trace_kprobe.c 50.9 KB
Newer Older
1
/*
2
 * Kprobes-based tracing events
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
 *
 * Created by Masami Hiramatsu <mhiramat@redhat.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

#include <linux/module.h>
#include <linux/uaccess.h>
#include <linux/kprobes.h>
#include <linux/seq_file.h>
#include <linux/slab.h>
#include <linux/smp.h>
#include <linux/debugfs.h>
#include <linux/types.h>
#include <linux/string.h>
#include <linux/ctype.h>
#include <linux/ptrace.h>
31
#include <linux/perf_event.h>
32
#include <linux/stringify.h>
33
#include <linux/limits.h>
34
#include <asm/bitsperlong.h>
35 36 37 38

#include "trace.h"
#include "trace_output.h"

39
#define MAX_TRACE_ARGS 128
40
#define MAX_ARGSTR_LEN 63
41
#define MAX_EVENT_NAME_LEN 64
42
#define MAX_STRING_SIZE PATH_MAX
43
#define KPROBE_EVENT_SYSTEM "kprobes"
44

45
/* Reserved field names */
46 47 48
#define FIELD_STRING_IP "__probe_ip"
#define FIELD_STRING_RETIP "__probe_ret_ip"
#define FIELD_STRING_FUNC "__probe_func"
49 50 51 52 53 54 55 56 57 58 59 60

const char *reserved_field_names[] = {
	"common_type",
	"common_flags",
	"common_preempt_count",
	"common_pid",
	"common_tgid",
	FIELD_STRING_IP,
	FIELD_STRING_RETIP,
	FIELD_STRING_FUNC,
};

61
/* Printing function type */
62 63
typedef int (*print_type_func_t)(struct trace_seq *, const char *, void *,
				 void *);
64 65 66 67 68 69
#define PRINT_TYPE_FUNC_NAME(type)	print_type_##type
#define PRINT_TYPE_FMT_NAME(type)	print_type_format_##type

/* Printing  in basic type function template */
#define DEFINE_BASIC_PRINT_TYPE_FUNC(type, fmt, cast)			\
static __kprobes int PRINT_TYPE_FUNC_NAME(type)(struct trace_seq *s,	\
70 71
						const char *name,	\
						void *data, void *ent)\
72 73 74 75 76 77 78 79 80 81 82 83 84 85
{									\
	return trace_seq_printf(s, " %s=" fmt, name, (cast)*(type *)data);\
}									\
static const char PRINT_TYPE_FMT_NAME(type)[] = fmt;

DEFINE_BASIC_PRINT_TYPE_FUNC(u8, "%x", unsigned int)
DEFINE_BASIC_PRINT_TYPE_FUNC(u16, "%x", unsigned int)
DEFINE_BASIC_PRINT_TYPE_FUNC(u32, "%lx", unsigned long)
DEFINE_BASIC_PRINT_TYPE_FUNC(u64, "%llx", unsigned long long)
DEFINE_BASIC_PRINT_TYPE_FUNC(s8, "%d", int)
DEFINE_BASIC_PRINT_TYPE_FUNC(s16, "%d", int)
DEFINE_BASIC_PRINT_TYPE_FUNC(s32, "%ld", long)
DEFINE_BASIC_PRINT_TYPE_FUNC(s64, "%lld", long long)

86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128
/* data_rloc: data relative location, compatible with u32 */
#define make_data_rloc(len, roffs)	\
	(((u32)(len) << 16) | ((u32)(roffs) & 0xffff))
#define get_rloc_len(dl)	((u32)(dl) >> 16)
#define get_rloc_offs(dl)	((u32)(dl) & 0xffff)

static inline void *get_rloc_data(u32 *dl)
{
	return (u8 *)dl + get_rloc_offs(*dl);
}

/* For data_loc conversion */
static inline void *get_loc_data(u32 *dl, void *ent)
{
	return (u8 *)ent + get_rloc_offs(*dl);
}

/*
 * Convert data_rloc to data_loc:
 *  data_rloc stores the offset from data_rloc itself, but data_loc
 *  stores the offset from event entry.
 */
#define convert_rloc_to_loc(dl, offs)	((u32)(dl) + (offs))

/* For defining macros, define string/string_size types */
typedef u32 string;
typedef u32 string_size;

