builtin-trace.c 92.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
/*
 * builtin-trace.c
 *
 * Builtin 'trace' command:
 *
 * Display a continuously updated trace of any workload, CPU, specific PID,
 * system wide, etc.  Default format is loosely strace like, but any other
 * event may be specified using --event.
 *
 * Copyright (C) 2012, 2013, 2014, 2015 Red Hat Inc, Arnaldo Carvalho de Melo <acme@redhat.com>
 *
 * Initially based on the 'trace' prototype by Thomas Gleixner:
 *
 * http://lwn.net/Articles/415728/ ("Announcing a new utility: 'trace'")
 *
 * Released under the GPL v2. (and only v2, not any later version)
 */

19
#include <traceevent/event-parse.h>
20
#include <api/fs/tracing_path.h>
A
Arnaldo Carvalho de Melo 已提交
21
#include "builtin.h"
22
#include "util/color.h"
23
#include "util/debug.h"
A
Arnaldo Carvalho de Melo 已提交
24
#include "util/evlist.h"
25
#include <subcmd/exec-cmd.h>
26
#include "util/machine.h"
27
#include "util/session.h"
28
#include "util/thread.h"
29
#include <subcmd/parse-options.h>
30
#include "util/strlist.h"
31
#include "util/intlist.h"
A
Arnaldo Carvalho de Melo 已提交
32
#include "util/thread_map.h"
33
#include "util/stat.h"
34
#include "trace-event.h"
35
#include "util/parse-events.h"
36
#include "util/bpf-loader.h"
37
#include "callchain.h"
38
#include "syscalltbl.h"
A
Arnaldo Carvalho de Melo 已提交
39

40
#include <libaudit.h> /* FIXME: Still needed for audit_errno_to_name */
A
Arnaldo Carvalho de Melo 已提交
41
#include <stdlib.h>
42
#include <sys/mman.h>
43
#include <linux/futex.h>
44
#include <linux/err.h>
45 46 47 48
#include <linux/seccomp.h>
#include <linux/filter.h>
#include <linux/audit.h>
#include <sys/ptrace.h>
49
#include <linux/random.h>
A
Arnaldo Carvalho de Melo 已提交
50

51 52 53 54 55 56 57
/* For older distros: */
#ifndef MAP_STACK
# define MAP_STACK		0x20000
#endif

#ifndef MADV_HWPOISON
# define MADV_HWPOISON		100
58

59 60 61 62 63 64 65 66 67 68
#endif

#ifndef MADV_MERGEABLE
# define MADV_MERGEABLE		12
#endif

#ifndef MADV_UNMERGEABLE
# define MADV_UNMERGEABLE	13
#endif

69 70 71 72
#ifndef EFD_SEMAPHORE
# define EFD_SEMAPHORE		1
#endif

73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
#ifndef EFD_NONBLOCK
# define EFD_NONBLOCK		00004000
#endif

#ifndef EFD_CLOEXEC
# define EFD_CLOEXEC		02000000
#endif

#ifndef O_CLOEXEC
# define O_CLOEXEC		02000000
#endif

#ifndef SOCK_DCCP
# define SOCK_DCCP		6
#endif

#ifndef SOCK_CLOEXEC
# define SOCK_CLOEXEC		02000000
#endif

#ifndef SOCK_NONBLOCK
# define SOCK_NONBLOCK		00004000
#endif

#ifndef MSG_CMSG_CLOEXEC
# define MSG_CMSG_CLOEXEC	0x40000000
#endif

101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116
#ifndef PERF_FLAG_FD_NO_GROUP
# define PERF_FLAG_FD_NO_GROUP		(1UL << 0)
#endif

#ifndef PERF_FLAG_FD_OUTPUT
# define PERF_FLAG_FD_OUTPUT		(1UL << 1)
#endif

#ifndef PERF_FLAG_PID_CGROUP
# define PERF_FLAG_PID_CGROUP		(1UL << 2) /* pid=cgroup id, per-cpu mode only */
#endif

#ifndef PERF_FLAG_FD_CLOEXEC
# define PERF_FLAG_FD_CLOEXEC		(1UL << 3) /* O_CLOEXEC */
#endif

117 118
struct trace {
	struct perf_tool	tool;
119
	struct syscalltbl	*sctbl;
120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161
	struct {
		int		max;
		struct syscall  *table;
		struct {
			struct perf_evsel *sys_enter,
					  *sys_exit;
		}		events;
	} syscalls;
	struct record_opts	opts;
	struct perf_evlist	*evlist;
	struct machine		*host;
	struct thread		*current;
	u64			base_time;
	FILE			*output;
	unsigned long		nr_events;
	struct strlist		*ev_qualifier;
	struct {
		size_t		nr;
		int		*entries;
	}			ev_qualifier_ids;
	struct intlist		*tid_list;
	struct intlist		*pid_list;
	struct {
		size_t		nr;
		pid_t		*entries;
	}			filter_pids;
	double			duration_filter;
	double			runtime_ms;
	struct {
		u64		vfs_getname,
				proc_getname;
	} stats;
	bool			not_ev_qualifier;
	bool			live;
	bool			full_time;
	bool			sched;
	bool			multiple_threads;
	bool			summary;
	bool			summary_only;
	bool			show_comm;
	bool			show_tool_stats;
	bool			trace_syscalls;
162
	bool			kernel_syscallchains;
163 164 165
	bool			force;
	bool			vfs_getname;
	int			trace_pgfaults;
166
	int			open_id;
167
};
168

169 170 171 172 173 174 175 176 177 178 179
struct tp_field {
	int offset;
	union {
		u64 (*integer)(struct tp_field *field, struct perf_sample *sample);
		void *(*pointer)(struct tp_field *field, struct perf_sample *sample);
	};
};

#define TP_UINT_FIELD(bits) \
static u64 tp_field__u##bits(struct tp_field *field, struct perf_sample *sample) \
{ \
180 181 182
	u##bits value; \
	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
	return value;  \
183 184 185 186 187 188 189 190 191 192
}

TP_UINT_FIELD(8);
TP_UINT_FIELD(16);
TP_UINT_FIELD(32);
TP_UINT_FIELD(64);

#define TP_UINT_FIELD__SWAPPED(bits) \
static u64 tp_field__swapped_u##bits(struct tp_field *field, struct perf_sample *sample) \
{ \
193 194
	u##bits value; \
	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280
	return bswap_##bits(value);\
}

TP_UINT_FIELD__SWAPPED(16);
TP_UINT_FIELD__SWAPPED(32);
TP_UINT_FIELD__SWAPPED(64);

static int tp_field__init_uint(struct tp_field *field,
			       struct format_field *format_field,
			       bool needs_swap)
{
	field->offset = format_field->offset;

	switch (format_field->size) {
	case 1:
		field->integer = tp_field__u8;
		break;
	case 2:
		field->integer = needs_swap ? tp_field__swapped_u16 : tp_field__u16;
		break;
	case 4:
		field->integer = needs_swap ? tp_field__swapped_u32 : tp_field__u32;
		break;
	case 8:
		field->integer = needs_swap ? tp_field__swapped_u64 : tp_field__u64;
		break;
	default:
		return -1;
	}

	return 0;
}

static void *tp_field__ptr(struct tp_field *field, struct perf_sample *sample)
{
	return sample->raw_data + field->offset;
}

static int tp_field__init_ptr(struct tp_field *field, struct format_field *format_field)
{
	field->offset = format_field->offset;
	field->pointer = tp_field__ptr;
	return 0;
}

struct syscall_tp {
	struct tp_field id;
	union {
		struct tp_field args, ret;
	};
};

static int perf_evsel__init_tp_uint_field(struct perf_evsel *evsel,
					  struct tp_field *field,
					  const char *name)
{
	struct format_field *format_field = perf_evsel__field(evsel, name);

	if (format_field == NULL)
		return -1;

	return tp_field__init_uint(field, format_field, evsel->needs_swap);
}

#define perf_evsel__init_sc_tp_uint_field(evsel, name) \
	({ struct syscall_tp *sc = evsel->priv;\
	   perf_evsel__init_tp_uint_field(evsel, &sc->name, #name); })

static int perf_evsel__init_tp_ptr_field(struct perf_evsel *evsel,
					 struct tp_field *field,
					 const char *name)
{
	struct format_field *format_field = perf_evsel__field(evsel, name);

	if (format_field == NULL)
		return -1;

	return tp_field__init_ptr(field, format_field);
}

#define perf_evsel__init_sc_tp_ptr_field(evsel, name) \
	({ struct syscall_tp *sc = evsel->priv;\
	   perf_evsel__init_tp_ptr_field(evsel, &sc->name, #name); })

static void perf_evsel__delete_priv(struct perf_evsel *evsel)
{
281
	zfree(&evsel->priv);
282 283 284
	perf_evsel__delete(evsel);
}

285 286 287 288 289 290 291 292 293 294 295 296 297 298
static int perf_evsel__init_syscall_tp(struct perf_evsel *evsel, void *handler)
{
	evsel->priv = malloc(sizeof(struct syscall_tp));
	if (evsel->priv != NULL) {
		if (perf_evsel__init_sc_tp_uint_field(evsel, id))
			goto out_delete;

		evsel->handler = handler;
		return 0;
	}

	return -ENOMEM;

out_delete:
299
	zfree(&evsel->priv);
300 301 302
	return -ENOENT;
}

303
static struct perf_evsel *perf_evsel__syscall_newtp(const char *direction, void *handler)
304
{
305
	struct perf_evsel *evsel = perf_evsel__newtp("raw_syscalls", direction);
306

307
	/* older kernel (e.g., RHEL6) use syscalls:{enter,exit} */
308
	if (IS_ERR(evsel))
309 310
		evsel = perf_evsel__newtp("syscalls", direction);

311 312 313 314 315
	if (IS_ERR(evsel))
		return NULL;

	if (perf_evsel__init_syscall_tp(evsel, handler))
		goto out_delete;
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331

	return evsel;

out_delete:
	perf_evsel__delete_priv(evsel);
	return NULL;
}

#define perf_evsel__sc_tp_uint(evsel, name, sample) \
	({ struct syscall_tp *fields = evsel->priv; \
	   fields->name.integer(&fields->name, sample); })

#define perf_evsel__sc_tp_ptr(evsel, name, sample) \
	({ struct syscall_tp *fields = evsel->priv; \
	   fields->name.pointer(&fields->name, sample); })

332 333
struct syscall_arg {
	unsigned long val;
334 335
	struct thread *thread;
	struct trace  *trace;
336
	void	      *parm;
337 338 339 340
	u8	      idx;
	u8	      mask;
};

341
struct strarray {
342
	int	    offset;
343 344 345 346 347 348 349 350 351
	int	    nr_entries;
	const char **entries;
};

#define DEFINE_STRARRAY(array) struct strarray strarray__##array = { \
	.nr_entries = ARRAY_SIZE(array), \
	.entries = array, \
}

352 353 354 355 356 357
#define DEFINE_STRARRAY_OFFSET(array, off) struct strarray strarray__##array = { \
	.offset	    = off, \
	.nr_entries = ARRAY_SIZE(array), \
	.entries = array, \
}

358 359 360
static size_t __syscall_arg__scnprintf_strarray(char *bf, size_t size,
						const char *intfmt,
					        struct syscall_arg *arg)
361 362
{
	struct strarray *sa = arg->parm;
363
	int idx = arg->val - sa->offset;
364 365

	if (idx < 0 || idx >= sa->nr_entries)
366
		return scnprintf(bf, size, intfmt, arg->val);
367 368 369 370

	return scnprintf(bf, size, "%s", sa->entries[idx]);
}

371 372 373 374 375 376
static size_t syscall_arg__scnprintf_strarray(char *bf, size_t size,
					      struct syscall_arg *arg)
{
	return __syscall_arg__scnprintf_strarray(bf, size, "%d", arg);
}

377 378
#define SCA_STRARRAY syscall_arg__scnprintf_strarray

379 380 381 382 383
#if defined(__i386__) || defined(__x86_64__)
/*
 * FIXME: Make this available to all arches as soon as the ioctl beautifier
 * 	  gets rewritten to support all arches.
 */
384 385 386 387 388 389 390
static size_t syscall_arg__scnprintf_strhexarray(char *bf, size_t size,
						 struct syscall_arg *arg)
{
	return __syscall_arg__scnprintf_strarray(bf, size, "%#x", arg);
}

#define SCA_STRHEXARRAY syscall_arg__scnprintf_strhexarray
391
#endif /* defined(__i386__) || defined(__x86_64__) */
392

393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415
static size_t syscall_arg__scnprintf_fd(char *bf, size_t size,
					struct syscall_arg *arg);

#define SCA_FD syscall_arg__scnprintf_fd

static size_t syscall_arg__scnprintf_fd_at(char *bf, size_t size,
					   struct syscall_arg *arg)
{
	int fd = arg->val;

	if (fd == AT_FDCWD)
		return scnprintf(bf, size, "CWD");

	return syscall_arg__scnprintf_fd(bf, size, arg);
}

#define SCA_FDAT syscall_arg__scnprintf_fd_at

static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
					      struct syscall_arg *arg);

#define SCA_CLOSE_FD syscall_arg__scnprintf_close_fd

416
static size_t syscall_arg__scnprintf_hex(char *bf, size_t size,
417
					 struct syscall_arg *arg)
418
{
419
	return scnprintf(bf, size, "%#lx", arg->val);
420 421
}

422 423
#define SCA_HEX syscall_arg__scnprintf_hex

424 425 426 427 428 429 430 431
static size_t syscall_arg__scnprintf_int(char *bf, size_t size,
					 struct syscall_arg *arg)
{
	return scnprintf(bf, size, "%d", arg->val);
}

#define SCA_INT syscall_arg__scnprintf_int

432
static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
433
					       struct syscall_arg *arg)
434
{
435
	int printed = 0, prot = arg->val;
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

	if (prot == PROT_NONE)
		return scnprintf(bf, size, "NONE");
#define	P_MMAP_PROT(n) \
	if (prot & PROT_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		prot &= ~PROT_##n; \
	}

	P_MMAP_PROT(EXEC);
	P_MMAP_PROT(READ);
	P_MMAP_PROT(WRITE);
#ifdef PROT_SEM
	P_MMAP_PROT(SEM);
#endif
	P_MMAP_PROT(GROWSDOWN);
	P_MMAP_PROT(GROWSUP);
#undef P_MMAP_PROT

	if (prot)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", prot);

	return printed;
}

#define SCA_MMAP_PROT syscall_arg__scnprintf_mmap_prot

463
static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
464
						struct syscall_arg *arg)
465
{
466
	int printed = 0, flags = arg->val;
467 468 469 470 471 472 473 474 475

#define	P_MMAP_FLAG(n) \
	if (flags & MAP_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~MAP_##n; \
	}

	P_MMAP_FLAG(SHARED);
	P_MMAP_FLAG(PRIVATE);
476
#ifdef MAP_32BIT
477
	P_MMAP_FLAG(32BIT);
478
#endif
479 480 481 482 483 484
	P_MMAP_FLAG(ANONYMOUS);
	P_MMAP_FLAG(DENYWRITE);
	P_MMAP_FLAG(EXECUTABLE);
	P_MMAP_FLAG(FILE);
	P_MMAP_FLAG(FIXED);
	P_MMAP_FLAG(GROWSDOWN);
485
#ifdef MAP_HUGETLB
486
	P_MMAP_FLAG(HUGETLB);
487
#endif
488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
	P_MMAP_FLAG(LOCKED);
	P_MMAP_FLAG(NONBLOCK);
	P_MMAP_FLAG(NORESERVE);
	P_MMAP_FLAG(POPULATE);
	P_MMAP_FLAG(STACK);
#ifdef MAP_UNINITIALIZED
	P_MMAP_FLAG(UNINITIALIZED);
#endif
#undef P_MMAP_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_MMAP_FLAGS syscall_arg__scnprintf_mmap_flags

