builtin-trace.c 85.2 KB
Newer Older
1
#include <traceevent/event-parse.h>
A
Arnaldo Carvalho de Melo 已提交
2
#include "builtin.h"
3
#include "util/color.h"
4
#include "util/debug.h"
A
Arnaldo Carvalho de Melo 已提交
5
#include "util/evlist.h"
6
#include "util/exec_cmd.h"
7
#include "util/machine.h"
8
#include "util/session.h"
9
#include "util/thread.h"
A
Arnaldo Carvalho de Melo 已提交
10
#include "util/parse-options.h"
11
#include "util/strlist.h"
12
#include "util/intlist.h"
A
Arnaldo Carvalho de Melo 已提交
13
#include "util/thread_map.h"
14
#include "util/stat.h"
15
#include "trace-event.h"
16
#include "util/parse-events.h"
A
Arnaldo Carvalho de Melo 已提交
17 18 19

#include <libaudit.h>
#include <stdlib.h>
20
#include <sys/mman.h>
21
#include <linux/futex.h>
A
Arnaldo Carvalho de Melo 已提交
22

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
/* For older distros: */
#ifndef MAP_STACK
# define MAP_STACK		0x20000
#endif

#ifndef MADV_HWPOISON
# define MADV_HWPOISON		100
#endif

#ifndef MADV_MERGEABLE
# define MADV_MERGEABLE		12
#endif

#ifndef MADV_UNMERGEABLE
# define MADV_UNMERGEABLE	13
#endif

40 41 42 43
#ifndef EFD_SEMAPHORE
# define EFD_SEMAPHORE		1
#endif

44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
#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

72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
#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


89 90 91 92 93 94 95 96 97 98 99
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) \
{ \
100 101 102
	u##bits value; \
	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
	return value;  \
103 104 105 106 107 108 109 110 111 112
}

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) \
{ \
113 114
	u##bits value; \
	memcpy(&value, sample->raw_data + field->offset, sizeof(value)); \
115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
	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)
{
201
	zfree(&evsel->priv);
202 203 204
	perf_evsel__delete(evsel);
}

205 206 207 208 209 210 211 212 213 214 215 216 217 218
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:
219
	zfree(&evsel->priv);
220 221 222
	return -ENOENT;
}

223
static struct perf_evsel *perf_evsel__syscall_newtp(const char *direction, void *handler)
224
{
225
	struct perf_evsel *evsel = perf_evsel__newtp("raw_syscalls", direction);
226

227 228 229 230
	/* older kernel (e.g., RHEL6) use syscalls:{enter,exit} */
	if (evsel == NULL)
		evsel = perf_evsel__newtp("syscalls", direction);

231
	if (evsel) {
232
		if (perf_evsel__init_syscall_tp(evsel, handler))
233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250
			goto out_delete;
	}

	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); })

251 252
struct syscall_arg {
	unsigned long val;
253 254
	struct thread *thread;
	struct trace  *trace;
255
	void	      *parm;
256 257 258 259
	u8	      idx;
	u8	      mask;
};

260
struct strarray {
261
	int	    offset;
262 263 264 265 266 267 268 269 270
	int	    nr_entries;
	const char **entries;
};

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

271 272 273 274 275 276
#define DEFINE_STRARRAY_OFFSET(array, off) struct strarray strarray__##array = { \
	.offset	    = off, \
	.nr_entries = ARRAY_SIZE(array), \
	.entries = array, \
}

277 278 279
static size_t __syscall_arg__scnprintf_strarray(char *bf, size_t size,
						const char *intfmt,
					        struct syscall_arg *arg)
280 281
{
	struct strarray *sa = arg->parm;
282
	int idx = arg->val - sa->offset;
283 284

	if (idx < 0 || idx >= sa->nr_entries)
285
		return scnprintf(bf, size, intfmt, arg->val);
286 287 288 289

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

290 291 292 293 294 295
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);
}

296 297
#define SCA_STRARRAY syscall_arg__scnprintf_strarray

298 299 300 301 302
#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.
 */
303 304 305 306 307 308 309
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
310
#endif /* defined(__i386__) || defined(__x86_64__) */
311

312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
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

335
static size_t syscall_arg__scnprintf_hex(char *bf, size_t size,
336
					 struct syscall_arg *arg)
337
{
338
	return scnprintf(bf, size, "%#lx", arg->val);
339 340
}

341 342
#define SCA_HEX syscall_arg__scnprintf_hex

343 344 345 346 347 348 349 350
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

351
static size_t syscall_arg__scnprintf_mmap_prot(char *bf, size_t size,
352
					       struct syscall_arg *arg)
353
{
354
	int printed = 0, prot = arg->val;
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381

	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

382
static size_t syscall_arg__scnprintf_mmap_flags(char *bf, size_t size,
383
						struct syscall_arg *arg)
384
{
385
	int printed = 0, flags = arg->val;
386 387 388 389 390 391 392 393 394

#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);
395
#ifdef MAP_32BIT
396
	P_MMAP_FLAG(32BIT);
397
#endif
398 399 400 401 402 403
	P_MMAP_FLAG(ANONYMOUS);
	P_MMAP_FLAG(DENYWRITE);
	P_MMAP_FLAG(EXECUTABLE);
	P_MMAP_FLAG(FILE);
	P_MMAP_FLAG(FIXED);
	P_MMAP_FLAG(GROWSDOWN);
404
#ifdef MAP_HUGETLB
405
	P_MMAP_FLAG(HUGETLB);
406
#endif
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
	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

425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
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

450
static size_t syscall_arg__scnprintf_madvise_behavior(char *bf, size_t size,
451
						      struct syscall_arg *arg)
452
{
453
	int behavior = arg->val;
454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470

	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);
471
#ifdef MADV_HUGEPAGE
472
	P_MADV_BHV(HUGEPAGE);
473 474
#endif
#ifdef MADV_NOHUGEPAGE
475
	P_MADV_BHV(NOHUGEPAGE);
476
#endif
477 478 479 480 481 482 483 484 485 486 487 488 489 490 491
#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

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
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

523
static size_t syscall_arg__scnprintf_futex_op(char *bf, size_t size, struct syscall_arg *arg)
524 525 526 527 528 529 530 531 532
{
	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),
	};
533
	int op = arg->val;
534 535 536 537 538
	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);
539 540 541 542 543 544
	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;
545
	P_FUTEX_OP(WAKE_OP);							  break;
546 547 548 549 550
	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;
551 552 553 554 555 556 557 558 559 560 561 562 563
	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;
}

564 565
#define SCA_FUTEX_OP  syscall_arg__scnprintf_futex_op

566 567
static const char *epoll_ctl_ops[] = { "ADD", "DEL", "MOD", };
static DEFINE_STRARRAY_OFFSET(epoll_ctl_ops, 1);
568

569 570 571
static const char *itimers[] = { "REAL", "VIRTUAL", "PROF", };
static DEFINE_STRARRAY(itimers);

572 573 574 575 576 577 578 579 580
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);

581 582 583 584 585 586 587 588 589
static const char *whences[] = { "SET", "CUR", "END",
#ifdef SEEK_DATA
"DATA",
#endif
#ifdef SEEK_HOLE
"HOLE",
#endif
};
static DEFINE_STRARRAY(whences);
590

591 592 593 594 595 596 597 598
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);

599 600 601 602 603 604 605
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);

606 607 608
static const char *sighow[] = { "BLOCK", "UNBLOCK", "SETMASK", };
static DEFINE_STRARRAY(sighow);

609 610
static const char *clockid[] = {
	"REALTIME", "MONOTONIC", "PROCESS_CPUTIME_ID", "THREAD_CPUTIME_ID",
611 612
	"MONOTONIC_RAW", "REALTIME_COARSE", "MONOTONIC_COARSE", "BOOTTIME",
	"REALTIME_ALARM", "BOOTTIME_ALARM", "SGI_CYCLE", "TAI"
613 614 615
};
static DEFINE_STRARRAY(clockid);

616 617 618 619 620 621 622 623 624 625
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);

626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672
#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

673 674 675
#ifndef MSG_PROBE
#define MSG_PROBE	     0x10
#endif
676 677 678
#ifndef MSG_WAITFORONE
#define MSG_WAITFORONE	0x10000
#endif
679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729
#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

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
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

757 758 759 760 761
static size_t syscall_arg__scnprintf_filename(char *bf, size_t size,
					      struct syscall_arg *arg);

#define SCA_FILENAME syscall_arg__scnprintf_filename

762
static size_t syscall_arg__scnprintf_open_flags(char *bf, size_t size,
763
					       struct syscall_arg *arg)
764
{
765
	int printed = 0, flags = arg->val;
766 767

	if (!(flags & O_CREAT))
768
		arg->mask |= 1 << (arg->idx + 1); /* Mask the mode parm */
769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817

	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

818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845
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

846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871
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

872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894
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

895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
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);
931 932 933 934 935 936 937 938 939
#ifdef SIGEMT
	P_SIGNUM(EMT);
#endif
#ifdef SIGSTKFLT
	P_SIGNUM(STKFLT);
#endif
#ifdef SIGSWI
	P_SIGNUM(SWI);
#endif
940 941 942 943 944 945 946 947
	default: break;
	}