/* Print type function for string type */
static __kprobes int PRINT_TYPE_FUNC_NAME(string)(struct trace_seq *s,
						  const char *name,
						  void *data, void *ent)
{
	int len = *(u32 *)data >> 16;

	if (!len)
		return trace_seq_printf(s, " %s=(fault)", name);
	else
		return trace_seq_printf(s, " %s=\"%s\"", name,
					(const char *)get_loc_data(data, ent));
}
static const char PRINT_TYPE_FMT_NAME(string)[] = "\\\"%s\\\"";

129 130 131 132 133
/* Data fetch function type */
typedef	void (*fetch_func_t)(struct pt_regs *, void *, void *);

struct fetch_param {
	fetch_func_t	fn;
134 135 136
	void *data;
};

137 138
static __kprobes void call_fetch(struct fetch_param *fprm,
				 struct pt_regs *regs, void *dest)
139
{
140
	return fprm->fn(regs, fprm->data, dest);
141 142
}

143
#define FETCH_FUNC_NAME(method, type)	fetch_##method##_##type
144 145 146 147
/*
 * Define macro for basic types - we don't need to define s* types, because
 * we have to care only about bitwidth at recording time.
 */
148 149 150 151 152 153 154 155 156 157 158 159 160 161
#define DEFINE_BASIC_FETCH_FUNCS(method) \
DEFINE_FETCH_##method(u8)		\
DEFINE_FETCH_##method(u16)		\
DEFINE_FETCH_##method(u32)		\
DEFINE_FETCH_##method(u64)

#define CHECK_FETCH_FUNCS(method, fn)			\
	(((FETCH_FUNC_NAME(method, u8) == fn) ||	\
	  (FETCH_FUNC_NAME(method, u16) == fn) ||	\
	  (FETCH_FUNC_NAME(method, u32) == fn) ||	\
	  (FETCH_FUNC_NAME(method, u64) == fn) ||	\
	  (FETCH_FUNC_NAME(method, string) == fn) ||	\
	  (FETCH_FUNC_NAME(method, string_size) == fn)) \
	 && (fn != NULL))
162 163 164 165

/* Data fetch function templates */
#define DEFINE_FETCH_reg(type)						\
static __kprobes void FETCH_FUNC_NAME(reg, type)(struct pt_regs *regs,	\
166
					void *offset, void *dest)	\
167 168 169
{									\
	*(type *)dest = (type)regs_get_register(regs,			\
				(unsigned int)((unsigned long)offset));	\
170
}
171
DEFINE_BASIC_FETCH_FUNCS(reg)
172 173 174
/* No string on the register */
#define fetch_reg_string NULL
#define fetch_reg_string_size NULL
175 176 177 178 179 180 181

#define DEFINE_FETCH_stack(type)					\
static __kprobes void FETCH_FUNC_NAME(stack, type)(struct pt_regs *regs,\
					  void *offset, void *dest)	\
{									\
	*(type *)dest = (type)regs_get_kernel_stack_nth(regs,		\
				(unsigned int)((unsigned long)offset));	\
182
}
183
DEFINE_BASIC_FETCH_FUNCS(stack)
184 185 186
/* No string on the stack entry */
#define fetch_stack_string NULL
#define fetch_stack_string_size NULL
187

188 189 190 191 192
#define DEFINE_FETCH_retval(type)					\
static __kprobes void FETCH_FUNC_NAME(retval, type)(struct pt_regs *regs,\
					  void *dummy, void *dest)	\
{									\
	*(type *)dest = (type)regs_return_value(regs);			\
193
}
194
DEFINE_BASIC_FETCH_FUNCS(retval)
195 196 197
/* No string on the retval */
#define fetch_retval_string NULL
#define fetch_retval_string_size NULL
198 199 200 201 202 203 204 205 206 207