506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530
static size_t syscall_arg__scnprintf_mremap_flags(char *bf, size_t size,
						  struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

#define P_MREMAP_FLAG(n) \
	if (flags & MREMAP_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~MREMAP_##n; \
	}

	P_MREMAP_FLAG(MAYMOVE);
#ifdef MREMAP_FIXED
	P_MREMAP_FLAG(FIXED);
#endif
#undef P_MREMAP_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_MREMAP_FLAGS syscall_arg__scnprintf_mremap_flags

531
static size_t syscall_arg__scnprintf_madvise_behavior(char *bf, size_t size,
532
						      struct syscall_arg *arg)
533
{
534
	int behavior = arg->val;
535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551

	switch (behavior) {
#define	P_MADV_BHV(n) case MADV_##n: return scnprintf(bf, size, #n)
	P_MADV_BHV(NORMAL);
	P_MADV_BHV(RANDOM);
	P_MADV_BHV(SEQUENTIAL);
	P_MADV_BHV(WILLNEED);
	P_MADV_BHV(DONTNEED);
	P_MADV_BHV(REMOVE);
	P_MADV_BHV(DONTFORK);
	P_MADV_BHV(DOFORK);
	P_MADV_BHV(HWPOISON);
#ifdef MADV_SOFT_OFFLINE
	P_MADV_BHV(SOFT_OFFLINE);
#endif
	P_MADV_BHV(MERGEABLE);
	P_MADV_BHV(UNMERGEABLE);
552
#ifdef MADV_HUGEPAGE
553
	P_MADV_BHV(HUGEPAGE);
554 555
#endif
#ifdef MADV_NOHUGEPAGE
556
	P_MADV_BHV(NOHUGEPAGE);
557
#endif
558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
#ifdef MADV_DONTDUMP
	P_MADV_BHV(DONTDUMP);
#endif
#ifdef MADV_DODUMP
	P_MADV_BHV(DODUMP);
#endif
#undef P_MADV_PHV
	default: break;
	}

	return scnprintf(bf, size, "%#x", behavior);
}

#define SCA_MADV_BHV syscall_arg__scnprintf_madvise_behavior

573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603
static size_t syscall_arg__scnprintf_flock(char *bf, size_t size,
					   struct syscall_arg *arg)
{
	int printed = 0, op = arg->val;

	if (op == 0)
		return scnprintf(bf, size, "NONE");
#define	P_CMD(cmd) \
	if ((op & LOCK_##cmd) == LOCK_##cmd) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #cmd); \
		op &= ~LOCK_##cmd; \
	}

	P_CMD(SH);
	P_CMD(EX);
	P_CMD(NB);
	P_CMD(UN);
	P_CMD(MAND);
	P_CMD(RW);
	P_CMD(READ);
	P_CMD(WRITE);
#undef P_OP

	if (op)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", op);

	return printed;
}

#define SCA_FLOCK syscall_arg__scnprintf_flock

604
static size_t syscall_arg__scnprintf_futex_op(char *bf, size_t size, struct syscall_arg *arg)
605 606 607 608 609 610 611 612 613
{
	enum syscall_futex_args {
		SCF_UADDR   = (1 << 0),
		SCF_OP	    = (1 << 1),
		SCF_VAL	    = (1 << 2),
		SCF_TIMEOUT = (1 << 3),
		SCF_UADDR2  = (1 << 4),
		SCF_VAL3    = (1 << 5),
	};
614
	int op = arg->val;
615 616 617 618 619
	int cmd = op & FUTEX_CMD_MASK;
	size_t printed = 0;

	switch (cmd) {
#define	P_FUTEX_OP(n) case FUTEX_##n: printed = scnprintf(bf, size, #n);
620 621 622 623 624 625
	P_FUTEX_OP(WAIT);	    arg->mask |= SCF_VAL3|SCF_UADDR2;		  break;
	P_FUTEX_OP(WAKE);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
	P_FUTEX_OP(FD);		    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
	P_FUTEX_OP(REQUEUE);	    arg->mask |= SCF_VAL3|SCF_TIMEOUT;	          break;
	P_FUTEX_OP(CMP_REQUEUE);    arg->mask |= SCF_TIMEOUT;			  break;
	P_FUTEX_OP(CMP_REQUEUE_PI); arg->mask |= SCF_TIMEOUT;			  break;
626
	P_FUTEX_OP(WAKE_OP);							  break;
627 628 629 630 631
	P_FUTEX_OP(LOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
	P_FUTEX_OP(UNLOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2|SCF_TIMEOUT; break;
	P_FUTEX_OP(TRYLOCK_PI);	    arg->mask |= SCF_VAL3|SCF_UADDR2;		  break;
	P_FUTEX_OP(WAIT_BITSET);    arg->mask |= SCF_UADDR2;			  break;
	P_FUTEX_OP(WAKE_BITSET);    arg->mask |= SCF_UADDR2;			  break;
632 633 634 635 636 637 638 639 640 641 642 643 644
	P_FUTEX_OP(WAIT_REQUEUE_PI);						  break;
	default: printed = scnprintf(bf, size, "%#x", cmd);			  break;
	}

	if (op & FUTEX_PRIVATE_FLAG)
		printed += scnprintf(bf + printed, size - printed, "|PRIV");

	if (op & FUTEX_CLOCK_REALTIME)
		printed += scnprintf(bf + printed, size - printed, "|CLKRT");

	return printed;
}

645 646
#define SCA_FUTEX_OP  syscall_arg__scnprintf_futex_op

647 648 649 650 651 652
static const char *bpf_cmd[] = {
	"MAP_CREATE", "MAP_LOOKUP_ELEM", "MAP_UPDATE_ELEM", "MAP_DELETE_ELEM",
	"MAP_GET_NEXT_KEY", "PROG_LOAD",
};
static DEFINE_STRARRAY(bpf_cmd);

653 654
static const char *epoll_ctl_ops[] = { "ADD", "DEL", "MOD", };
static DEFINE_STRARRAY_OFFSET(epoll_ctl_ops, 1);
655

656 657 658
static const char *itimers[] = { "REAL", "VIRTUAL", "PROF", };
static DEFINE_STRARRAY(itimers);

659 660 661 662 663 664 665 666 667
static const char *keyctl_options[] = {
	"GET_KEYRING_ID", "JOIN_SESSION_KEYRING", "UPDATE", "REVOKE", "CHOWN",
	"SETPERM", "DESCRIBE", "CLEAR", "LINK", "UNLINK", "SEARCH", "READ",
	"INSTANTIATE", "NEGATE", "SET_REQKEY_KEYRING", "SET_TIMEOUT",
	"ASSUME_AUTHORITY", "GET_SECURITY", "SESSION_TO_PARENT", "REJECT",
	"INSTANTIATE_IOV", "INVALIDATE", "GET_PERSISTENT",
};
static DEFINE_STRARRAY(keyctl_options);

668 669 670 671 672 673 674 675 676
static const char *whences[] = { "SET", "CUR", "END",
#ifdef SEEK_DATA
"DATA",
#endif
#ifdef SEEK_HOLE
"HOLE",
#endif
};
static DEFINE_STRARRAY(whences);
677

678 679 680 681 682 683 684 685
static const char *fcntl_cmds[] = {
	"DUPFD", "GETFD", "SETFD", "GETFL", "SETFL", "GETLK", "SETLK",
	"SETLKW", "SETOWN", "GETOWN", "SETSIG", "GETSIG", "F_GETLK64",
	"F_SETLK64", "F_SETLKW64", "F_SETOWN_EX", "F_GETOWN_EX",
	"F_GETOWNER_UIDS",
};
static DEFINE_STRARRAY(fcntl_cmds);

686 687 688 689 690 691 692
static const char *rlimit_resources[] = {
	"CPU", "FSIZE", "DATA", "STACK", "CORE", "RSS", "NPROC", "NOFILE",
	"MEMLOCK", "AS", "LOCKS", "SIGPENDING", "MSGQUEUE", "NICE", "RTPRIO",
	"RTTIME",
};
static DEFINE_STRARRAY(rlimit_resources);

693 694 695
static const char *sighow[] = { "BLOCK", "UNBLOCK", "SETMASK", };
static DEFINE_STRARRAY(sighow);

696 697
static const char *clockid[] = {
	"REALTIME", "MONOTONIC", "PROCESS_CPUTIME_ID", "THREAD_CPUTIME_ID",
698 699
	"MONOTONIC_RAW", "REALTIME_COARSE", "MONOTONIC_COARSE", "BOOTTIME",
	"REALTIME_ALARM", "BOOTTIME_ALARM", "SGI_CYCLE", "TAI"
700 701 702
};
static DEFINE_STRARRAY(clockid);

703 704 705 706 707 708 709 710 711 712
static const char *socket_families[] = {
	"UNSPEC", "LOCAL", "INET", "AX25", "IPX", "APPLETALK", "NETROM",
	"BRIDGE", "ATMPVC", "X25", "INET6", "ROSE", "DECnet", "NETBEUI",
	"SECURITY", "KEY", "NETLINK", "PACKET", "ASH", "ECONET", "ATMSVC",
	"RDS", "SNA", "IRDA", "PPPOX", "WANPIPE", "LLC", "IB", "CAN", "TIPC",
	"BLUETOOTH", "IUCV", "RXRPC", "ISDN", "PHONET", "IEEE802154", "CAIF",
	"ALG", "NFC", "VSOCK",
};
static DEFINE_STRARRAY(socket_families);

713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
#ifndef SOCK_TYPE_MASK
#define SOCK_TYPE_MASK 0xf
#endif

static size_t syscall_arg__scnprintf_socket_type(char *bf, size_t size,
						      struct syscall_arg *arg)
{
	size_t printed;
	int type = arg->val,
	    flags = type & ~SOCK_TYPE_MASK;

	type &= SOCK_TYPE_MASK;
	/*
 	 * Can't use a strarray, MIPS may override for ABI reasons.
 	 */
	switch (type) {
#define	P_SK_TYPE(n) case SOCK_##n: printed = scnprintf(bf, size, #n); break;
	P_SK_TYPE(STREAM);
	P_SK_TYPE(DGRAM);
	P_SK_TYPE(RAW);
	P_SK_TYPE(RDM);
	P_SK_TYPE(SEQPACKET);
	P_SK_TYPE(DCCP);
	P_SK_TYPE(PACKET);
#undef P_SK_TYPE
	default:
		printed = scnprintf(bf, size, "%#x", type);
	}

#define	P_SK_FLAG(n) \
	if (flags & SOCK_##n) { \
		printed += scnprintf(bf + printed, size - printed, "|%s", #n); \
		flags &= ~SOCK_##n; \
	}

	P_SK_FLAG(CLOEXEC);
	P_SK_FLAG(NONBLOCK);
#undef P_SK_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "|%#x", flags);

	return printed;
}

#define SCA_SK_TYPE syscall_arg__scnprintf_socket_type

760 761 762
#ifndef MSG_PROBE
#define MSG_PROBE	     0x10
#endif
763 764 765
#ifndef MSG_WAITFORONE
#define MSG_WAITFORONE	0x10000
#endif
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
#ifndef MSG_SENDPAGE_NOTLAST
#define MSG_SENDPAGE_NOTLAST 0x20000
#endif
#ifndef MSG_FASTOPEN
#define MSG_FASTOPEN	     0x20000000
#endif

static size_t syscall_arg__scnprintf_msg_flags(char *bf, size_t size,
					       struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

	if (flags == 0)
		return scnprintf(bf, size, "NONE");
#define	P_MSG_FLAG(n) \
	if (flags & MSG_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~MSG_##n; \
	}

	P_MSG_FLAG(OOB);
	P_MSG_FLAG(PEEK);
	P_MSG_FLAG(DONTROUTE);
	P_MSG_FLAG(TRYHARD);
	P_MSG_FLAG(CTRUNC);
	P_MSG_FLAG(PROBE);
	P_MSG_FLAG(TRUNC);
	P_MSG_FLAG(DONTWAIT);
	P_MSG_FLAG(EOR);
	P_MSG_FLAG(WAITALL);
	P_MSG_FLAG(FIN);
	P_MSG_FLAG(SYN);
	P_MSG_FLAG(CONFIRM);
	P_MSG_FLAG(RST);
	P_MSG_FLAG(ERRQUEUE);
	P_MSG_FLAG(NOSIGNAL);
	P_MSG_FLAG(MORE);
	P_MSG_FLAG(WAITFORONE);
	P_MSG_FLAG(SENDPAGE_NOTLAST);
	P_MSG_FLAG(FASTOPEN);
	P_MSG_FLAG(CMSG_CLOEXEC);
#undef P_MSG_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_MSG_FLAGS syscall_arg__scnprintf_msg_flags

817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843
static size_t syscall_arg__scnprintf_access_mode(char *bf, size_t size,
						 struct syscall_arg *arg)
{
	size_t printed = 0;
	int mode = arg->val;

	if (mode == F_OK) /* 0 */
		return scnprintf(bf, size, "F");
#define	P_MODE(n) \
	if (mode & n##_OK) { \
		printed += scnprintf(bf + printed, size - printed, "%s", #n); \
		mode &= ~n##_OK; \
	}

	P_MODE(R);
	P_MODE(W);
	P_MODE(X);
#undef P_MODE

	if (mode)
		printed += scnprintf(bf + printed, size - printed, "|%#x", mode);

	return printed;
}

#define SCA_ACCMODE syscall_arg__scnprintf_access_mode

844 845 846 847 848
static size_t syscall_arg__scnprintf_filename(char *bf, size_t size,
					      struct syscall_arg *arg);

#define SCA_FILENAME syscall_arg__scnprintf_filename

849
static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
850
					       struct syscall_arg *arg)
851
{
852
	int printed = 0, flags = arg->val;
853 854

	if (!(flags & O_CREAT))
855
		arg->mask |= 1 << (arg->idx + 1); /* Mask the mode parm */
856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904

	if (flags == 0)
		return scnprintf(bf, size, "RDONLY");
#define	P_FLAG(n) \
	if (flags & O_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~O_##n; \
	}

	P_FLAG(APPEND);
	P_FLAG(ASYNC);
	P_FLAG(CLOEXEC);
	P_FLAG(CREAT);
	P_FLAG(DIRECT);
	P_FLAG(DIRECTORY);
	P_FLAG(EXCL);
	P_FLAG(LARGEFILE);
	P_FLAG(NOATIME);
	P_FLAG(NOCTTY);
#ifdef O_NONBLOCK
	P_FLAG(NONBLOCK);
#elif O_NDELAY
	P_FLAG(NDELAY);
#endif
#ifdef O_PATH
	P_FLAG(PATH);
#endif
	P_FLAG(RDWR);
#ifdef O_DSYNC
	if ((flags & O_SYNC) == O_SYNC)
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", "SYNC");
	else {
		P_FLAG(DSYNC);
	}
#else
	P_FLAG(SYNC);
#endif
	P_FLAG(TRUNC);
	P_FLAG(WRONLY);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_OPEN_FLAGS syscall_arg__scnprintf_open_flags

905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932
static size_t syscall_arg__scnprintf_perf_flags(char *bf, size_t size,
						struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

	if (flags == 0)
		return 0;

#define	P_FLAG(n) \
	if (flags & PERF_FLAG_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~PERF_FLAG_##n; \
	}

	P_FLAG(FD_NO_GROUP);
	P_FLAG(FD_OUTPUT);
	P_FLAG(PID_CGROUP);
	P_FLAG(FD_CLOEXEC);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_PERF_FLAGS syscall_arg__scnprintf_perf_flags