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

#define SCA_SIGNUM syscall_arg__scnprintf_signum

948 949 950 951
#if defined(__i386__) || defined(__x86_64__)
/*
 * FIXME: Make this available to all arches.
 */
952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
#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);
973
#endif /* defined(__i386__) || defined(__x86_64__) */
974

975 976 977 978
#define STRARRAY(arg, name, array) \
	  .arg_scnprintf = { [arg] = SCA_STRARRAY, }, \
	  .arg_parm	 = { [arg] = &strarray__##array, }

A
Arnaldo Carvalho de Melo 已提交
979 980
static struct syscall_fmt {
	const char *name;
981
	const char *alias;
982
	size_t	   (*arg_scnprintf[6])(char *bf, size_t size, struct syscall_arg *arg);
983
	void	   *arg_parm[6];
A
Arnaldo Carvalho de Melo 已提交
984 985
	bool	   errmsg;
	bool	   timeout;
986
	bool	   hexret;
A
Arnaldo Carvalho de Melo 已提交
987
} syscall_fmts[] = {
988
	{ .name	    = "access",	    .errmsg = true,
989 990
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */
			     [1] = SCA_ACCMODE,  /* mode */ }, },
991
	{ .name	    = "arch_prctl", .errmsg = true, .alias = "prctl", },
992 993
	{ .name	    = "brk",	    .hexret = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* brk */ }, },
994 995 996 997 998 999
	{ .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 */ }, },
1000
	{ .name     = "clock_gettime",  .errmsg = true, STRARRAY(0, clk_id, clockid), },
1001
	{ .name	    = "close",	    .errmsg = true,
1002
	  .arg_scnprintf = { [0] = SCA_CLOSE_FD, /* fd */ }, },
1003
	{ .name	    = "connect",    .errmsg = true, },
1004 1005
	{ .name	    = "creat",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1006
	{ .name	    = "dup",	    .errmsg = true,
1007
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1008
	{ .name	    = "dup2",	    .errmsg = true,
1009
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1010
	{ .name	    = "dup3",	    .errmsg = true,
1011
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1012
	{ .name	    = "epoll_ctl",  .errmsg = true, STRARRAY(1, op, epoll_ctl_ops), },
1013 1014
	{ .name	    = "eventfd2",   .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_EFD_FLAGS, /* flags */ }, },
1015
	{ .name	    = "faccessat",  .errmsg = true,
1016 1017
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1018
	{ .name	    = "fadvise64",  .errmsg = true,
1019
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1020
	{ .name	    = "fallocate",  .errmsg = true,
1021
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1022
	{ .name	    = "fchdir",	    .errmsg = true,
1023
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1024
	{ .name	    = "fchmod",	    .errmsg = true,
1025
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1026
	{ .name	    = "fchmodat",   .errmsg = true,
1027 1028
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1029
	{ .name	    = "fchown",	    .errmsg = true,
1030
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1031
	{ .name	    = "fchownat",   .errmsg = true,
1032 1033
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1034 1035 1036 1037 1038
	{ .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,
1039
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1040
	{ .name	    = "flock",	    .errmsg = true,
1041 1042 1043
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [1] = SCA_FLOCK, /* cmd */ }, },
	{ .name	    = "fsetxattr",  .errmsg = true,
1044
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1045
	{ .name	    = "fstat",	    .errmsg = true, .alias = "newfstat",
1046
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1047
	{ .name	    = "fstatat",    .errmsg = true, .alias = "newfstatat",
1048 1049
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1050
	{ .name	    = "fstatfs",    .errmsg = true,
1051
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1052
	{ .name	    = "fsync",    .errmsg = true,
1053
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1054
	{ .name	    = "ftruncate", .errmsg = true,
1055
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1056 1057
	{ .name	    = "futex",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_FUTEX_OP, /* op */ }, },
1058
	{ .name	    = "futimesat", .errmsg = true,
1059 1060
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1061
	{ .name	    = "getdents",   .errmsg = true,
1062
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1063
	{ .name	    = "getdents64", .errmsg = true,
1064
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1065 1066
	{ .name	    = "getitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
	{ .name	    = "getrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
1067 1068 1069 1070
	{ .name	    = "getxattr",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "inotify_add_watch",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_FILENAME, /* pathname */ }, },
1071
	{ .name	    = "ioctl",	    .errmsg = true,
1072
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
1073 1074 1075 1076
#if defined(__i386__) || defined(__x86_64__)
/*
 * FIXME: Make this available to all arches.
 */
1077 1078 1079
			     [1] = SCA_STRHEXARRAY, /* cmd */
			     [2] = SCA_HEX, /* arg */ },
	  .arg_parm	 = { [1] = &strarray__tioctls, /* cmd */ }, },
1080 1081 1082
#else
			     [2] = SCA_HEX, /* arg */ }, },
#endif
1083
	{ .name	    = "keyctl",	    .errmsg = true, STRARRAY(0, option, keyctl_options), },
1084 1085
	{ .name	    = "kill",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
1086 1087 1088 1089
	{ .name	    = "lchown",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "lgetxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1090
	{ .name	    = "linkat",	    .errmsg = true,
1091
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */ }, },
1092 1093
	{ .name	    = "listxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1094 1095 1096 1097
	{ .name	    = "llistxattr", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "lremovexattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1098 1099 1100 1101
	{ .name	    = "lseek",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_STRARRAY, /* whence */ },
	  .arg_parm	 = { [2] = &strarray__whences, /* whence */ }, },
1102 1103
	{ .name	    = "lsetxattr",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1104 1105
	{ .name	    = "lstat",	    .errmsg = true, .alias = "newlstat",
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1106 1107
	{ .name	    = "lsxattr",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1108 1109 1110
	{ .name     = "madvise",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX,	 /* start */
			     [2] = SCA_MADV_BHV, /* behavior */ }, },
1111 1112
	{ .name	    = "mkdir",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1113
	{ .name	    = "mkdirat",    .errmsg = true,
1114 1115 1116 1117
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "mknod",      .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1118
	{ .name	    = "mknodat",    .errmsg = true,
1119 1120
	  .arg_scnprintf = { [0] = SCA_FDAT, /* fd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1121 1122 1123 1124
	{ .name	    = "mlock",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
	{ .name	    = "mlockall",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1125
	{ .name	    = "mmap",	    .hexret = true,
1126
	  .arg_scnprintf = { [0] = SCA_HEX,	  /* addr */
1127
			     [2] = SCA_MMAP_PROT, /* prot */
1128 1129
			     [3] = SCA_MMAP_FLAGS, /* flags */
			     [4] = SCA_FD, 	  /* fd */ }, },
1130
	{ .name	    = "mprotect",   .errmsg = true,
1131 1132
	  .arg_scnprintf = { [0] = SCA_HEX, /* start */
			     [2] = SCA_MMAP_PROT, /* prot */ }, },
1133 1134
	{ .name	    = "mq_unlink", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* u_name */ }, },
1135 1136
	{ .name	    = "mremap",	    .hexret = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */
1137
			     [3] = SCA_MREMAP_FLAGS, /* flags */
1138
			     [4] = SCA_HEX, /* new_addr */ }, },
1139 1140
	{ .name	    = "munlock",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1141 1142
	{ .name	    = "munmap",	    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_HEX, /* addr */ }, },
1143
	{ .name	    = "name_to_handle_at", .errmsg = true,
1144
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1145
	{ .name	    = "newfstatat", .errmsg = true,
1146 1147
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
1148
	{ .name	    = "open",	    .errmsg = true,
1149 1150
	  .arg_scnprintf = { [0] = SCA_FILENAME,   /* filename */
			     [1] = SCA_OPEN_FLAGS, /* flags */ }, },
1151
	{ .name	    = "open_by_handle_at", .errmsg = true,
1152 1153
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
1154
	{ .name	    = "openat",	    .errmsg = true,
1155
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
1156
			     [1] = SCA_FILENAME, /* filename */
1157
			     [2] = SCA_OPEN_FLAGS, /* flags */ }, },
1158 1159 1160 1161 1162
	{ .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 */ }, },
1163 1164
	{ .name	    = "pipe2",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_PIPE_FLAGS, /* flags */ }, },
1165 1166
	{ .name	    = "poll",	    .errmsg = true, .timeout = true, },
	{ .name	    = "ppoll",	    .errmsg = true, .timeout = true, },
1167
	{ .name	    = "pread",	    .errmsg = true, .alias = "pread64",
1168
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1169
	{ .name	    = "preadv",	    .errmsg = true, .alias = "pread",
1170
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1171
	{ .name	    = "prlimit64",  .errmsg = true, STRARRAY(1, resource, rlimit_resources), },
1172
	{ .name	    = "pwrite",	    .errmsg = true, .alias = "pwrite64",
1173
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1174
	{ .name	    = "pwritev",    .errmsg = true,
1175
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1176
	{ .name	    = "read",	    .errmsg = true,
1177
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1178 1179
	{ .name	    = "readlink",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
1180
	{ .name	    = "readlinkat", .errmsg = true,
1181 1182
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
1183
	{ .name	    = "readv",	    .errmsg = true,
1184
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1185
	{ .name	    = "recvfrom",   .errmsg = true,
1186 1187
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1188
	{ .name	    = "recvmmsg",   .errmsg = true,
1189 1190
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1191
	{ .name	    = "recvmsg",    .errmsg = true,
1192 1193
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_MSG_FLAGS, /* flags */ }, },
1194 1195
	{ .name	    = "removexattr", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1196
	{ .name	    = "renameat",   .errmsg = true,
1197
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1198 1199
	{ .name	    = "rmdir",    .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1200 1201
	{ .name	    = "rt_sigaction", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_SIGNUM, /* sig */ }, },
1202
	{ .name	    = "rt_sigprocmask",  .errmsg = true, STRARRAY(0, how, sighow), },
1203 1204 1205 1206
	{ .name	    = "rt_sigqueueinfo", .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
	{ .name	    = "rt_tgsigqueueinfo", .errmsg = true,
	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
1207
	{ .name	    = "select",	    .errmsg = true, .timeout = true, },
1208
	{ .name	    = "sendmmsg",    .errmsg = true,
1209 1210
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1211
	{ .name	    = "sendmsg",    .errmsg = true,
1212 1213
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [2] = SCA_MSG_FLAGS, /* flags */ }, },
1214
	{ .name	    = "sendto",	    .errmsg = true,
1215 1216
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */
			     [3] = SCA_MSG_FLAGS, /* flags */ }, },
1217 1218
	{ .name	    = "setitimer",  .errmsg = true, STRARRAY(0, which, itimers), },
	{ .name	    = "setrlimit",  .errmsg = true, STRARRAY(0, resource, rlimit_resources), },
1219 1220
	{ .name	    = "setxattr",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1221
	{ .name	    = "shutdown",   .errmsg = true,
1222
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1223
	{ .name	    = "socket",	    .errmsg = true,
1224 1225
	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
			     [1] = SCA_SK_TYPE, /* type */ },
1226 1227 1228 1229
	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
	{ .name	    = "socketpair", .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_STRARRAY, /* family */
			     [1] = SCA_SK_TYPE, /* type */ },
1230
	  .arg_parm	 = { [0] = &strarray__socket_families, /* family */ }, },
1231 1232
	{ .name	    = "stat",	    .errmsg = true, .alias = "newstat",
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* pathname */ }, },
1233 1234 1235 1236 1237 1238
	{ .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 */ }, },
1239
	{ .name	    = "symlinkat",  .errmsg = true,
1240
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */ }, },
1241 1242 1243 1244
	{ .name	    = "tgkill",	    .errmsg = true,
	  .arg_scnprintf = { [2] = SCA_SIGNUM, /* sig */ }, },
	{ .name	    = "tkill",	    .errmsg = true,
	  .arg_scnprintf = { [1] = SCA_SIGNUM, /* sig */ }, },
1245 1246
	{ .name	    = "truncate",   .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* path */ }, },
1247
	{ .name	    = "uname",	    .errmsg = true, .alias = "newuname", },
1248
	{ .name	    = "unlinkat",   .errmsg = true,
1249 1250 1251 1252
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dfd */
			     [1] = SCA_FILENAME, /* pathname */ }, },
	{ .name	    = "utime",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1253
	{ .name	    = "utimensat",  .errmsg = true,
1254 1255 1256 1257
	  .arg_scnprintf = { [0] = SCA_FDAT, /* dirfd */
			     [1] = SCA_FILENAME, /* filename */ }, },
	{ .name	    = "utimes",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FILENAME, /* filename */ }, },
1258 1259
	{ .name	    = "vmsplice",  .errmsg = true,
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1260
	{ .name	    = "write",	    .errmsg = true,
1261
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
1262
	{ .name	    = "writev",	    .errmsg = true,
1263
	  .arg_scnprintf = { [0] = SCA_FD, /* fd */ }, },
A
Arnaldo Carvalho de Melo 已提交
1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279
};

