machine.c 34.3 KB
Newer Older
1
#include "callchain.h"
2 3
#include "debug.h"
#include "event.h"
4 5
#include "evsel.h"
#include "hist.h"
6 7
#include "machine.h"
#include "map.h"
8
#include "sort.h"
9
#include "strlist.h"
10 11
#include "thread.h"
#include <stdbool.h>
12
#include <symbol/kallsyms.h>
13
#include "unwind.h"
14

15 16 17 18 19 20 21 22 23 24 25 26 27 28
int machine__init(struct machine *machine, const char *root_dir, pid_t pid)
{
	map_groups__init(&machine->kmaps);
	RB_CLEAR_NODE(&machine->rb_node);
	INIT_LIST_HEAD(&machine->user_dsos);
	INIT_LIST_HEAD(&machine->kernel_dsos);

	machine->threads = RB_ROOT;
	INIT_LIST_HEAD(&machine->dead_threads);
	machine->last_match = NULL;

	machine->kmaps.machine = machine;
	machine->pid = pid;

29
	machine->symbol_filter = NULL;
30
	machine->id_hdr_size = 0;
31

32 33 34 35 36
	machine->root_dir = strdup(root_dir);
	if (machine->root_dir == NULL)
		return -ENOMEM;

	if (pid != HOST_KERNEL_ID) {
37 38
		struct thread *thread = machine__findnew_thread(machine, 0,
								pid);
39 40 41 42 43 44
		char comm[64];

		if (thread == NULL)
			return -ENOMEM;

		snprintf(comm, sizeof(comm), "[guest/%d]", pid);
45
		thread__set_comm(thread, comm, 0);
46 47 48 49 50
	}

	return 0;
}

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
struct machine *machine__new_host(void)
{
	struct machine *machine = malloc(sizeof(*machine));

	if (machine != NULL) {
		machine__init(machine, "", HOST_KERNEL_ID);

		if (machine__create_kernel_maps(machine) < 0)
			goto out_delete;
	}

	return machine;
out_delete:
	free(machine);
	return NULL;
}

68 69 70 71 72 73 74 75 76 77
static void dsos__delete(struct list_head *dsos)
{
	struct dso *pos, *n;

	list_for_each_entry_safe(pos, n, dsos, node) {
		list_del(&pos->node);
		dso__delete(pos);
	}
}

78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
void machine__delete_dead_threads(struct machine *machine)
{
	struct thread *n, *t;

	list_for_each_entry_safe(t, n, &machine->dead_threads, node) {
		list_del(&t->node);
		thread__delete(t);
	}
}

void machine__delete_threads(struct machine *machine)
{
	struct rb_node *nd = rb_first(&machine->threads);

	while (nd) {
		struct thread *t = rb_entry(nd, struct thread, rb_node);

		rb_erase(&t->rb_node, &machine->threads);
		nd = rb_next(nd);
		thread__delete(t);
	}
}

101 102 103 104 105
void machine__exit(struct machine *machine)
{
	map_groups__exit(&machine->kmaps);
	dsos__delete(&machine->user_dsos);
	dsos__delete(&machine->kernel_dsos);
106
	zfree(&machine->root_dir);
107 108 109 110 111 112 113 114
}

void machine__delete(struct machine *machine)
{
	machine__exit(machine);
	free(machine);
}

115 116 117 118
void machines__init(struct machines *machines)
{
	machine__init(&machines->host, "", HOST_KERNEL_ID);
	machines->guests = RB_ROOT;
119
	machines->symbol_filter = NULL;
120 121 122 123 124 125 126 127 128
}

void machines__exit(struct machines *machines)
{
	machine__exit(&machines->host);
	/* XXX exit guest */
}

struct machine *machines__add(struct machines *machines, pid_t pid,
129 130
			      const char *root_dir)
{
131
	struct rb_node **p = &machines->guests.rb_node;
132 133 134 135 136 137 138 139 140 141 142
	struct rb_node *parent = NULL;
	struct machine *pos, *machine = malloc(sizeof(*machine));

	if (machine == NULL)
		return NULL;

	if (machine__init(machine, root_dir, pid) != 0) {
		free(machine);
		return NULL;
	}

143 144
	machine->symbol_filter = machines->symbol_filter;

145 146 147 148 149 150 151 152 153 154
	while (*p != NULL) {
		parent = *p;
		pos = rb_entry(parent, struct machine, rb_node);
		if (pid < pos->pid)
			p = &(*p)->rb_left;
		else
			p = &(*p)->rb_right;
	}

	rb_link_node(&machine->rb_node, parent, p);
155
	rb_insert_color(&machine->rb_node, &machines->guests);
156 157 158 159

	return machine;
}

160 161 162 163 164 165 166 167 168 169 170 171 172 173 174
void machines__set_symbol_filter(struct machines *machines,
				 symbol_filter_t symbol_filter)
{
	struct rb_node *nd;

	machines->symbol_filter = symbol_filter;
	machines->host.symbol_filter = symbol_filter;

	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
		struct machine *machine = rb_entry(nd, struct machine, rb_node);

		machine->symbol_filter = symbol_filter;
	}
}

175
struct machine *machines__find(struct machines *machines, pid_t pid)
176
{
177
	struct rb_node **p = &machines->guests.rb_node;
178 179 180 181
	struct rb_node *parent = NULL;
	struct machine *machine;
	struct machine *default_machine = NULL;

182 183 184
	if (pid == HOST_KERNEL_ID)
		return &machines->host;