933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958
static size_t syscall_arg__scnprintf_eventfd_flags(char *bf, size_t size,
						   struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

	if (flags == 0)
		return scnprintf(bf, size, "NONE");
#define	P_FLAG(n) \
	if (flags & EFD_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~EFD_##n; \
	}

	P_FLAG(SEMAPHORE);
	P_FLAG(CLOEXEC);
	P_FLAG(NONBLOCK);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_EFD_FLAGS syscall_arg__scnprintf_eventfd_flags

959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981
static size_t syscall_arg__scnprintf_pipe_flags(char *bf, size_t size,
						struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

#define	P_FLAG(n) \
	if (flags & O_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~O_##n; \
	}

	P_FLAG(CLOEXEC);
	P_FLAG(NONBLOCK);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_PIPE_FLAGS syscall_arg__scnprintf_pipe_flags

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 1017
static size_t syscall_arg__scnprintf_signum(char *bf, size_t size, struct syscall_arg *arg)
{
	int sig = arg->val;

	switch (sig) {
#define	P_SIGNUM(n) case SIG##n: return scnprintf(bf, size, #n)
	P_SIGNUM(HUP);
	P_SIGNUM(INT);
	P_SIGNUM(QUIT);
	P_SIGNUM(ILL);
	P_SIGNUM(TRAP);
	P_SIGNUM(ABRT);
	P_SIGNUM(BUS);
	P_SIGNUM(FPE);
	P_SIGNUM(KILL);
	P_SIGNUM(USR1);
	P_SIGNUM(SEGV);
	P_SIGNUM(USR2);
	P_SIGNUM(PIPE);
	P_SIGNUM(ALRM);
	P_SIGNUM(TERM);
	P_SIGNUM(CHLD);
	P_SIGNUM(CONT);
	P_SIGNUM(STOP);
	P_SIGNUM(TSTP);
	P_SIGNUM(TTIN);
	P_SIGNUM(TTOU);
	P_SIGNUM(URG);
	P_SIGNUM(XCPU);
	P_SIGNUM(XFSZ);
	P_SIGNUM(VTALRM);
	P_SIGNUM(PROF);
	P_SIGNUM(WINCH);
	P_SIGNUM(IO);
	P_SIGNUM(PWR);
	P_SIGNUM(SYS);
1018 1019 1020 1021 1022 1023 1024 1025 1026
#ifdef SIGEMT
	P_SIGNUM(EMT);
#endif
#ifdef SIGSTKFLT
	P_SIGNUM(STKFLT);
#endif
#ifdef SIGSWI
	P_SIGNUM(SWI);
#endif
1027 1028 1029 1030 1031 1032 1033 1034
	default: break;
	}

	return scnprintf(bf, size, "%#x", sig);
}

#define SCA_SIGNUM syscall_arg__scnprintf_signum

1035 1036 1037 1038
#if defined(__i386__) || defined(__x86_64__)
/*
 * FIXME: Make this available to all arches.
 */
1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059
#define TCGETS		0x5401

static const char *tioctls[] = {
	"TCGETS", "TCSETS", "TCSETSW", "TCSETSF", "TCGETA", "TCSETA", "TCSETAW",
	"TCSETAF", "TCSBRK", "TCXONC", "TCFLSH", "TIOCEXCL", "TIOCNXCL",
	"TIOCSCTTY", "TIOCGPGRP", "TIOCSPGRP", "TIOCOUTQ", "TIOCSTI",
	"TIOCGWINSZ", "TIOCSWINSZ", "TIOCMGET", "TIOCMBIS", "TIOCMBIC",
	"TIOCMSET", "TIOCGSOFTCAR", "TIOCSSOFTCAR", "FIONREAD", "TIOCLINUX",
	"TIOCCONS", "TIOCGSERIAL", "TIOCSSERIAL", "TIOCPKT", "FIONBIO",
	"TIOCNOTTY", "TIOCSETD", "TIOCGETD", "TCSBRKP", [0x27] = "TIOCSBRK",
	"TIOCCBRK", "TIOCGSID", "TCGETS2", "TCSETS2", "TCSETSW2", "TCSETSF2",
	"TIOCGRS485", "TIOCSRS485", "TIOCGPTN", "TIOCSPTLCK",
	"TIOCGDEV||TCGETX", "TCSETX", "TCSETXF", "TCSETXW", "TIOCSIG",
	"TIOCVHANGUP", "TIOCGPKT", "TIOCGPTLCK", "TIOCGEXCL",
	[0x50] = "FIONCLEX", "FIOCLEX", "FIOASYNC", "TIOCSERCONFIG",
	"TIOCSERGWILD", "TIOCSERSWILD", "TIOCGLCKTRMIOS", "TIOCSLCKTRMIOS",
	"TIOCSERGSTRUCT", "TIOCSERGETLSR", "TIOCSERGETMULTI", "TIOCSERSETMULTI",
	"TIOCMIWAIT", "TIOCGICOUNT", [0x60] = "FIOQSIZE",
};

static DEFINE_STRARRAY_OFFSET(tioctls, 0x5401);
1060
#endif /* defined(__i386__) || defined(__x86_64__) */
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
static size_t syscall_arg__scnprintf_seccomp_op(char *bf, size_t size, struct syscall_arg *arg)
{
	int op = arg->val;
	size_t printed = 0;

	switch (op) {
#define	P_SECCOMP_SET_MODE_OP(n) case SECCOMP_SET_MODE_##n: printed = scnprintf(bf, size, #n); break
	P_SECCOMP_SET_MODE_OP(STRICT);
	P_SECCOMP_SET_MODE_OP(FILTER);
#undef P_SECCOMP_SET_MODE_OP
	default: printed = scnprintf(bf, size, "%#x", op);			  break;
	}

	return printed;
}

#define SCA_SECCOMP_OP  syscall_arg__scnprintf_seccomp_op

static size_t syscall_arg__scnprintf_seccomp_flags(char *bf, size_t size,
						   struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

#define	P_FLAG(n) \
	if (flags & SECCOMP_FILTER_FLAG_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~SECCOMP_FILTER_FLAG_##n; \
	}

	P_FLAG(TSYNC);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_SECCOMP_FLAGS syscall_arg__scnprintf_seccomp_flags

1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124
static size_t syscall_arg__scnprintf_getrandom_flags(char *bf, size_t size,
						   struct syscall_arg *arg)
{
	int printed = 0, flags = arg->val;

#define	P_FLAG(n) \
	if (flags & GRND_##n) { \
		printed += scnprintf(bf + printed, size - printed, "%s%s", printed ? "|" : "", #n); \
		flags &= ~GRND_##n; \
	}

	P_FLAG(RANDOM);
	P_FLAG(NONBLOCK);
#undef P_FLAG

	if (flags)
		printed += scnprintf(bf + printed, size - printed, "%s%#x", printed ? "|" : "", flags);

	return printed;
}

#define SCA_GETRANDOM_FLAGS syscall_arg__scnprintf_getrandom_flags

1125 1126 1127 1128
#define STRARRAY(arg, name, array) \
	  .arg_scnprintf = { [arg] = SCA_STRARRAY, }, \
	  .arg_parm	 = { [arg] = &strarray__##array, }

1129
#include "trace/beauty/pid.c"
1130
#include "trace/beauty/mode_t.c"
1131
#include "trace/beauty/sched_policy.c"
1132
#include "trace/beauty/waitid_options.c"
1133

A
Arnaldo Carvalho de Melo 已提交
1134 1135
static struct syscall_fmt {
	const char *name;
1136
	const char *alias;
1137
	size_t	   (*arg_scnprintf[6])(char *bf, size_t size, struct syscall_arg *arg);
1138
	void	   *arg_parm[6];
A
Arnaldo Carvalho de Melo 已提交
1139
	bool	   errmsg;
1140
	bool	   errpid;
A
Arnaldo Carvalho de Melo 已提交
1141
	bool	   timeout;
1142
	bool	   hexret;
A
Arnaldo Carvalho de Melo 已提交
1143
} syscall_fmts[] = {
1144
	{ .name	    = "access",	    .errmsg = true,
1145 1146
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */
			     [1] = SCA_ACCMODE,  /* mode */ }, },
1147
	{ .name	    = "arch_prctl", .errmsg = true, .alias = "prctl", },
1148
	{ .name	    = "bpf",	    .errmsg = true, STRARRAY(0, cmd, bpf_cmd), },
1149 1150
	{ .name	    = "brk",	    .hexret = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* brk */ }, },
1151 1152 1153 1154 1155 1156
	{ .name	    = "chdir",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "chmod",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "chroot",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1157
	{ .name     = "clock_gettime",  .errmsg = true, STRARRAY(0, clk_id, clockid), },
1158
	{ .name	    = "clone",	    .errpid = true, },
1159
	{ .name	    = "close",	    .errmsg = true,
1160
	  .arg_scnprintf = { [0] = SCA_CLOSE_FD, /* fd */ }, },
1161
	{ .name	    = "connect",    .errmsg = true, },
1162 1163
	{ .name	    = "creat",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1164
	{ .name	    = "dup",	    .errmsg = true,
1165
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1166
	{ .name	    = "dup2",	    .errmsg = true,
1167
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1168
	{ .name	    = "dup3",	    .errmsg = true,
1169
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1170
	{ .name	    = "epoll_ctl",  .errmsg = true, STRARRAY(1, op, epoll_ctl_ops), },
1171 1172
	{ .name	    = "eventfd2",   .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_EFD_FLAGS, /* flags */ }, },
1173
	{ .name	    = "faccessat",  .errmsg = true,
1174 1175
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1176
	{ .name	    = "fadvise64",  .errmsg = true,
1177
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1178
	{ .name	    = "fallocate",  .errmsg = true,
1179
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1180
	{ .name	    = "fchdir",	    .errmsg = true,
1181
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1182
	{ .name	    = "fchmod",	    .errmsg = true,
1183
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1184
	{ .name	    = "fchmodat",   .errmsg = true,
1185 1186
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1187
	{ .name	    = "fchown",	    .errmsg = true,
1188
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1189
	{ .name	    = "fchownat",   .errmsg = true,
1190 1191
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1192 1193 1194 1195 1196
	{ .name	    = "fcntl",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [1] = SCA_STRARRAY, /* cmd */ },
	  .arg_parm	 = { [1] = &strarray__fcntl_cmds, /* cmd */ }, },
	{ .name	    = "fdatasync",  .errmsg = true,
1197
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1198
	{ .name	    = "flock",	    .errmsg = true,
1199 1200 1201
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [1] = SCA_FLOCK, /* cmd */ }, },
	{ .name	    = "fsetxattr",  .errmsg = true,
1202
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1203
	{ .name	    = "fstat",	    .errmsg = true, .alias = "newfstat",
1204
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1205
	{ .name	    = "fstatat",    .errmsg = true, .alias = "newfstatat",
1206 1207
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1208
	{ .name	    = "fstatfs",    .errmsg = true,
1209
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1210
	{ .name	    = "fsync",    .errmsg = true,
1211
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1212
	{ .name	    = "ftruncate", .errmsg = true,
1213
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1214 1215
	{ .name	    = "futex",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_FUTEX_OP, /* op */ }, },
1216
	{ .name	    = "futimesat", .errmsg = true,
1217 1218
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1219
	{ .name	    = "getdents",   .errmsg = true,
1220
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1221
	{ .name	    = "getdents64", .errmsg = true,
1222
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1223
	{ .name	    = "getitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
1224
	{ .name	    = "getpid",	    .errpid = true, },
1225
	{ .name	    = "getpgid",    .errpid = true, },
1226
	{ .name	    = "getppid",    .errpid = true, },
1227 1228
	{ .name	    = "getrandom",  .errmsg = true,
	  .arg_scnprintf = { [2] = SCA_GETRANDOM_FLAGS, /* flags */ }, },
1229
	{ .name	    = "getrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
1230 1231 1232 1233
	{ .name	    = "getxattr",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "inotify_add_watch",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_FILENAME, /* pathname */ }, },
1234
	{ .name	    = "ioctl",	    .errmsg = true,
1235
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
1236 1237 1238 1239
#if defined(__i386__) || defined(__x86_64__)
/*
 * FIXME: Make this available to all arches.
 */
1240 1241 1242
			     [1] = SCA_STRHEXARRAY, /* cmd */
			     [2] = SCA_HEX, /* arg */ },
	  .arg_parm	 = { [1] = &strarray__tioctls, /* cmd */ }, },
1243 1244 1245
#else
			     [2] = SCA_HEX, /* arg */ }, },
#endif
1246
	{ .name	    = "keyctl",	    .errmsg = true, STRARRAY(0, option, keyctl_options), },
1247 1248
	{ .name	    = "kill",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
1249 1250 1251 1252
	{ .name	    = "lchown",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "lgetxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1253
	{ .name	    = "linkat",	    .errmsg = true,
1254
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
1255 1256
	{ .name	    = "listxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1257 1258 1259 1260
	{ .name	    = "llistxattr", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "lremovexattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1261 1262 1263 1264
	{ .name	    = "lseek",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_STRARRAY, /* whence */ },
	  .arg_parm	 = { [2] = &strarray__whences, /* whence */ }, },
1265 1266
	{ .name	    = "lsetxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1267 1268
	{ .name	    = "lstat",	    .errmsg = true, .alias = "newlstat",
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1269 1270
	{ .name	    = "lsxattr",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1271 1272 1273
	{ .name     = "madvise",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX,	 /* start */
			     [2] = SCA_MADV_BHV, /* behavior */ }, },
1274 1275
	{ .name	    = "mkdir",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1276
	{ .name	    = "mkdirat",    .errmsg = true,
1277 1278 1279 1280
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "mknod",      .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1281
	{ .name	    = "mknodat",    .errmsg = true,
1282 1283
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1284 1285 1286 1287
	{ .name	    = "mlock",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
	{ .name	    = "mlockall",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1288
	{ .name	    = "mmap",	    .hexret = true,
1289
	  .arg_scnprintf = { [0] = SCA_HEX,	  /* addr */
1290
			     [2] = SCA_MMAP_PROT, /* prot */
1291 1292
			     [3] = SCA_MMAP_FLAGS, /* flags */
			     [4] = SCA_FD, 	  /* fd */ }, },
1293
	{ .name	    = "mprotect",   .errmsg = true,
1294 1295
	  .arg_scnprintf = { [0] = SCA_HEX, /* start */
			     [2] = SCA_MMAP_PROT, /* prot */ }, },
1296 1297
	{ .name	    = "mq_unlink", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* u_name */ }, },
1298 1299
	{ .name	    = "mremap",	    .hexret = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */
1300
			     [3] = SCA_MREMAP_FLAGS, /* flags */
1301
			     [4] = SCA_HEX, /* new_addr */ }, },
1302 1303
	{ .name	    = "munlock",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1304 1305
	{ .name	    = "munmap",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1306
	{ .name	    = "name_to_handle_at", .errmsg = true,
1307
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1308
	{ .name	    = "newfstatat", .errmsg = true,
1309 1310
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1311
	{ .name	    = "open",	    .errmsg = true,
1312 1313
	  .arg_scnprintf = { [0] = SCA_FILENAME,   /* filename */
			     [1] = SCA_OPEN_FLAGS, /* flags */ }, },
1314
	{ .name	    = "open_by_handle_at", .errmsg = true,
1315 1316
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
1317
	{ .name	    = "openat",	    .errmsg = true,
1318
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
1319
			     [1] = SCA_FILENAME, /* filename */
1320
			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
1321 1322 1323 1324 1325
	{ .name	    = "perf_event_open", .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_INT, /* pid */
			     [2] = SCA_INT, /* cpu */
			     [3] = SCA_FD,  /* group_fd */
			     [4] = SCA_PERF_FLAGS,  /* flags */ }, },
1326 1327
	{ .name	    = "pipe2",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_PIPE_FLAGS, /* flags */ }, },
1328 1329
	{ .name	    = "poll",	    .errmsg = true, .timeout = true, },
	{ .name	    = "ppoll",	    .errmsg = true, .timeout = true, },
1330
	{ .name	    = "pread",	    .errmsg = true, .alias = "pread64",
1331
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1332
	{ .name	    = "preadv",	    .errmsg = true, .alias = "pread",
1333
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1334
	{ .name	    = "prlimit64",  .errmsg = true, STRARRAY(1, resource, rlimit_resources), },
1335
	{ .name	    = "pwrite",	    .errmsg = true, .alias = "pwrite64",
1336
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1337
	{ .name	    = "pwritev",    .errmsg = true,
1338
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1339
	{ .name	    = "read",	    .errmsg = true,
1340
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1341 1342
	{ .name	    = "readlink",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
1343
	{ .name	    = "readlinkat", .errmsg = true,
1344 1345
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
1346
	{ .name	    = "readv",	    .errmsg = true,
1347
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1348
	{ .name	    = "recvfrom",   .errmsg = true,
1349 1350
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1351
	{ .name	    = "recvmmsg",   .errmsg = true,
1352 1353
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1354
	{ .name	    = "recvmsg",    .errmsg = true,
1355 1356
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_MSG_FLAGS, /* flags */ }, },
1357 1358
	{ .name	    = "removexattr", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1359
	{ .name	    = "renameat",   .errmsg = true,
1360
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1361 1362
	{ .name	    = "rmdir",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1363 1364
	{ .name	    = "rt_sigaction", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_SIGNUM, /* sig */ }, },
1365
	{ .name	    = "rt_sigprocmask",  .errmsg = true, STRARRAY(0, how, sighow), },
1366 1367 1368 1369
	{ .name	    = "rt_sigqueueinfo", .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
	{ .name	    = "rt_tgsigqueueinfo", .errmsg = true,
	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
1370 1371
	{ .name	    = "sched_setscheduler",   .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SCHED_POLICY, /* policy */ }, },
1372 1373 1374
	{ .name	    = "seccomp", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_SECCOMP_OP, /* op */
			     [1] = SCA_SECCOMP_FLAGS, /* flags */ }, },
1375
	{ .name	    = "select",	    .errmsg = true, .timeout = true, },
1376
	{ .name	    = "sendmmsg",    .errmsg = true,
1377 1378
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1379
	{ .name	    = "sendmsg",    .errmsg = true,
1380 1381
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_MSG_FLAGS, /* flags */ }, },
1382
	{ .name	    = "sendto",	    .errmsg = true,
1383 1384
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1385
	{ .name	    = "set_tid_address", .errpid = true, },
1386
	{ .name	    = "setitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
1387
	{ .name	    = "setpgid",    .errmsg = true, },
1388
	{ .name	    = "setrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
1389 1390
	{ .name	    = "setxattr",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1391
	{ .name	    = "shutdown",   .errmsg = true,
1392
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1393
	{ .name	    = "socket",	    .errmsg = true,
1394 1395
	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
			     [1] = SCA_SK_TYPE, /* type */ },
1396 1397 1398 1399
	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
	{ .name	    = "socketpair", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
			     [1] = SCA_SK_TYPE, /* type */ },
1400
	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
1401 1402
	{ .name	    = "stat",	    .errmsg = true, .alias = "newstat",
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1403 1404 1405 1406 1407 1408
	{ .name	    = "statfs",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "swapoff",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* specialfile */ }, },
	{ .name	    = "swapon",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* specialfile */ }, },
1409
	{ .name	    = "symlinkat",  .errmsg = true,
1410
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1411 1412 1413 1414
	{ .name	    = "tgkill",	    .errmsg = true,
	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
	{ .name	    = "tkill",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
1415 1416
	{ .name	    = "truncate",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
1417
	{ .name	    = "uname",	    .errmsg = true, .alias = "newuname", },
1418
	{ .name	    = "unlinkat",   .errmsg = true,
1419 1420 1421 1422
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "utime",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1423
	{ .name	    = "utimensat",  .errmsg = true,
1424 1425 1426 1427
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dirfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "utimes",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1428 1429
	{ .name	    = "vmsplice",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1430
	{ .name	    = "wait4",	    .errpid = true,
1431
	  .arg_scnprintf = { [2] = SCA_WAITID_OPTIONS, /* options */ }, },
1432
	{ .name	    = "waitid",	    .errpid = true,
1433
	  .arg_scnprintf = { [3] = SCA_WAITID_OPTIONS, /* options */ }, },
1434
	{ .name	    = "write",	    .errmsg = true,
1435
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1436
	{ .name	    = "writev",	    .errmsg = true,
1437
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
A
Arnaldo Carvalho de Melo 已提交
1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453
};

