
1037 1038 1039 1040 1041 1042 1043 1044

		ret = do_write(fd, &pmu->type, sizeof(pmu->type));
		if (ret < 0)
			return ret;

		ret = do_write_string(fd, pmu->name);
		if (ret < 0)
			return ret;
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055
	}

	if (pwrite(fd, &pmu_num, sizeof(pmu_num), offset) != sizeof(pmu_num)) {
		/* discard all */
		lseek(fd, offset, SEEK_SET);
		return -1;
	}

	return 0;
}

1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
/*
 * File format:
 *
 * struct group_descs {
 *	u32	nr_groups;
 *	struct group_desc {
 *		char	name[];
 *		u32	leader_idx;
 *		u32	nr_members;
 *	}[nr_groups];
 * };
 */
static int write_group_desc(int fd, struct perf_header *h __maybe_unused,
			    struct perf_evlist *evlist)
{
	u32 nr_groups = evlist->nr_groups;
	struct perf_evsel *evsel;
	int ret;

	ret = do_write(fd, &nr_groups, sizeof(nr_groups));
	if (ret < 0)
		return ret;

	list_for_each_entry(evsel, &evlist->entries, node) {
		if (perf_evsel__is_group_leader(evsel) &&
		    evsel->nr_members > 1) {
			const char *name = evsel->group_name ?: "{anon_group}";
			u32 leader_idx = evsel->idx;
			u32 nr_members = evsel->nr_members;

			ret = do_write_string(fd, name);
			if (ret < 0)
				return ret;

			ret = do_write(fd, &leader_idx, sizeof(leader_idx));
			if (ret < 0)
				return ret;

			ret = do_write(fd, &nr_members, sizeof(nr_members));
			if (ret < 0)
				return ret;
		}
	}
	return 0;
}

1102 1103 1104 1105
/*
 * default get_cpuid(): nothing gets recorded
 * actual implementation must be in arch/$(ARCH)/util/header.c
 */
1106 1107
int __attribute__ ((weak)) get_cpuid(char *buffer __maybe_unused,
				     size_t sz __maybe_unused)
1108 1109 1110 1111
{
	return -1;
}

1112 1113
static int write_cpuid(int fd, struct perf_header *h __maybe_unused,
		       struct perf_evlist *evlist __maybe_unused)
1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126
{
	char buffer[64];
	int ret;

	ret = get_cpuid(buffer, sizeof(buffer));
	if (!ret)
		goto write_it;

	return -1;
write_it:
	return do_write_string(fd, buffer);
}

1127 1128 1129
static int write_branch_stack(int fd __maybe_unused,
			      struct perf_header *h __maybe_unused,
		       struct perf_evlist *evlist __maybe_unused)
1130 1131 1132 1133
{
	return 0;
}

1134 1135
static void print_hostname(struct perf_header *ph, int fd __maybe_unused,
			   FILE *fp)
1136
{
1137
	fprintf(fp, "# hostname : %s\n", ph->env.hostname);
1138 1139
}

1140 1141
static void print_osrelease(struct perf_header *ph, int fd __maybe_unused,
			    FILE *fp)
1142
{
1143
	fprintf(fp, "# os release : %s\n", ph->env.os_release);
1144 1145
}

1146
static void print_arch(struct perf_header *ph, int fd __maybe_unused, FILE *fp)
1147
{
1148
	fprintf(fp, "# arch : %s\n", ph->env.arch);
1149 1150
}

1151 1152
static void print_cpudesc(struct perf_header *ph, int fd __maybe_unused,
			  FILE *fp)
1153
{
1154
	fprintf(fp, "# cpudesc : %s\n", ph->env.cpu_desc);
1155 1156
}

1157 1158
static void print_nrcpus(struct perf_header *ph, int fd __maybe_unused,
			 FILE *fp)
1159
{
1160 1161
	fprintf(fp, "# nrcpus online : %u\n", ph->env.nr_cpus_online);
	fprintf(fp, "# nrcpus avail : %u\n", ph->env.nr_cpus_avail);
1162 1163
}

1164 1165
static void print_version(struct perf_header *ph, int fd __maybe_unused,
			  FILE *fp)
1166
{
1167
	fprintf(fp, "# perf version : %s\n", ph->env.version);
1168 1169
}

1170 1171
static void print_cmdline(struct perf_header *ph, int fd __maybe_unused,
			  FILE *fp)
1172
{
1173
	int nr, i;
1174 1175
	char *str;

1176 1177
	nr = ph->env.nr_cmdline;
	str = ph->env.cmdline;
1178 1179 1180 1181 1182

	fprintf(fp, "# cmdline : ");

	for (i = 0; i < nr; i++) {
		fprintf(fp, "%s ", str);
1183
		str += strlen(str) + 1;
1184 1185 1186 1187
	}
	fputc('\n', fp);
}

1188 1189
static void print_cpu_topology(struct perf_header *ph, int fd __maybe_unused,
			       FILE *fp)
1190
{
1191
	int nr, i;
1192 1193
	char *str;

1194 1195
	nr = ph->env.nr_sibling_cores;
	str = ph->env.sibling_cores;
1196 1197 1198

	for (i = 0; i < nr; i++) {
		fprintf(fp, "# sibling cores   : %s\n", str);
1199
		str += strlen(str) + 1;
1200 1201
	}

1202 1203
	nr = ph->env.nr_sibling_threads;
	str = ph->env.sibling_threads;
1204 1205 1206

	for (i = 0; i < nr; i++) {
		fprintf(fp, "# sibling threads : %s\n", str);
1207
		str += strlen(str) + 1;
1208 1209 1210
	}
}

1211
static void free_event_desc(struct perf_evsel *events)
1212
{
1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232
	struct perf_evsel *evsel;

	if (!events)
		return;

	for (evsel = events; evsel->attr.size; evsel++) {
		if (evsel->name)
			free(evsel->name);
		if (evsel->id)
			free(evsel->id);
	}

	free(events);
}

static struct perf_evsel *
read_event_desc(struct perf_header *ph, int fd)
{
	struct perf_evsel *evsel, *events = NULL;
	u64 *id;
1233
	void *buf = NULL;
1234 1235 1236
	u32 nre, sz, nr, i, j;
	ssize_t ret;
	size_t msz;
1237 1238

	/* number of events */
1239
	ret = readn(fd, &nre, sizeof(nre));
1240 1241 1242 1243 1244 1245
	if (ret != (ssize_t)sizeof(nre))
		goto error;

	if (ph->needs_swap)
		nre = bswap_32(nre);

1246
	ret = readn(fd, &sz, sizeof(sz));
1247 1248 1249 1250 1251 1252
	if (ret != (ssize_t)sizeof(sz))
		goto error;

	if (ph->needs_swap)
		sz = bswap_32(sz);

1253
	/* buffer to hold on file attr struct */
1254 1255 1256 1257
	buf = malloc(sz);
	if (!buf)
		goto error;

1258 1259 1260 1261 1262 1263
	/* the last event terminates with evsel->attr.size == 0: */
	events = calloc(nre + 1, sizeof(*events));
	if (!events)
		goto error;

	msz = sizeof(evsel->attr);
1264
	if (sz < msz)
1265 1266
		msz = sz;

1267 1268
	for (i = 0, evsel = events; i < nre; evsel++, i++) {
		evsel->idx = i;
1269

1270 1271 1272 1273
		/*
		 * must read entire on-file attr struct to
		 * sync up with layout.
		 */
1274
		ret = readn(fd, buf, sz);
1275 1276 1277 1278 1279 1280
		if (ret != (ssize_t)sz)
			goto error;

		if (ph->needs_swap)
			perf_event__attr_swap(buf);

1281
		memcpy(&evsel->attr, buf, msz);
1282

1283
		ret = readn(fd, &nr, sizeof(nr));
1284 1285 1286
		if (ret != (ssize_t)sizeof(nr))
			goto error;

1287
		if (ph->needs_swap) {
1288
			nr = bswap_32(nr);
1289 1290
			evsel->needs_swap = true;
		}
1291