185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
	while (*p != NULL) {
		parent = *p;
		machine = rb_entry(parent, struct machine, rb_node);
		if (pid < machine->pid)
			p = &(*p)->rb_left;
		else if (pid > machine->pid)
			p = &(*p)->rb_right;
		else
			return machine;
		if (!machine->pid)
			default_machine = machine;
	}

	return default_machine;
}

201
struct machine *machines__findnew(struct machines *machines, pid_t pid)
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
{
	char path[PATH_MAX];
	const char *root_dir = "";
	struct machine *machine = machines__find(machines, pid);

	if (machine && (machine->pid == pid))
		goto out;

	if ((pid != HOST_KERNEL_ID) &&
	    (pid != DEFAULT_GUEST_KERNEL_ID) &&
	    (symbol_conf.guestmount)) {
		sprintf(path, "%s/%d", symbol_conf.guestmount, pid);
		if (access(path, R_OK)) {
			static struct strlist *seen;

			if (!seen)
				seen = strlist__new(true, NULL);

			if (!strlist__has_entry(seen, path)) {
				pr_err("Can't access file %s\n", path);
				strlist__add(seen, path);
			}
			machine = NULL;
			goto out;
		}
		root_dir = path;
	}

	machine = machines__add(machines, pid, root_dir);
out:
	return machine;
}

235 236
void machines__process_guests(struct machines *machines,
			      machine__process_t process, void *data)
237 238 239
{
	struct rb_node *nd;

240
	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		process(pos, data);
	}
}

char *machine__mmap_name(struct machine *machine, char *bf, size_t size)
{
	if (machine__is_host(machine))
		snprintf(bf, size, "[%s]", "kernel.kallsyms");
	else if (machine__is_default_guest(machine))
		snprintf(bf, size, "[%s]", "guest.kernel.kallsyms");
	else {
		snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms",
			 machine->pid);
	}

	return bf;
}

260
void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size)
261 262 263 264
{
	struct rb_node *node;
	struct machine *machine;

265 266 267
	machines->host.id_hdr_size = id_hdr_size;

	for (node = rb_first(&machines->guests); node; node = rb_next(node)) {
268 269 270 271 272 273 274
		machine = rb_entry(node, struct machine, rb_node);
		machine->id_hdr_size = id_hdr_size;
	}

	return;
}

275 276
static struct thread *__machine__findnew_thread(struct machine *machine,
						pid_t pid, pid_t tid,
277 278 279 280 281 282 283
						bool create)
{
	struct rb_node **p = &machine->threads.rb_node;
	struct rb_node *parent = NULL;
	struct thread *th;

	/*
284
	 * Front-end cache - TID lookups come in blocks,
285 286 287
	 * so most of the time we dont have to look up
	 * the full rbtree:
	 */
288 289 290
	if (machine->last_match && machine->last_match->tid == tid) {
		if (pid && pid != machine->last_match->pid_)
			machine->last_match->pid_ = pid;
291
		return machine->last_match;
292
	}
293 294 295 296 297

	while (*p != NULL) {
		parent = *p;
		th = rb_entry(parent, struct thread, rb_node);

298
		if (th->tid == tid) {
299
			machine->last_match = th;
300 301
			if (pid && pid != th->pid_)
				th->pid_ = pid;
302 303 304
			return th;
		}

305
		if (tid < th->tid)
306 307 308 309 310 311 312 313
			p = &(*p)->rb_left;
		else
			p = &(*p)->rb_right;
	}

	if (!create)
		return NULL;

314
	th = thread__new(pid, tid);
315 316 317 318
	if (th != NULL) {
		rb_link_node(&th->rb_node, parent, p);
		rb_insert_color(&th->rb_node, &machine->threads);
		machine->last_match = th;
319 320 321 322 323 324 325 326 327 328 329

		/*
		 * We have to initialize map_groups separately
		 * after rb tree is updated.
		 *
		 * The reason is that we call machine__findnew_thread
		 * within thread__init_map_groups to find the thread
		 * leader and that would screwed the rb tree.
		 */
		if (thread__init_map_groups(th, machine))
			return NULL;
330 331 332 333 334
	}

	return th;
}

335 336
struct thread *machine__findnew_thread(struct machine *machine, pid_t pid,
				       pid_t tid)
337
{
338
	return __machine__findnew_thread(machine, pid, tid, true);
339 340
}

341 342
struct thread *machine__find_thread(struct machine *machine, pid_t pid,
				    pid_t tid)
343
{
344
	return __machine__findnew_thread(machine, pid, tid, false);
345
}
346

347 348
int machine__process_comm_event(struct machine *machine, union perf_event *event,
				struct perf_sample *sample)
349
{
350 351 352
	struct thread *thread = machine__findnew_thread(machine,
							event->comm.pid,
							event->comm.tid);
353 354 355 356

	if (dump_trace)
		perf_event__fprintf_comm(event, stdout);

357
	if (thread == NULL || thread__set_comm(thread, event->comm.comm, sample->time)) {
358 359 360 361 362 363 364 365
		dump_printf("problem processing PERF_RECORD_COMM, skipping event.\n");
		return -1;
	}

	return 0;
}

int machine__process_lost_event(struct machine *machine __maybe_unused,
366
				union perf_event *event, struct perf_sample *sample __maybe_unused)
367 368 369 370 371 372
{
	dump_printf(": id:%" PRIu64 ": lost:%" PRIu64 "\n",
		    event->lost.id, event->lost.lost);
	return 0;
}