#define DEFINE_FETCH_memory(type)					\
static __kprobes void FETCH_FUNC_NAME(memory, type)(struct pt_regs *regs,\
					  void *addr, void *dest)	\
{									\
	type retval;							\
	if (probe_kernel_address(addr, retval))				\
		*(type *)dest = 0;					\
	else								\
		*(type *)dest = retval;					\
208
}
209
DEFINE_BASIC_FETCH_FUNCS(memory)
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265
/*
 * Fetch a null-terminated string. Caller MUST set *(u32 *)dest with max
 * length and relative data location.
 */
static __kprobes void FETCH_FUNC_NAME(memory, string)(struct pt_regs *regs,
						      void *addr, void *dest)
{
	long ret;
	int maxlen = get_rloc_len(*(u32 *)dest);
	u8 *dst = get_rloc_data(dest);
	u8 *src = addr;
	mm_segment_t old_fs = get_fs();
	if (!maxlen)
		return;
	/*
	 * Try to get string again, since the string can be changed while
	 * probing.
	 */
	set_fs(KERNEL_DS);
	pagefault_disable();
	do
		ret = __copy_from_user_inatomic(dst++, src++, 1);
	while (dst[-1] && ret == 0 && src - (u8 *)addr < maxlen);
	dst[-1] = '\0';
	pagefault_enable();
	set_fs(old_fs);

	if (ret < 0) {	/* Failed to fetch string */
		((u8 *)get_rloc_data(dest))[0] = '\0';
		*(u32 *)dest = make_data_rloc(0, get_rloc_offs(*(u32 *)dest));
	} else
		*(u32 *)dest = make_data_rloc(src - (u8 *)addr,
					      get_rloc_offs(*(u32 *)dest));
}
/* Return the length of string -- including null terminal byte */
static __kprobes void FETCH_FUNC_NAME(memory, string_size)(struct pt_regs *regs,
							void *addr, void *dest)
{
	int ret, len = 0;
	u8 c;
	mm_segment_t old_fs = get_fs();

	set_fs(KERNEL_DS);
	pagefault_disable();
	do {
		ret = __copy_from_user_inatomic(&c, (u8 *)addr + len, 1);
		len++;
	} while (c && ret == 0 && len < MAX_STRING_SIZE);
	pagefault_enable();
	set_fs(old_fs);

	if (ret < 0)	/* Failed to check the length */
		*(u32 *)dest = 0;
	else
		*(u32 *)dest = len;
}
266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308

/* Memory fetching by symbol */
struct symbol_cache {
	char *symbol;
	long offset;
	unsigned long addr;
};

static unsigned long update_symbol_cache(struct symbol_cache *sc)
{
	sc->addr = (unsigned long)kallsyms_lookup_name(sc->symbol);
	if (sc->addr)
		sc->addr += sc->offset;
	return sc->addr;
}

static void free_symbol_cache(struct symbol_cache *sc)
{
	kfree(sc->symbol);
	kfree(sc);
}

static struct symbol_cache *alloc_symbol_cache(const char *sym, long offset)
{
	struct symbol_cache *sc;

	if (!sym || strlen(sym) == 0)
		return NULL;
	sc = kzalloc(sizeof(struct symbol_cache), GFP_KERNEL);
	if (!sc)
		return NULL;

	sc->symbol = kstrdup(sym, GFP_KERNEL);
	if (!sc->symbol) {
		kfree(sc);
		return NULL;
	}
	sc->offset = offset;

	update_symbol_cache(sc);
	return sc;
}

309 310 311 312 313 314 315 316 317
#define DEFINE_FETCH_symbol(type)					\
static __kprobes void FETCH_FUNC_NAME(symbol, type)(struct pt_regs *regs,\
					  void *data, void *dest)	\
{									\
	struct symbol_cache *sc = data;					\
	if (sc->addr)							\
		fetch_memory_##type(regs, (void *)sc->addr, dest);	\
	else								\
		*(type *)dest = 0;					\
318
}
319
DEFINE_BASIC_FETCH_FUNCS(symbol)
320 321
DEFINE_FETCH_symbol(string)
DEFINE_FETCH_symbol(string_size)
322

323 324 325
/* Dereference memory access function */
struct deref_fetch_param {
	struct fetch_param orig;
326 327 328
	long offset;
};

329 330 331 332 333 334 335 336 337 338 339 340
#define DEFINE_FETCH_deref(type)					\
static __kprobes void FETCH_FUNC_NAME(deref, type)(struct pt_regs *regs,\
					    void *data, void *dest)	\
{									\
	struct deref_fetch_param *dprm = data;				\
	unsigned long addr;						\
	call_fetch(&dprm->orig, regs, &addr);				\
	if (addr) {							\
		addr += dprm->offset;					\
		fetch_memory_##type(regs, (void *)addr, dest);		\
	} else								\
		*(type *)dest = 0;					\
341
}
342
DEFINE_BASIC_FETCH_FUNCS(deref)
343 344
DEFINE_FETCH_deref(string)
DEFINE_FETCH_deref(string_size)
345