1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303
		evsel->name = do_read_string(fd, ph);

		if (!nr)
			continue;

		id = calloc(nr, sizeof(*id));
		if (!id)
			goto error;
		evsel->ids = nr;
		evsel->id = id;

		for (j = 0 ; j < nr; j++) {
1304
			ret = readn(fd, id, sizeof(*id));
1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
			if (ret != (ssize_t)sizeof(*id))
				goto error;
			if (ph->needs_swap)
				*id = bswap_64(*id);
			id++;
		}
	}
out:
	if (buf)
		free(buf);
	return events;
error:
	if (events)
		free_event_desc(events);
	events = NULL;
	goto out;
}

static void print_event_desc(struct perf_header *ph, int fd, FILE *fp)
{
	struct perf_evsel *evsel, *events = read_event_desc(ph, fd);
	u32 j;
	u64 *id;

	if (!events) {
		fprintf(fp, "# event desc: not available or unable to read\n");
		return;
	}

	for (evsel = events; evsel->attr.size; evsel++) {
		fprintf(fp, "# event : name = %s, ", evsel->name);
1336 1337 1338

		fprintf(fp, "type = %d, config = 0x%"PRIx64
			    ", config1 = 0x%"PRIx64", config2 = 0x%"PRIx64,
1339 1340 1341 1342
				evsel->attr.type,
				(u64)evsel->attr.config,
				(u64)evsel->attr.config1,
				(u64)evsel->attr.config2);
1343 1344

		fprintf(fp, ", excl_usr = %d, excl_kern = %d",
1345 1346
				evsel->attr.exclude_user,
				evsel->attr.exclude_kernel);
1347

1348
		fprintf(fp, ", excl_host = %d, excl_guest = %d",
1349 1350
				evsel->attr.exclude_host,
				evsel->attr.exclude_guest);
1351

1352
		fprintf(fp, ", precise_ip = %d", evsel->attr.precise_ip);
1353

1354
		if (evsel->ids) {
1355
			fprintf(fp, ", id = {");
1356 1357 1358 1359 1360
			for (j = 0, id = evsel->id; j < evsel->ids; j++, id++) {
				if (j)
					fputc(',', fp);
				fprintf(fp, " %"PRIu64, *id);
			}
1361
			fprintf(fp, " }");
1362 1363
		}

1364 1365
		fputc('\n', fp);
	}
1366 1367

	free_event_desc(events);
1368 1369
}

1370
static void print_total_mem(struct perf_header *ph, int fd __maybe_unused,
1371
			    FILE *fp)
1372
{
1373
	fprintf(fp, "# total memory : %Lu kB\n", ph->env.total_mem);
1374 1375
}

1376
static void print_numa_topology(struct perf_header *ph, int fd __maybe_unused,
1377
				FILE *fp)
1378 1379
{
	u32 nr, c, i;
1380
	char *str, *tmp;
1381 1382 1383
	uint64_t mem_total, mem_free;

	/* nr nodes */
1384 1385
	nr = ph->env.nr_numa_nodes;
	str = ph->env.numa_nodes;
1386 1387 1388

	for (i = 0; i < nr; i++) {
		/* node number */
1389 1390
		c = strtoul(str, &tmp, 0);
		if (*tmp != ':')
1391 1392
			goto error;

1393 1394 1395
		str = tmp + 1;
		mem_total = strtoull(str, &tmp, 0);
		if (*tmp != ':')
1396 1397
			goto error;

1398 1399 1400
		str = tmp + 1;
		mem_free = strtoull(str, &tmp, 0);
		if (*tmp != ':')
1401 1402 1403 1404
			goto error;

		fprintf(fp, "# node%u meminfo  : total = %"PRIu64" kB,"
			    " free = %"PRIu64" kB\n",
1405
			c, mem_total, mem_free);
1406

1407
		str = tmp + 1;
1408
		fprintf(fp, "# node%u cpu list : %s\n", c, str);
1409 1410

		str += strlen(str) + 1;
1411 1412 1413 1414 1415 1416
	}
	return;
error:
	fprintf(fp, "# numa topology : not available\n");
}

1417
static void print_cpuid(struct perf_header *ph, int fd __maybe_unused, FILE *fp)
1418
{
1419
	fprintf(fp, "# cpuid : %s\n", ph->env.cpuid);
1420 1421
}

1422
static void print_branch_stack(struct perf_header *ph __maybe_unused,
1423
			       int fd __maybe_unused, FILE *fp)
1424 1425 1426 1427
{
	fprintf(fp, "# contains samples with branch stack\n");
}

1428 1429
static void print_pmu_mappings(struct perf_header *ph, int fd __maybe_unused,
			       FILE *fp)
1430 1431
{
	const char *delimiter = "# pmu mappings: ";
1432
	char *str, *tmp;
1433 1434 1435
	u32 pmu_num;
	u32 type;

1436
	pmu_num = ph->env.nr_pmu_mappings;
1437 1438 1439 1440 1441
	if (!pmu_num) {
		fprintf(fp, "# pmu mappings: not available\n");
		return;
	}

1442 1443
	str = ph->env.pmu_mappings;

1444
	while (pmu_num) {
1445 1446 1447 1448 1449 1450
		type = strtoul(str, &tmp, 0);
		if (*tmp != ':')
			goto error;

		str = tmp + 1;
		fprintf(fp, "%s%s = %" PRIu32, delimiter, str, type);
1451

1452
		delimiter = ", ";
1453 1454
		str += strlen(str) + 1;
		pmu_num--;
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464
	}

	fprintf(fp, "\n");

	if (!pmu_num)
		return;
error:
	fprintf(fp, "# pmu mappings: unable to read\n");
}

1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
static void print_group_desc(struct perf_header *ph, int fd __maybe_unused,
			     FILE *fp)
{
	struct perf_session *session;
	struct perf_evsel *evsel;
	u32 nr = 0;

	session = container_of(ph, struct perf_session, header);

	list_for_each_entry(evsel, &session->evlist->entries, node) {
		if (perf_evsel__is_group_leader(evsel) &&
		    evsel->nr_members > 1) {
			fprintf(fp, "# group: %s{%s", evsel->group_name ?: "",
				perf_evsel__name(evsel));

			nr = evsel->nr_members - 1;
		} else if (nr) {
			fprintf(fp, ",%s", perf_evsel__name(evsel));

			if (--nr == 0)
				fprintf(fp, "}\n");
		}
	}
}

1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
static int __event_process_build_id(struct build_id_event *bev,
				    char *filename,
				    struct perf_session *session)
{
	int err = -1;
	struct list_head *head;
	struct machine *machine;
	u16 misc;
	struct dso *dso;
	enum dso_kernel_type dso_type;

	machine = perf_session__findnew_machine(session, bev->pid);
	if (!machine)
		goto out;

	misc = bev->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;

	switch (misc) {
	case PERF_RECORD_MISC_KERNEL:
		dso_type = DSO_TYPE_KERNEL;
		head = &machine->kernel_dsos;
		break;
	case PERF_RECORD_MISC_GUEST_KERNEL:
		dso_type = DSO_TYPE_GUEST_KERNEL;
		head = &machine->kernel_dsos;
		break;
	case PERF_RECORD_MISC_USER:
	case PERF_RECORD_MISC_GUEST_USER:
		dso_type = DSO_TYPE_USER;
		head = &machine->user_dsos;
		break;
	default:
		goto out;
	}

	dso = __dsos__findnew(head, filename);
	if (dso != NULL) {
		char sbuild_id[BUILD_ID_SIZE * 2 + 1];

		dso__set_build_id(dso, &bev->build_id);

		if (filename[0] == '[')
			dso->kernel = dso_type;

		build_id__sprintf(dso->build_id, sizeof(dso->build_id),
				  sbuild_id);
		pr_debug("build id event received for %s: %s\n",
			 dso->long_name, sbuild_id);
	}

	err = 0;
out:
	return err;
}

static int perf_header__read_build_ids_abi_quirk(struct perf_header *header,
						 int input, u64 offset, u64 size)
{
	struct perf_session *session = container_of(header, struct perf_session, header);
	struct {
		struct perf_event_header   header;
1551
		u8			   build_id[PERF_ALIGN(BUILD_ID_SIZE, sizeof(u64))];
1552 1553 1554 1555 1556 1557 1558 1559 1560
		char			   filename[0];
	} old_bev;
	struct build_id_event bev;
	char filename[PATH_MAX];
	u64 limit = offset + size;

	while (offset < limit) {
		ssize_t len;