373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393
struct map *machine__new_module(struct machine *machine, u64 start,
				const char *filename)
{
	struct map *map;
	struct dso *dso = __dsos__findnew(&machine->kernel_dsos, filename);

	if (dso == NULL)
		return NULL;

	map = map__new2(start, dso, MAP__FUNCTION);
	if (map == NULL)
		return NULL;

	if (machine__is_host(machine))
		dso->symtab_type = DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE;
	else
		dso->symtab_type = DSO_BINARY_TYPE__GUEST_KMODULE;
	map_groups__insert(&machine->kmaps, map);
	return map;
}

394
size_t machines__fprintf_dsos(struct machines *machines, FILE *fp)
395 396
{
	struct rb_node *nd;
397 398
	size_t ret = __dsos__fprintf(&machines->host.kernel_dsos, fp) +
		     __dsos__fprintf(&machines->host.user_dsos, fp);
399

400
	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		ret += __dsos__fprintf(&pos->kernel_dsos, fp);
		ret += __dsos__fprintf(&pos->user_dsos, fp);
	}

	return ret;
}

size_t machine__fprintf_dsos_buildid(struct machine *machine, FILE *fp,
				     bool (skip)(struct dso *dso, int parm), int parm)
{
	return __dsos__fprintf_buildid(&machine->kernel_dsos, fp, skip, parm) +
	       __dsos__fprintf_buildid(&machine->user_dsos, fp, skip, parm);
}

416
size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp,
417 418 419
				     bool (skip)(struct dso *dso, int parm), int parm)
{
	struct rb_node *nd;
420
	size_t ret = machine__fprintf_dsos_buildid(&machines->host, fp, skip, parm);
421

422
	for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) {
423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498
		struct machine *pos = rb_entry(nd, struct machine, rb_node);
		ret += machine__fprintf_dsos_buildid(pos, fp, skip, parm);
	}
	return ret;
}

size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp)
{
	int i;
	size_t printed = 0;
	struct dso *kdso = machine->vmlinux_maps[MAP__FUNCTION]->dso;

	if (kdso->has_build_id) {
		char filename[PATH_MAX];
		if (dso__build_id_filename(kdso, filename, sizeof(filename)))
			printed += fprintf(fp, "[0] %s\n", filename);
	}

	for (i = 0; i < vmlinux_path__nr_entries; ++i)
		printed += fprintf(fp, "[%d] %s\n",
				   i + kdso->has_build_id, vmlinux_path[i]);

	return printed;
}

size_t machine__fprintf(struct machine *machine, FILE *fp)
{
	size_t ret = 0;
	struct rb_node *nd;

	for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) {
		struct thread *pos = rb_entry(nd, struct thread, rb_node);

		ret += thread__fprintf(pos, fp);
	}

	return ret;
}

static struct dso *machine__get_kernel(struct machine *machine)
{
	const char *vmlinux_name = NULL;
	struct dso *kernel;

	if (machine__is_host(machine)) {
		vmlinux_name = symbol_conf.vmlinux_name;
		if (!vmlinux_name)
			vmlinux_name = "[kernel.kallsyms]";

		kernel = dso__kernel_findnew(machine, vmlinux_name,
					     "[kernel]",
					     DSO_TYPE_KERNEL);
	} else {
		char bf[PATH_MAX];

		if (machine__is_default_guest(machine))
			vmlinux_name = symbol_conf.default_guest_vmlinux_name;
		if (!vmlinux_name)
			vmlinux_name = machine__mmap_name(machine, bf,
							  sizeof(bf));

		kernel = dso__kernel_findnew(machine, vmlinux_name,
					     "[guest.kernel]",
					     DSO_TYPE_GUEST_KERNEL);
	}

	if (kernel != NULL && (!kernel->has_build_id))
		dso__read_running_kernel_build_id(kernel, machine);

	return kernel;
}

struct process_args {
	u64 start;
};

499 500 501 502 503 504 505 506 507
static void machine__get_kallsyms_filename(struct machine *machine, char *buf,
					   size_t bufsz)
{
	if (machine__is_default_guest(machine))
		scnprintf(buf, bufsz, "%s", symbol_conf.default_guest_kallsyms);
	else
		scnprintf(buf, bufsz, "%s/proc/kallsyms", machine->root_dir);
}

508 509 510 511 512 513 514 515
const char *ref_reloc_sym_names[] = {"_text", "_stext", NULL};

/* Figure out the start address of kernel map from /proc/kallsyms.
 * Returns the name of the start symbol in *symbol_name. Pass in NULL as
 * symbol_name if it's not that important.
 */
static u64 machine__get_kernel_start_addr(struct machine *machine,
					  const char **symbol_name)
516
{
517
	char filename[PATH_MAX];
518 519 520
	int i;
	const char *name;
	u64 addr = 0;
521

522
	machine__get_kallsyms_filename(machine, filename, PATH_MAX);
523 524 525 526

	if (symbol__restricted_filename(filename, "/proc/kallsyms"))
		return 0;