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;
1280 1281
	int		    nr_args;
	struct format_field *args;
A
Arnaldo Carvalho de Melo 已提交
1282
	const char	    *name;
1283
	bool		    is_exit;
A
Arnaldo Carvalho de Melo 已提交
1284
	struct syscall_fmt  *fmt;
1285
	size_t		    (**arg_scnprintf)(char *bf, size_t size, struct syscall_arg *arg);
1286
	void		    **arg_parm;
A
Arnaldo Carvalho de Melo 已提交
1287 1288
};

1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299
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);
1300
	return printed + fprintf(fp, "): ");
1301 1302
}

1303 1304 1305 1306 1307
/**
 * 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.
 */
1308 1309 1310 1311
struct thread_trace {
	u64		  entry_time;
	u64		  exit_time;
	bool		  entry_pending;
1312
	unsigned long	  nr_events;
1313
	unsigned long	  pfmaj, pfmin;
1314
	char		  *entry_str;
1315
	double		  runtime_ms;
1316 1317 1318 1319
        struct {
		unsigned long ptr;
		int           entry_str_pos;
	} filename;
1320 1321 1322 1323
	struct {
		int	  max;
		char	  **table;
	} paths;
1324 1325

	struct intlist *syscall_stats;
1326 1327 1328 1329
};

static struct thread_trace *thread_trace__new(void)
{
1330 1331 1332 1333 1334
	struct thread_trace *ttrace =  zalloc(sizeof(struct thread_trace));

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

1335 1336
	ttrace->syscall_stats = intlist__new(NULL);

1337
	return ttrace;
1338 1339
}

1340
static struct thread_trace *thread__trace(struct thread *thread, FILE *fp)
1341
{
1342 1343
	struct thread_trace *ttrace;

1344 1345 1346
	if (thread == NULL)
		goto fail;

1347 1348
	if (thread__priv(thread) == NULL)
		thread__set_priv(thread, thread_trace__new());
1349

1350
	if (thread__priv(thread) == NULL)
1351 1352
		goto fail;

1353
	ttrace = thread__priv(thread);
1354 1355 1356
	++ttrace->nr_events;

	return ttrace;
1357
fail:
1358
	color_fprintf(fp, PERF_COLOR_RED,
1359 1360 1361 1362
		      "WARNING: not enough memory, dropping samples!\n");
	return NULL;
}

1363 1364 1365
#define TRACE_PFMAJ		(1 << 0)
#define TRACE_PFMIN		(1 << 1)

1366 1367
static const size_t trace__entry_str_size = 2048;

A
Arnaldo Carvalho de Melo 已提交
1368
struct trace {
1369
	struct perf_tool	tool;
1370 1371 1372 1373
	struct {
		int		machine;
		int		open_id;
	}			audit;
A
Arnaldo Carvalho de Melo 已提交
1374 1375 1376
	struct {
		int		max;
		struct syscall  *table;
1377
		struct {
1378 1379
			struct perf_evsel *sys_enter,
					  *sys_exit;
1380
		}		events;
A
Arnaldo Carvalho de Melo 已提交
1381
	} syscalls;
1382
	struct record_opts	opts;
1383
	struct perf_evlist	*evlist;
1384
	struct machine		*host;
1385
	struct thread		*current;
1386
	u64			base_time;
1387
	FILE			*output;
1388
	unsigned long		nr_events;
1389
	struct strlist		*ev_qualifier;
1390 1391 1392 1393
	struct {
		size_t		nr;
		int		*entries;
	}			ev_qualifier_ids;
1394
	const char 		*last_vfs_getname;
1395 1396
	struct intlist		*tid_list;
	struct intlist		*pid_list;
1397 1398 1399 1400
	struct {
		size_t		nr;
		pid_t		*entries;
	}			filter_pids;
1401 1402 1403 1404 1405 1406 1407 1408 1409
	double			duration_filter;
	double			runtime_ms;
	struct {
		u64		vfs_getname,
				proc_getname;
	} stats;
	bool			not_ev_qualifier;
	bool			live;
	bool			full_time;
1410
	bool			sched;
1411
	bool			multiple_threads;
1412
	bool			summary;
D
David Ahern 已提交
1413
	bool			summary_only;
1414
	bool			show_comm;
1415
	bool			show_tool_stats;
1416
	bool			trace_syscalls;
1417
	bool			force;
1418
	bool			vfs_getname;
1419
	int			trace_pgfaults;
A
Arnaldo Carvalho de Melo 已提交
1420 1421
};

1422
static int trace__set_fd_pathname(struct thread *thread, int fd, const char *pathname)
1423
{
1424
	struct thread_trace *ttrace = thread__priv(thread);
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447

	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;
}

1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
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);
}

1474 1475
static const char *thread__fd_path(struct thread *thread, int fd,
				   struct trace *trace)
1476
{
1477
	struct thread_trace *ttrace = thread__priv(thread);
1478 1479 1480 1481 1482 1483 1484

	if (ttrace == NULL)
		return NULL;

	if (fd < 0)
		return NULL;