1561
		if (readn(input, &old_bev, sizeof(old_bev)) != sizeof(old_bev))
1562 1563 1564 1565 1566 1567
			return -1;

		if (header->needs_swap)
			perf_event_header__bswap(&old_bev.header);

		len = old_bev.header.size - sizeof(old_bev);
1568
		if (readn(input, filename, len) != len)
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
			return -1;

		bev.header = old_bev.header;

		/*
		 * As the pid is the missing value, we need to fill
		 * it properly. The header.misc value give us nice hint.
		 */
		bev.pid	= HOST_KERNEL_ID;
		if (bev.header.misc == PERF_RECORD_MISC_GUEST_USER ||
		    bev.header.misc == PERF_RECORD_MISC_GUEST_KERNEL)
			bev.pid	= DEFAULT_GUEST_KERNEL_ID;

		memcpy(bev.build_id, old_bev.build_id, sizeof(bev.build_id));
		__event_process_build_id(&bev, filename, session);

		offset += bev.header.size;
	}

	return 0;
}

static int perf_header__read_build_ids(struct perf_header *header,
				       int input, u64 offset, u64 size)
{
	struct perf_session *session = container_of(header, struct perf_session, header);
	struct build_id_event bev;
	char filename[PATH_MAX];
	u64 limit = offset + size, orig_offset = offset;
	int err = -1;

	while (offset < limit) {
		ssize_t len;

1603
		if (readn(input, &bev, sizeof(bev)) != sizeof(bev))
1604 1605 1606 1607 1608 1609
			goto out;

		if (header->needs_swap)
			perf_event_header__bswap(&bev.header);

		len = bev.header.size - sizeof(bev);
1610
		if (readn(input, filename, len) != len)
1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639
			goto out;
		/*
		 * The a1645ce1 changeset:
		 *
		 * "perf: 'perf kvm' tool for monitoring guest performance from host"
		 *
		 * Added a field to struct build_id_event that broke the file
		 * format.
		 *
		 * Since the kernel build-id is the first entry, process the
		 * table using the old format if the well known
		 * '[kernel.kallsyms]' string for the kernel build-id has the
		 * first 4 characters chopped off (where the pid_t sits).
		 */
		if (memcmp(filename, "nel.kallsyms]", 13) == 0) {
			if (lseek(input, orig_offset, SEEK_SET) == (off_t)-1)
				return -1;
			return perf_header__read_build_ids_abi_quirk(header, input, offset, size);
		}

		__event_process_build_id(&bev, filename, session);

		offset += bev.header.size;
	}
	err = 0;
out:
	return err;
}

1640 1641 1642
static int process_tracing_data(struct perf_file_section *section __maybe_unused,
				struct perf_header *ph __maybe_unused,
				int fd, void *data)
1643
{
1644 1645
	ssize_t ret = trace_report(fd, data, false);
	return ret < 0 ? -1 : 0;
1646 1647 1648
}

static int process_build_id(struct perf_file_section *section,
1649
			    struct perf_header *ph, int fd,
1650
			    void *data __maybe_unused)
1651 1652 1653 1654 1655 1656
{
	if (perf_header__read_build_ids(ph, fd, section->offset, section->size))
		pr_debug("Failed to read buildids, continuing...\n");
	return 0;
}

1657
static int process_hostname(struct perf_file_section *section __maybe_unused,
1658 1659
			    struct perf_header *ph, int fd,
			    void *data __maybe_unused)
1660 1661 1662 1663 1664 1665
{
	ph->env.hostname = do_read_string(fd, ph);
	return ph->env.hostname ? 0 : -ENOMEM;
}

static int process_osrelease(struct perf_file_section *section __maybe_unused,
1666 1667
			     struct perf_header *ph, int fd,
			     void *data __maybe_unused)
1668 1669 1670 1671 1672 1673
{
	ph->env.os_release = do_read_string(fd, ph);
	return ph->env.os_release ? 0 : -ENOMEM;
}

static int process_version(struct perf_file_section *section __maybe_unused,
1674 1675
			   struct perf_header *ph, int fd,
			   void *data __maybe_unused)
1676 1677 1678 1679 1680 1681
{
	ph->env.version = do_read_string(fd, ph);
	return ph->env.version ? 0 : -ENOMEM;
}

static int process_arch(struct perf_file_section *section __maybe_unused,
1682 1683
			struct perf_header *ph,	int fd,
			void *data __maybe_unused)
1684 1685 1686 1687 1688 1689
{
	ph->env.arch = do_read_string(fd, ph);
	return ph->env.arch ? 0 : -ENOMEM;
}

static int process_nrcpus(struct perf_file_section *section __maybe_unused,
1690 1691
			  struct perf_header *ph, int fd,
			  void *data __maybe_unused)
1692 1693 1694 1695
{
	size_t ret;
	u32 nr;

1696
	ret = readn(fd, &nr, sizeof(nr));
1697 1698 1699 1700 1701 1702 1703 1704
	if (ret != sizeof(nr))
		return -1;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_cpus_online = nr;

1705
	ret = readn(fd, &nr, sizeof(nr));
1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716
	if (ret != sizeof(nr))
		return -1;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_cpus_avail = nr;
	return 0;
}

static int process_cpudesc(struct perf_file_section *section __maybe_unused,
1717 1718
			   struct perf_header *ph, int fd,
			   void *data __maybe_unused)
1719 1720 1721 1722 1723 1724
{
	ph->env.cpu_desc = do_read_string(fd, ph);
	return ph->env.cpu_desc ? 0 : -ENOMEM;
}

static int process_cpuid(struct perf_file_section *section __maybe_unused,
1725 1726
			 struct perf_header *ph,  int fd,
			 void *data __maybe_unused)
1727 1728 1729 1730 1731 1732
{
	ph->env.cpuid = do_read_string(fd, ph);
	return ph->env.cpuid ? 0 : -ENOMEM;
}

static int process_total_mem(struct perf_file_section *section __maybe_unused,
1733 1734
			     struct perf_header *ph, int fd,
			     void *data __maybe_unused)
1735 1736 1737 1738
{
	uint64_t mem;
	size_t ret;

1739
	ret = readn(fd, &mem, sizeof(mem));
1740 1741 1742 1743 1744 1745 1746 1747 1748 1749
	if (ret != sizeof(mem))
		return -1;

	if (ph->needs_swap)
		mem = bswap_64(mem);

	ph->env.total_mem = mem;
	return 0;
}

1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763
static struct perf_evsel *
perf_evlist__find_by_index(struct perf_evlist *evlist, int idx)
{
	struct perf_evsel *evsel;

	list_for_each_entry(evsel, &evlist->entries, node) {
		if (evsel->idx == idx)
			return evsel;
	}

	return NULL;
}

static void
1764 1765
perf_evlist__set_event_name(struct perf_evlist *evlist,
			    struct perf_evsel *event)
1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782
{
	struct perf_evsel *evsel;

	if (!event->name)
		return;

	evsel = perf_evlist__find_by_index(evlist, event->idx);
	if (!evsel)
		return;

	if (evsel->name)
		return;

	evsel->name = strdup(event->name);
}

static int
1783
process_event_desc(struct perf_file_section *section __maybe_unused,
1784
		   struct perf_header *header, int fd,
1785
		   void *data __maybe_unused)
1786
{
1787
	struct perf_session *session;
1788 1789 1790 1791 1792
	struct perf_evsel *evsel, *events = read_event_desc(header, fd);

	if (!events)
		return 0;

1793
	session = container_of(header, struct perf_session, header);
1794 1795 1796 1797 1798 1799 1800 1801
	for (evsel = events; evsel->attr.size; evsel++)
		perf_evlist__set_event_name(session->evlist, evsel);

	free_event_desc(events);

	return 0;
}

1802
static int process_cmdline(struct perf_file_section *section __maybe_unused,
1803 1804
			   struct perf_header *ph, int fd,
			   void *data __maybe_unused)
1805 1806 1807 1808 1809 1810
{
	size_t ret;
	char *str;
	u32 nr, i;
	struct strbuf sb;

1811
	ret = readn(fd, &nr, sizeof(nr));
1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838
	if (ret != sizeof(nr))
		return -1;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_cmdline = nr;
	strbuf_init(&sb, 128);

	for (i = 0; i < nr; i++) {
		str = do_read_string(fd, ph);
		if (!str)
			goto error;