527 528 529 530 531 532 533 534
	for (i = 0; (name = ref_reloc_sym_names[i]) != NULL; i++) {
		addr = kallsyms__get_function_start(filename, name);
		if (addr)
			break;
	}

	if (symbol_name)
		*symbol_name = name;
535

536
	return addr;
537 538 539 540 541
}

int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel)
{
	enum map_type type;
542
	u64 start = machine__get_kernel_start_addr(machine, NULL);
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581

	for (type = 0; type < MAP__NR_TYPES; ++type) {
		struct kmap *kmap;

		machine->vmlinux_maps[type] = map__new2(start, kernel, type);
		if (machine->vmlinux_maps[type] == NULL)
			return -1;

		machine->vmlinux_maps[type]->map_ip =
			machine->vmlinux_maps[type]->unmap_ip =
				identity__map_ip;
		kmap = map__kmap(machine->vmlinux_maps[type]);
		kmap->kmaps = &machine->kmaps;
		map_groups__insert(&machine->kmaps,
				   machine->vmlinux_maps[type]);
	}

	return 0;
}

void machine__destroy_kernel_maps(struct machine *machine)
{
	enum map_type type;

	for (type = 0; type < MAP__NR_TYPES; ++type) {
		struct kmap *kmap;

		if (machine->vmlinux_maps[type] == NULL)
			continue;

		kmap = map__kmap(machine->vmlinux_maps[type]);
		map_groups__remove(&machine->kmaps,
				   machine->vmlinux_maps[type]);
		if (kmap->ref_reloc_sym) {
			/*
			 * ref_reloc_sym is shared among all maps, so free just
			 * on one of them.
			 */
			if (type == MAP__FUNCTION) {
582 583 584 585
				zfree((char **)&kmap->ref_reloc_sym->name);
				zfree(&kmap->ref_reloc_sym);
			} else
				kmap->ref_reloc_sym = NULL;
586 587 588 589 590 591 592
		}

		map__delete(machine->vmlinux_maps[type]);
		machine->vmlinux_maps[type] = NULL;
	}
}

593
int machines__create_guest_kernel_maps(struct machines *machines)
594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641
{
	int ret = 0;
	struct dirent **namelist = NULL;
	int i, items = 0;
	char path[PATH_MAX];
	pid_t pid;
	char *endp;

	if (symbol_conf.default_guest_vmlinux_name ||
	    symbol_conf.default_guest_modules ||
	    symbol_conf.default_guest_kallsyms) {
		machines__create_kernel_maps(machines, DEFAULT_GUEST_KERNEL_ID);
	}

	if (symbol_conf.guestmount) {
		items = scandir(symbol_conf.guestmount, &namelist, NULL, NULL);
		if (items <= 0)
			return -ENOENT;
		for (i = 0; i < items; i++) {
			if (!isdigit(namelist[i]->d_name[0])) {
				/* Filter out . and .. */
				continue;
			}
			pid = (pid_t)strtol(namelist[i]->d_name, &endp, 10);
			if ((*endp != '\0') ||
			    (endp == namelist[i]->d_name) ||
			    (errno == ERANGE)) {
				pr_debug("invalid directory (%s). Skipping.\n",
					 namelist[i]->d_name);
				continue;
			}
			sprintf(path, "%s/%s/proc/kallsyms",
				symbol_conf.guestmount,
				namelist[i]->d_name);
			ret = access(path, R_OK);
			if (ret) {
				pr_debug("Can't access file %s\n", path);
				goto failure;
			}
			machines__create_kernel_maps(machines, pid);
		}
failure:
		free(namelist);
	}

	return ret;
}

642
void machines__destroy_kernel_maps(struct machines *machines)
643
{
644 645 646
	struct rb_node *next = rb_first(&machines->guests);

	machine__destroy_kernel_maps(&machines->host);
647 648 649 650 651

	while (next) {
		struct machine *pos = rb_entry(next, struct machine, rb_node);

		next = rb_next(&pos->rb_node);
652
		rb_erase(&pos->rb_node, &machines->guests);
653 654 655 656
		machine__delete(pos);
	}
}

657
int machines__create_kernel_maps(struct machines *machines, pid_t pid)
658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691
{
	struct machine *machine = machines__findnew(machines, pid);

	if (machine == NULL)
		return -1;

	return machine__create_kernel_maps(machine);
}

int machine__load_kallsyms(struct machine *machine, const char *filename,
			   enum map_type type, symbol_filter_t filter)
{
	struct map *map = machine->vmlinux_maps[type];
	int ret = dso__load_kallsyms(map->dso, filename, map, filter);

	if (ret > 0) {
		dso__set_loaded(map->dso, type);
		/*
		 * Since /proc/kallsyms will have multiple sessions for the
		 * kernel, with modules between them, fixup the end of all
		 * sections.
		 */
		__map_groups__fixup_end(&machine->kmaps, type);
	}

	return ret;
}

int machine__load_vmlinux_path(struct machine *machine, enum map_type type,
			       symbol_filter_t filter)
{
	struct map *map = machine->vmlinux_maps[type];
	int ret = dso__load_vmlinux_path(map->dso, map, filter);

692
	if (ret > 0)
693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
		dso__set_loaded(map->dso, type);

	return ret;
}

static void map_groups__fixup_end(struct map_groups *mg)
{
	int i;
	for (i = 0; i < MAP__NR_TYPES; ++i)
		__map_groups__fixup_end(mg, i);
}

static char *get_kernel_version(const char *root_dir)
{
	char version[PATH_MAX];
	FILE *file;
	char *name, *tmp;
	const char *prefix = "Linux version ";

	sprintf(version, "%s/proc/version", root_dir);
	file = fopen(version, "r");
	if (!file)
		return NULL;

	version[0] = '\0';
	tmp = fgets(version, sizeof(version), file);
	fclose(file);

	name = strstr(version, prefix);
	if (!name)
		return NULL;
	name += strlen(prefix);
	tmp = strchr(name, ' ');
	if (tmp)
		*tmp = '\0';

	return strdup(name);
}

static int map_groups__set_modules_path_dir(struct map_groups *mg,
733
				const char *dir_name, int depth)
734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
{
	struct dirent *dent;
	DIR *dir = opendir(dir_name);
	int ret = 0;

	if (!dir) {
		pr_debug("%s: cannot open %s dir\n", __func__, dir_name);
		return -1;
	}

	while ((dent = readdir(dir)) != NULL) {
		char path[PATH_MAX];
		struct stat st;