1485
	if ((fd > ttrace->paths.max || ttrace->paths.table[fd] == NULL)) {
1486 1487 1488
		if (!trace->live)
			return NULL;
		++trace->stats.proc_getname;
1489
		if (thread__read_fd_path(thread, fd))
1490 1491
			return NULL;
	}
1492 1493 1494 1495 1496 1497 1498 1499 1500

	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);
1501
	const char *path = thread__fd_path(arg->thread, fd, arg->trace);
1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513

	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);
1514
	struct thread_trace *ttrace = thread__priv(arg->thread);
1515

1516 1517
	if (ttrace && fd >= 0 && fd <= ttrace->paths.max)
		zfree(&ttrace->paths.table[fd]);
1518 1519 1520 1521

	return printed;
}

1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542
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;
}

1543 1544 1545 1546 1547
static bool trace__filter_duration(struct trace *trace, double t)
{
	return t < (trace->duration_filter * NSEC_PER_MSEC);
}

1548 1549 1550 1551
static size_t trace__fprintf_tstamp(struct trace *trace, u64 tstamp, FILE *fp)
{
	double ts = (double)(tstamp - trace->base_time) / NSEC_PER_MSEC;

1552
	return fprintf(fp, "%10.3f ", ts);
1553 1554
}

1555
static bool done = false;
1556
static bool interrupted = false;
1557

1558
static void sig_handler(int sig)
1559 1560
{
	done = true;
1561
	interrupted = sig == SIGINT;
1562 1563
}

1564
static size_t trace__fprintf_entry_head(struct trace *trace, struct thread *thread,
1565
					u64 duration, u64 tstamp, FILE *fp)
1566 1567
{
	size_t printed = trace__fprintf_tstamp(trace, tstamp, fp);
1568
	printed += fprintf_duration(duration, fp);
1569

1570 1571
	if (trace->multiple_threads) {
		if (trace->show_comm)
1572
			printed += fprintf(fp, "%.14s/", thread__comm_str(thread));
1573
		printed += fprintf(fp, "%d ", thread->tid);
1574
	}
1575 1576 1577 1578

	return printed;
}

1579
static int trace__process_event(struct trace *trace, struct machine *machine,
1580
				union perf_event *event, struct perf_sample *sample)
1581 1582 1583 1584 1585
{
	int ret = 0;

	switch (event->header.type) {
	case PERF_RECORD_LOST:
1586
		color_fprintf(trace->output, PERF_COLOR_RED,
1587
			      "LOST %" PRIu64 " events!\n", event->lost.lost);
1588
		ret = machine__process_lost_event(machine, event, sample);
1589
	default:
1590
		ret = machine__process_event(machine, event, sample);
1591 1592 1593 1594 1595 1596
		break;
	}

	return ret;
}

1597
static int trace__tool_process(struct perf_tool *tool,
1598
			       union perf_event *event,
1599
			       struct perf_sample *sample,
1600 1601
			       struct machine *machine)
{
1602
	struct trace *trace = container_of(tool, struct trace, tool);
1603
	return trace__process_event(trace, machine, event, sample);
1604 1605 1606 1607
}

static int trace__symbols_init(struct trace *trace, struct perf_evlist *evlist)
{
1608
	int err = symbol__init(NULL);
1609 1610 1611 1612

	if (err)
		return err;

1613 1614 1615
	trace->host = machine__new_host();
	if (trace->host == NULL)
		return -ENOMEM;
1616

1617
	if (trace_event__register_resolver(trace->host, machine__resolve_kernel_addr) < 0)
1618 1619
		return -errno;

1620
	err = __machine__synthesize_threads(trace->host, &trace->tool, &trace->opts.target,
1621 1622
					    evlist->threads, trace__tool_process, false,
					    trace->opts.proc_map_timeout);
1623 1624 1625 1626 1627 1628
	if (err)
		symbol__exit();

	return err;
}

1629 1630 1631 1632 1633
static int syscall__set_arg_fmts(struct syscall *sc)
{
	struct format_field *field;
	int idx = 0;

1634
	sc->arg_scnprintf = calloc(sc->nr_args, sizeof(void *));
1635 1636 1637
	if (sc->arg_scnprintf == NULL)
		return -1;

1638 1639 1640
	if (sc->fmt)
		sc->arg_parm = sc->fmt->arg_parm;

1641
	for (field = sc->args; field; field = field->next) {
1642 1643 1644
		if (sc->fmt && sc->fmt->arg_scnprintf[idx])
			sc->arg_scnprintf[idx] = sc->fmt->arg_scnprintf[idx];
		else if (field->flags & FIELD_IS_POINTER)
1645 1646 1647 1648 1649 1650 1651
			sc->arg_scnprintf[idx] = syscall_arg__scnprintf_hex;
		++idx;
	}

	return 0;
}

A
Arnaldo Carvalho de Melo 已提交
1652 1653 1654 1655
static int trace__read_syscall_info(struct trace *trace, int id)
{
	char tp_name[128];
	struct syscall *sc;
1656
	const char *name = audit_syscall_to_name(id, trace->audit.machine);
1657 1658 1659

	if (name == NULL)
		return -1;
A
Arnaldo Carvalho de Melo 已提交
1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678

	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;
1679
	sc->name = name;
1680

1681
	sc->fmt  = syscall_fmt__find(sc->name);
A
Arnaldo Carvalho de Melo 已提交
1682

1683
	snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->name);
1684
	sc->tp_format = trace_event__tp_format("syscalls", tp_name);
1685 1686 1687

	if (sc->tp_format == NULL && sc->fmt && sc->fmt->alias) {
		snprintf(tp_name, sizeof(tp_name), "sys_enter_%s", sc->fmt->alias);
1688
		sc->tp_format = trace_event__tp_format("syscalls", tp_name);
1689
	}
A
Arnaldo Carvalho de Melo 已提交
1690

1691 1692 1693
	if (sc->tp_format == NULL)
		return -1;

1694 1695 1696 1697 1698 1699 1700 1701
	sc->args = sc->tp_format->format.fields;
	sc->nr_args = sc->tp_format->format.nr_fields;
	/* drop nr field - not relevant here; does not exist on older kernels */
	if (sc->args && strcmp(sc->args->name, "nr") == 0) {
		sc->args = sc->args->next;
		--sc->nr_args;
	}

1702 1703
	sc->is_exit = !strcmp(name, "exit_group") || !strcmp(name, "exit");

1704
	return syscall__set_arg_fmts(sc);
A
Arnaldo Carvalho de Melo 已提交
1705 1706
}

1707 1708
static int trace__validate_ev_qualifier(struct trace *trace)
{
1709
	int err = 0, i;
1710 1711
	struct str_node *pos;

1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724
	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;

1725 1726
	strlist__for_each(pos, trace->ev_qualifier) {
		const char *sc = pos->s;
1727
		int id = audit_name_to_syscall(sc, trace->audit.machine);
1728

1729
		if (id < 0) {
1730 1731 1732 1733 1734 1735 1736 1737 1738
			if (err == 0) {
				fputs("Error:\tInvalid syscall ", trace->output);
				err = -EINVAL;
			} else {
				fputs(", ", trace->output);
			}

			fputs(sc, trace->output);
		}
1739 1740

		trace->ev_qualifier_ids.entries[i++] = id;
1741 1742 1743 1744 1745
	}

	if (err < 0) {
		fputs("\nHint:\ttry 'perf list syscalls:sys_enter_*'"
		      "\nHint:\tand: 'man syscalls'\n", trace->output);
1746 1747
		zfree(&trace->ev_qualifier_ids.entries);
		trace->ev_qualifier_ids.nr = 0;
1748
	}
1749
out:
1750 1751 1752
	return err;
}

1753 1754 1755 1756 1757 1758 1759 1760 1761
/*
 * 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
 */

1762
static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size,
1763
				      unsigned char *args, struct trace *trace,
1764
				      struct thread *thread)
A
Arnaldo Carvalho de Melo 已提交
1765 1766
{
	size_t printed = 0;
1767 1768
	unsigned char *p;
	unsigned long val;
A
Arnaldo Carvalho de Melo 已提交
1769

1770
	if (sc->args != NULL) {
A
Arnaldo Carvalho de Melo 已提交
1771
		struct format_field *field;
1772 1773
		u8 bit = 1;
		struct syscall_arg arg = {
1774 1775 1776 1777
			.idx	= 0,
			.mask	= 0,
			.trace  = trace,
			.thread = thread,
1778
		};
1779

1780
		for (field = sc->args; field;
1781 1782
		     field = field->next, ++arg.idx, bit <<= 1) {
			if (arg.mask & bit)
1783
				continue;
1784 1785 1786 1787 1788

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

1789 1790 1791 1792 1793
			/*
 			 * Suppress this argument if its value is zero and
 			 * and we don't have a string associated in an
 			 * strarray for it.
 			 */
1794
			if (val == 0 &&
1795 1796 1797
			    !(sc->arg_scnprintf &&
			      sc->arg_scnprintf[arg.idx] == SCA_STRARRAY &&
			      sc->arg_parm[arg.idx]))
1798 1799
				continue;