346
static __kprobes void free_deref_fetch_param(struct deref_fetch_param *data)
347
{
348
	if (CHECK_FETCH_FUNCS(deref, data->orig.fn))
349
		free_deref_fetch_param(data->orig.data);
350
	else if (CHECK_FETCH_FUNCS(symbol, data->orig.fn))
351 352 353 354
		free_symbol_cache(data->orig.data);
	kfree(data);
}

355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391
/* Bitfield fetch function */
struct bitfield_fetch_param {
	struct fetch_param orig;
	unsigned char hi_shift;
	unsigned char low_shift;
};

#define DEFINE_FETCH_bitfield(type)					\
static __kprobes void FETCH_FUNC_NAME(bitfield, type)(struct pt_regs *regs,\
					    void *data, void *dest)	\
{									\
	struct bitfield_fetch_param *bprm = data;			\
	type buf = 0;							\
	call_fetch(&bprm->orig, regs, &buf);				\
	if (buf) {							\
		buf <<= bprm->hi_shift;					\
		buf >>= bprm->low_shift;				\
	}								\
	*(type *)dest = buf;						\
}
DEFINE_BASIC_FETCH_FUNCS(bitfield)
#define fetch_bitfield_string NULL
#define fetch_bitfield_string_size NULL

static __kprobes void
free_bitfield_fetch_param(struct bitfield_fetch_param *data)
{
	/*
	 * Don't check the bitfield itself, because this must be the
	 * last fetch function.
	 */
	if (CHECK_FETCH_FUNCS(deref, data->orig.fn))
		free_deref_fetch_param(data->orig.data);
	else if (CHECK_FETCH_FUNCS(symbol, data->orig.fn))
		free_symbol_cache(data->orig.data);
	kfree(data);
}
392 393 394 395 396 397
/* Default (unsigned long) fetch type */
#define __DEFAULT_FETCH_TYPE(t) u##t
#define _DEFAULT_FETCH_TYPE(t) __DEFAULT_FETCH_TYPE(t)
#define DEFAULT_FETCH_TYPE _DEFAULT_FETCH_TYPE(BITS_PER_LONG)
#define DEFAULT_FETCH_TYPE_STR __stringify(DEFAULT_FETCH_TYPE)

398 399 400 401 402 403 404 405
/* Fetch types */
enum {
	FETCH_MTD_reg = 0,
	FETCH_MTD_stack,
	FETCH_MTD_retval,
	FETCH_MTD_memory,
	FETCH_MTD_symbol,
	FETCH_MTD_deref,
406
	FETCH_MTD_bitfield,
407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426
	FETCH_MTD_END,
};

#define ASSIGN_FETCH_FUNC(method, type)	\
	[FETCH_MTD_##method] = FETCH_FUNC_NAME(method, type)

#define __ASSIGN_FETCH_TYPE(_name, ptype, ftype, _size, sign, _fmttype)	\
	{.name = _name,				\
	 .size = _size,					\
	 .is_signed = sign,				\
	 .print = PRINT_TYPE_FUNC_NAME(ptype),		\
	 .fmt = PRINT_TYPE_FMT_NAME(ptype),		\
	 .fmttype = _fmttype,				\
	 .fetch = {					\
ASSIGN_FETCH_FUNC(reg, ftype),				\
ASSIGN_FETCH_FUNC(stack, ftype),			\
ASSIGN_FETCH_FUNC(retval, ftype),			\
ASSIGN_FETCH_FUNC(memory, ftype),			\
ASSIGN_FETCH_FUNC(symbol, ftype),			\
ASSIGN_FETCH_FUNC(deref, ftype),			\
427
ASSIGN_FETCH_FUNC(bitfield, ftype),			\
428
	  }						\
429 430
	}

431 432 433 434 435 436
#define ASSIGN_FETCH_TYPE(ptype, ftype, sign)			\
	__ASSIGN_FETCH_TYPE(#ptype, ptype, ftype, sizeof(ftype), sign, #ptype)