		/*sshfs might return bad dent->d_type, so we have to stat*/
		snprintf(path, sizeof(path), "%s/%s", dir_name, dent->d_name);
		if (stat(path, &st))
			continue;

		if (S_ISDIR(st.st_mode)) {
			if (!strcmp(dent->d_name, ".") ||
			    !strcmp(dent->d_name, ".."))
				continue;

758 759 760 761 762 763 764 765 766
			/* Do not follow top-level source and build symlinks */
			if (depth == 0) {
				if (!strcmp(dent->d_name, "source") ||
				    !strcmp(dent->d_name, "build"))
					continue;
			}

			ret = map_groups__set_modules_path_dir(mg, path,
							       depth + 1);
767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790
			if (ret < 0)
				goto out;
		} else {
			char *dot = strrchr(dent->d_name, '.'),
			     dso_name[PATH_MAX];
			struct map *map;
			char *long_name;

			if (dot == NULL || strcmp(dot, ".ko"))
				continue;
			snprintf(dso_name, sizeof(dso_name), "[%.*s]",
				 (int)(dot - dent->d_name), dent->d_name);

			strxfrchar(dso_name, '-', '_');
			map = map_groups__find_by_name(mg, MAP__FUNCTION,
						       dso_name);
			if (map == NULL)
				continue;

			long_name = strdup(path);
			if (long_name == NULL) {
				ret = -1;
				goto out;
			}
791
			dso__set_long_name(map->dso, long_name, true);
792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809
			dso__kernel_module_get_build_id(map->dso, "");
		}
	}

out:
	closedir(dir);
	return ret;
}

static int machine__set_modules_path(struct machine *machine)
{
	char *version;
	char modules_path[PATH_MAX];

	version = get_kernel_version(machine->root_dir);
	if (!version)
		return -1;

810
	snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s",
811 812 813
		 machine->root_dir, version);
	free(version);

814
	return map_groups__set_modules_path_dir(&machine->kmaps, modules_path, 0);
815 816
}

817
static int machine__create_module(void *arg, const char *name, u64 start)
818
{
819
	struct machine *machine = arg;
820
	struct map *map;
821 822 823 824 825 826 827 828 829 830 831 832

	map = machine__new_module(machine, start, name);
	if (map == NULL)
		return -1;

	dso__kernel_module_get_build_id(map->dso, machine->root_dir);

	return 0;
}

static int machine__create_modules(struct machine *machine)
{
833 834 835
	const char *modules;
	char path[PATH_MAX];

836
	if (machine__is_default_guest(machine)) {
837
		modules = symbol_conf.default_guest_modules;
838 839
	} else {
		snprintf(path, PATH_MAX, "%s/proc/modules", machine->root_dir);
840 841 842
		modules = path;
	}

843
	if (symbol__restricted_filename(modules, "/proc/modules"))
844 845
		return -1;

846
	if (modules__parse(modules, machine, machine__create_module))
847 848
		return -1;

849 850
	if (!machine__set_modules_path(machine))
		return 0;
851

852
	pr_debug("Problems setting modules path maps, continuing anyway...\n");
853

854
	return 0;
855 856 857 858 859
}

int machine__create_kernel_maps(struct machine *machine)
{
	struct dso *kernel = machine__get_kernel(machine);
860
	const char *name;
861
	u64 addr = machine__get_kernel_start_addr(machine, &name);
862 863
	if (!addr)
		return -1;
864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881

	if (kernel == NULL ||
	    __machine__create_kernel_maps(machine, kernel) < 0)
		return -1;

	if (symbol_conf.use_modules && machine__create_modules(machine) < 0) {
		if (machine__is_host(machine))
			pr_debug("Problems creating module maps, "
				 "continuing anyway...\n");
		else
			pr_debug("Problems creating module maps for guest %d, "
				 "continuing anyway...\n", machine->pid);
	}

	/*
	 * Now that we have all the maps created, just set the ->end of them:
	 */
	map_groups__fixup_end(&machine->kmaps);
882 883 884 885 886 887 888

	if (maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, name,
					     addr)) {
		machine__destroy_kernel_maps(machine);
		return -1;
	}

889 890 891
	return 0;
}

892 893 894
static void machine__set_kernel_mmap_len(struct machine *machine,
					 union perf_event *event)
{
895 896 897 898 899 900 901 902 903 904 905 906 907
	int i;

	for (i = 0; i < MAP__NR_TYPES; i++) {
		machine->vmlinux_maps[i]->start = event->mmap.start;
		machine->vmlinux_maps[i]->end   = (event->mmap.start +
						   event->mmap.len);
		/*
		 * Be a bit paranoid here, some perf.data file came with
		 * a zero sized synthesized MMAP event for the kernel.
		 */
		if (machine->vmlinux_maps[i]->end == 0)
			machine->vmlinux_maps[i]->end = ~0ULL;
	}
908 909
}

910 911 912 913 914 915 916 917 918 919 920 921
static bool machine__uses_kcore(struct machine *machine)
{
	struct dso *dso;

	list_for_each_entry(dso, &machine->kernel_dsos, node) {
		if (dso__is_kcore(dso))
			return true;
	}

	return false;
}

922 923 924 925 926 927 928 929
static int machine__process_kernel_mmap_event(struct machine *machine,
					      union perf_event *event)
{
	struct map *map;
	char kmmap_prefix[PATH_MAX];
	enum dso_kernel_type kernel_type;
	bool is_kernel_mmap;