static int syscall_fmt__cmp(const void *name, const void *fmtp)
{
	const struct syscall_fmt *fmt = fmtp;
	return strcmp(name, fmt->name);
}

static struct syscall_fmt *syscall_fmt__find(const char *name)
{
	const int nmemb = ARRAY_SIZE(syscall_fmts);
	return bsearch(name, syscall_fmts, nmemb, sizeof(struct syscall_fmt), syscall_fmt__cmp);
}

struct syscall {
	struct event_format *tp_format;
1454 1455
	int		    nr_args;
	struct format_field *args;
A
Arnaldo Carvalho de Melo 已提交
1456
	const char	    *name;
1457
	bool		    is_exit;
A
Arnaldo Carvalho de Melo 已提交
1458
	struct syscall_fmt  *fmt;
1459
	size_t		    (**arg_scnprintf)(char *bf, size_t size, struct syscall_arg *arg);
1460
	void		    **arg_parm;
A
Arnaldo Carvalho de Melo 已提交
1461 1462
};

1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
static size_t fprintf_duration(unsigned long t, FILE *fp)
{
	double duration = (double)t / NSEC_PER_MSEC;
	size_t printed = fprintf(fp, "(");

	if (duration >= 1.0)
		printed += color_fprintf(fp, PERF_COLOR_RED, "%6.3f ms", duration);
	else if (duration >= 0.01)
		printed += color_fprintf(fp, PERF_COLOR_YELLOW, "%6.3f ms", duration);
	else
		printed += color_fprintf(fp, PERF_COLOR_NORMAL, "%6.3f ms", duration);
1474
	return printed + fprintf(fp, "): ");
1475 1476
}

1477 1478 1479 1480 1481
/**
 * filename.ptr: The filename char pointer that will be vfs_getname'd
 * filename.entry_str_pos: Where to insert the string translated from
 *                         filename.ptr by the vfs_getname tracepoint/kprobe.
 */
1482 1483 1484 1485
struct thread_trace {
	u64		  entry_time;
	u64		  exit_time;
	bool		  entry_pending;
1486
	unsigned long	  nr_events;
1487
	unsigned long	  pfmaj, pfmin;
1488
	char		  *entry_str;
1489
	double		  runtime_ms;
1490 1491
        struct {
		unsigned long ptr;
1492 1493 1494 1495
		short int     entry_str_pos;
		bool	      pending_open;
		unsigned int  namelen;
		char	      *name;
1496
	} filename;
1497 1498 1499 1500
	struct {
		int	  max;
		char	  **table;
	} paths;
1501 1502

	struct intlist *syscall_stats;
1503 1504 1505 1506
};

static struct thread_trace *thread_trace__new(void)
{
1507 1508 1509 1510 1511
	struct thread_trace *ttrace =  zalloc(sizeof(struct thread_trace));

	if (ttrace)
		ttrace->paths.max = -1;

1512 1513
	ttrace->syscall_stats = intlist__new(NULL);

1514
	return ttrace;
1515 1516
}

1517
static struct thread_trace *thread__trace(struct thread *thread, FILE *fp)
1518
{
1519 1520
	struct thread_trace *ttrace;

1521 1522 1523
	if (thread == NULL)
		goto fail;

1524 1525
	if (thread__priv(thread) == NULL)
		thread__set_priv(thread, thread_trace__new());
1526

1527
	if (thread__priv(thread) == NULL)
1528 1529
		goto fail;

1530
	ttrace = thread__priv(thread);
1531 1532 1533
	++ttrace->nr_events;

	return ttrace;
1534
fail:
1535
	color_fprintf(fp, PERF_COLOR_RED,
1536 1537 1538 1539
		      "WARNING: not enough memory, dropping samples!\n");
	return NULL;
}

1540 1541 1542
#define TRACE_PFMAJ		(1 << 0)
#define TRACE_PFMIN		(1 << 1)

1543 1544
static const size_t trace__entry_str_size = 2048;

1545
static int trace__set_fd_pathname(struct thread *thread, int fd, const char *pathname)
1546
{
1547
	struct thread_trace *ttrace = thread__priv(thread);
1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570

	if (fd > ttrace->paths.max) {
		char **npath = realloc(ttrace->paths.table, (fd + 1) * sizeof(char *));

		if (npath == NULL)
			return -1;

		if (ttrace->paths.max != -1) {
			memset(npath + ttrace->paths.max + 1, 0,
			       (fd - ttrace->paths.max) * sizeof(char *));
		} else {
			memset(npath, 0, (fd + 1) * sizeof(char *));
		}

		ttrace->paths.table = npath;
		ttrace->paths.max   = fd;
	}

	ttrace->paths.table[fd] = strdup(pathname);

	return ttrace->paths.table[fd] != NULL ? 0 : -1;
}

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
static int thread__read_fd_path(struct thread *thread, int fd)
{
	char linkname[PATH_MAX], pathname[PATH_MAX];
	struct stat st;
	int ret;

	if (thread->pid_ == thread->tid) {
		scnprintf(linkname, sizeof(linkname),
			  "/proc/%d/fd/%d", thread->pid_, fd);
	} else {
		scnprintf(linkname, sizeof(linkname),
			  "/proc/%d/task/%d/fd/%d", thread->pid_, thread->tid, fd);
	}

	if (lstat(linkname, &st) < 0 || st.st_size + 1 > (off_t)sizeof(pathname))
		return -1;

	ret = readlink(linkname, pathname, sizeof(pathname));

	if (ret < 0 || ret > st.st_size)
		return -1;

	pathname[ret] = '\0';
	return trace__set_fd_pathname(thread, fd, pathname);
}

1597 1598
static const char *thread__fd_path(struct thread *thread, int fd,
				   struct trace *trace)
1599
{
1600
	struct thread_trace *ttrace = thread__priv(thread);
1601 1602 1603 1604 1605 1606 1607

	if (ttrace == NULL)
		return NULL;

	if (fd < 0)
		return NULL;

1608
	if ((fd > ttrace->paths.max || ttrace->paths.table[fd] == NULL)) {
1609 1610 1611
		if (!trace->live)
			return NULL;
		++trace->stats.proc_getname;
1612
		if (thread__read_fd_path(thread, fd))
1613 1614
			return NULL;
	}
1615 1616 1617 1618 1619 1620 1621 1622 1623

	return ttrace->paths.table[fd];
}

static size_t syscall_arg__scnprintf_fd(char *bf, size_t size,
					struct syscall_arg *arg)
{
	int fd = arg->val;
	size_t printed = scnprintf(bf, size, "%d", fd);
1624
	const char *path = thread__fd_path(arg->thread, fd, arg->trace);
1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636

	if (path)
		printed += scnprintf(bf + printed, size - printed, "<%s>", path);

	return printed;
}

static size_t syscall_arg__scnprintf_close_fd(char *bf, size_t size,
					      struct syscall_arg *arg)
{
	int fd = arg->val;
	size_t printed = syscall_arg__scnprintf_fd(bf, size, arg);
1637
	struct thread_trace *ttrace = thread__priv(arg->thread);
1638

1639 1640
	if (ttrace && fd >= 0 && fd <= ttrace->paths.max)
		zfree(&ttrace->paths.table[fd]);
1641 1642 1643 1644

	return printed;
}

1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665
static void thread__set_filename_pos(struct thread *thread, const char *bf,
				     unsigned long ptr)
{
	struct thread_trace *ttrace = thread__priv(thread);

	ttrace->filename.ptr = ptr;
	ttrace->filename.entry_str_pos = bf - ttrace->entry_str;
}

static size_t syscall_arg__scnprintf_filename(char *bf, size_t size,
					      struct syscall_arg *arg)
{
	unsigned long ptr = arg->val;

	if (!arg->trace->vfs_getname)
		return scnprintf(bf, size, "%#x", ptr);

	thread__set_filename_pos(arg->thread, bf, ptr);
	return 0;
}

1666 1667 1668 1669 1670
static bool trace__filter_duration(struct trace *trace, double t)
{
	return t < (trace->duration_filter * NSEC_PER_MSEC);
}

1671 1672 1673 1674
static size_t trace__fprintf_tstamp(struct trace *trace, u64 tstamp, FILE *fp)
{
	double ts = (double)(tstamp - trace->base_time) / NSEC_PER_MSEC;

1675
	return fprintf(fp, "%10.3f ", ts);
1676 1677
}

1678
static bool done = false;
1679
static bool interrupted = false;
1680

1681
static void sig_handler(int sig)
1682 1683
{
	done = true;
1684
	interrupted = sig == SIGINT;
1685 1686
}

1687
static size_t trace__fprintf_entry_head(struct trace *trace, struct thread *thread,
1688
					u64 duration, u64 tstamp, FILE *fp)
1689 1690
{
	size_t printed = trace__fprintf_tstamp(trace, tstamp, fp);
1691
	printed += fprintf_duration(duration, fp);
1692

1693 1694
	if (trace->multiple_threads) {
		if (trace->show_comm)
1695
			printed += fprintf(fp, "%.14s/", thread__comm_str(thread));
1696
		printed += fprintf(fp, "%d ", thread->tid);
1697
	}
1698 1699 1700 1701

	return printed;
}

1702
static int trace__process_event(struct trace *trace, struct machine *machine,
1703
				union perf_event *event, struct perf_sample *sample)
1704 1705 1706 1707 1708
{
	int ret = 0;

	switch (event->header.type) {
	case PERF_RECORD_LOST:
1709
		color_fprintf(trace->output, PERF_COLOR_RED,
1710
			      "LOST %" PRIu64 " events!\n", event->lost.lost);
1711
		ret = machine__process_lost_event(machine, event, sample);
1712
		break;
1713
	default:
1714
		ret = machine__process_event(machine, event, sample);
1715 1716 1717 1718 1719 1720
		break;
	}

	return ret;
}

1721
static int trace__tool_process(struct perf_tool *tool,
1722
			       union perf_event *event,
1723
			       struct perf_sample *sample,
1724 1725
			       struct machine *machine)
{
1726
	struct trace *trace = container_of(tool, struct trace, tool);
1727
	return trace__process_event(trace, machine, event, sample);
1728 1729 1730 1731
}

static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist)
{
1732
	int err = symbol__init(NULL);
1733 1734 1735 1736

	if (err)
		return err;

1737 1738 1739
	trace->host = machine__new_host();
	if (trace->host == NULL)
		return -ENOMEM;
1740

1741
	if (trace_event__register_resolver(trace->host, machine__resolve_kernel_addr) < 0)
1742 1743
		return -errno;

1744
	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
1745 1746
					    evlist->threads, trace__tool_process, false,
					    trace->opts.proc_map_timeout);
1747 1748 1749 1750 1751 1752
	if (err)
		symbol__exit();

	return err;
}

1753 1754 1755 1756 1757
static int syscall__set_arg_fmts(struct syscall *sc)
{
	struct format_field *field;
	int idx = 0;

1758
	sc->arg_scnprintf = calloc(sc->nr_args, sizeof(void *));
1759 1760 1761
	if (sc->arg_scnprintf == NULL)
		return -1;

1762 1763 1764
	if (sc->fmt)
		sc->arg_parm = sc->fmt->arg_parm;