#define FETCH_TYPE_STRING 0
#define FETCH_TYPE_STRSIZE 1

437 438 439 440 441 442 443
/* Fetch type information table */
static const struct fetch_type {
	const char	*name;		/* Name of type */
	size_t		size;		/* Byte size of type */
	int		is_signed;	/* Signed flag */
	print_type_func_t	print;	/* Print functions */
	const char	*fmt;		/* Fromat string */
444
	const char	*fmttype;	/* Name in format file */
445
	/* Fetch functions */
446
	fetch_func_t	fetch[FETCH_MTD_END];
447
} fetch_type_table[] = {
448 449 450 451 452 453
	/* Special types */
	[FETCH_TYPE_STRING] = __ASSIGN_FETCH_TYPE("string", string, string,
					sizeof(u32), 1, "__data_loc char[]"),
	[FETCH_TYPE_STRSIZE] = __ASSIGN_FETCH_TYPE("string_size", u32,
					string_size, sizeof(u32), 0, "u32"),
	/* Basic types */
454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
	ASSIGN_FETCH_TYPE(u8,  u8,  0),
	ASSIGN_FETCH_TYPE(u16, u16, 0),
	ASSIGN_FETCH_TYPE(u32, u32, 0),
	ASSIGN_FETCH_TYPE(u64, u64, 0),
	ASSIGN_FETCH_TYPE(s8,  u8,  1),
	ASSIGN_FETCH_TYPE(s16, u16, 1),
	ASSIGN_FETCH_TYPE(s32, u32, 1),
	ASSIGN_FETCH_TYPE(s64, u64, 1),
};

static const struct fetch_type *find_fetch_type(const char *type)
{
	int i;

	if (!type)
		type = DEFAULT_FETCH_TYPE_STR;

471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
	/* Special case: bitfield */
	if (*type == 'b') {
		unsigned long bs;
		type = strchr(type, '/');
		if (!type)
			goto fail;
		type++;
		if (strict_strtoul(type, 0, &bs))
			goto fail;
		switch (bs) {
		case 8:
			return find_fetch_type("u8");
		case 16:
			return find_fetch_type("u16");
		case 32:
			return find_fetch_type("u32");
		case 64:
			return find_fetch_type("u64");
		default:
			goto fail;
		}
	}

494 495 496
	for (i = 0; i < ARRAY_SIZE(fetch_type_table); i++)
		if (strcmp(type, fetch_type_table[i].name) == 0)
			return &fetch_type_table[i];
497
fail:
498 499 500 501 502 503 504 505 506 507
	return NULL;
}

/* Special function : only accept unsigned long */
static __kprobes void fetch_stack_address(struct pt_regs *regs,
					  void *dummy, void *dest)
{
	*(unsigned long *)dest = kernel_stack_pointer(regs);
}

508 509 510 511 512 513 514 515 516 517 518 519 520 521 522
static fetch_func_t get_fetch_size_function(const struct fetch_type *type,
					    fetch_func_t orig_fn)
{
	int i;

	if (type != &fetch_type_table[FETCH_TYPE_STRING])
		return NULL;	/* Only string type needs size function */
	for (i = 0; i < FETCH_MTD_END; i++)
		if (type->fetch[i] == orig_fn)
			return fetch_type_table[FETCH_TYPE_STRSIZE].fetch[i];

	WARN_ON(1);	/* This should not happen */
	return NULL;
}

523
/**
524
 * Kprobe event core functions
525 526
 */

527
struct probe_arg {
528
	struct fetch_param	fetch;
529
	struct fetch_param	fetch_size;
530 531 532 533
	unsigned int		offset;	/* Offset from argument entry */
	const char		*name;	/* Name of this argument */
	const char		*comm;	/* Command of this argument */
	const struct fetch_type	*type;	/* Type of this argument */
534 535
};

536 537 538
/* Flags for trace_probe */
#define TP_FLAG_TRACE	1
#define TP_FLAG_PROFILE	2
539
#define TP_FLAG_REGISTERED 4
540

541 542
struct trace_probe {
	struct list_head	list;
543
	struct kretprobe	rp;	/* Use rp.kp for kprobe use */
544
	unsigned long 		nhit;
545
	unsigned int		flags;	/* For TP_FLAG_* */
546
	const char		*symbol;	/* symbol name */
547
	struct ftrace_event_class	class;
548
	struct ftrace_event_call	call;
549
	ssize_t			size;		/* trace entry size */
550
	unsigned int		nr_args;
551
	struct probe_arg	args[];
552 553
};