930 931 932 933
	/* If we have maps from kcore then we do not need or want any others */
	if (machine__uses_kcore(machine))
		return 0;

934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
	machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix));
	if (machine__is_host(machine))
		kernel_type = DSO_TYPE_KERNEL;
	else
		kernel_type = DSO_TYPE_GUEST_KERNEL;

	is_kernel_mmap = memcmp(event->mmap.filename,
				kmmap_prefix,
				strlen(kmmap_prefix) - 1) == 0;
	if (event->mmap.filename[0] == '/' ||
	    (!is_kernel_mmap && event->mmap.filename[0] == '[')) {

		char short_module_name[1024];
		char *name, *dot;

		if (event->mmap.filename[0] == '/') {
			name = strrchr(event->mmap.filename, '/');
			if (name == NULL)
				goto out_problem;

			++name; /* skip / */
			dot = strrchr(name, '.');
			if (dot == NULL)
				goto out_problem;
			snprintf(short_module_name, sizeof(short_module_name),
					"[%.*s]", (int)(dot - name), name);
			strxfrchar(short_module_name, '-', '_');
		} else
			strcpy(short_module_name, event->mmap.filename);

		map = machine__new_module(machine, event->mmap.start,
					  event->mmap.filename);
		if (map == NULL)
			goto out_problem;

		name = strdup(short_module_name);
		if (name == NULL)
			goto out_problem;

973
		dso__set_short_name(map->dso, name, true);
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 1008 1009 1010 1011 1012 1013 1014 1015 1016
		map->end = map->start + event->mmap.len;
	} else if (is_kernel_mmap) {
		const char *symbol_name = (event->mmap.filename +
				strlen(kmmap_prefix));
		/*
		 * Should be there already, from the build-id table in
		 * the header.
		 */
		struct dso *kernel = __dsos__findnew(&machine->kernel_dsos,
						     kmmap_prefix);
		if (kernel == NULL)
			goto out_problem;

		kernel->kernel = kernel_type;
		if (__machine__create_kernel_maps(machine, kernel) < 0)
			goto out_problem;

		machine__set_kernel_mmap_len(machine, event);

		/*
		 * Avoid using a zero address (kptr_restrict) for the ref reloc
		 * symbol. Effectively having zero here means that at record
		 * time /proc/sys/kernel/kptr_restrict was non zero.
		 */
		if (event->mmap.pgoff != 0) {
			maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps,
							 symbol_name,
							 event->mmap.pgoff);
		}

		if (machine__is_default_guest(machine)) {
			/*
			 * preload dso of guest kernel and modules
			 */
			dso__load(kernel, machine->vmlinux_maps[MAP__FUNCTION],
				  NULL);
		}
	}
	return 0;
out_problem:
	return -1;
}

1017
int machine__process_mmap2_event(struct machine *machine,
1018 1019
				 union perf_event *event,
				 struct perf_sample *sample __maybe_unused)
1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038
{
	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
	struct thread *thread;
	struct map *map;
	enum map_type type;
	int ret = 0;

	if (dump_trace)
		perf_event__fprintf_mmap2(event, stdout);

	if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
	    cpumode == PERF_RECORD_MISC_KERNEL) {
		ret = machine__process_kernel_mmap_event(machine, event);
		if (ret < 0)
			goto out_problem;
		return 0;
	}

	thread = machine__findnew_thread(machine, event->mmap2.pid,
1039
					event->mmap2.tid);
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052
	if (thread == NULL)
		goto out_problem;

	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
		type = MAP__VARIABLE;
	else
		type = MAP__FUNCTION;

	map = map__new(&machine->user_dsos, event->mmap2.start,
			event->mmap2.len, event->mmap2.pgoff,
			event->mmap2.pid, event->mmap2.maj,
			event->mmap2.min, event->mmap2.ino,
			event->mmap2.ino_generation,
1053 1054
			event->mmap2.prot,
			event->mmap2.flags,
1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067
			event->mmap2.filename, type);

	if (map == NULL)
		goto out_problem;

	thread__insert_map(thread, map);
	return 0;

out_problem:
	dump_printf("problem processing PERF_RECORD_MMAP2, skipping event.\n");
	return 0;
}

1068 1069
int machine__process_mmap_event(struct machine *machine, union perf_event *event,
				struct perf_sample *sample __maybe_unused)
1070 1071 1072 1073
{
	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
	struct thread *thread;
	struct map *map;
1074
	enum map_type type;
1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
	int ret = 0;

	if (dump_trace)
		perf_event__fprintf_mmap(event, stdout);

	if (cpumode == PERF_RECORD_MISC_GUEST_KERNEL ||
	    cpumode == PERF_RECORD_MISC_KERNEL) {
		ret = machine__process_kernel_mmap_event(machine, event);
		if (ret < 0)
			goto out_problem;
		return 0;
	}

1088
	thread = machine__findnew_thread(machine, event->mmap.pid,
1089
					 event->mmap.tid);
1090 1091
	if (thread == NULL)
		goto out_problem;
1092 1093 1094 1095 1096 1097

	if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA)
		type = MAP__VARIABLE;
	else
		type = MAP__FUNCTION;

1098 1099
	map = map__new(&machine->user_dsos, event->mmap.start,
			event->mmap.len, event->mmap.pgoff,
1100
			event->mmap.pid, 0, 0, 0, 0, 0, 0,
1101
			event->mmap.filename,
1102 1103
			type);

1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114
	if (map == NULL)
		goto out_problem;

	thread__insert_map(thread, map);
	return 0;

out_problem:
	dump_printf("problem processing PERF_RECORD_MMAP, skipping event.\n");
	return 0;
}