		/* include a NULL character at the end */
		strbuf_add(&sb, str, strlen(str) + 1);
		free(str);
	}
	ph->env.cmdline = strbuf_detach(&sb, NULL);
	return 0;

error:
	strbuf_release(&sb);
	return -1;
}

static int process_cpu_topology(struct perf_file_section *section __maybe_unused,
1839 1840
				struct perf_header *ph, int fd,
				void *data __maybe_unused)
1841 1842 1843 1844 1845 1846
{
	size_t ret;
	u32 nr, i;
	char *str;
	struct strbuf sb;

1847
	ret = readn(fd, &nr, sizeof(nr));
1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867
	if (ret != sizeof(nr))
		return -1;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_sibling_cores = nr;
	strbuf_init(&sb, 128);

	for (i = 0; i < nr; i++) {
		str = do_read_string(fd, ph);
		if (!str)
			goto error;

		/* include a NULL character at the end */
		strbuf_add(&sb, str, strlen(str) + 1);
		free(str);
	}
	ph->env.sibling_cores = strbuf_detach(&sb, NULL);

1868
	ret = readn(fd, &nr, sizeof(nr));
1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894
	if (ret != sizeof(nr))
		return -1;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_sibling_threads = nr;

	for (i = 0; i < nr; i++) {
		str = do_read_string(fd, ph);
		if (!str)
			goto error;

		/* include a NULL character at the end */
		strbuf_add(&sb, str, strlen(str) + 1);
		free(str);
	}
	ph->env.sibling_threads = strbuf_detach(&sb, NULL);
	return 0;

error:
	strbuf_release(&sb);
	return -1;
}

static int process_numa_topology(struct perf_file_section *section __maybe_unused,
1895 1896
				 struct perf_header *ph, int fd,
				 void *data __maybe_unused)
1897 1898 1899 1900 1901 1902 1903 1904
{
	size_t ret;
	u32 nr, node, i;
	char *str;
	uint64_t mem_total, mem_free;
	struct strbuf sb;

	/* nr nodes */
1905
	ret = readn(fd, &nr, sizeof(nr));
1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916
	if (ret != sizeof(nr))
		goto error;

	if (ph->needs_swap)
		nr = bswap_32(nr);

	ph->env.nr_numa_nodes = nr;
	strbuf_init(&sb, 256);

	for (i = 0; i < nr; i++) {
		/* node number */
1917
		ret = readn(fd, &node, sizeof(node));
1918 1919 1920
		if (ret != sizeof(node))
			goto error;

1921
		ret = readn(fd, &mem_total, sizeof(u64));
1922 1923 1924
		if (ret != sizeof(u64))
			goto error;

1925
		ret = readn(fd, &mem_free, sizeof(u64));
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
		if (ret != sizeof(u64))
			goto error;

		if (ph->needs_swap) {
			node = bswap_32(node);
			mem_total = bswap_64(mem_total);
			mem_free = bswap_64(mem_free);
		}

		strbuf_addf(&sb, "%u:%"PRIu64":%"PRIu64":",
			    node, mem_total, mem_free);

		str = do_read_string(fd, ph);
		if (!str)
			goto error;

		/* include a NULL character at the end */
		strbuf_add(&sb, str, strlen(str) + 1);
		free(str);
	}
	ph->env.numa_nodes = strbuf_detach(&sb, NULL);
	return 0;

error:
	strbuf_release(&sb);
	return -1;
}

static int process_pmu_mappings(struct perf_file_section *section __maybe_unused,
1955 1956
				struct perf_header *ph, int fd,
				void *data __maybe_unused)
1957 1958 1959 1960 1961 1962 1963
{
	size_t ret;
	char *name;
	u32 pmu_num;
	u32 type;
	struct strbuf sb;

1964
	ret = readn(fd, &pmu_num, sizeof(pmu_num));
1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979
	if (ret != sizeof(pmu_num))
		return -1;

	if (ph->needs_swap)
		pmu_num = bswap_32(pmu_num);

	if (!pmu_num) {
		pr_debug("pmu mappings not available\n");
		return 0;
	}

	ph->env.nr_pmu_mappings = pmu_num;
	strbuf_init(&sb, 128);

	while (pmu_num) {
1980
		if (readn(fd, &type, sizeof(type)) != sizeof(type))
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
			goto error;
		if (ph->needs_swap)
			type = bswap_32(type);

		name = do_read_string(fd, ph);
		if (!name)
			goto error;

		strbuf_addf(&sb, "%u:%s", type, name);
		/* include a NULL character at the end */
		strbuf_add(&sb, "", 1);

		free(name);
		pmu_num--;
	}
	ph->env.pmu_mappings = strbuf_detach(&sb, NULL);
	return 0;

error:
	strbuf_release(&sb);
	return -1;
}

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 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 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095
static int process_group_desc(struct perf_file_section *section __maybe_unused,
			      struct perf_header *ph, int fd,
			      void *data __maybe_unused)
{
	size_t ret = -1;
	u32 i, nr, nr_groups;
	struct perf_session *session;
	struct perf_evsel *evsel, *leader = NULL;
	struct group_desc {
		char *name;
		u32 leader_idx;
		u32 nr_members;
	} *desc;

	if (readn(fd, &nr_groups, sizeof(nr_groups)) != sizeof(nr_groups))
		return -1;

	if (ph->needs_swap)
		nr_groups = bswap_32(nr_groups);

	ph->env.nr_groups = nr_groups;
	if (!nr_groups) {
		pr_debug("group desc not available\n");
		return 0;
	}

	desc = calloc(nr_groups, sizeof(*desc));
	if (!desc)
		return -1;

	for (i = 0; i < nr_groups; i++) {
		desc[i].name = do_read_string(fd, ph);
		if (!desc[i].name)
			goto out_free;

		if (readn(fd, &desc[i].leader_idx, sizeof(u32)) != sizeof(u32))
			goto out_free;

		if (readn(fd, &desc[i].nr_members, sizeof(u32)) != sizeof(u32))
			goto out_free;

		if (ph->needs_swap) {
			desc[i].leader_idx = bswap_32(desc[i].leader_idx);
			desc[i].nr_members = bswap_32(desc[i].nr_members);
		}
	}

	/*
	 * Rebuild group relationship based on the group_desc
	 */
	session = container_of(ph, struct perf_session, header);
	session->evlist->nr_groups = nr_groups;

	i = nr = 0;
	list_for_each_entry(evsel, &session->evlist->entries, node) {
		if (evsel->idx == (int) desc[i].leader_idx) {
			evsel->leader = evsel;
			/* {anon_group} is a dummy name */
			if (strcmp(desc[i].name, "{anon_group}"))
				evsel->group_name = desc[i].name;
			evsel->nr_members = desc[i].nr_members;

			if (i >= nr_groups || nr > 0) {
				pr_debug("invalid group desc\n");
				goto out_free;
			}

			leader = evsel;
			nr = evsel->nr_members - 1;
			i++;
		} else if (nr) {
			/* This is a group member */
			evsel->leader = leader;

			nr--;
		}
	}

	if (i != nr_groups || nr != 0) {
		pr_debug("invalid group desc\n");
		goto out_free;
	}

	ret = 0;
out_free:
	while ((int) --i >= 0)
		free(desc[i].name);
	free(desc);

	return ret;
}

2096 2097 2098
struct feature_ops {
	int (*write)(int fd, struct perf_header *h, struct perf_evlist *evlist);
	void (*print)(struct perf_header *h, int fd, FILE *fp);
2099
	int (*process)(struct perf_file_section *section,
2100
		       struct perf_header *h, int fd, void *data);
2101 2102 2103 2104
	const char *name;
	bool full_only;
};

2105 2106
#define FEAT_OPA(n, func) \
	[n] = { .name = #n, .write = write_##func, .print = print_##func }
2107 2108 2109
#define FEAT_OPP(n, func) \
	[n] = { .name = #n, .write = write_##func, .print = print_##func, \
		.process = process_##func }
2110
#define FEAT_OPF(n, func) \
2111
	[n] = { .name = #n, .write = write_##func, .print = print_##func, \
2112
		.process = process_##func, .full_only = true }
2113 2114