554 555
#define SIZEOF_TRACE_PROBE(n)			\
	(offsetof(struct trace_probe, args) +	\
556
	(sizeof(struct probe_arg) * (n)))
557

558

559
static __kprobes int trace_probe_is_return(struct trace_probe *tp)
560
{
561
	return tp->rp.handler != NULL;
562 563
}

564
static __kprobes const char *trace_probe_symbol(struct trace_probe *tp)
565 566 567 568
{
	return tp->symbol ? tp->symbol : "unknown";
}

569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601
static __kprobes unsigned long trace_probe_offset(struct trace_probe *tp)
{
	return tp->rp.kp.offset;
}

static __kprobes bool trace_probe_is_enabled(struct trace_probe *tp)
{
	return !!(tp->flags & (TP_FLAG_TRACE | TP_FLAG_PROFILE));
}

static __kprobes bool trace_probe_is_registered(struct trace_probe *tp)
{
	return !!(tp->flags & TP_FLAG_REGISTERED);
}

static __kprobes bool trace_probe_has_gone(struct trace_probe *tp)
{
	return !!(kprobe_gone(&tp->rp.kp));
}

static __kprobes bool trace_probe_within_module(struct trace_probe *tp,
						struct module *mod)
{
	int len = strlen(mod->name);
	const char *name = trace_probe_symbol(tp);
	return strncmp(mod->name, name, len) == 0 && name[len] == ':';
}

static __kprobes bool trace_probe_is_on_module(struct trace_probe *tp)
{
	return !!strchr(trace_probe_symbol(tp), ':');
}

602 603 604 605 606 607
static int register_probe_event(struct trace_probe *tp);
static void unregister_probe_event(struct trace_probe *tp);

static DEFINE_MUTEX(probe_lock);
static LIST_HEAD(probe_list);

608 609 610 611
static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs);
static int kretprobe_dispatcher(struct kretprobe_instance *ri,
				struct pt_regs *regs);

612 613
/* Check the name is good for event/group/fields */
static int is_good_name(const char *name)
614 615 616 617 618 619 620 621 622 623
{
	if (!isalpha(*name) && *name != '_')
		return 0;
	while (*++name != '\0') {
		if (!isalpha(*name) && !isdigit(*name) && *name != '_')
			return 0;
	}
	return 1;
}

624 625 626
/*
 * Allocate new trace_probe and initialize it (including kprobes).
 */
627 628
static struct trace_probe *alloc_trace_probe(const char *group,
					     const char *event,
629 630 631 632
					     void *addr,
					     const char *symbol,
					     unsigned long offs,
					     int nargs, int is_return)
633 634
{
	struct trace_probe *tp;
635
	int ret = -ENOMEM;
636

637
	tp = kzalloc(SIZEOF_TRACE_PROBE(nargs), GFP_KERNEL);
638
	if (!tp)
639
		return ERR_PTR(ret);
640 641 642 643 644

	if (symbol) {
		tp->symbol = kstrdup(symbol, GFP_KERNEL);
		if (!tp->symbol)
			goto error;
645 646 647 648 649 650
		tp->rp.kp.symbol_name = tp->symbol;
		tp->rp.kp.offset = offs;
	} else
		tp->rp.kp.addr = addr;

	if (is_return)
651
		tp->rp.handler = kretprobe_dispatcher;
652
	else
653
		tp->rp.kp.pre_handler = kprobe_dispatcher;
654

655
	if (!event || !is_good_name(event)) {
656
		ret = -EINVAL;
657
		goto error;
658 659
	}

660
	tp->call.class = &tp->class;
661 662 663
	tp->call.name = kstrdup(event, GFP_KERNEL);
	if (!tp->call.name)
		goto error;
664

665
	if (!group || !is_good_name(group)) {
666
		ret = -EINVAL;
667
		goto error;
668 669
	}

670 671
	tp->class.system = kstrdup(group, GFP_KERNEL);
	if (!tp->class.system)
672 673
		goto error;