1800
			printed += scnprintf(bf + printed, size - printed,
1801
					     "%s%s: ", printed ? ", " : "", field->name);
1802
			if (sc->arg_scnprintf && sc->arg_scnprintf[arg.idx]) {
1803
				arg.val = val;
1804 1805
				if (sc->arg_parm)
					arg.parm = sc->arg_parm[arg.idx];
1806 1807
				printed += sc->arg_scnprintf[arg.idx](bf + printed,
								      size - printed, &arg);
1808
			} else {
1809
				printed += scnprintf(bf + printed, size - printed,
1810
						     "%ld", val);
1811
			}
A
Arnaldo Carvalho de Melo 已提交
1812 1813
		}
	} else {
1814 1815
		int i = 0;

A
Arnaldo Carvalho de Melo 已提交
1816
		while (i < 6) {
1817 1818 1819
			/* special care for unaligned accesses */
			p = args + sizeof(unsigned long) * i;
			memcpy(&val, p, sizeof(val));
1820 1821
			printed += scnprintf(bf + printed, size - printed,
					     "%sarg%d: %ld",
1822
					     printed ? ", " : "", i, val);
A
Arnaldo Carvalho de Melo 已提交
1823 1824 1825 1826 1827 1828 1829
			++i;
		}
	}

	return printed;
}

1830
typedef int (*tracepoint_handler)(struct trace *trace, struct perf_evsel *evsel,
1831
				  union perf_event *event,
1832 1833 1834
				  struct perf_sample *sample);

static struct syscall *trace__syscall_info(struct trace *trace,
1835
					   struct perf_evsel *evsel, int id)
1836 1837 1838
{

	if (id < 0) {
1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854

		/*
		 * 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);
		}
1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867
		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:
1868 1869 1870 1871 1872 1873
	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);
	}
1874 1875 1876
	return NULL;
}

1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902
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);
}

1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925
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;
}

1926
static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
1927
			    union perf_event *event __maybe_unused,
1928 1929
			    struct perf_sample *sample)
{
1930
	char *msg;
1931
	void *args;
1932
	size_t printed = 0;
1933
	struct thread *thread;
1934
	int id = perf_evsel__sc_tp_uint(evsel, id, sample), err = -1;
1935
	struct syscall *sc = trace__syscall_info(trace, evsel, id);
1936 1937 1938 1939
	struct thread_trace *ttrace;

	if (sc == NULL)
		return -1;
1940

1941
	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
1942
	ttrace = thread__trace(thread, trace->output);
1943
	if (ttrace == NULL)
1944
		goto out_put;
1945

1946
	args = perf_evsel__sc_tp_ptr(evsel, args, sample);
1947 1948

	if (ttrace->entry_str == NULL) {
1949
		ttrace->entry_str = malloc(trace__entry_str_size);
1950
		if (!ttrace->entry_str)
1951
			goto out_put;
1952 1953
	}

1954
	if (!trace->summary_only)
1955
		trace__printf_interrupted_entry(trace, sample);
1956

1957 1958
	ttrace->entry_time = sample->time;
	msg = ttrace->entry_str;
1959
	printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
1960

1961
	printed += syscall__scnprintf_args(sc, msg + printed, trace__entry_str_size - printed,
1962
					   args, trace, thread);
1963

1964
	if (sc->is_exit) {
D
David Ahern 已提交
1965
		if (!trace->duration_filter && !trace->summary_only) {
1966 1967
			trace__fprintf_entry_head(trace, thread, 1, sample->time, trace->output);
			fprintf(trace->output, "%-70s\n", ttrace->entry_str);
1968
		}
1969 1970
	} else
		ttrace->entry_pending = true;
1971

1972 1973 1974 1975
	if (trace->current != thread) {
		thread__put(trace->current);
		trace->current = thread__get(thread);
	}
1976 1977 1978 1979
	err = 0;
out_put:
	thread__put(thread);
	return err;
1980 1981 1982
}

static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel,
1983
			   union perf_event *event __maybe_unused,
1984 1985
			   struct perf_sample *sample)
{
1986
	long ret;
1987
	u64 duration = 0;
1988
	struct thread *thread;
1989
	int id = perf_evsel__sc_tp_uint(evsel, id, sample), err = -1;
1990
	struct syscall *sc = trace__syscall_info(trace, evsel, id);
1991 1992 1993 1994
	struct thread_trace *ttrace;

	if (sc == NULL)
		return -1;
1995

1996
	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
1997
	ttrace = thread__trace(thread, trace->output);
1998
	if (ttrace == NULL)
1999
		goto out_put;
2000

2001 2002 2003
	if (trace->summary)
		thread__update_stats(ttrace, id, sample);

2004
	ret = perf_evsel__sc_tp_uint(evsel, ret, sample);
2005

2006 2007 2008 2009 2010 2011
	if (id == trace->audit.open_id && ret >= 0 && trace->last_vfs_getname) {
		trace__set_fd_pathname(thread, ret, trace->last_vfs_getname);
		trace->last_vfs_getname = NULL;
		++trace->stats.vfs_getname;
	}

2012 2013
	ttrace->exit_time = sample->time;

2014
	if (ttrace->entry_time) {
2015
		duration = sample->time - ttrace->entry_time;
2016 2017 2018 2019
		if (trace__filter_duration(trace, duration))
			goto out;
	} else if (trace->duration_filter)
		goto out;
2020

D
David Ahern 已提交
2021 2022 2023
	if (trace->summary_only)
		goto out;

2024
	trace__fprintf_entry_head(trace, thread, duration, sample->time, trace->output);
2025 2026

	if (ttrace->entry_pending) {
2027
		fprintf(trace->output, "%-70s", ttrace->entry_str);
2028
	} else {
2029 2030 2031
		fprintf(trace->output, " ... [");
		color_fprintf(trace->output, PERF_COLOR_YELLOW, "continued");
		fprintf(trace->output, "]: %s()", sc->name);
2032 2033
	}

2034 2035
	if (sc->fmt == NULL) {
signed_print:
2036
		fprintf(trace->output, ") = %ld", ret);
2037
	} else if (ret < 0 && sc->fmt->errmsg) {
2038
		char bf[STRERR_BUFSIZE];
2039 2040 2041
		const char *emsg = strerror_r(-ret, bf, sizeof(bf)),
			   *e = audit_errno_to_name(-ret);

2042
		fprintf(trace->output, ") = -1 %s %s", e, emsg);
2043
	} else if (ret == 0 && sc->fmt->timeout)
2044
		fprintf(trace->output, ") = 0 Timeout");
2045
	else if (sc->fmt->hexret)
2046
		fprintf(trace->output, ") = %#lx", ret);
2047
	else
2048
		goto signed_print;
2049

2050
	fputc('\n', trace->output);
2051
out:
2052
	ttrace->entry_pending = false;
2053 2054 2055 2056
	err = 0;
out_put:
	thread__put(thread);
	return err;
2057 2058
}

2059
static int trace__vfs_getname(struct trace *trace, struct perf_evsel *evsel,
2060
			      union perf_event *event __maybe_unused,
2061 2062
			      struct perf_sample *sample)
{
2063 2064 2065 2066 2067 2068 2069
	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;
	const char *filename;

2070
	trace->last_vfs_getname = perf_evsel__rawptr(evsel, sample, "pathname");
2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101

	if (!thread)
		goto out;

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

	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;

	filename = trace->last_vfs_getname;
	filename_len = strlen(filename);
	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:
2102 2103 2104
	return 0;
}

2105
static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evsel,
2106
				     union perf_event *event __maybe_unused,
2107 2108 2109 2110
				     struct perf_sample *sample)
{
        u64 runtime = perf_evsel__intval(evsel, sample, "runtime");
	double runtime_ms = (double)runtime / NSEC_PER_MSEC;
2111
	struct thread *thread = machine__findnew_thread(trace->host,
2112 2113
							sample->pid,
							sample->tid);
2114
	struct thread_trace *ttrace = thread__trace(thread, trace->output);
2115 2116 2117 2118 2119 2120

	if (ttrace == NULL)
		goto out_dump;

	ttrace->runtime_ms += runtime_ms;
	trace->runtime_ms += runtime_ms;
2121
	thread__put(thread);
2122 2123 2124
	return 0;

out_dump:
2125
	fprintf(trace->output, "%s: comm=%s,pid=%u,runtime=%" PRIu64 ",vruntime=%" PRIu64 ")\n",
2126 2127 2128 2129 2130
	       evsel->name,
	       perf_evsel__strval(evsel, sample, "comm"),
	       (pid_t)perf_evsel__intval(evsel, sample, "pid"),
	       runtime,
	       perf_evsel__intval(evsel, sample, "vruntime"));
2131
	thread__put(thread);
2132 2133 2134
	return 0;
}

2135 2136 2137 2138 2139 2140
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);
2141 2142 2143 2144 2145

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

	fprintf(trace->output, "%s:", evsel->name);
2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156

	if (evsel->tp_format) {
		event_format__fprintf(evsel->tp_format, sample->cpu,
				      sample->raw_data, sample->raw_size,
				      trace->output);
	}