/* feature_ops not implemented: */
2115 2116
#define print_tracing_data	NULL
#define print_build_id		NULL
2117 2118

static const struct feature_ops feat_ops[HEADER_LAST_FEATURE] = {
2119
	FEAT_OPP(HEADER_TRACING_DATA,	tracing_data),
2120
	FEAT_OPP(HEADER_BUILD_ID,	build_id),
2121 2122 2123 2124 2125 2126
	FEAT_OPP(HEADER_HOSTNAME,	hostname),
	FEAT_OPP(HEADER_OSRELEASE,	osrelease),
	FEAT_OPP(HEADER_VERSION,	version),
	FEAT_OPP(HEADER_ARCH,		arch),
	FEAT_OPP(HEADER_NRCPUS,		nrcpus),
	FEAT_OPP(HEADER_CPUDESC,	cpudesc),
2127
	FEAT_OPP(HEADER_CPUID,		cpuid),
2128
	FEAT_OPP(HEADER_TOTAL_MEM,	total_mem),
2129
	FEAT_OPP(HEADER_EVENT_DESC,	event_desc),
2130
	FEAT_OPP(HEADER_CMDLINE,	cmdline),
2131 2132
	FEAT_OPF(HEADER_CPU_TOPOLOGY,	cpu_topology),
	FEAT_OPF(HEADER_NUMA_TOPOLOGY,	numa_topology),
2133
	FEAT_OPA(HEADER_BRANCH_STACK,	branch_stack),
2134
	FEAT_OPP(HEADER_PMU_MAPPINGS,	pmu_mappings),
2135
	FEAT_OPP(HEADER_GROUP_DESC,	group_desc),
2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153
};

struct header_print_data {
	FILE *fp;
	bool full; /* extended list of headers */
};

static int perf_file_section__fprintf_info(struct perf_file_section *section,
					   struct perf_header *ph,
					   int feat, int fd, void *data)
{
	struct header_print_data *hd = data;

	if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) {
		pr_debug("Failed to lseek to %" PRIu64 " offset for feature "
				"%d, continuing...\n", section->offset, feat);
		return 0;
	}
2154
	if (feat >= HEADER_LAST_FEATURE) {
2155
		pr_warning("unknown feature %d\n", feat);
2156
		return 0;
2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190
	}
	if (!feat_ops[feat].print)
		return 0;

	if (!feat_ops[feat].full_only || hd->full)
		feat_ops[feat].print(ph, fd, hd->fp);
	else
		fprintf(hd->fp, "# %s info available, use -I to display\n",
			feat_ops[feat].name);

	return 0;
}

int perf_header__fprintf_info(struct perf_session *session, FILE *fp, bool full)
{
	struct header_print_data hd;
	struct perf_header *header = &session->header;
	int fd = session->fd;
	hd.fp = fp;
	hd.full = full;

	perf_header__process_sections(header, fd, &hd,
				      perf_file_section__fprintf_info);
	return 0;
}

static int do_write_feat(int fd, struct perf_header *h, int type,
			 struct perf_file_section **p,
			 struct perf_evlist *evlist)
{
	int err;
	int ret = 0;

	if (perf_header__has_feat(h, type)) {
2191 2192
		if (!feat_ops[type].write)
			return -1;
2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210

		(*p)->offset = lseek(fd, 0, SEEK_CUR);

		err = feat_ops[type].write(fd, h, evlist);
		if (err < 0) {
			pr_debug("failed to write feature %d\n", type);

			/* undo anything written */
			lseek(fd, (*p)->offset, SEEK_SET);

			return -1;
		}
		(*p)->size = lseek(fd, 0, SEEK_CUR) - (*p)->offset;
		(*p)++;
	}
	return ret;
}

2211
static int perf_header__adds_write(struct perf_header *header,
2212
				   struct perf_evlist *evlist, int fd)
2213
{
2214
	int nr_sections;
2215
	struct perf_file_section *feat_sec, *p;
2216 2217
	int sec_size;
	u64 sec_start;
2218
	int feat;
2219
	int err;
2220

2221
	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
2222
	if (!nr_sections)
2223
		return 0;
2224

2225
	feat_sec = p = calloc(nr_sections, sizeof(*feat_sec));
2226 2227
	if (feat_sec == NULL)
		return -ENOMEM;
2228 2229 2230

	sec_size = sizeof(*feat_sec) * nr_sections;

2231
	sec_start = header->feat_offset;
2232
	lseek(fd, sec_start + sec_size, SEEK_SET);
2233

2234 2235 2236 2237
	for_each_set_bit(feat, header->adds_features, HEADER_FEAT_BITS) {
		if (do_write_feat(fd, header, feat, &p, evlist))
			perf_header__clear_feat(header, feat);
	}
2238

2239
	lseek(fd, sec_start, SEEK_SET);
2240 2241 2242 2243
	/*
	 * may write more than needed due to dropped feature, but
	 * this is okay, reader will skip the mising entries
	 */
2244 2245 2246
	err = do_write(fd, feat_sec, sec_size);
	if (err < 0)
		pr_debug("failed to write feature section\n");
2247
	free(feat_sec);
2248
	return err;
2249
}
2250

2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269
int perf_header__write_pipe(int fd)
{
	struct perf_pipe_file_header f_header;
	int err;

	f_header = (struct perf_pipe_file_header){
		.magic	   = PERF_MAGIC,
		.size	   = sizeof(f_header),
	};

	err = do_write(fd, &f_header, sizeof(f_header));
	if (err < 0) {
		pr_debug("failed to write perf pipe header\n");
		return err;
	}

	return 0;
}

2270 2271 2272
int perf_session__write_header(struct perf_session *session,
			       struct perf_evlist *evlist,
			       int fd, bool at_exit)
2273 2274 2275
{
	struct perf_file_header f_header;
	struct perf_file_attr   f_attr;
2276
	struct perf_header *header = &session->header;
2277
	struct perf_evsel *evsel;
2278
	u64 attr_offset;
2279
	int err;
2280 2281 2282

	lseek(fd, sizeof(f_header), SEEK_SET);

2283 2284 2285
	list_for_each_entry(evsel, &evlist->entries, node) {
		evsel->id_offset = lseek(fd, 0, SEEK_CUR);
		err = do_write(fd, evsel->id, evsel->ids * sizeof(u64));
2286 2287 2288 2289
		if (err < 0) {
			pr_debug("failed to write perf header\n");
			return err;
		}
2290 2291
	}

2292
	attr_offset = lseek(fd, 0, SEEK_CUR);
2293

2294
	list_for_each_entry(evsel, &evlist->entries, node) {
2295
		f_attr = (struct perf_file_attr){
2296
			.attr = evsel->attr,
2297
			.ids  = {
2298 2299
				.offset = evsel->id_offset,
				.size   = evsel->ids * sizeof(u64),
2300 2301
			}
		};
2302 2303 2304 2305 2306
		err = do_write(fd, &f_attr, sizeof(f_attr));
		if (err < 0) {
			pr_debug("failed to write perf header attribute\n");
			return err;
		}
2307 2308
	}

2309
	header->data_offset = lseek(fd, 0, SEEK_CUR);
2310
	header->feat_offset = header->data_offset + header->data_size;
2311

2312
	if (at_exit) {
2313
		err = perf_header__adds_write(header, evlist, fd);
2314 2315 2316
		if (err < 0)
			return err;
	}
2317

2318 2319 2320 2321 2322
	f_header = (struct perf_file_header){
		.magic	   = PERF_MAGIC,
		.size	   = sizeof(f_header),
		.attr_size = sizeof(f_attr),
		.attrs = {
2323
			.offset = attr_offset,
2324
			.size   = evlist->nr_entries * sizeof(f_attr),
2325 2326
		},
		.data = {
2327 2328
			.offset = header->data_offset,
			.size	= header->data_size,
2329
		},
2330
		/* event_types is ignored, store zeros */
2331 2332
	};

2333
	memcpy(&f_header.adds_features, &header->adds_features, sizeof(header->adds_features));
2334

2335
	lseek(fd, 0, SEEK_SET);
2336 2337 2338 2339 2340
	err = do_write(fd, &f_header, sizeof(f_header));
	if (err < 0) {
		pr_debug("failed to write perf header\n");
		return err;
	}
2341
	lseek(fd, header->data_offset + header->data_size, SEEK_SET);
2342

2343
	return 0;
2344 2345
}

2346
static int perf_header__getbuffer64(struct perf_header *header,
2347 2348
				    int fd, void *buf, size_t size)
{
2349
	if (readn(fd, buf, size) <= 0)
2350 2351
		return -1;