674 675 676
	INIT_LIST_HEAD(&tp->list);
	return tp;
error:
677
	kfree(tp->call.name);
678 679
	kfree(tp->symbol);
	kfree(tp);
680
	return ERR_PTR(ret);
681 682
}

683 684
static void free_probe_arg(struct probe_arg *arg)
{
685 686 687
	if (CHECK_FETCH_FUNCS(bitfield, arg->fetch.fn))
		free_bitfield_fetch_param(arg->fetch.data);
	else if (CHECK_FETCH_FUNCS(deref, arg->fetch.fn))
688
		free_deref_fetch_param(arg->fetch.data);
689
	else if (CHECK_FETCH_FUNCS(symbol, arg->fetch.fn))
690 691
		free_symbol_cache(arg->fetch.data);
	kfree(arg->name);
692
	kfree(arg->comm);
693 694
}

695 696 697 698 699
static void free_trace_probe(struct trace_probe *tp)
{
	int i;

	for (i = 0; i < tp->nr_args; i++)
700
		free_probe_arg(&tp->args[i]);
701

702
	kfree(tp->call.class->system);
703 704 705 706 707
	kfree(tp->call.name);
	kfree(tp->symbol);
	kfree(tp);
}

708
static struct trace_probe *find_trace_probe(const char *event,
709
					    const char *group)
710 711 712 713
{
	struct trace_probe *tp;

	list_for_each_entry(tp, &probe_list, list)
714
		if (strcmp(tp->call.name, event) == 0 &&
715
		    strcmp(tp->call.class->system, group) == 0)
716 717 718 719
			return tp;
	return NULL;
}

720 721 722 723 724 725
/* Enable trace_probe - @flag must be TP_FLAG_TRACE or TP_FLAG_PROFILE */
static int enable_trace_probe(struct trace_probe *tp, int flag)
{
	int ret = 0;

	tp->flags |= flag;
726 727
	if (trace_probe_is_enabled(tp) && trace_probe_is_registered(tp) &&
	    !trace_probe_has_gone(tp)) {
728 729 730 731 732 733 734 735 736 737 738 739 740
		if (trace_probe_is_return(tp))
			ret = enable_kretprobe(&tp->rp);
		else
			ret = enable_kprobe(&tp->rp.kp);
	}

	return ret;
}

/* Disable trace_probe - @flag must be TP_FLAG_TRACE or TP_FLAG_PROFILE */
static void disable_trace_probe(struct trace_probe *tp, int flag)
{
	tp->flags &= ~flag;
741
	if (!trace_probe_is_enabled(tp) && trace_probe_is_registered(tp)) {
742 743 744 745 746 747 748
		if (trace_probe_is_return(tp))
			disable_kretprobe(&tp->rp);
		else
			disable_kprobe(&tp->rp.kp);
	}
}

749 750
/* Internal register function - just handle k*probes and flags */
static int __register_trace_probe(struct trace_probe *tp)
751
{
752 753 754 755 756 757 758 759 760 761 762
	int ret;

	if (trace_probe_is_registered(tp))
		return -EINVAL;

	/* Set/clear disabled flag according to tp->flag */
	if (trace_probe_is_enabled(tp))
		tp->rp.kp.flags &= ~KPROBE_FLAG_DISABLED;
	else
		tp->rp.kp.flags |= KPROBE_FLAG_DISABLED;

763
	if (trace_probe_is_return(tp))
764
		ret = register_kretprobe(&tp->rp);
765
	else
766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
		ret = register_kprobe(&tp->rp.kp);

	if (ret == 0)
		tp->flags |= TP_FLAG_REGISTERED;
	else {
		pr_warning("Could not insert probe at %s+%lu: %d\n",
			   trace_probe_symbol(tp), trace_probe_offset(tp), ret);
		if (ret == -ENOENT && trace_probe_is_on_module(tp)) {
			pr_warning("This probe might be able to register after"
				   "target module is loaded. Continue.\n");
			ret = 0;
		} else if (ret == -EILSEQ) {
			pr_warning("Probing address(0x%p) is not an "
				   "instruction boundary.\n",
				   tp->rp.kp.addr);
			ret = -EINVAL;
		}
	}

	return ret;
}

/* Internal unregister function - just handle k*probes and flags */
static void __unregister_trace_probe(struct trace_probe *tp)
{
	if (trace_probe_is_registered(tp)) {
		if (trace_probe_is_return(tp))
			unregister_kretprobe(&tp->rp);
		else
			unregister_kprobe(&tp->rp.kp);
		tp->flags &= ~TP_FLAG_REGISTERED;
		/* Cleanup kprobe for reuse */
		if (tp->rp.kp.symbol_name)
			tp->rp.kp.addr = NULL;
	}
}