1765
	for (field = sc->args; field; field = field->next) {
1766 1767 1768
		if (sc->fmt && sc->fmt->arg_scnprintf[idx])
			sc->arg_scnprintf[idx] = sc->fmt->arg_scnprintf[idx];
		else if (field->flags & FIELD_IS_POINTER)
1769
			sc->arg_scnprintf[idx] = syscall_arg__scnprintf_hex;
1770 1771
		else if (strcmp(field->type, "pid_t") == 0)
			sc->arg_scnprintf[idx] = SCA_PID;
1772 1773
		else if (strcmp(field->type, "umode_t") == 0)
			sc->arg_scnprintf[idx] = SCA_MODE_T;
1774 1775 1776 1777 1778 1779
		++idx;
	}

	return 0;
}

A
Arnaldo Carvalho de Melo 已提交
1780 1781 1782 1783
static int trace__read_syscall_info(struct trace *trace, int id)
{
	char tp_name[128];
	struct syscall *sc;
1784
	const char *name = syscalltbl__name(trace->sctbl, id);
1785 1786 1787

	if (name == NULL)
		return -1;
A
Arnaldo Carvalho de Melo 已提交
1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806

	if (id > trace->syscalls.max) {
		struct syscall *nsyscalls = realloc(trace->syscalls.table, (id + 1) * sizeof(*sc));

		if (nsyscalls == NULL)
			return -1;

		if (trace->syscalls.max != -1) {
			memset(nsyscalls + trace->syscalls.max + 1, 0,
			       (id - trace->syscalls.max) * sizeof(*sc));
		} else {
			memset(nsyscalls, 0, (id + 1) * sizeof(*sc));
		}

		trace->syscalls.table = nsyscalls;
		trace->syscalls.max   = id;
	}

	sc = trace->syscalls.table + id;
1807
	sc->name = name;
1808

1809
	sc->fmt  = syscall_fmt__find(sc->name);
A
Arnaldo Carvalho de Melo 已提交
1810

1811
	snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->name);
1812
	sc->tp_format = trace_event__tp_format("syscalls", tp_name);
1813

1814
	if (IS_ERR(sc->tp_format) && sc->fmt && sc->fmt->alias) {
1815
		snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->fmt->alias);
1816
		sc->tp_format = trace_event__tp_format("syscalls", tp_name);
1817
	}
A
Arnaldo Carvalho de Melo 已提交
1818

1819
	if (IS_ERR(sc->tp_format))
1820 1821
		return -1;

1822 1823
	sc->args = sc->tp_format->format.fields;
	sc->nr_args = sc->tp_format->format.nr_fields;
1824 1825 1826 1827 1828 1829
	/*
	 * We need to check and discard the first variable '__syscall_nr'
	 * or 'nr' that mean the syscall number. It is needless here.
	 * So drop '__syscall_nr' or 'nr' field but does not exist on older kernels.
	 */
	if (sc->args && (!strcmp(sc->args->name, "__syscall_nr") || !strcmp(sc->args->name, "nr"))) {
1830 1831 1832 1833
		sc->args = sc->args->next;
		--sc->nr_args;
	}

1834 1835
	sc->is_exit = !strcmp(name, "exit_group") || !strcmp(name, "exit");

1836
	return syscall__set_arg_fmts(sc);
A
Arnaldo Carvalho de Melo 已提交
1837 1838
}

1839 1840
static int trace__validate_ev_qualifier(struct trace *trace)
{
1841
	int err = 0, i;
1842 1843
	struct str_node *pos;

1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856
	trace->ev_qualifier_ids.nr = strlist__nr_entries(trace->ev_qualifier);
	trace->ev_qualifier_ids.entries = malloc(trace->ev_qualifier_ids.nr *
						 sizeof(trace->ev_qualifier_ids.entries[0]));

	if (trace->ev_qualifier_ids.entries == NULL) {
		fputs("Error:\tNot enough memory for allocating events qualifier ids\n",
		       trace->output);
		err = -EINVAL;
		goto out;
	}

	i = 0;

1857 1858
	strlist__for_each(pos, trace->ev_qualifier) {
		const char *sc = pos->s;
1859
		int id = syscalltbl__id(trace->sctbl, sc);
1860

1861
		if (id < 0) {
1862 1863 1864 1865 1866 1867 1868 1869 1870
			if (err == 0) {
				fputs("Error:\tInvalid syscall ", trace->output);
				err = -EINVAL;
			} else {
				fputs(", ", trace->output);
			}

			fputs(sc, trace->output);
		}
1871 1872

		trace->ev_qualifier_ids.entries[i++] = id;
1873 1874 1875 1876 1877
	}

	if (err < 0) {
		fputs("\nHint:\ttry 'perf list syscalls:sys_enter_*'"
		      "\nHint:\tand: 'man syscalls'\n", trace->output);
1878 1879
		zfree(&trace->ev_qualifier_ids.entries);
		trace->ev_qualifier_ids.nr = 0;
1880
	}
1881
out:
1882 1883 1884
	return err;
}

1885 1886 1887 1888 1889 1890 1891 1892 1893
/*
 * args is to be interpreted as a series of longs but we need to handle
 * 8-byte unaligned accesses. args points to raw_data within the event
 * and raw_data is guaranteed to be 8-byte unaligned because it is
 * preceded by raw_size which is a u32. So we need to copy args to a temp
 * variable to read it. Most notably this avoids extended load instructions
 * on unaligned addresses
 */

1894
static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size,
1895
				      unsigned char *args, struct trace *trace,
1896
				      struct thread *thread)
A
Arnaldo Carvalho de Melo 已提交
1897 1898
{
	size_t printed = 0;
1899 1900
	unsigned char *p;
	unsigned long val;
A
Arnaldo Carvalho de Melo 已提交
1901

1902
	if (sc->args != NULL) {
A
Arnaldo Carvalho de Melo 已提交
1903
		struct format_field *field;
1904 1905
		u8 bit = 1;
		struct syscall_arg arg = {
1906 1907 1908 1909
			.idx	= 0,
			.mask	= 0,
			.trace  = trace,
			.thread = thread,
1910
		};
1911

1912
		for (field = sc->args; field;
1913 1914
		     field = field->next, ++arg.idx, bit <<= 1) {
			if (arg.mask & bit)
1915
				continue;
1916 1917 1918 1919 1920

			/* special care for unaligned accesses */
			p = args + sizeof(unsigned long) * arg.idx;
			memcpy(&val, p, sizeof(val));

1921 1922 1923 1924 1925
			/*
 			 * Suppress this argument if its value is zero and
 			 * and we don't have a string associated in an
 			 * strarray for it.
 			 */
1926
			if (val == 0 &&
1927 1928 1929
			    !(sc->arg_scnprintf &&
			      sc->arg_scnprintf[arg.idx] == SCA_STRARRAY &&
			      sc->arg_parm[arg.idx]))
1930 1931
				continue;

1932
			printed += scnprintf(bf + printed, size - printed,
1933
					     "%s%s: ", printed ? ", " : "", field->name);
1934
			if (sc->arg_scnprintf && sc->arg_scnprintf[arg.idx]) {
1935
				arg.val = val;
1936 1937
				if (sc->arg_parm)
					arg.parm = sc->arg_parm[arg.idx];
1938 1939
				printed += sc->arg_scnprintf[arg.idx](bf + printed,
								      size - printed, &arg);
1940
			} else {
1941
				printed += scnprintf(bf + printed, size - printed,
1942
						     "%ld", val);
1943
			}
A
Arnaldo Carvalho de Melo 已提交
1944 1945
		}
	} else {
1946 1947
		int i = 0;

A
Arnaldo Carvalho de Melo 已提交
1948
		while (i < 6) {
1949 1950 1951
			/* special care for unaligned accesses */
			p = args + sizeof(unsigned long) * i;
			memcpy(&val, p, sizeof(val));
1952 1953
			printed += scnprintf(bf + printed, size - printed,
					     "%sarg%d: %ld",
1954
					     printed ? ", " : "", i, val);
A
Arnaldo Carvalho de Melo 已提交
1955 1956 1957 1958 1959 1960 1961
			++i;
		}
	}

	return printed;
}

1962
typedef int (*tracepoint_handler)(struct trace *trace, struct perf_evsel *evsel,
1963
				  union perf_event *event,
1964 1965 1966
				  struct perf_sample *sample);

static struct syscall *trace__syscall_info(struct trace *trace,
1967
					   struct perf_evsel *evsel, int id)
1968 1969 1970
{

	if (id < 0) {
1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

		/*
		 * XXX: Noticed on x86_64, reproduced as far back as 3.0.36, haven't tried
		 * before that, leaving at a higher verbosity level till that is
		 * explained. Reproduced with plain ftrace with:
		 *
		 * echo 1 > /t/events/raw_syscalls/sys_exit/enable
		 * grep "NR -1 " /t/trace_pipe
		 *
		 * After generating some load on the machine.
 		 */
		if (verbose > 1) {
			static u64 n;
			fprintf(trace->output, "Invalid syscall %d id, skipping (%s, %" PRIu64 ") ...\n",
				id, perf_evsel__name(evsel), ++n);
		}
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
		return NULL;
	}

	if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL) &&
	    trace__read_syscall_info(trace, id))
		goto out_cant_read;

	if ((id > trace->syscalls.max || trace->syscalls.table[id].name == NULL))
		goto out_cant_read;

	return &trace->syscalls.table[id];

out_cant_read:
2000 2001 2002 2003 2004 2005
	if (verbose) {
		fprintf(trace->output, "Problems reading syscall %d", id);
		if (id <= trace->syscalls.max && trace->syscalls.table[id].name != NULL)
			fprintf(trace->output, "(%s)", trace->syscalls.table[id].name);
		fputs(" information\n", trace->output);
	}
2006 2007 2008
	return NULL;
}

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
static void thread__update_stats(struct thread_trace *ttrace,
				 int id, struct perf_sample *sample)
{
	struct int_node *inode;
	struct stats *stats;
	u64 duration = 0;

	inode = intlist__findnew(ttrace->syscall_stats, id);
	if (inode == NULL)
		return;

	stats = inode->priv;
	if (stats == NULL) {
		stats = malloc(sizeof(struct stats));
		if (stats == NULL)
			return;
		init_stats(stats);
		inode->priv = stats;
	}

	if (ttrace->entry_time && sample->time > ttrace->entry_time)
		duration = sample->time - ttrace->entry_time;

	update_stats(stats, duration);
}

2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057
static int trace__printf_interrupted_entry(struct trace *trace, struct perf_sample *sample)
{
	struct thread_trace *ttrace;
	u64 duration;
	size_t printed;

	if (trace->current == NULL)
		return 0;

	ttrace = thread__priv(trace->current);

	if (!ttrace->entry_pending)
		return 0;

	duration = sample->time - ttrace->entry_time;

	printed  = trace__fprintf_entry_head(trace, trace->current, duration, sample->time, trace->output);
	printed += fprintf(trace->output, "%-70s) ...\n", ttrace->entry_str);
	ttrace->entry_pending = false;

	return printed;
}

2058
static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
2059
			    union perf_event *event __maybe_unused,
2060 2061
			    struct perf_sample *sample)
{
2062
	char *msg;
2063
	void *args;
2064
	size_t printed = 0;
2065
	struct thread *thread;
2066
	int id = perf_evsel__sc_tp_uint(evsel, id, sample), err = -1;
2067
	struct syscall *sc = trace__syscall_info(trace, evsel, id);
2068 2069 2070 2071
	struct thread_trace *ttrace;

	if (sc == NULL)
		return -1;
2072

2073
	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
2074
	ttrace = thread__trace(thread, trace->output);
2075
	if (ttrace == NULL)
2076
		goto out_put;
2077

2078
	args = perf_evsel__sc_tp_ptr(evsel, args, sample);
2079 2080

	if (ttrace->entry_str == NULL) {
2081
		ttrace->entry_str = malloc(trace__entry_str_size);
2082
		if (!ttrace->entry_str)
2083
			goto out_put;
2084 2085
	}

2086
	if (!trace->summary_only)
2087
		trace__printf_interrupted_entry(trace, sample);
2088

2089 2090
	ttrace->entry_time = sample->time;
	msg = ttrace->entry_str;
2091
	printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
2092

2093
	printed += syscall__scnprintf_args(sc, msg + printed, trace__entry_str_size - printed,
2094
					   args, trace, thread);
2095

2096
	if (sc->is_exit) {
D
David Ahern 已提交
2097
		if (!trace->duration_filter && !trace->summary_only) {
2098 2099
			trace__fprintf_entry_head(trace, thread, 1, sample->time, trace->output);
			fprintf(trace->output, "%-70s\n", ttrace->entry_str);
2100
		}
2101
	} else {
2102
		ttrace->entry_pending = true;
2103 2104 2105
		/* See trace__vfs_getname & trace__sys_exit */
		ttrace->filename.pending_open = false;
	}
2106

2107 2108 2109 2110
	if (trace->current != thread) {
		thread__put(trace->current);
		trace->current = thread__get(thread);
	}
2111 2112 2113 2114
	err = 0;
out_put:
	thread__put(thread);
	return err;
2115 2116 2117
}

static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
2118
			   union perf_event *event __maybe_unused,
2119 2120
			   struct perf_sample *sample)
{
2121
	long ret;
2122
	u64 duration = 0;
2123
	struct thread *thread;
2124
	int id = perf_evsel__sc_tp_uint(evsel, id, sample), err = -1;
2125
	struct syscall *sc = trace__syscall_info(trace, evsel, id);
2126 2127 2128 2129
	struct thread_trace *ttrace;

	if (sc == NULL)
		return -1;
2130

2131
	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
2132
	ttrace = thread__trace(thread, trace->output);
2133
	if (ttrace == NULL)
2134
		goto out_put;
2135

2136 2137 2138
	if (trace->summary)
		thread__update_stats(ttrace, id, sample);

2139
	ret = perf_evsel__sc_tp_uint(evsel, ret, sample);
2140

2141
	if (id == trace->open_id && ret >= 0 && ttrace->filename.pending_open) {
2142 2143
		trace__set_fd_pathname(thread, ret, ttrace->filename.name);
		ttrace->filename.pending_open = false;
2144 2145 2146
		++trace->stats.vfs_getname;
	}

2147 2148
	ttrace->exit_time = sample->time;

2149
	if (ttrace->entry_time) {
2150
		duration = sample->time - ttrace->entry_time;
2151 2152 2153 2154
		if (trace__filter_duration(trace, duration))
			goto out;
	} else if (trace->duration_filter)
		goto out;
2155

D
David Ahern 已提交
2156 2157 2158
	if (trace->summary_only)
		goto out;

2159
	trace__fprintf_entry_head(trace, thread, duration, sample->time, trace->output);
2160 2161

	if (ttrace->entry_pending) {
2162
		fprintf(trace->output, "%-70s", ttrace->entry_str);
2163
	} else {
2164 2165 2166
		fprintf(trace->output, " ... [");
		color_fprintf(trace->output, PERF_COLOR_YELLOW, "continued");
		fprintf(trace->output, "]: %s()", sc->name);
2167 2168
	}

2169 2170
	if (sc->fmt == NULL) {
signed_print:
2171
		fprintf(trace->output, ") = %ld", ret);
2172
	} else if (ret < 0 && (sc->fmt->errmsg || sc->fmt->errpid)) {
2173
		char bf[STRERR_BUFSIZE];
2174 2175 2176
		const char *emsg = strerror_r(-ret, bf, sizeof(bf)),
			   *e = audit_errno_to_name(-ret);