2352
	if (header->needs_swap)
2353 2354 2355 2356 2357
		mem_bswap_64(buf, size);

	return 0;
}

2358
int perf_header__process_sections(struct perf_header *header, int fd,
2359
				  void *data,
2360
				  int (*process)(struct perf_file_section *section,
2361 2362
						 struct perf_header *ph,
						 int feat, int fd, void *data))
2363
{
2364
	struct perf_file_section *feat_sec, *sec;
2365 2366
	int nr_sections;
	int sec_size;
2367 2368
	int feat;
	int err;
2369

2370
	nr_sections = bitmap_weight(header->adds_features, HEADER_FEAT_BITS);
2371
	if (!nr_sections)
2372
		return 0;
2373

2374
	feat_sec = sec = calloc(nr_sections, sizeof(*feat_sec));
2375
	if (!feat_sec)
2376
		return -1;
2377 2378 2379

	sec_size = sizeof(*feat_sec) * nr_sections;

2380
	lseek(fd, header->feat_offset, SEEK_SET);
2381

2382 2383
	err = perf_header__getbuffer64(header, fd, feat_sec, sec_size);
	if (err < 0)
2384
		goto out_free;
2385

2386 2387 2388 2389
	for_each_set_bit(feat, header->adds_features, HEADER_LAST_FEATURE) {
		err = process(sec++, header, feat, fd, data);
		if (err < 0)
			goto out_free;
2390
	}
2391
	err = 0;
2392
out_free:
2393 2394
	free(feat_sec);
	return err;
2395
}
2396

2397 2398 2399
static const int attr_file_abi_sizes[] = {
	[0] = PERF_ATTR_SIZE_VER0,
	[1] = PERF_ATTR_SIZE_VER1,
2400
	[2] = PERF_ATTR_SIZE_VER2,
2401
	[3] = PERF_ATTR_SIZE_VER3,
2402 2403 2404 2405 2406 2407 2408 2409 2410 2411
	0,
};

/*
 * In the legacy file format, the magic number is not used to encode endianness.
 * hdr_sz was used to encode endianness. But given that hdr_sz can vary based
 * on ABI revisions, we need to try all combinations for all endianness to
 * detect the endianness.
 */
static int try_all_file_abis(uint64_t hdr_sz, struct perf_header *ph)
2412
{
2413 2414
	uint64_t ref_size, attr_size;
	int i;
2415

2416 2417 2418 2419 2420 2421 2422
	for (i = 0 ; attr_file_abi_sizes[i]; i++) {
		ref_size = attr_file_abi_sizes[i]
			 + sizeof(struct perf_file_section);
		if (hdr_sz != ref_size) {
			attr_size = bswap_64(hdr_sz);
			if (attr_size != ref_size)
				continue;
2423

2424 2425 2426 2427 2428 2429 2430 2431 2432 2433
			ph->needs_swap = true;
		}
		pr_debug("ABI%d perf.data file detected, need_swap=%d\n",
			 i,
			 ph->needs_swap);
		return 0;
	}
	/* could not determine endianness */
	return -1;
}
2434

2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458
#define PERF_PIPE_HDR_VER0	16

static const size_t attr_pipe_abi_sizes[] = {
	[0] = PERF_PIPE_HDR_VER0,
	0,
};

/*
 * In the legacy pipe format, there is an implicit assumption that endiannesss
 * between host recording the samples, and host parsing the samples is the
 * same. This is not always the case given that the pipe output may always be
 * redirected into a file and analyzed on a different machine with possibly a
 * different endianness and perf_event ABI revsions in the perf tool itself.
 */
static int try_all_pipe_abis(uint64_t hdr_sz, struct perf_header *ph)
{
	u64 attr_size;
	int i;

	for (i = 0 ; attr_pipe_abi_sizes[i]; i++) {
		if (hdr_sz != attr_pipe_abi_sizes[i]) {
			attr_size = bswap_64(hdr_sz);
			if (attr_size != hdr_sz)
				continue;
2459 2460 2461

			ph->needs_swap = true;
		}
2462
		pr_debug("Pipe ABI%d perf.data file detected\n", i);
2463 2464
		return 0;
	}
2465 2466 2467
	return -1;
}

2468 2469 2470 2471 2472 2473 2474 2475 2476 2477
bool is_perf_magic(u64 magic)
{
	if (!memcmp(&magic, __perf_magic1, sizeof(magic))
		|| magic == __perf_magic2
		|| magic == __perf_magic2_sw)
		return true;

	return false;
}

2478 2479 2480 2481 2482 2483 2484 2485
static int check_magic_endian(u64 magic, uint64_t hdr_sz,
			      bool is_pipe, struct perf_header *ph)
{
	int ret;

	/* check for legacy format */
	ret = memcmp(&magic, __perf_magic1, sizeof(magic));
	if (ret == 0) {
2486
		ph->version = PERF_HEADER_VERSION_1;
2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497
		pr_debug("legacy perf.data format\n");
		if (is_pipe)
			return try_all_pipe_abis(hdr_sz, ph);

		return try_all_file_abis(hdr_sz, ph);
	}
	/*
	 * the new magic number serves two purposes:
	 * - unique number to identify actual perf.data files
	 * - encode endianness of file
	 */
2498

2499 2500
	/* check magic number with one endianness */
	if (magic == __perf_magic2)
2501 2502
		return 0;

2503 2504
	/* check magic number with opposite endianness */
	if (magic != __perf_magic2_sw)
2505 2506 2507
		return -1;

	ph->needs_swap = true;
2508
	ph->version = PERF_HEADER_VERSION_2;
2509 2510 2511 2512

	return 0;
}

2513
int perf_file_header__read(struct perf_file_header *header,
2514 2515
			   struct perf_header *ph, int fd)
{
2516 2517
	int ret;

2518 2519
	lseek(fd, 0, SEEK_SET);

2520 2521
	ret = readn(fd, header, sizeof(*header));
	if (ret <= 0)
2522 2523
		return -1;

2524 2525 2526
	if (check_magic_endian(header->magic,
			       header->attr_size, false, ph) < 0) {
		pr_debug("magic/endian check failed\n");
2527
		return -1;
2528
	}
2529

2530
	if (ph->needs_swap) {
2531
		mem_bswap_64(header, offsetof(struct perf_file_header,
2532
			     adds_features));
2533 2534
	}

2535
	if (header->size != sizeof(*header)) {
2536
		/* Support the previous format */
2537 2538
		if (header->size == offsetof(typeof(*header), adds_features))
			bitmap_zero(header->adds_features, HEADER_FEAT_BITS);
2539 2540
		else
			return -1;
2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556
	} else if (ph->needs_swap) {
		/*
		 * feature bitmap is declared as an array of unsigned longs --
		 * not good since its size can differ between the host that
		 * generated the data file and the host analyzing the file.
		 *
		 * We need to handle endianness, but we don't know the size of
		 * the unsigned long where the file was generated. Take a best
		 * guess at determining it: try 64-bit swap first (ie., file
		 * created on a 64-bit host), and check if the hostname feature
		 * bit is set (this feature bit is forced on as of fbe96f2).
		 * If the bit is not, undo the 64-bit swap and try a 32-bit
		 * swap. If the hostname bit is still not set (e.g., older data
		 * file), punt and fallback to the original behavior --
		 * clearing all feature bits and setting buildid.
		 */
2557 2558
		mem_bswap_64(&header->adds_features,
			    BITS_TO_U64(HEADER_FEAT_BITS));
2559 2560

		if (!test_bit(HEADER_HOSTNAME, header->adds_features)) {
2561 2562 2563 2564 2565 2566 2567
			/* unswap as u64 */
			mem_bswap_64(&header->adds_features,
				    BITS_TO_U64(HEADER_FEAT_BITS));

			/* unswap as u32 */
			mem_bswap_32(&header->adds_features,
				    BITS_TO_U32(HEADER_FEAT_BITS));
2568 2569 2570 2571 2572 2573
		}

		if (!test_bit(HEADER_HOSTNAME, header->adds_features)) {
			bitmap_zero(header->adds_features, HEADER_FEAT_BITS);
			set_bit(HEADER_BUILD_ID, header->adds_features);
		}
2574
	}
2575