1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125
static void machine__remove_thread(struct machine *machine, struct thread *th)
{
	machine->last_match = NULL;
	rb_erase(&th->rb_node, &machine->threads);
	/*
	 * We may have references to this thread, for instance in some hist_entry
	 * instances, so just move them to a separate list.
	 */
	list_add_tail(&th->node, &machine->dead_threads);
}

1126 1127
int machine__process_fork_event(struct machine *machine, union perf_event *event,
				struct perf_sample *sample)
1128
{
1129 1130 1131
	struct thread *thread = machine__find_thread(machine,
						     event->fork.pid,
						     event->fork.tid);
1132 1133 1134
	struct thread *parent = machine__findnew_thread(machine,
							event->fork.ppid,
							event->fork.ptid);
1135

1136 1137 1138 1139
	/* if a thread currently exists for the thread id remove it */
	if (thread != NULL)
		machine__remove_thread(machine, thread);

1140 1141
	thread = machine__findnew_thread(machine, event->fork.pid,
					 event->fork.tid);
1142 1143 1144 1145
	if (dump_trace)
		perf_event__fprintf_task(event, stdout);

	if (thread == NULL || parent == NULL ||
1146
	    thread__fork(thread, parent, sample->time) < 0) {
1147 1148 1149 1150 1151 1152 1153
		dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n");
		return -1;
	}

	return 0;
}

1154 1155
int machine__process_exit_event(struct machine *machine, union perf_event *event,
				struct perf_sample *sample __maybe_unused)
1156
{
1157 1158 1159
	struct thread *thread = machine__find_thread(machine,
						     event->fork.pid,
						     event->fork.tid);
1160 1161 1162 1163 1164

	if (dump_trace)
		perf_event__fprintf_task(event, stdout);

	if (thread != NULL)
1165
		thread__exited(thread);
1166 1167 1168 1169

	return 0;
}

1170 1171
int machine__process_event(struct machine *machine, union perf_event *event,
			   struct perf_sample *sample)
1172 1173 1174 1175 1176
{
	int ret;

	switch (event->header.type) {
	case PERF_RECORD_COMM:
1177
		ret = machine__process_comm_event(machine, event, sample); break;
1178
	case PERF_RECORD_MMAP:
1179
		ret = machine__process_mmap_event(machine, event, sample); break;
1180
	case PERF_RECORD_MMAP2:
1181
		ret = machine__process_mmap2_event(machine, event, sample); break;
1182
	case PERF_RECORD_FORK:
1183
		ret = machine__process_fork_event(machine, event, sample); break;
1184
	case PERF_RECORD_EXIT:
1185
		ret = machine__process_exit_event(machine, event, sample); break;
1186
	case PERF_RECORD_LOST:
1187
		ret = machine__process_lost_event(machine, event, sample); break;
1188 1189 1190 1191 1192 1193 1194
	default:
		ret = -1;
		break;
	}

	return ret;
}
1195

1196
static bool symbol__match_regex(struct symbol *sym, regex_t *regex)
1197
{
1198
	if (sym->name && !regexec(regex, sym->name, 0, NULL, 0))
1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209
		return 1;
	return 0;
}

static void ip__resolve_ams(struct machine *machine, struct thread *thread,
			    struct addr_map_symbol *ams,
			    u64 ip)
{
	struct addr_location al;

	memset(&al, 0, sizeof(al));
1210 1211 1212 1213 1214 1215 1216 1217
	/*
	 * We cannot use the header.misc hint to determine whether a
	 * branch stack address is user, kernel, guest, hypervisor.
	 * Branches may straddle the kernel/user/hypervisor boundaries.
	 * Thus, we have to try consecutively until we find a match
	 * or else, the symbol is unknown
	 */
	thread__find_cpumode_addr_location(thread, machine, MAP__FUNCTION, ip, &al);
1218 1219 1220 1221 1222 1223 1224

	ams->addr = ip;
	ams->al_addr = al.addr;
	ams->sym = al.sym;
	ams->map = al.map;
}

1225 1226 1227 1228 1229 1230 1231
static void ip__resolve_data(struct machine *machine, struct thread *thread,
			     u8 m, struct addr_map_symbol *ams, u64 addr)
{
	struct addr_location al;

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

1232 1233
	thread__find_addr_location(thread, machine, m, MAP__VARIABLE, addr,
				   &al);
1234 1235 1236 1237 1238 1239
	ams->addr = addr;
	ams->al_addr = al.addr;
	ams->sym = al.sym;
	ams->map = al.map;
}

1240 1241
struct mem_info *sample__resolve_mem(struct perf_sample *sample,
				     struct addr_location *al)
1242 1243 1244 1245 1246 1247
{
	struct mem_info *mi = zalloc(sizeof(*mi));

	if (!mi)
		return NULL;