2177
		fprintf(trace->output, ") = -1 %s %s", e, emsg);
2178
	} else if (ret == 0 && sc->fmt->timeout)
2179
		fprintf(trace->output, ") = 0 Timeout");
2180
	else if (sc->fmt->hexret)
2181
		fprintf(trace->output, ") = %#lx", ret);
2182 2183 2184 2185 2186 2187 2188 2189 2190 2191
	else if (sc->fmt->errpid) {
		struct thread *child = machine__find_thread(trace->host, ret, ret);

		if (child != NULL) {
			fprintf(trace->output, ") = %ld", ret);
			if (child->comm_set)
				fprintf(trace->output, " (%s)", thread__comm_str(child));
			thread__put(child);
		}
	} else
2192
		goto signed_print;
2193

2194
	fputc('\n', trace->output);
2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206

	if (sample->callchain) {
		struct addr_location al;
		/* TODO: user-configurable print_opts */
		const unsigned int print_opts = PRINT_IP_OPT_SYM
					      | PRINT_IP_OPT_DSO;

		if (machine__resolve(trace->host, &al, sample) < 0) {
			pr_err("problem processing %d event, skipping it.\n",
			       event->header.type);
			goto out_put;
		}
2207 2208
		perf_evsel__fprintf_callchain(evsel, sample, &al, 38, print_opts,
					      scripting_max_stack, trace->output);
2209
	}
2210
out:
2211
	ttrace->entry_pending = false;
2212 2213 2214 2215
	err = 0;
out_put:
	thread__put(thread);
	return err;
2216 2217
}

2218
static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
2219
			      union perf_event *event __maybe_unused,
2220 2221
			      struct perf_sample *sample)
{
2222 2223 2224 2225 2226
	struct thread *thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
	struct thread_trace *ttrace;
	size_t filename_len, entry_str_len, to_move;
	ssize_t remaining_space;
	char *pos;
2227
	const char *filename = perf_evsel__rawptr(evsel, sample, "pathname");
2228 2229 2230 2231 2232 2233 2234 2235

	if (!thread)
		goto out;

	ttrace = thread__priv(thread);
	if (!ttrace)
		goto out;

2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250
	filename_len = strlen(filename);

	if (ttrace->filename.namelen < filename_len) {
		char *f = realloc(ttrace->filename.name, filename_len + 1);

		if (f == NULL)
				goto out;

		ttrace->filename.namelen = filename_len;
		ttrace->filename.name = f;
	}

	strcpy(ttrace->filename.name, filename);
	ttrace->filename.pending_open = true;

2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271
	if (!ttrace->filename.ptr)
		goto out;

	entry_str_len = strlen(ttrace->entry_str);
	remaining_space = trace__entry_str_size - entry_str_len - 1; /* \0 */
	if (remaining_space <= 0)
		goto out;

	if (filename_len > (size_t)remaining_space) {
		filename += filename_len - remaining_space;
		filename_len = remaining_space;
	}

	to_move = entry_str_len - ttrace->filename.entry_str_pos + 1; /* \0 */
	pos = ttrace->entry_str + ttrace->filename.entry_str_pos;
	memmove(pos + filename_len, pos, to_move);
	memcpy(pos, filename, filename_len);

	ttrace->filename.ptr = 0;
	ttrace->filename.entry_str_pos = 0;
out:
2272 2273 2274
	return 0;
}

2275
static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evsel,
2276
				     union perf_event *event __maybe_unused,
2277 2278 2279 2280
				     struct perf_sample *sample)
{
        u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
	double runtime_ms = (double)runtime / NSEC_PER_MSEC;
2281
	struct thread *thread = machine__findnew_thread(trace->host,
2282 2283
							sample->pid,
							sample->tid);
2284
	struct thread_trace *ttrace = thread__trace(thread, trace->output);
2285 2286 2287 2288 2289 2290

	if (ttrace == NULL)
		goto out_dump;

	ttrace->runtime_ms += runtime_ms;
	trace->runtime_ms += runtime_ms;
2291
	thread__put(thread);
2292 2293 2294
	return 0;

out_dump:
2295
	fprintf(trace->output, "%s: comm=%s,pid=%u,runtime=%" PRIu64 ",vruntime=%" PRIu64 ")\n",
2296 2297 2298 2299 2300
	       evsel->name,
	       perf_evsel__strval(evsel, sample, "comm"),
	       (pid_t)perf_evsel__intval(evsel, sample, "pid"),
	       runtime,
	       perf_evsel__intval(evsel, sample, "vruntime"));
2301
	thread__put(thread);
2302 2303 2304
	return 0;
}

2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335
static void bpf_output__printer(enum binary_printer_ops op,
				unsigned int val, void *extra)
{
	FILE *output = extra;
	unsigned char ch = (unsigned char)val;

	switch (op) {
	case BINARY_PRINT_CHAR_DATA:
		fprintf(output, "%c", isprint(ch) ? ch : '.');
		break;
	case BINARY_PRINT_DATA_BEGIN:
	case BINARY_PRINT_LINE_BEGIN:
	case BINARY_PRINT_ADDR:
	case BINARY_PRINT_NUM_DATA:
	case BINARY_PRINT_NUM_PAD:
	case BINARY_PRINT_SEP:
	case BINARY_PRINT_CHAR_PAD:
	case BINARY_PRINT_LINE_END:
	case BINARY_PRINT_DATA_END:
	default:
		break;
	}
}

static void bpf_output__fprintf(struct trace *trace,
				struct perf_sample *sample)
{
	print_binary(sample->raw_data, sample->raw_size, 8,
		     bpf_output__printer, trace->output);
}

2336 2337 2338 2339 2340 2341
static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel,
				union perf_event *event __maybe_unused,
				struct perf_sample *sample)
{
	trace__printf_interrupted_entry(trace, sample);
	trace__fprintf_tstamp(trace, sample->time, trace->output);
2342 2343 2344 2345 2346

	if (trace->trace_syscalls)
		fprintf(trace->output, "(         ): ");

	fprintf(trace->output, "%s:", evsel->name);
2347

2348 2349 2350
	if (perf_evsel__is_bpf_output(evsel)) {
		bpf_output__fprintf(trace, sample);
	} else if (evsel->tp_format) {
2351 2352 2353 2354 2355 2356 2357 2358 2359
		event_format__fprintf(evsel->tp_format, sample->cpu,
				      sample->raw_data, sample->raw_size,
				      trace->output);
	}

	fprintf(trace->output, ")\n");
	return 0;
}

2360 2361 2362 2363 2364 2365 2366 2367 2368
static void print_location(FILE *f, struct perf_sample *sample,
			   struct addr_location *al,
			   bool print_dso, bool print_sym)
{

	if ((verbose || print_dso) && al->map)
		fprintf(f, "%s@", al->map->dso->long_name);

	if ((verbose || print_sym) && al->sym)
2369
		fprintf(f, "%s+0x%" PRIx64, al->sym->name,
2370 2371
			al->addr - al->sym->start);
	else if (al->map)
2372
		fprintf(f, "0x%" PRIx64, al->addr);
2373
	else
2374
		fprintf(f, "0x%" PRIx64, sample->addr);
2375 2376 2377 2378
}

static int trace__pgfault(struct trace *trace,
			  struct perf_evsel *evsel,
2379
			  union perf_event *event __maybe_unused,
2380 2381 2382 2383 2384
			  struct perf_sample *sample)
{
	struct thread *thread;
	struct addr_location al;
	char map_type = 'd';
2385
	struct thread_trace *ttrace;
2386
	int err = -1;
2387 2388

	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
2389 2390
	ttrace = thread__trace(thread, trace->output);
	if (ttrace == NULL)
2391
		goto out_put;
2392 2393 2394 2395 2396 2397 2398

	if (evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ)
		ttrace->pfmaj++;
	else
		ttrace->pfmin++;

	if (trace->summary_only)
2399
		goto out;
2400

2401
	thread__find_addr_location(thread, sample->cpumode, MAP__FUNCTION,
2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413
			      sample->ip, &al);

	trace__fprintf_entry_head(trace, thread, 0, sample->time, trace->output);

	fprintf(trace->output, "%sfault [",
		evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ ?
		"maj" : "min");

	print_location(trace->output, sample, &al, false, true);

	fprintf(trace->output, "] => ");

2414
	thread__find_addr_location(thread, sample->cpumode, MAP__VARIABLE,
2415 2416 2417
				   sample->addr, &al);

	if (!al.map) {
2418
		thread__find_addr_location(thread, sample->cpumode,
2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429
					   MAP__FUNCTION, sample->addr, &al);

		if (al.map)
			map_type = 'x';
		else
			map_type = '?';
	}

	print_location(trace->output, sample, &al, true, false);

	fprintf(trace->output, " (%c%c)\n", map_type, al.level);
2430 2431 2432 2433 2434
out:
	err = 0;
out_put:
	thread__put(thread);
	return err;
2435 2436
}

2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448
static bool skip_sample(struct trace *trace, struct perf_sample *sample)
{
	if ((trace->pid_list && intlist__find(trace->pid_list, sample->pid)) ||
	    (trace->tid_list && intlist__find(trace->tid_list, sample->tid)))
		return false;

	if (trace->pid_list || trace->tid_list)
		return true;

	return false;
}

2449
static void trace__set_base_time(struct trace *trace,
2450
				 struct perf_evsel *evsel,
2451 2452
				 struct perf_sample *sample)
{
2453 2454 2455 2456 2457 2458 2459 2460 2461 2462
	/*
	 * BPF events were not setting PERF_SAMPLE_TIME, so be more robust
	 * and don't use sample->time unconditionally, we may end up having
	 * some other event in the future without PERF_SAMPLE_TIME for good
	 * reason, i.e. we may not be interested in its timestamps, just in
	 * it taking place, picking some piece of information when it
	 * appears in our event stream (vfs_getname comes to mind).
	 */
	if (trace->base_time == 0 && !trace->full_time &&
	    (evsel->attr.sample_type & PERF_SAMPLE_TIME))
2463 2464 2465
		trace->base_time = sample->time;
}

2466
static int trace__process_sample(struct perf_tool *tool,
2467
				 union perf_event *event,
2468 2469 2470 2471 2472 2473 2474
				 struct perf_sample *sample,
				 struct perf_evsel *evsel,
				 struct machine *machine __maybe_unused)
{
	struct trace *trace = container_of(tool, struct trace, tool);
	int err = 0;

2475
	tracepoint_handler handler = evsel->handler;
2476

2477 2478 2479
	if (skip_sample(trace, sample))
		return 0;

2480
	trace__set_base_time(trace, evsel, sample);
2481

2482 2483
	if (handler) {
		++trace->nr_events;
2484
		handler(trace, evsel, event, sample);
2485
	}
2486 2487 2488 2489

	return err;
}

2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510
static int parse_target_str(struct trace *trace)
{
	if (trace->opts.target.pid) {
		trace->pid_list = intlist__new(trace->opts.target.pid);
		if (trace->pid_list == NULL) {
			pr_err("Error parsing process id string\n");
			return -EINVAL;
		}
	}

	if (trace->opts.target.tid) {
		trace->tid_list = intlist__new(trace->opts.target.tid);
		if (trace->tid_list == NULL) {
			pr_err("Error parsing thread id string\n");
			return -EINVAL;
		}
	}

	return 0;
}

2511
static int trace__record(struct trace *trace, int argc, const char **argv)
D
David Ahern 已提交
2512 2513 2514 2515 2516 2517 2518 2519 2520 2521
{
	unsigned int rec_argc, i, j;
	const char **rec_argv;
	const char * const record_args[] = {
		"record",
		"-R",
		"-m", "1024",
		"-c", "1",
	};

2522 2523 2524 2525 2526 2527 2528
	const char * const sc_args[] = { "-e", };
	unsigned int sc_args_nr = ARRAY_SIZE(sc_args);
	const char * const majpf_args[] = { "-e", "major-faults" };
	unsigned int majpf_args_nr = ARRAY_SIZE(majpf_args);
	const char * const minpf_args[] = { "-e", "minor-faults" };
	unsigned int minpf_args_nr = ARRAY_SIZE(minpf_args);

2529
	/* +1 is for the event string below */
2530 2531
	rec_argc = ARRAY_SIZE(record_args) + sc_args_nr + 1 +
		majpf_args_nr + minpf_args_nr + argc;
D
David Ahern 已提交
2532 2533 2534 2535 2536
	rec_argv = calloc(rec_argc + 1, sizeof(char *));

	if (rec_argv == NULL)
		return -ENOMEM;

2537
	j = 0;
D
David Ahern 已提交
2538
	for (i = 0; i < ARRAY_SIZE(record_args); i++)
2539 2540
		rec_argv[j++] = record_args[i];

2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553
	if (trace->trace_syscalls) {
		for (i = 0; i < sc_args_nr; i++)
			rec_argv[j++] = sc_args[i];

		/* event string may be different for older kernels - e.g., RHEL6 */
		if (is_valid_tracepoint("raw_syscalls:sys_enter"))
			rec_argv[j++] = "raw_syscalls:sys_enter,raw_syscalls:sys_exit";
		else if (is_valid_tracepoint("syscalls:sys_enter"))
			rec_argv[j++] = "syscalls:sys_enter,syscalls:sys_exit";
		else {
			pr_err("Neither raw_syscalls nor syscalls events exist.\n");
			return -1;
		}
2554 2555
	}

2556 2557 2558 2559 2560 2561 2562 2563 2564 2565
	if (trace->trace_pgfaults & TRACE_PFMAJ)
		for (i = 0; i < majpf_args_nr; i++)
			rec_argv[j++] = majpf_args[i];

	if (trace->trace_pgfaults & TRACE_PFMIN)
		for (i = 0; i < minpf_args_nr; i++)
			rec_argv[j++] = minpf_args[i];

	for (i = 0; i < (unsigned int)argc; i++)
		rec_argv[j++] = argv[i];
D
David Ahern 已提交
2566

2567
	return cmd_record(j, rec_argv, NULL);
D
David Ahern 已提交
2568 2569
}

2570 2571
static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp);

2572
static bool perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
2573
{
2574
	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname");
2575 2576

	if (IS_ERR(evsel))
2577
		return false;
2578 2579 2580

	if (perf_evsel__field(evsel, "pathname") == NULL) {
		perf_evsel__delete(evsel);
2581
		return false;
2582 2583
	}

2584
	evsel->handler = trace__vfs_getname;
2585
	perf_evlist__add(evlist, evsel);
2586
	return true;
2587 2588
}

2589 2590 2591 2592 2593 2594 2595 2596 2597 2598
static int perf_evlist__add_pgfault(struct perf_evlist *evlist,
				    u64 config)
{
	struct perf_evsel *evsel;
	struct perf_event_attr attr = {
		.type = PERF_TYPE_SOFTWARE,
		.mmap_data = 1,
	};

	attr.config = config;
2599
	attr.sample_period = 1;
2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612

	event_attr_init(&attr);

	evsel = perf_evsel__new(&attr);
	if (!evsel)
		return -ENOMEM;

	evsel->handler = trace__pgfault;
	perf_evlist__add(evlist, evsel);

	return 0;
}

2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628
static void trace__handle_event(struct trace *trace, union perf_event *event, struct perf_sample *sample)
{
	const u32 type = event->header.type;
	struct perf_evsel *evsel;

	if (type != PERF_RECORD_SAMPLE) {
		trace__process_event(trace, trace->host, event, sample);
		return;
	}

	evsel = perf_evlist__id2evsel(trace->evlist, sample->id);
	if (evsel == NULL) {
		fprintf(trace->output, "Unknown tp ID %" PRIu64 ", skipping...\n", sample->id);
		return;
	}

2629 2630
	trace__set_base_time(trace, evsel, sample);

2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641
	if (evsel->attr.type == PERF_TYPE_TRACEPOINT &&
	    sample->raw_data == NULL) {
		fprintf(trace->output, "%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n",
		       perf_evsel__name(evsel), sample->tid,
		       sample->cpu, sample->raw_size);
	} else {
		tracepoint_handler handler = evsel->handler;
		handler(trace, evsel, event, sample);
	}
}