2576
	memcpy(&ph->adds_features, &header->adds_features,
2577
	       sizeof(ph->adds_features));
2578

2579 2580
	ph->data_offset  = header->data.offset;
	ph->data_size	 = header->data.size;
2581
	ph->feat_offset  = header->data.offset + header->data.size;
2582 2583 2584
	return 0;
}

2585
static int perf_file_section__process(struct perf_file_section *section,
2586
				      struct perf_header *ph,
2587
				      int feat, int fd, void *data)
2588
{
2589
	if (lseek(fd, section->offset, SEEK_SET) == (off_t)-1) {
2590
		pr_debug("Failed to lseek to %" PRIu64 " offset for feature "
2591
			  "%d, continuing...\n", section->offset, feat);
2592 2593 2594
		return 0;
	}

2595 2596 2597 2598 2599
	if (feat >= HEADER_LAST_FEATURE) {
		pr_debug("unknown feature %d, continuing...\n", feat);
		return 0;
	}

2600 2601
	if (!feat_ops[feat].process)
		return 0;
2602

2603
	return feat_ops[feat].process(section, ph, fd, data);
2604
}
2605

2606
static int perf_file_header__read_pipe(struct perf_pipe_file_header *header,
Tom Zanussi's avatar
Tom Zanussi committed
2607 2608
				       struct perf_header *ph, int fd,
				       bool repipe)
2609
{
2610 2611 2612 2613 2614 2615
	int ret;

	ret = readn(fd, header, sizeof(*header));
	if (ret <= 0)
		return -1;

2616 2617
	if (check_magic_endian(header->magic, header->size, true, ph) < 0) {
		pr_debug("endian/magic failed\n");
2618
		return -1;
2619 2620 2621 2622
	}

	if (ph->needs_swap)
		header->size = bswap_64(header->size);
2623

2624
	if (repipe && do_write(STDOUT_FILENO, header, sizeof(*header)) < 0)
Tom Zanussi's avatar
Tom Zanussi committed
2625 2626
		return -1;

2627 2628 2629
	return 0;
}

2630
static int perf_header__read_pipe(struct perf_session *session)
2631
{
2632
	struct perf_header *header = &session->header;
2633 2634
	struct perf_pipe_file_header f_header;

2635
	if (perf_file_header__read_pipe(&f_header, header, session->fd,
Tom Zanussi's avatar
Tom Zanussi committed
2636
					session->repipe) < 0) {
2637 2638 2639 2640 2641 2642 2643
		pr_debug("incompatible file format\n");
		return -EINVAL;
	}

	return 0;
}

2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663
static int read_attr(int fd, struct perf_header *ph,
		     struct perf_file_attr *f_attr)
{
	struct perf_event_attr *attr = &f_attr->attr;
	size_t sz, left;
	size_t our_sz = sizeof(f_attr->attr);
	int ret;

	memset(f_attr, 0, sizeof(*f_attr));

	/* read minimal guaranteed structure */
	ret = readn(fd, attr, PERF_ATTR_SIZE_VER0);
	if (ret <= 0) {
		pr_debug("cannot read %d bytes of header attr\n",
			 PERF_ATTR_SIZE_VER0);
		return -1;
	}

	/* on file perf_event_attr size */
	sz = attr->size;
2664

2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689
	if (ph->needs_swap)
		sz = bswap_32(sz);

	if (sz == 0) {
		/* assume ABI0 */
		sz =  PERF_ATTR_SIZE_VER0;
	} else if (sz > our_sz) {
		pr_debug("file uses a more recent and unsupported ABI"
			 " (%zu bytes extra)\n", sz - our_sz);
		return -1;
	}
	/* what we have not yet read and that we know about */
	left = sz - PERF_ATTR_SIZE_VER0;
	if (left) {
		void *ptr = attr;
		ptr += PERF_ATTR_SIZE_VER0;

		ret = readn(fd, ptr, left);
	}
	/* read perf_file_section, ids are read in caller */
	ret = readn(fd, &f_attr->ids, sizeof(f_attr->ids));

	return ret <= 0 ? -1 : 0;
}

2690 2691
static int perf_evsel__prepare_tracepoint_event(struct perf_evsel *evsel,
						struct pevent *pevent)
2692
{
2693
	struct event_format *event;
2694 2695
	char bf[128];

2696 2697 2698 2699
	/* already prepared */
	if (evsel->tp_format)
		return 0;

2700 2701 2702 2703 2704
	if (pevent == NULL) {
		pr_debug("broken or missing trace data\n");
		return -1;
	}

2705
	event = pevent_find_event(pevent, evsel->attr.config);
2706 2707 2708
	if (event == NULL)
		return -1;

2709 2710 2711 2712 2713 2714
	if (!evsel->name) {
		snprintf(bf, sizeof(bf), "%s:%s", event->system, event->name);
		evsel->name = strdup(bf);
		if (evsel->name == NULL)
			return -1;
	}
2715

2716
	evsel->tp_format = event;
2717 2718 2719
	return 0;
}

2720 2721
static int perf_evlist__prepare_tracepoint_events(struct perf_evlist *evlist,
						  struct pevent *pevent)
2722 2723 2724 2725
{
	struct perf_evsel *pos;

	list_for_each_entry(pos, &evlist->entries, node) {
2726 2727
		if (pos->attr.type == PERF_TYPE_TRACEPOINT &&
		    perf_evsel__prepare_tracepoint_event(pos, pevent))
2728 2729 2730 2731 2732 2733
			return -1;
	}

	return 0;
}

2734
int perf_session__read_header(struct perf_session *session)
2735
{
2736
	struct perf_header *header = &session->header;
2737
	struct perf_file_header	f_header;
2738 2739 2740
	struct perf_file_attr	f_attr;
	u64			f_id;
	int nr_attrs, nr_ids, i, j;
2741
	int fd = session->fd;
2742

2743
	session->evlist = perf_evlist__new();
2744 2745 2746
	if (session->evlist == NULL)
		return -ENOMEM;

2747
	if (session->fd_pipe)
2748
		return perf_header__read_pipe(session);
2749

2750
	if (perf_file_header__read(&f_header, header, fd) < 0)
2751
		return -EINVAL;
2752

2753
	nr_attrs = f_header.attrs.size / f_header.attr_size;
2754 2755 2756
	lseek(fd, f_header.attrs.offset, SEEK_SET);

	for (i = 0; i < nr_attrs; i++) {
2757
		struct perf_evsel *evsel;
2758
		off_t tmp;
2759

2760
		if (read_attr(fd, header, &f_attr) < 0)
2761
			goto out_errno;
2762

2763 2764 2765
		if (header->needs_swap)
			perf_event__attr_swap(&f_attr.attr);

2766
		tmp = lseek(fd, 0, SEEK_CUR);
2767
		evsel = perf_evsel__new(&f_attr.attr, i);
2768

2769 2770
		if (evsel == NULL)
			goto out_delete_evlist;
2771 2772

		evsel->needs_swap = header->needs_swap;
2773 2774 2775 2776 2777
		/*
		 * Do it before so that if perf_evsel__alloc_id fails, this
		 * entry gets purged too at perf_evlist__delete().
		 */
		perf_evlist__add(session->evlist, evsel);
2778 2779

		nr_ids = f_attr.ids.size / sizeof(u64);
2780 2781 2782 2783 2784 2785 2786 2787
		/*
		 * We don't have the cpu and thread maps on the header, so
		 * for allocating the perf_sample_id table we fake 1 cpu and
		 * hattr->ids threads.
		 */
		if (perf_evsel__alloc_id(evsel, 1, nr_ids))
			goto out_delete_evlist;

2788 2789 2790
		lseek(fd, f_attr.ids.offset, SEEK_SET);

		for (j = 0; j < nr_ids; j++) {
2791
			if (perf_header__getbuffer64(header, fd, &f_id, sizeof(f_id)))
2792
				goto out_errno;
2793

2794
			perf_evlist__id_add(session->evlist, evsel, 0, j, f_id);
2795
		}
2796

2797 2798 2799
		lseek(fd, tmp, SEEK_SET);
	}

2800 2801
	symbol_conf.nr_events = nr_attrs;

2802
	perf_header__process_sections(header, fd, &session->pevent,
2803
				      perf_file_section__process);
2804

2805 2806
	if (perf_evlist__prepare_tracepoint_events(session->evlist,
						   session->pevent))
2807 2808
		goto out_delete_evlist;