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

2157 2158 2159 2160 2161 2162 2163 2164 2165
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)
2166
		fprintf(f, "%s+0x%" PRIx64, al->sym->name,
2167 2168
			al->addr - al->sym->start);
	else if (al->map)
2169
		fprintf(f, "0x%" PRIx64, al->addr);
2170
	else
2171
		fprintf(f, "0x%" PRIx64, sample->addr);
2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182
}

static int trace__pgfault(struct trace *trace,
			  struct perf_evsel *evsel,
			  union perf_event *event,
			  struct perf_sample *sample)
{
	struct thread *thread;
	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
	struct addr_location al;
	char map_type = 'd';
2183
	struct thread_trace *ttrace;
2184
	int err = -1;
2185 2186

	thread = machine__findnew_thread(trace->host, sample->pid, sample->tid);
2187 2188
	ttrace = thread__trace(thread, trace->output);
	if (ttrace == NULL)
2189
		goto out_put;
2190 2191 2192 2193 2194 2195 2196

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

	if (trace->summary_only)
2197
		goto out;
2198

2199
	thread__find_addr_location(thread, cpumode, MAP__FUNCTION,
2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211
			      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, "] => ");

2212
	thread__find_addr_location(thread, cpumode, MAP__VARIABLE,
2213 2214 2215
				   sample->addr, &al);

	if (!al.map) {
2216
		thread__find_addr_location(thread, cpumode,
2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227
					   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);
2228 2229 2230 2231 2232
out:
	err = 0;
out_put:
	thread__put(thread);
	return err;
2233 2234
}

2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246
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;
}

2247
static int trace__process_sample(struct perf_tool *tool,
2248
				 union perf_event *event,
2249 2250 2251 2252 2253 2254 2255
				 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;

2256
	tracepoint_handler handler = evsel->handler;
2257

2258 2259 2260
	if (skip_sample(trace, sample))
		return 0;

2261
	if (!trace->full_time && trace->base_time == 0)
2262 2263
		trace->base_time = sample->time;

2264 2265
	if (handler) {
		++trace->nr_events;
2266
		handler(trace, evsel, event, sample);
2267
	}
2268 2269 2270 2271

	return err;
}

2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292
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;
}

2293
static int trace__record(struct trace *trace, int argc, const char **argv)
D
David Ahern 已提交
2294 2295 2296 2297 2298 2299 2300 2301 2302 2303
{
	unsigned int rec_argc, i, j;
	const char **rec_argv;
	const char * const record_args[] = {
		"record",
		"-R",
		"-m", "1024",
		"-c", "1",
	};

2304 2305 2306 2307 2308 2309 2310
	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);

2311
	/* +1 is for the event string below */
2312 2313
	rec_argc = ARRAY_SIZE(record_args) + sc_args_nr + 1 +
		majpf_args_nr + minpf_args_nr + argc;
D
David Ahern 已提交
2314 2315 2316 2317 2318
	rec_argv = calloc(rec_argc + 1, sizeof(char *));

	if (rec_argv == NULL)
		return -ENOMEM;

2319
	j = 0;
D
David Ahern 已提交
2320
	for (i = 0; i < ARRAY_SIZE(record_args); i++)
2321 2322
		rec_argv[j++] = record_args[i];

2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335
	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;
		}
2336 2337
	}

2338 2339 2340 2341 2342 2343 2344 2345 2346 2347
	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
2348

2349
	return cmd_record(j, rec_argv, NULL);
D
David Ahern 已提交
2350 2351
}

2352 2353
static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp);

2354
static bool perf_evlist__add_vfs_getname(struct perf_evlist *evlist)
2355
{
2356
	struct perf_evsel *evsel = perf_evsel__newtp("probe", "vfs_getname");
2357
	if (evsel == NULL)
2358
		return false;
2359 2360 2361

	if (perf_evsel__field(evsel, "pathname") == NULL) {
		perf_evsel__delete(evsel);
2362
		return false;
2363 2364
	}

2365
	evsel->handler = trace__vfs_getname;
2366
	perf_evlist__add(evlist, evsel);
2367
	return true;
2368 2369
}

2370 2371 2372 2373 2374 2375 2376 2377 2378 2379
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;
2380
	attr.sample_period = 1;
2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393

	event_attr_init(&attr);

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

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

	return 0;
}

2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423
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 (!trace->full_time && trace->base_time == 0)
		trace->base_time = sample->time;

	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;
	}

	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);
	}
}

2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446
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);

2447 2448
	trace->syscalls.events.sys_enter = sys_enter;
	trace->syscalls.events.sys_exit  = sys_exit;
2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460

	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;
}

2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480
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;
}
2481

2482
static int trace__run(struct trace *trace, int argc, const char **argv)
A
Arnaldo Carvalho de Melo 已提交
2483
{
2484
	struct perf_evlist *evlist = trace->evlist;
2485
	struct perf_evsel *evsel;
2486 2487
	int err = -1, i;
	unsigned long before;
2488
	const bool forks = argc > 0;
2489
	bool draining = false;
A
Arnaldo Carvalho de Melo 已提交
2490

2491 2492
	trace->live = true;

2493
	if (trace->trace_syscalls && trace__add_syscall_newtp(trace))
2494
		goto out_error_raw_syscalls;
A
Arnaldo Carvalho de Melo 已提交
2495

2496
	if (trace->trace_syscalls)
2497
		trace->vfs_getname = perf_evlist__add_vfs_getname(evlist);
2498

2499
	if ((trace->trace_pgfaults & TRACE_PFMAJ) &&
2500
	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MAJ)) {
2501
		goto out_error_mem;
2502
	}
2503 2504 2505

	if ((trace->trace_pgfaults & TRACE_PFMIN) &&
	    perf_evlist__add_pgfault(evlist, PERF_COUNT_SW_PAGE_FAULTS_MIN))
2506
		goto out_error_mem;
2507

2508
	if (trace->sched &&
2509 2510 2511
	    perf_evlist__add_newtp(evlist, "sched", "sched_stat_runtime",
				   trace__sched_stat_runtime))
		goto out_error_sched_stat_runtime;
2512

A
Arnaldo Carvalho de Melo 已提交
2513 2514
	err = perf_evlist__create_maps(evlist, &trace->opts.target);
	if (err < 0) {
2515
		fprintf(trace->output, "Problems parsing the target to trace, check your options!\n");
A
Arnaldo Carvalho de Melo 已提交
2516 2517 2518
		goto out_delete_evlist;
	}

2519 2520
	err = trace__symbols_init(trace, evlist);
	if (err < 0) {
2521
		fprintf(trace->output, "Problems initializing symbol libraries!\n");
2522
		goto out_delete_evlist;
2523 2524
	}

2525
	perf_evlist__config(evlist, &trace->opts);
A
Arnaldo Carvalho de Melo 已提交
2526

2527 2528 2529 2530
	signal(SIGCHLD, sig_handler);
	signal(SIGINT, sig_handler);

	if (forks) {
2531
		err = perf_evlist__prepare_workload(evlist, &trace->opts.target,
2532
						    argv, false, NULL);
2533
		if (err < 0) {
2534
			fprintf(trace->output, "Couldn't run the workload!\n");
2535
			goto out_delete_evlist;
2536 2537 2538
		}
	}

A
Arnaldo Carvalho de Melo 已提交
2539
	err = perf_evlist__open(evlist);
2540 2541
	if (err < 0)
		goto out_error_open;
A
Arnaldo Carvalho de Melo 已提交
2542

2543 2544 2545 2546 2547 2548
	/*
	 * 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.
	 */
2549 2550
	if (trace->filter_pids.nr > 0)
		err = perf_evlist__set_filter_pids(evlist, trace->filter_pids.nr, trace->filter_pids.entries);
2551
	else if (thread_map__pid(evlist->threads, 0) == -1)
2552 2553
		err = perf_evlist__set_filter_pid(evlist, getpid());

2554 2555 2556
	if (err < 0)
		goto out_error_mem;

2557 2558 2559 2560 2561
	if (trace->ev_qualifier_ids.nr > 0) {
		err = trace__set_ev_qualifier_filter(trace);
		if (err < 0)
			goto out_errno;

2562 2563 2564
		pr_debug("event qualifier tracepoint filter: %s\n",
			 trace->syscalls.events.sys_exit->filter);
	}
2565

2566 2567 2568
	err = perf_evlist__apply_filters(evlist, &evsel);
	if (err < 0)
		goto out_error_apply_filters;
2569

J
Jiri Olsa 已提交
2570
	err = perf_evlist__mmap(evlist, trace->opts.mmap_pages, false);
2571 2572
	if (err < 0)
		goto out_error_mmap;
A
Arnaldo Carvalho de Melo 已提交
2573

2574 2575 2576
	if (!target__none(&trace->opts.target))
		perf_evlist__enable(evlist);

2577 2578 2579
	if (forks)
		perf_evlist__start_workload(evlist);