2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664
static int trace__add_syscall_newtp(struct trace *trace)
{
	int ret = -1;
	struct perf_evlist *evlist = trace->evlist;
	struct perf_evsel *sys_enter, *sys_exit;

	sys_enter = perf_evsel__syscall_newtp("sys_enter", trace__sys_enter);
	if (sys_enter == NULL)
		goto out;

	if (perf_evsel__init_sc_tp_ptr_field(sys_enter, args))
		goto out_delete_sys_enter;

	sys_exit = perf_evsel__syscall_newtp("sys_exit", trace__sys_exit);
	if (sys_exit == NULL)
		goto out_delete_sys_enter;

	if (perf_evsel__init_sc_tp_uint_field(sys_exit, ret))
		goto out_delete_sys_exit;

	perf_evlist__add(evlist, sys_enter);
	perf_evlist__add(evlist, sys_exit);

2665 2666 2667 2668 2669 2670 2671 2672 2673
	if (trace->opts.callgraph_set && !trace->kernel_syscallchains) {
		/*
		 * We're interested only in the user space callchain
		 * leading to the syscall, allow overriding that for
		 * debugging reasons using --kernel_syscall_callchains
		 */
		sys_exit->attr.exclude_callchain_kernel = 1;
	}

2674 2675
	trace->syscalls.events.sys_enter = sys_enter;
	trace->syscalls.events.sys_exit  = sys_exit;
2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687

	ret = 0;
out:
	return ret;

out_delete_sys_exit:
	perf_evsel__delete_priv(sys_exit);
out_delete_sys_enter:
	perf_evsel__delete_priv(sys_enter);
	goto out;
}

2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 2702 2703 2704 2705 2706 2707
static int trace__set_ev_qualifier_filter(struct trace *trace)
{
	int err = -1;
	char *filter = asprintf_expr_inout_ints("id", !trace->not_ev_qualifier,
						trace->ev_qualifier_ids.nr,
						trace->ev_qualifier_ids.entries);

	if (filter == NULL)
		goto out_enomem;

	if (!perf_evsel__append_filter(trace->syscalls.events.sys_enter, "&&", filter))
		err = perf_evsel__append_filter(trace->syscalls.events.sys_exit, "&&", filter);

	free(filter);
out:
	return err;
out_enomem:
	errno = ENOMEM;
	goto out;
}
2708

2709
static int trace__run(struct trace *trace, int argc, const char **argv)
A
Arnaldo Carvalho de Melo 已提交
2710
{
2711
	struct perf_evlist *evlist = trace->evlist;
2712
	struct perf_evsel *evsel;
2713 2714
	int err = -1, i;
	unsigned long before;
2715
	const bool forks = argc > 0;
2716
	bool draining = false;
A
Arnaldo Carvalho de Melo 已提交
2717

2718 2719
	trace->live = true;

2720
	if (trace->trace_syscalls && trace__add_syscall_newtp(trace))
2721
		goto out_error_raw_syscalls;
A
Arnaldo Carvalho de Melo 已提交
2722

2723
	if (trace->trace_syscalls)
2724
		trace->vfs_getname = perf_evlist__add_vfs_getname(evlist);
2725

2726
	if ((trace->trace_pgfaults & TRACE_PFMAJ) &&
2727
	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MAJ)) {
2728
		goto out_error_mem;
2729
	}
2730 2731 2732

	if ((trace->trace_pgfaults & TRACE_PFMIN) &&
	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MIN))
2733
		goto out_error_mem;
2734

2735
	if (trace->sched &&
2736 2737 2738
	    perf_evlist__add_newtp(evlist, "sched", "sched_stat_runtime",
				   trace__sched_stat_runtime))
		goto out_error_sched_stat_runtime;
2739

A
Arnaldo Carvalho de Melo 已提交
2740 2741
	err = perf_evlist__create_maps(evlist, &trace->opts.target);
	if (err < 0) {
2742
		fprintf(trace->output, "Problems parsing the target to trace, check your options!\n");
A
Arnaldo Carvalho de Melo 已提交
2743 2744 2745
		goto out_delete_evlist;
	}

2746 2747
	err = trace__symbols_init(trace, evlist);
	if (err < 0) {
2748
		fprintf(trace->output, "Problems initializing symbol libraries!\n");
2749
		goto out_delete_evlist;
2750 2751
	}

2752
	perf_evlist__config(evlist, &trace->opts, &callchain_param);
A
Arnaldo Carvalho de Melo 已提交
2753

2754 2755 2756 2757
	signal(SIGCHLD, sig_handler);
	signal(SIGINT, sig_handler);

	if (forks) {
2758
		err = perf_evlist__prepare_workload(evlist, &trace->opts.target,
2759
						    argv, false, NULL);
2760
		if (err < 0) {
2761
			fprintf(trace->output, "Couldn't run the workload!\n");
2762
			goto out_delete_evlist;
2763 2764 2765
		}
	}

A
Arnaldo Carvalho de Melo 已提交
2766
	err = perf_evlist__open(evlist);
2767 2768
	if (err < 0)
		goto out_error_open;
A
Arnaldo Carvalho de Melo 已提交
2769

2770 2771 2772 2773 2774 2775 2776 2777 2778 2779
	err = bpf__apply_obj_config();
	if (err) {
		char errbuf[BUFSIZ];

		bpf__strerror_apply_obj_config(err, errbuf, sizeof(errbuf));
		pr_err("ERROR: Apply config to BPF failed: %s\n",
			 errbuf);
		goto out_error_open;
	}

2780 2781 2782 2783 2784 2785
	/*
	 * Better not use !target__has_task() here because we need to cover the
	 * case where no threads were specified in the command line, but a
	 * workload was, and in that case we will fill in the thread_map when
	 * we fork the workload in perf_evlist__prepare_workload.
	 */
2786 2787
	if (trace->filter_pids.nr > 0)
		err = perf_evlist__set_filter_pids(evlist, trace->filter_pids.nr, trace->filter_pids.entries);
2788
	else if (thread_map__pid(evlist->threads, 0) == -1)
2789 2790
		err = perf_evlist__set_filter_pid(evlist, getpid());

2791 2792 2793
	if (err < 0)
		goto out_error_mem;

2794 2795 2796 2797 2798
	if (trace->ev_qualifier_ids.nr > 0) {
		err = trace__set_ev_qualifier_filter(trace);
		if (err < 0)
			goto out_errno;

2799 2800 2801
		pr_debug("event qualifier tracepoint filter: %s\n",
			 trace->syscalls.events.sys_exit->filter);
	}
2802

2803 2804 2805
	err = perf_evlist__apply_filters(evlist, &evsel);
	if (err < 0)
		goto out_error_apply_filters;
2806

J
Jiri Olsa 已提交
2807
	err = perf_evlist__mmap(evlist, trace->opts.mmap_pages, false);
2808 2809
	if (err < 0)
		goto out_error_mmap;
A
Arnaldo Carvalho de Melo 已提交
2810

2811 2812 2813
	if (!target__none(&trace->opts.target))
		perf_evlist__enable(evlist);

2814 2815 2816
	if (forks)
		perf_evlist__start_workload(evlist);

2817
	trace->multiple_threads = thread_map__pid(evlist->threads, 0) == -1 ||
2818 2819
				  evlist->threads->nr > 1 ||
				  perf_evlist__first(evlist)->attr.inherit;
A
Arnaldo Carvalho de Melo 已提交
2820
again:
2821
	before = trace->nr_events;
A
Arnaldo Carvalho de Melo 已提交
2822 2823 2824 2825 2826 2827 2828

	for (i = 0; i < evlist->nr_mmaps; i++) {
		union perf_event *event;

		while ((event = perf_evlist__mmap_read(evlist, i)) != NULL) {
			struct perf_sample sample;

2829
			++trace->nr_events;
A
Arnaldo Carvalho de Melo 已提交
2830 2831 2832

			err = perf_evlist__parse_sample(evlist, event, &sample);
			if (err) {
2833
				fprintf(trace->output, "Can't parse sample, err = %d, skipping...\n", err);
2834
				goto next_event;
A
Arnaldo Carvalho de Melo 已提交
2835 2836
			}

2837
			trace__handle_event(trace, event, &sample);
2838 2839
next_event:
			perf_evlist__mmap_consume(evlist, i);
2840

2841 2842
			if (interrupted)
				goto out_disable;
2843 2844 2845 2846 2847

			if (done && !draining) {
				perf_evlist__disable(evlist);
				draining = true;
			}
A
Arnaldo Carvalho de Melo 已提交
2848 2849 2850
		}
	}

2851
	if (trace->nr_events == before) {
2852
		int timeout = done ? 100 : -1;
2853

2854 2855 2856 2857
		if (!draining && perf_evlist__poll(evlist, timeout) > 0) {
			if (perf_evlist__filter_pollfd(evlist, POLLERR | POLLHUP) == 0)
				draining = true;

2858
			goto again;
2859
		}
2860 2861
	} else {
		goto again;
2862 2863
	}

2864
out_disable:
2865 2866
	thread__zput(trace->current);

2867
	perf_evlist__disable(evlist);
A
Arnaldo Carvalho de Melo 已提交
2868

2869 2870 2871 2872 2873 2874 2875 2876 2877 2878 2879 2880
	if (!err) {
		if (trace->summary)
			trace__fprintf_thread_summary(trace, trace->output);

		if (trace->show_tool_stats) {
			fprintf(trace->output, "Stats:\n "
					       " vfs_getname : %" PRIu64 "\n"
					       " proc_getname: %" PRIu64 "\n",
				trace->stats.vfs_getname,
				trace->stats.proc_getname);
		}
	}
2881

A
Arnaldo Carvalho de Melo 已提交
2882 2883
out_delete_evlist:
	perf_evlist__delete(evlist);
2884
	trace->evlist = NULL;
2885
	trace->live = false;
A
Arnaldo Carvalho de Melo 已提交
2886
	return err;
2887 2888
{
	char errbuf[BUFSIZ];
2889

2890
out_error_sched_stat_runtime:
2891
	tracing_path__strerror_open_tp(errno, errbuf, sizeof(errbuf), "sched", "sched_stat_runtime");
2892 2893
	goto out_error;

2894
out_error_raw_syscalls:
2895
	tracing_path__strerror_open_tp(errno, errbuf, sizeof(errbuf), "raw_syscalls", "sys_(enter|exit)");
2896 2897
	goto out_error;

2898 2899 2900 2901
out_error_mmap:
	perf_evlist__strerror_mmap(evlist, errno, errbuf, sizeof(errbuf));
	goto out_error;

2902 2903 2904 2905
out_error_open:
	perf_evlist__strerror_open(evlist, errno, errbuf, sizeof(errbuf));

out_error:
2906
	fprintf(trace->output, "%s\n", errbuf);
2907
	goto out_delete_evlist;
2908 2909 2910 2911 2912 2913 2914

out_error_apply_filters:
	fprintf(trace->output,
		"Failed to set filter \"%s\" on event %s with %d (%s)\n",
		evsel->filter, perf_evsel__name(evsel), errno,
		strerror_r(errno, errbuf, sizeof(errbuf)));
	goto out_delete_evlist;
A
Arnaldo Carvalho de Melo 已提交
2915
}
2916 2917 2918
out_error_mem:
	fprintf(trace->output, "Not enough memory to run!\n");
	goto out_delete_evlist;
2919 2920 2921 2922

out_errno:
	fprintf(trace->output, "errno=%d,%s\n", errno, strerror(errno));
	goto out_delete_evlist;
2923
}
A
Arnaldo Carvalho de Melo 已提交
2924

2925 2926 2927
static int trace__replay(struct trace *trace)
{
	const struct perf_evsel_str_handler handlers[] = {
2928
		{ "probe:vfs_getname",	     trace__vfs_getname, },
2929
	};
2930 2931 2932
	struct perf_data_file file = {
		.path  = input_name,
		.mode  = PERF_DATA_MODE_READ,
2933
		.force = trace->force,
2934
	};
2935
	struct perf_session *session;
2936
	struct perf_evsel *evsel;
2937 2938 2939 2940
	int err = -1;

	trace->tool.sample	  = trace__process_sample;
	trace->tool.mmap	  = perf_event__process_mmap;
D
David Ahern 已提交
2941
	trace->tool.mmap2	  = perf_event__process_mmap2;
2942 2943 2944 2945 2946 2947 2948
	trace->tool.comm	  = perf_event__process_comm;
	trace->tool.exit	  = perf_event__process_exit;
	trace->tool.fork	  = perf_event__process_fork;
	trace->tool.attr	  = perf_event__process_attr;
	trace->tool.tracing_data = perf_event__process_tracing_data;
	trace->tool.build_id	  = perf_event__process_build_id;

2949
	trace->tool.ordered_events = true;
2950 2951 2952 2953 2954
	trace->tool.ordering_requires_timestamps = true;

	/* add tid to output */
	trace->multiple_threads = true;

2955
	session = perf_session__new(&file, false, &trace->tool);
2956
	if (session == NULL)
2957
		return -1;
2958

2959
	if (symbol__init(&session->header.env) < 0)
2960 2961
		goto out;

2962 2963
	trace->host = &session->machines.host;

2964 2965 2966 2967
	err = perf_session__set_tracepoints_handlers(session, handlers);
	if (err)
		goto out;

2968 2969
	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
						     "raw_syscalls:sys_enter");
2970 2971 2972 2973
	/* older kernels have syscalls tp versus raw_syscalls */
	if (evsel == NULL)
		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
							     "syscalls:sys_enter");
2974

2975 2976 2977
	if (evsel &&
	    (perf_evsel__init_syscall_tp(evsel, trace__sys_enter) < 0 ||
	    perf_evsel__init_sc_tp_ptr_field(evsel, args))) {
2978 2979 2980 2981 2982 2983
		pr_err("Error during initialize raw_syscalls:sys_enter event\n");
		goto out;
	}

	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
						     "raw_syscalls:sys_exit");
2984 2985 2986
	if (evsel == NULL)
		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
							     "syscalls:sys_exit");
2987 2988 2989
	if (evsel &&
	    (perf_evsel__init_syscall_tp(evsel, trace__sys_exit) < 0 ||
	    perf_evsel__init_sc_tp_uint_field(evsel, ret))) {
2990
		pr_err("Error during initialize raw_syscalls:sys_exit event\n");
2991 2992 2993
		goto out;
	}

2994 2995 2996 2997 2998 2999 3000 3001
	evlist__for_each(session->evlist, evsel) {
		if (evsel->attr.type == PERF_TYPE_SOFTWARE &&
		    (evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MAJ ||
		     evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS_MIN ||
		     evsel->attr.config == PERF_COUNT_SW_PAGE_FAULTS))
			evsel->handler = trace__pgfault;
	}

3002 3003 3004 3005
	err = parse_target_str(trace);
	if (err != 0)
		goto out;

3006 3007
	setup_pager();

3008
	err = perf_session__process_events(session);
3009 3010 3011
	if (err)
		pr_err("Failed to process events, error %d", err);

3012 3013 3014
	else if (trace->summary)
		trace__fprintf_thread_summary(trace, trace->output);

3015 3016 3017 3018 3019 3020
out:
	perf_session__delete(session);

	return err;
}

3021 3022 3023 3024
static size_t trace__fprintf_threads_header(FILE *fp)
{
	size_t printed;

3025
	printed  = fprintf(fp, "\n Summary of events:\n\n");
3026 3027 3028 3029 3030 3031 3032 3033 3034 3035 3036 3037 3038 3039 3040 3041 3042

	return printed;
}

static size_t thread__dump_stats(struct thread_trace *ttrace,
				 struct trace *trace, FILE *fp)
{
	struct stats *stats;
	size_t printed = 0;
	struct syscall *sc;
	struct int_node *inode = intlist__first(ttrace->syscall_stats);

	if (inode == NULL)
		return 0;

	printed += fprintf(fp, "\n");