2809
	return 0;
2810 2811
out_errno:
	return -errno;
2812 2813 2814 2815 2816

out_delete_evlist:
	perf_evlist__delete(session->evlist);
	session->evlist = NULL;
	return -ENOMEM;
2817
}
2818

2819
int perf_event__synthesize_attr(struct perf_tool *tool,
2820
				struct perf_event_attr *attr, u32 ids, u64 *id,
2821
				perf_event__handler_t process)
2822
{
2823
	union perf_event *ev;
2824 2825 2826 2827
	size_t size;
	int err;

	size = sizeof(struct perf_event_attr);
2828
	size = PERF_ALIGN(size, sizeof(u64));
2829 2830 2831 2832 2833
	size += sizeof(struct perf_event_header);
	size += ids * sizeof(u64);

	ev = malloc(size);

2834 2835 2836
	if (ev == NULL)
		return -ENOMEM;

2837 2838 2839 2840
	ev->attr.attr = *attr;
	memcpy(ev->attr.id, id, ids * sizeof(u64));

	ev->attr.header.type = PERF_RECORD_HEADER_ATTR;
2841
	ev->attr.header.size = (u16)size;
2842

2843 2844 2845 2846
	if (ev->attr.header.size == size)
		err = process(tool, ev, NULL, NULL);
	else
		err = -E2BIG;
2847 2848 2849 2850 2851 2852

	free(ev);

	return err;
}

2853
int perf_event__synthesize_attrs(struct perf_tool *tool,
2854
				   struct perf_session *session,
2855
				   perf_event__handler_t process)
2856
{
2857
	struct perf_evsel *evsel;
2858
	int err = 0;
2859

2860 2861 2862
	list_for_each_entry(evsel, &session->evlist->entries, node) {
		err = perf_event__synthesize_attr(tool, &evsel->attr, evsel->ids,
						  evsel->id, process);
2863 2864 2865 2866 2867 2868 2869 2870 2871
		if (err) {
			pr_debug("failed to create perf header attribute\n");
			return err;
		}
	}

	return err;
}

2872 2873
int perf_event__process_attr(struct perf_tool *tool __maybe_unused,
			     union perf_event *event,
2874
			     struct perf_evlist **pevlist)
2875
{
2876
	u32 i, ids, n_ids;
2877
	struct perf_evsel *evsel;
2878
	struct perf_evlist *evlist = *pevlist;
2879

2880
	if (evlist == NULL) {
2881
		*pevlist = evlist = perf_evlist__new();
2882
		if (evlist == NULL)
2883 2884 2885
			return -ENOMEM;
	}

2886
	evsel = perf_evsel__new(&event->attr.attr, evlist->nr_entries);
2887
	if (evsel == NULL)
2888 2889
		return -ENOMEM;

2890
	perf_evlist__add(evlist, evsel);
2891

2892 2893
	ids = event->header.size;
	ids -= (void *)&event->attr.id - (void *)event;
2894
	n_ids = ids / sizeof(u64);
2895 2896 2897 2898 2899 2900 2901
	/*
	 * We don't have the cpu and thread maps on the header, so
	 * for allocating the perf_sample_id table we fake 1 cpu and
	 * hattr->ids threads.
	 */
	if (perf_evsel__alloc_id(evsel, 1, n_ids))
		return -ENOMEM;
2902 2903

	for (i = 0; i < n_ids; i++) {
2904
		perf_evlist__id_add(evlist, evsel, 0, i, event->attr.id[i]);
2905 2906
	}

2907 2908
	symbol_conf.nr_events = evlist->nr_entries;

2909 2910
	return 0;
}
2911

2912
int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd,
2913
					struct perf_evlist *evlist,
2914
					perf_event__handler_t process)
2915
{
2916
	union perf_event ev;
2917
	struct tracing_data *tdata;
2918
	ssize_t size = 0, aligned_size = 0, padding;
2919
	int err __maybe_unused = 0;
2920

2921 2922 2923 2924 2925 2926 2927 2928 2929 2930 2931 2932 2933 2934 2935
	/*
	 * We are going to store the size of the data followed
	 * by the data contents. Since the fd descriptor is a pipe,
	 * we cannot seek back to store the size of the data once
	 * we know it. Instead we:
	 *
	 * - write the tracing data to the temp file
	 * - get/write the data size to pipe
	 * - write the tracing data from the temp file
	 *   to the pipe
	 */
	tdata = tracing_data_get(&evlist->entries, fd, true);
	if (!tdata)
		return -1;

2936 2937 2938
	memset(&ev, 0, sizeof(ev));

	ev.tracing_data.header.type = PERF_RECORD_HEADER_TRACING_DATA;
2939
	size = tdata->size;
2940
	aligned_size = PERF_ALIGN(size, sizeof(u64));
2941 2942 2943 2944
	padding = aligned_size - size;
	ev.tracing_data.header.size = sizeof(ev.tracing_data);
	ev.tracing_data.size = aligned_size;

2945
	process(tool, &ev, NULL, NULL);
2946

2947 2948 2949 2950 2951 2952
	/*
	 * The put function will copy all the tracing data
	 * stored in temp file to the pipe.
	 */
	tracing_data_put(tdata);

2953 2954 2955 2956 2957
	write_padded(fd, NULL, 0, padding);

	return aligned_size;
}

2958 2959
int perf_event__process_tracing_data(struct perf_tool *tool __maybe_unused,
				     union perf_event *event,
2960
				     struct perf_session *session)
2961
{
2962
	ssize_t size_read, padding, size = event->tracing_data.size;
2963 2964 2965 2966 2967 2968 2969
	off_t offset = lseek(session->fd, 0, SEEK_CUR);
	char buf[BUFSIZ];

	/* setup for reading amidst mmap */
	lseek(session->fd, offset + sizeof(struct tracing_data_event),
	      SEEK_SET);

2970 2971
	size_read = trace_report(session->fd, &session->pevent,
				 session->repipe);
2972
	padding = PERF_ALIGN(size_read, sizeof(u64)) - size_read;
2973

2974 2975 2976 2977
	if (readn(session->fd, buf, padding) < 0) {
		pr_err("%s: reading input file", __func__);
		return -1;
	}
Tom Zanussi's avatar
Tom Zanussi committed
2978 2979
	if (session->repipe) {
		int retw = write(STDOUT_FILENO, buf, padding);
2980 2981 2982 2983
		if (retw <= 0 || retw != padding) {
			pr_err("%s: repiping tracing data padding", __func__);
			return -1;
		}
Tom Zanussi's avatar
Tom Zanussi committed
2984
	}
2985

2986 2987 2988 2989
	if (size_read + padding != size) {
		pr_err("%s: tracing data size mismatch", __func__);
		return -1;
	}
2990

2991 2992
	perf_evlist__prepare_tracepoint_events(session->evlist,
					       session->pevent);
2993

2994 2995
	return size_read + padding;
}
2996

2997
int perf_event__synthesize_build_id(struct perf_tool *tool,
2998
				    struct dso *pos, u16 misc,
2999
				    perf_event__handler_t process,
3000
				    struct machine *machine)
3001
{
3002
	union perf_event ev;
3003 3004 3005 3006 3007 3008 3009 3010 3011
	size_t len;
	int err = 0;

	if (!pos->hit)
		return err;

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

	len = pos->long_name_len + 1;
3012
	len = PERF_ALIGN(len, NAME_ALIGN);
3013 3014 3015
	memcpy(&ev.build_id.build_id, pos->build_id, sizeof(pos->build_id));
	ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID;
	ev.build_id.header.misc = misc;
3016
	ev.build_id.pid = machine->pid;
3017 3018 3019
	ev.build_id.header.size = sizeof(ev.build_id) + len;
	memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len);

3020
	err = process(tool, &ev, NULL, machine);
3021 3022 3023 3024

	return err;
}

3025
int perf_event__process_build_id(struct perf_tool *tool __maybe_unused,
3026
				 union perf_event *event,
3027
				 struct perf_session *session)
3028
{
3029 3030
	__event_process_build_id(&event->build_id,
				 event->build_id.filename,
3031
				 session);
3032 3033
	return 0;
}
3034 3035 3036 3037 3038

void disable_buildid_cache(void)
{
	no_buildid_cache = true;
}