2580
	trace->multiple_threads = thread_map__pid(evlist->threads, 0) == -1 ||
2581 2582
				  evlist->threads->nr > 1 ||
				  perf_evlist__first(evlist)->attr.inherit;
A
Arnaldo Carvalho de Melo 已提交
2583
again:
2584
	before = trace->nr_events;
A
Arnaldo Carvalho de Melo 已提交
2585 2586 2587 2588 2589 2590 2591

	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;

2592
			++trace->nr_events;
A
Arnaldo Carvalho de Melo 已提交
2593 2594 2595

			err = perf_evlist__parse_sample(evlist, event, &sample);
			if (err) {
2596
				fprintf(trace->output, "Can't parse sample, err = %d, skipping...\n", err);
2597
				goto next_event;
A
Arnaldo Carvalho de Melo 已提交
2598 2599
			}

2600
			trace__handle_event(trace, event, &sample);
2601 2602
next_event:
			perf_evlist__mmap_consume(evlist, i);
2603

2604 2605
			if (interrupted)
				goto out_disable;
2606 2607 2608 2609 2610

			if (done && !draining) {
				perf_evlist__disable(evlist);
				draining = true;
			}
A
Arnaldo Carvalho de Melo 已提交
2611 2612 2613
		}
	}

2614
	if (trace->nr_events == before) {
2615
		int timeout = done ? 100 : -1;
2616

2617 2618 2619 2620
		if (!draining && perf_evlist__poll(evlist, timeout) > 0) {
			if (perf_evlist__filter_pollfd(evlist, POLLERR | POLLHUP) == 0)
				draining = true;

2621
			goto again;
2622
		}
2623 2624
	} else {
		goto again;
2625 2626
	}

2627
out_disable:
2628 2629
	thread__zput(trace->current);

2630
	perf_evlist__disable(evlist);
A
Arnaldo Carvalho de Melo 已提交
2631

2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643
	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);
		}
	}
2644

A
Arnaldo Carvalho de Melo 已提交
2645 2646
out_delete_evlist:
	perf_evlist__delete(evlist);
2647
	trace->evlist = NULL;
2648
	trace->live = false;
A
Arnaldo Carvalho de Melo 已提交
2649
	return err;
2650 2651
{
	char errbuf[BUFSIZ];
2652

2653 2654 2655 2656
out_error_sched_stat_runtime:
	debugfs__strerror_open_tp(errno, errbuf, sizeof(errbuf), "sched", "sched_stat_runtime");
	goto out_error;

2657
out_error_raw_syscalls:
2658
	debugfs__strerror_open_tp(errno, errbuf, sizeof(errbuf), "raw_syscalls", "sys_(enter|exit)");
2659 2660
	goto out_error;

2661 2662 2663 2664
out_error_mmap:
	perf_evlist__strerror_mmap(evlist, errno, errbuf, sizeof(errbuf));
	goto out_error;

2665 2666 2667 2668
out_error_open:
	perf_evlist__strerror_open(evlist, errno, errbuf, sizeof(errbuf));

out_error:
2669
	fprintf(trace->output, "%s\n", errbuf);
2670
	goto out_delete_evlist;
2671 2672 2673 2674 2675 2676 2677

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
2678
}
2679 2680 2681
out_error_mem:
	fprintf(trace->output, "Not enough memory to run!\n");
	goto out_delete_evlist;
2682 2683 2684 2685

out_errno:
	fprintf(trace->output, "errno=%d,%s\n", errno, strerror(errno));
	goto out_delete_evlist;
2686
}
A
Arnaldo Carvalho de Melo 已提交
2687

2688 2689 2690
static int trace__replay(struct trace *trace)
{
	const struct perf_evsel_str_handler handlers[] = {
2691
		{ "probe:vfs_getname",	     trace__vfs_getname, },
2692
	};
2693 2694 2695
	struct perf_data_file file = {
		.path  = input_name,
		.mode  = PERF_DATA_MODE_READ,
2696
		.force = trace->force,
2697
	};
2698
	struct perf_session *session;
2699
	struct perf_evsel *evsel;
2700 2701 2702 2703
	int err = -1;

	trace->tool.sample	  = trace__process_sample;
	trace->tool.mmap	  = perf_event__process_mmap;
D
David Ahern 已提交
2704
	trace->tool.mmap2	  = perf_event__process_mmap2;
2705 2706 2707 2708 2709 2710 2711
	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;

2712
	trace->tool.ordered_events = true;
2713 2714 2715 2716 2717
	trace->tool.ordering_requires_timestamps = true;

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

2718
	session = perf_session__new(&file, false, &trace->tool);
2719
	if (session == NULL)
2720
		return -1;
2721

2722
	if (symbol__init(&session->header.env) < 0)
2723 2724
		goto out;

2725 2726
	trace->host = &session->machines.host;

2727 2728 2729 2730
	err = perf_session__set_tracepoints_handlers(session, handlers);
	if (err)
		goto out;

2731 2732
	evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
						     "raw_syscalls:sys_enter");
2733 2734 2735 2736
	/* older kernels have syscalls tp versus raw_syscalls */
	if (evsel == NULL)
		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
							     "syscalls:sys_enter");
2737

2738 2739 2740
	if (evsel &&
	    (perf_evsel__init_syscall_tp(evsel, trace__sys_enter) < 0 ||
	    perf_evsel__init_sc_tp_ptr_field(evsel, args))) {
2741 2742 2743 2744 2745 2746
		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");
2747 2748 2749
	if (evsel == NULL)
		evsel = perf_evlist__find_tracepoint_by_name(session->evlist,
							     "syscalls:sys_exit");
2750 2751 2752
	if (evsel &&
	    (perf_evsel__init_syscall_tp(evsel, trace__sys_exit) < 0 ||
	    perf_evsel__init_sc_tp_uint_field(evsel, ret))) {
2753
		pr_err("Error during initialize raw_syscalls:sys_exit event\n");
2754 2755 2756
		goto out;
	}

2757 2758 2759 2760 2761 2762 2763 2764
	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;
	}

2765 2766 2767 2768
	err = parse_target_str(trace);
	if (err != 0)
		goto out;

2769 2770
	setup_pager();

2771
	err = perf_session__process_events(session);
2772 2773 2774
	if (err)
		pr_err("Failed to process events, error %d", err);

2775 2776 2777
	else if (trace->summary)
		trace__fprintf_thread_summary(trace, trace->output);

2778 2779 2780 2781 2782 2783
out:
	perf_session__delete(session);

	return err;
}

2784 2785 2786 2787
static size_t trace__fprintf_threads_header(FILE *fp)
{
	size_t printed;

2788
	printed  = fprintf(fp, "\n Summary of events:\n\n");
2789 2790 2791 2792 2793 2794 2795 2796 2797 2798 2799 2800 2801 2802 2803 2804 2805

	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");

2806 2807 2808
	printed += fprintf(fp, "   syscall            calls    total       min       avg       max      stddev\n");
	printed += fprintf(fp, "                               (msec)    (msec)    (msec)    (msec)        (%%)\n");
	printed += fprintf(fp, "   --------------- -------- --------- --------- --------- ---------     ------\n");
2809

2810 2811 2812 2813 2814 2815 2816 2817 2818 2819 2820 2821 2822 2823
	/* 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];
2824
			printed += fprintf(fp, "   %-15s", sc->name);
2825 2826
			printed += fprintf(fp, " %8" PRIu64 " %9.3f %9.3f %9.3f",
					   n, avg * n, min, avg);
P
Pekka Enberg 已提交
2827
			printed += fprintf(fp, " %9.3f %9.2f%%\n", max, pct);
2828 2829 2830 2831 2832 2833
		}

		inode = intlist__next(inode);
	}

	printed += fprintf(fp, "\n\n");
2834 2835 2836 2837

	return printed;
}

2838 2839 2840 2841 2842 2843 2844 2845 2846 2847 2848 2849 2850
/* 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;
2851
	struct thread_trace *ttrace = thread__priv(thread);
2852 2853 2854 2855 2856 2857 2858
	double ratio;

	if (ttrace == NULL)
		return 0;

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

2859
	printed += fprintf(fp, " %s (%d), ", thread__comm_str(thread), thread->tid);
2860
	printed += fprintf(fp, "%lu events, ", ttrace->nr_events);
2861
	printed += fprintf(fp, "%.1f%%", ratio);
2862 2863 2864 2865
	if (ttrace->pfmaj)
		printed += fprintf(fp, ", %lu majfaults", ttrace->pfmaj);
	if (ttrace->pfmin)
		printed += fprintf(fp, ", %lu minfaults", ttrace->pfmin);
2866
	printed += fprintf(fp, ", %.3f msec\n", ttrace->runtime_ms);
2867
	printed += thread__dump_stats(ttrace, trace, fp);
2868 2869 2870 2871 2872 2873

	data->printed += printed;

	return 0;
}

2874 2875
static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp)
{
2876 2877 2878 2879 2880
	struct summary_data data = {
		.fp = fp,
		.trace = trace
	};
	data.printed = trace__fprintf_threads_header(fp);
2881