3043 3044 3045
	printed += fprintf(fp, "   syscall            calls    total       min       avg       max      stddev\n");
	printed += fprintf(fp, "                               (msec)    (msec)    (msec)    (msec)        (%%)\n");
	printed += fprintf(fp, "   --------------- -------- --------- --------- --------- ---------     ------\n");
3046

3047 3048 3049 3050 3051 3052 3053 3054 3055 3056 3057 3058 3059 3060
	/* each int_node is a syscall */
	while (inode) {
		stats = inode->priv;
		if (stats) {
			double min = (double)(stats->min) / NSEC_PER_MSEC;
			double max = (double)(stats->max) / NSEC_PER_MSEC;
			double avg = avg_stats(stats);
			double pct;
			u64 n = (u64) stats->n;

			pct = avg ? 100.0 * stddev_stats(stats)/avg : 0.0;
			avg /= NSEC_PER_MSEC;

			sc = &trace->syscalls.table[inode->i];
3061
			printed += fprintf(fp, "   %-15s", sc->name);
3062 3063
			printed += fprintf(fp, " %8" PRIu64 " %9.3f %9.3f %9.3f",
					   n, avg * n, min, avg);
P
Pekka Enberg 已提交
3064
			printed += fprintf(fp, " %9.3f %9.2f%%\n", max, pct);
3065 3066 3067 3068 3069 3070
		}

		inode = intlist__next(inode);
	}

	printed += fprintf(fp, "\n\n");
3071 3072 3073 3074

	return printed;
}

3075 3076 3077 3078 3079 3080 3081 3082 3083 3084 3085 3086 3087
/* struct used to pass data to per-thread function */
struct summary_data {
	FILE *fp;
	struct trace *trace;
	size_t printed;
};

static int trace__fprintf_one_thread(struct thread *thread, void *priv)
{
	struct summary_data *data = priv;
	FILE *fp = data->fp;
	size_t printed = data->printed;
	struct trace *trace = data->trace;
3088
	struct thread_trace *ttrace = thread__priv(thread);
3089 3090 3091 3092 3093 3094 3095
	double ratio;

	if (ttrace == NULL)
		return 0;

	ratio = (double)ttrace->nr_events / trace->nr_events * 100.0;

3096
	printed += fprintf(fp, " %s (%d), ", thread__comm_str(thread), thread->tid);
3097
	printed += fprintf(fp, "%lu events, ", ttrace->nr_events);
3098
	printed += fprintf(fp, "%.1f%%", ratio);
3099 3100 3101 3102
	if (ttrace->pfmaj)
		printed += fprintf(fp, ", %lu majfaults", ttrace->pfmaj);
	if (ttrace->pfmin)
		printed += fprintf(fp, ", %lu minfaults", ttrace->pfmin);
3103
	printed += fprintf(fp, ", %.3f msec\n", ttrace->runtime_ms);
3104
	printed += thread__dump_stats(ttrace, trace, fp);
3105 3106 3107 3108 3109 3110

	data->printed += printed;

	return 0;
}

3111 3112
static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp)
{
3113 3114 3115 3116 3117
	struct summary_data data = {
		.fp = fp,
		.trace = trace
	};
	data.printed = trace__fprintf_threads_header(fp);
3118

3119 3120 3121
	machine__for_each_thread(trace->host, trace__fprintf_one_thread, &data);

	return data.printed;
3122 3123
}

3124 3125 3126 3127 3128 3129 3130 3131 3132
static int trace__set_duration(const struct option *opt, const char *str,
			       int unset __maybe_unused)
{
	struct trace *trace = opt->value;

	trace->duration_filter = atof(str);
	return 0;
}

3133 3134 3135 3136 3137 3138 3139 3140 3141 3142 3143 3144 3145 3146 3147 3148 3149 3150 3151 3152 3153 3154 3155 3156 3157 3158 3159 3160 3161 3162 3163 3164
static int trace__set_filter_pids(const struct option *opt, const char *str,
				  int unset __maybe_unused)
{
	int ret = -1;
	size_t i;
	struct trace *trace = opt->value;
	/*
	 * FIXME: introduce a intarray class, plain parse csv and create a
	 * { int nr, int entries[] } struct...
	 */
	struct intlist *list = intlist__new(str);

	if (list == NULL)
		return -1;

	i = trace->filter_pids.nr = intlist__nr_entries(list) + 1;
	trace->filter_pids.entries = calloc(i, sizeof(pid_t));

	if (trace->filter_pids.entries == NULL)
		goto out;

	trace->filter_pids.entries[0] = getpid();

	for (i = 1; i < trace->filter_pids.nr; ++i)
		trace->filter_pids.entries[i] = intlist__entry(list, i - 1)->i;

	intlist__delete(list);
	ret = 0;
out:
	return ret;
}

3165 3166 3167 3168 3169 3170 3171 3172 3173 3174 3175 3176 3177 3178 3179 3180 3181
static int trace__open_output(struct trace *trace, const char *filename)
{
	struct stat st;

	if (!stat(filename, &st) && st.st_size) {
		char oldname[PATH_MAX];

		scnprintf(oldname, sizeof(oldname), "%s.old", filename);
		unlink(oldname);
		rename(filename, oldname);
	}

	trace->output = fopen(filename, "w");

	return trace->output == NULL ? -errno : 0;
}

3182 3183 3184 3185 3186 3187 3188 3189 3190 3191 3192 3193 3194 3195 3196 3197 3198
static int parse_pagefaults(const struct option *opt, const char *str,
			    int unset __maybe_unused)
{
	int *trace_pgfaults = opt->value;

	if (strcmp(str, "all") == 0)
		*trace_pgfaults |= TRACE_PFMAJ | TRACE_PFMIN;
	else if (strcmp(str, "maj") == 0)
		*trace_pgfaults |= TRACE_PFMAJ;
	else if (strcmp(str, "min") == 0)
		*trace_pgfaults |= TRACE_PFMIN;
	else
		return -1;

	return 0;
}

3199 3200 3201 3202 3203 3204 3205 3206
static void evlist__set_evsel_handler(struct perf_evlist *evlist, void *handler)
{
	struct perf_evsel *evsel;

	evlist__for_each(evlist, evsel)
		evsel->handler = handler;
}

A
Arnaldo Carvalho de Melo 已提交
3207 3208
int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
{
3209
	const char *trace_usage[] = {
3210 3211
		"perf trace [<options>] [<command>]",
		"perf trace [<options>] -- <command> [<options>]",
D
David Ahern 已提交
3212 3213
		"perf trace record [<options>] [<command>]",
		"perf trace record [<options>] -- <command> [<options>]",
A
Arnaldo Carvalho de Melo 已提交
3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226
		NULL
	};
	struct trace trace = {
		.syscalls = {
			. max = -1,
		},
		.opts = {
			.target = {
				.uid	   = UINT_MAX,
				.uses_mmap = true,
			},
			.user_freq     = UINT_MAX,
			.user_interval = ULLONG_MAX,
3227
			.no_buffering  = true,
3228
			.mmap_pages    = UINT_MAX,
3229
			.proc_map_timeout  = 500,
A
Arnaldo Carvalho de Melo 已提交
3230
		},
3231
		.output = stderr,
3232
		.show_comm = true,
3233
		.trace_syscalls = true,
3234
		.kernel_syscallchains = false,
A
Arnaldo Carvalho de Melo 已提交
3235
	};
3236
	const char *output_name = NULL;
3237
	const char *ev_qualifier_str = NULL;
A
Arnaldo Carvalho de Melo 已提交
3238
	const struct option trace_options[] = {
3239 3240 3241
	OPT_CALLBACK(0, "event", &trace.evlist, "event",
		     "event selector. use 'perf list' to list available events",
		     parse_events_option),
3242 3243
	OPT_BOOLEAN(0, "comm", &trace.show_comm,
		    "show the thread COMM next to its id"),
3244
	OPT_BOOLEAN(0, "tool_stats", &trace.show_tool_stats, "show tool stats"),
3245
	OPT_STRING('e', "expr", &ev_qualifier_str, "expr", "list of syscalls to trace"),
3246
	OPT_STRING('o', "output", &output_name, "file", "output file name"),
3247
	OPT_STRING('i', "input", &input_name, "file", "Analyze events in file"),
A
Arnaldo Carvalho de Melo 已提交
3248 3249
	OPT_STRING('p', "pid", &trace.opts.target.pid, "pid",
		    "trace events on existing process id"),
3250
	OPT_STRING('t', "tid", &trace.opts.target.tid, "tid",
A
Arnaldo Carvalho de Melo 已提交
3251
		    "trace events on existing thread id"),
3252 3253
	OPT_CALLBACK(0, "filter-pids", &trace, "CSV list of pids",
		     "pids to filter (by the kernel)", trace__set_filter_pids),
3254
	OPT_BOOLEAN('a', "all-cpus", &trace.opts.target.system_wide,
A
Arnaldo Carvalho de Melo 已提交
3255
		    "system-wide collection from all CPUs"),
3256
	OPT_STRING('C', "cpu", &trace.opts.target.cpu_list, "cpu",
A
Arnaldo Carvalho de Melo 已提交
3257
		    "list of cpus to monitor"),
3258
	OPT_BOOLEAN(0, "no-inherit", &trace.opts.no_inherit,
A
Arnaldo Carvalho de Melo 已提交
3259
		    "child tasks do not inherit counters"),
3260 3261 3262
	OPT_CALLBACK('m', "mmap-pages", &trace.opts.mmap_pages, "pages",
		     "number of mmap data pages",
		     perf_evlist__parse_mmap_pages),
3263
	OPT_STRING('u', "uid", &trace.opts.target.uid_str, "user",
A
Arnaldo Carvalho de Melo 已提交
3264
		   "user to profile"),
3265 3266 3267
	OPT_CALLBACK(0, "duration", &trace, "float",
		     "show only events with duration > N.M ms",
		     trace__set_duration),
3268
	OPT_BOOLEAN(0, "sched", &trace.sched, "show blocking scheduler events"),
3269
	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
3270 3271
	OPT_BOOLEAN('T', "time", &trace.full_time,
		    "Show full timestamp, not time relative to first start"),
D
David Ahern 已提交
3272 3273 3274 3275
	OPT_BOOLEAN('s', "summary", &trace.summary_only,
		    "Show only syscall summary with statistics"),
	OPT_BOOLEAN('S', "with-summary", &trace.summary,
		    "Show all syscalls and summary with statistics"),
3276 3277
	OPT_CALLBACK_DEFAULT('F', "pf", &trace.trace_pgfaults, "all|maj|min",
		     "Trace pagefaults", parse_pagefaults, "maj"),
3278
	OPT_BOOLEAN(0, "syscalls", &trace.trace_syscalls, "Trace syscalls"),
3279
	OPT_BOOLEAN('f', "force", &trace.force, "don't complain, do it"),
3280 3281 3282
	OPT_CALLBACK(0, "call-graph", &trace.opts,
		     "record_mode[,record_size]", record_callchain_help,
		     &record_parse_callchain_opt),
3283 3284
	OPT_BOOLEAN(0, "kernel-syscall-graph", &trace.kernel_syscallchains,
		    "Show the kernel callchains on the syscall exit path"),
3285 3286
	OPT_UINTEGER(0, "proc-map-timeout", &trace.opts.proc_map_timeout,
			"per thread proc mmap processing timeout in ms"),
A
Arnaldo Carvalho de Melo 已提交
3287 3288
	OPT_END()
	};
3289
	const char * const trace_subcommands[] = { "record", NULL };
A
Arnaldo Carvalho de Melo 已提交
3290
	int err;
3291
	char bf[BUFSIZ];
A
Arnaldo Carvalho de Melo 已提交
3292

3293 3294 3295
	signal(SIGSEGV, sighandler_dump_stack);
	signal(SIGFPE, sighandler_dump_stack);

3296
	trace.evlist = perf_evlist__new();
3297
	trace.sctbl = syscalltbl__new();
3298

3299
	if (trace.evlist == NULL || trace.sctbl == NULL) {
3300
		pr_err("Not enough memory to run!\n");
3301
		err = -ENOMEM;
3302 3303 3304
		goto out;
	}

3305 3306
	argc = parse_options_subcommand(argc, argv, trace_options, trace_subcommands,
				 trace_usage, PARSE_OPT_STOP_AT_NON_OPTION);
D
David Ahern 已提交
3307

3308 3309 3310 3311 3312 3313 3314
	err = bpf__setup_stdout(trace.evlist);
	if (err) {
		bpf__strerror_setup_stdout(trace.evlist, err, bf, sizeof(bf));
		pr_err("ERROR: Setup BPF stdout failed: %s\n", bf);
		goto out;
	}

3315 3316 3317 3318 3319
	if (trace.trace_pgfaults) {
		trace.opts.sample_address = true;
		trace.opts.sample_time = true;
	}

3320 3321 3322
	if (trace.opts.callgraph_set)
		symbol_conf.use_callchain = true;

3323 3324 3325
	if (trace.evlist->nr_entries > 0)
		evlist__set_evsel_handler(trace.evlist, trace__event_handler);

3326 3327 3328 3329 3330 3331 3332
	if ((argc >= 1) && (strcmp(argv[0], "record") == 0))
		return trace__record(&trace, argc-1, &argv[1]);

	/* summary_only implies summary option, but don't overwrite summary if set */
	if (trace.summary_only)
		trace.summary = trace.summary_only;

3333 3334
	if (!trace.trace_syscalls && !trace.trace_pgfaults &&
	    trace.evlist->nr_entries == 0 /* Was --events used? */) {
3335 3336 3337 3338
		pr_err("Please specify something to trace.\n");
		return -1;
	}

3339 3340 3341 3342 3343 3344 3345 3346
	if (output_name != NULL) {
		err = trace__open_output(&trace, output_name);
		if (err < 0) {
			perror("failed to create output file");
			goto out;
		}
	}

3347 3348
	trace.open_id = syscalltbl__id(trace.sctbl, "open");

3349
	if (ev_qualifier_str != NULL) {
3350
		const char *s = ev_qualifier_str;
3351 3352 3353
		struct strlist_config slist_config = {
			.dirname = system_path(STRACE_GROUPS_DIR),
		};
3354 3355 3356 3357

		trace.not_ev_qualifier = *s == '!';
		if (trace.not_ev_qualifier)
			++s;
3358
		trace.ev_qualifier = strlist__new(s, &slist_config);
3359
		if (trace.ev_qualifier == NULL) {
3360 3361 3362 3363
			fputs("Not enough memory to parse event qualifier",
			      trace.output);
			err = -ENOMEM;
			goto out_close;
3364
		}
3365 3366 3367 3368

		err = trace__validate_ev_qualifier(&trace);
		if (err)
			goto out_close;
3369 3370
	}

3371
	err = target__validate(&trace.opts.target);
3372
	if (err) {
3373
		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
3374 3375
		fprintf(trace.output, "%s", bf);
		goto out_close;
3376 3377
	}

3378
	err = target__parse_uid(&trace.opts.target);
A
Arnaldo Carvalho de Melo 已提交
3379
	if (err) {
3380
		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
3381 3382
		fprintf(trace.output, "%s", bf);
		goto out_close;
A
Arnaldo Carvalho de Melo 已提交
3383 3384
	}

3385
	if (!argc && target__none(&trace.opts.target))
3386 3387
		trace.opts.target.system_wide = true;

3388 3389 3390 3391
	if (input_name)
		err = trace__replay(&trace);
	else
		err = trace__run(&trace, argc, argv);
3392

3393 3394 3395 3396
out_close:
	if (output_name != NULL)
		fclose(trace.output);
out:
3397
	return err;
A
Arnaldo Carvalho de Melo 已提交
3398
}