1248 1249 1250
	ip__resolve_ams(al->machine, al->thread, &mi->iaddr, sample->ip);
	ip__resolve_data(al->machine, al->thread, al->cpumode,
			 &mi->daddr, sample->addr);
1251 1252 1253 1254 1255
	mi->data_src.val = sample->data_src;

	return mi;
}

1256 1257
struct branch_info *sample__resolve_bstack(struct perf_sample *sample,
					   struct addr_location *al)
1258 1259
{
	unsigned int i;
1260 1261
	const struct branch_stack *bs = sample->branch_stack;
	struct branch_info *bi = calloc(bs->nr, sizeof(struct branch_info));
1262 1263 1264 1265 1266

	if (!bi)
		return NULL;

	for (i = 0; i < bs->nr; i++) {
1267 1268
		ip__resolve_ams(al->machine, al->thread, &bi[i].to, bs->entries[i].to);
		ip__resolve_ams(al->machine, al->thread, &bi[i].from, bs->entries[i].from);
1269 1270 1271 1272 1273 1274 1275 1276
		bi[i].flags = bs->entries[i].flags;
	}
	return bi;
}

static int machine__resolve_callchain_sample(struct machine *machine,
					     struct thread *thread,
					     struct ip_callchain *chain,
1277
					     struct symbol **parent,
1278 1279
					     struct addr_location *root_al,
					     int max_stack)
1280 1281
{
	u8 cpumode = PERF_RECORD_MISC_USER;
1282 1283
	int chain_nr = min(max_stack, (int)chain->nr);
	int i;
1284
	int j;
1285
	int err;
1286
	int skip_idx __maybe_unused;
1287 1288 1289 1290 1291 1292 1293 1294

	callchain_cursor_reset(&callchain_cursor);

	if (chain->nr > PERF_MAX_STACK_DEPTH) {
		pr_warning("corrupted callchain. skipping...\n");
		return 0;
	}

1295 1296 1297 1298 1299 1300
	/*
	 * Based on DWARF debug information, some architectures skip
	 * a callchain entry saved by the kernel.
	 */
	skip_idx = arch_skip_callchain_idx(machine, thread, chain);

1301
	for (i = 0; i < chain_nr; i++) {
1302 1303 1304 1305
		u64 ip;
		struct addr_location al;

		if (callchain_param.order == ORDER_CALLEE)
1306
			j = i;
1307
		else
1308 1309 1310 1311 1312 1313 1314
			j = chain->nr - i - 1;

#ifdef HAVE_SKIP_CALLCHAIN_IDX
		if (j == skip_idx)
			continue;
#endif
		ip = chain->ips[j];
1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339

		if (ip >= PERF_CONTEXT_MAX) {
			switch (ip) {
			case PERF_CONTEXT_HV:
				cpumode = PERF_RECORD_MISC_HYPERVISOR;
				break;
			case PERF_CONTEXT_KERNEL:
				cpumode = PERF_RECORD_MISC_KERNEL;
				break;
			case PERF_CONTEXT_USER:
				cpumode = PERF_RECORD_MISC_USER;
				break;
			default:
				pr_debug("invalid callchain context: "
					 "%"PRId64"\n", (s64) ip);
				/*
				 * It seems the callchain is corrupted.
				 * Discard all.
				 */
				callchain_cursor_reset(&callchain_cursor);
				return 0;
			}
			continue;
		}

1340
		al.filtered = 0;
1341
		thread__find_addr_location(thread, machine, cpumode,
1342
					   MAP__FUNCTION, ip, &al);
1343 1344
		if (al.sym != NULL) {
			if (sort__has_parent && !*parent &&
1345
			    symbol__match_regex(al.sym, &parent_regex))
1346
				*parent = al.sym;
1347 1348 1349 1350 1351 1352 1353
			else if (have_ignore_callees && root_al &&
			  symbol__match_regex(al.sym, &ignore_callees_regex)) {
				/* Treat this symbol as the root,
				   forgetting its callees. */
				*root_al = al;
				callchain_cursor_reset(&callchain_cursor);
			}
1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375
		}

		err = callchain_cursor_append(&callchain_cursor,
					      ip, al.map, al.sym);
		if (err)
			return err;
	}

	return 0;
}

static int unwind_entry(struct unwind_entry *entry, void *arg)
{
	struct callchain_cursor *cursor = arg;
	return callchain_cursor_append(cursor, entry->ip,
				       entry->map, entry->sym);
}

int machine__resolve_callchain(struct machine *machine,
			       struct perf_evsel *evsel,
			       struct thread *thread,
			       struct perf_sample *sample,
1376
			       struct symbol **parent,
1377 1378
			       struct addr_location *root_al,
			       int max_stack)
1379 1380 1381 1382
{
	int ret;

	ret = machine__resolve_callchain_sample(machine, thread,
1383 1384
						sample->callchain, parent,
						root_al, max_stack);
1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398
	if (ret)
		return ret;

	/* Can we do dwarf post unwind? */
	if (!((evsel->attr.sample_type & PERF_SAMPLE_REGS_USER) &&
	      (evsel->attr.sample_type & PERF_SAMPLE_STACK_USER)))
		return 0;

	/* Bail out if nothing was captured. */
	if ((!sample->user_regs.regs) ||
	    (!sample->user_stack.size))
		return 0;

	return unwind__get_entries(unwind_entry, &callchain_cursor, machine,
1399
				   thread, sample, max_stack);
1400 1401

}
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424

int machine__for_each_thread(struct machine *machine,
			     int (*fn)(struct thread *thread, void *p),
			     void *priv)
{
	struct rb_node *nd;
	struct thread *thread;
	int rc = 0;

	for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) {
		thread = rb_entry(nd, struct thread, rb_node);
		rc = fn(thread, priv);
		if (rc != 0)
			return rc;
	}

	list_for_each_entry(thread, &machine->dead_threads, node) {
		rc = fn(thread, priv);
		if (rc != 0)
			return rc;
	}
	return rc;
}
1425

1426
int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool,
1427
				  struct target *target, struct thread_map *threads,
1428
				  perf_event__handler_t process, bool data_mmap)
1429
{
1430
	if (target__has_task(target))
1431
		return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap);
1432
	else if (target__has_cpu(target))
1433 1434 1435 1436
		return perf_event__synthesize_threads(tool, process, machine, data_mmap);
	/* command specified */
	return 0;
}