2882 2883 2884
	machine__for_each_thread(trace->host, trace__fprintf_one_thread, &data);

	return data.printed;
2885 2886
}

2887 2888 2889 2890 2891 2892 2893 2894 2895
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;
}

2896 2897 2898 2899 2900 2901 2902 2903 2904 2905 2906 2907 2908 2909 2910 2911 2912 2913 2914 2915 2916 2917 2918 2919 2920 2921 2922 2923 2924 2925 2926 2927
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;
}

2928 2929 2930 2931 2932 2933 2934 2935 2936 2937 2938 2939 2940 2941 2942 2943 2944
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;
}

2945 2946 2947 2948 2949 2950 2951 2952 2953 2954 2955 2956 2957 2958 2959 2960 2961
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;
}

2962 2963 2964 2965 2966 2967 2968 2969
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
2970 2971
int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)
{
2972
	const char *trace_usage[] = {
2973 2974
		"perf trace [<options>] [<command>]",
		"perf trace [<options>] -- <command> [<options>]",
D
David Ahern 已提交
2975 2976
		"perf trace record [<options>] [<command>]",
		"perf trace record [<options>] -- <command> [<options>]",
A
Arnaldo Carvalho de Melo 已提交
2977 2978 2979
		NULL
	};
	struct trace trace = {
2980 2981 2982 2983
		.audit = {
			.machine = audit_detect_machine(),
			.open_id = audit_name_to_syscall("open", trace.audit.machine),
		},
A
Arnaldo Carvalho de Melo 已提交
2984 2985 2986 2987 2988 2989 2990 2991 2992 2993
		.syscalls = {
			. max = -1,
		},
		.opts = {
			.target = {
				.uid	   = UINT_MAX,
				.uses_mmap = true,
			},
			.user_freq     = UINT_MAX,
			.user_interval = ULLONG_MAX,
2994
			.no_buffering  = true,
2995
			.mmap_pages    = UINT_MAX,
2996
			.proc_map_timeout  = 500,
A
Arnaldo Carvalho de Melo 已提交
2997
		},
2998
		.output = stderr,
2999
		.show_comm = true,
3000
		.trace_syscalls = true,
A
Arnaldo Carvalho de Melo 已提交
3001
	};
3002
	const char *output_name = NULL;
3003
	const char *ev_qualifier_str = NULL;
A
Arnaldo Carvalho de Melo 已提交
3004
	const struct option trace_options[] = {
3005 3006 3007
	OPT_CALLBACK(0, "event", &trace.evlist, "event",
		     "event selector. use 'perf list' to list available events",
		     parse_events_option),
3008 3009
	OPT_BOOLEAN(0, "comm", &trace.show_comm,
		    "show the thread COMM next to its id"),
3010
	OPT_BOOLEAN(0, "tool_stats", &trace.show_tool_stats, "show tool stats"),
3011
	OPT_STRING('e', "expr", &ev_qualifier_str, "expr", "list of syscalls to trace"),
3012
	OPT_STRING('o', "output", &output_name, "file", "output file name"),
3013
	OPT_STRING('i', "input", &input_name, "file", "Analyze events in file"),
A
Arnaldo Carvalho de Melo 已提交
3014 3015
	OPT_STRING('p', "pid", &trace.opts.target.pid, "pid",
		    "trace events on existing process id"),
3016
	OPT_STRING('t', "tid", &trace.opts.target.tid, "tid",
A
Arnaldo Carvalho de Melo 已提交
3017
		    "trace events on existing thread id"),
3018 3019
	OPT_CALLBACK(0, "filter-pids", &trace, "CSV list of pids",
		     "pids to filter (by the kernel)", trace__set_filter_pids),
3020
	OPT_BOOLEAN('a', "all-cpus", &trace.opts.target.system_wide,
A
Arnaldo Carvalho de Melo 已提交
3021
		    "system-wide collection from all CPUs"),
3022
	OPT_STRING('C', "cpu", &trace.opts.target.cpu_list, "cpu",
A
Arnaldo Carvalho de Melo 已提交
3023
		    "list of cpus to monitor"),
3024
	OPT_BOOLEAN(0, "no-inherit", &trace.opts.no_inherit,
A
Arnaldo Carvalho de Melo 已提交
3025
		    "child tasks do not inherit counters"),
3026 3027 3028
	OPT_CALLBACK('m', "mmap-pages", &trace.opts.mmap_pages, "pages",
		     "number of mmap data pages",
		     perf_evlist__parse_mmap_pages),
3029
	OPT_STRING('u', "uid", &trace.opts.target.uid_str, "user",
A
Arnaldo Carvalho de Melo 已提交
3030
		   "user to profile"),
3031 3032 3033
	OPT_CALLBACK(0, "duration", &trace, "float",
		     "show only events with duration > N.M ms",
		     trace__set_duration),
3034
	OPT_BOOLEAN(0, "sched", &trace.sched, "show blocking scheduler events"),
3035
	OPT_INCR('v', "verbose", &verbose, "be more verbose"),
3036 3037
	OPT_BOOLEAN('T', "time", &trace.full_time,
		    "Show full timestamp, not time relative to first start"),
D
David Ahern 已提交
3038 3039 3040 3041
	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"),
3042 3043
	OPT_CALLBACK_DEFAULT('F', "pf", &trace.trace_pgfaults, "all|maj|min",
		     "Trace pagefaults", parse_pagefaults, "maj"),
3044
	OPT_BOOLEAN(0, "syscalls", &trace.trace_syscalls, "Trace syscalls"),
3045
	OPT_BOOLEAN('f', "force", &trace.force, "don't complain, do it"),
3046 3047
	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
3048 3049
	OPT_END()
	};
3050
	const char * const trace_subcommands[] = { "record", NULL };
A
Arnaldo Carvalho de Melo 已提交
3051
	int err;
3052
	char bf[BUFSIZ];
A
Arnaldo Carvalho de Melo 已提交
3053

3054 3055 3056
	signal(SIGSEGV, sighandler_dump_stack);
	signal(SIGFPE, sighandler_dump_stack);

3057 3058 3059 3060
	trace.evlist = perf_evlist__new();

	if (trace.evlist == NULL) {
		pr_err("Not enough memory to run!\n");
3061
		err = -ENOMEM;
3062 3063 3064
		goto out;
	}

3065 3066
	argc = parse_options_subcommand(argc, argv, trace_options, trace_subcommands,
				 trace_usage, PARSE_OPT_STOP_AT_NON_OPTION);
D
David Ahern 已提交
3067

3068 3069 3070 3071 3072
	if (trace.trace_pgfaults) {
		trace.opts.sample_address = true;
		trace.opts.sample_time = true;
	}

3073 3074 3075
	if (trace.evlist->nr_entries > 0)
		evlist__set_evsel_handler(trace.evlist, trace__event_handler);

3076 3077 3078 3079 3080 3081 3082
	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;

3083 3084
	if (!trace.trace_syscalls && !trace.trace_pgfaults &&
	    trace.evlist->nr_entries == 0 /* Was --events used? */) {
3085 3086 3087 3088
		pr_err("Please specify something to trace.\n");
		return -1;
	}

3089 3090 3091 3092 3093 3094 3095 3096
	if (output_name != NULL) {
		err = trace__open_output(&trace, output_name);
		if (err < 0) {
			perror("failed to create output file");
			goto out;
		}
	}

3097
	if (ev_qualifier_str != NULL) {
3098
		const char *s = ev_qualifier_str;
3099 3100 3101
		struct strlist_config slist_config = {
			.dirname = system_path(STRACE_GROUPS_DIR),
		};
3102 3103 3104 3105

		trace.not_ev_qualifier = *s == '!';
		if (trace.not_ev_qualifier)
			++s;
3106
		trace.ev_qualifier = strlist__new(s, &slist_config);
3107
		if (trace.ev_qualifier == NULL) {
3108 3109 3110 3111
			fputs("Not enough memory to parse event qualifier",
			      trace.output);
			err = -ENOMEM;
			goto out_close;
3112
		}
3113 3114 3115 3116

		err = trace__validate_ev_qualifier(&trace);
		if (err)
			goto out_close;
3117 3118
	}

3119
	err = target__validate(&trace.opts.target);
3120
	if (err) {
3121
		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
3122 3123
		fprintf(trace.output, "%s", bf);
		goto out_close;
3124 3125
	}

3126
	err = target__parse_uid(&trace.opts.target);
A
Arnaldo Carvalho de Melo 已提交
3127
	if (err) {
3128
		target__strerror(&trace.opts.target, err, bf, sizeof(bf));
3129 3130
		fprintf(trace.output, "%s", bf);
		goto out_close;
A
Arnaldo Carvalho de Melo 已提交
3131 3132
	}

3133
	if (!argc && target__none(&trace.opts.target))
3134 3135
		trace.opts.target.system_wide = true;

3136 3137 3138 3139
	if (input_name)
		err = trace__replay(&trace);
	else
		err = trace__run(&trace, argc, argv);
3140

3141 3142 3143 3144
out_close:
	if (output_name != NULL)
		fclose(trace.output);
out:
3145
	return err;
A
Arnaldo Carvalho de Melo 已提交
3146
}