/* Unregister a trace_probe and probe_event: call with locking probe_lock */
static void unregister_trace_probe(struct trace_probe *tp)
{
	__unregister_trace_probe(tp);
807
	list_del(&tp->list);
808
	unregister_probe_event(tp);
809 810 811 812 813 814 815 816 817 818
}

/* Register a trace_probe and probe_event */
static int register_trace_probe(struct trace_probe *tp)
{
	struct trace_probe *old_tp;
	int ret;

	mutex_lock(&probe_lock);

819
	/* Delete old (same name) event if exist */
820
	old_tp = find_trace_probe(tp->call.name, tp->call.class->system);
821 822 823 824
	if (old_tp) {
		unregister_trace_probe(old_tp);
		free_trace_probe(old_tp);
	}
825 826

	/* Register new event */
827 828
	ret = register_probe_event(tp);
	if (ret) {
P
Paul Bolle 已提交
829
		pr_warning("Failed to register probe event(%d)\n", ret);
830 831 832
		goto end;
	}

833 834 835
	/* Register k*probe */
	ret = __register_trace_probe(tp);
	if (ret < 0)
836
		unregister_probe_event(tp);
837
	else
838
		list_add_tail(&tp->list, &probe_list);
839

840 841 842 843 844
end:
	mutex_unlock(&probe_lock);
	return ret;
}

845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877
/* Module notifier call back, checking event on the module */
static int trace_probe_module_callback(struct notifier_block *nb,
				       unsigned long val, void *data)
{
	struct module *mod = data;
	struct trace_probe *tp;
	int ret;

	if (val != MODULE_STATE_COMING)
		return NOTIFY_DONE;

	/* Update probes on coming module */
	mutex_lock(&probe_lock);
	list_for_each_entry(tp, &probe_list, list) {
		if (trace_probe_within_module(tp, mod)) {
			__unregister_trace_probe(tp);
			ret = __register_trace_probe(tp);
			if (ret)
				pr_warning("Failed to re-register probe %s on"
					   "%s: %d\n",
					   tp->call.name, mod->name, ret);
		}
	}
	mutex_unlock(&probe_lock);

	return NOTIFY_DONE;
}

static struct notifier_block trace_probe_module_nb = {
	.notifier_call = trace_probe_module_callback,
	.priority = 1	/* Invoked after kprobe module callback */
};

878
/* Split symbol and offset. */
879
static int split_symbol_offset(char *symbol, unsigned long *offset)
880 881 882 883 884 885 886 887 888 889
{
	char *tmp;
	int ret;

	if (!offset)
		return -EINVAL;

	tmp = strchr(symbol, '+');
	if (tmp) {
		/* skip sign because strict_strtol doesn't accept '+' */
890
		ret = strict_strtoul(tmp + 1, 0, offset);
891 892 893 894 895 896 897 898 899 900 901
		if (ret)
			return ret;
		*tmp = '\0';
	} else
		*offset = 0;
	return 0;
}

#define PARAM_MAX_ARGS 16
#define PARAM_MAX_STACK (THREAD_SIZE / sizeof(unsigned long))

902 903
static int parse_probe_vars(char *arg, const struct fetch_type *t,
			    struct fetch_param *f, int is_return)
904 905 906 907
{
	int ret = 0;
	unsigned long param;

908
	if (strcmp(arg, "retval") == 0) {
909
		if (is_return)
910
			f->fn = t->fetch[FETCH_MTD_retval];
911
		else
912
			ret = -EINVAL;
913 914
	} else if (strncmp(arg, "stack", 5) == 0) {
		if (arg[5] == '\0') {
915 916 917 918
			if (strcmp(t->name, DEFAULT_FETCH_TYPE_STR) == 0)
				f->fn = fetch_stack_address;
			else
				ret = -EINVAL;
919 920
		} else if (isdigit(arg[5])) {
			ret = strict_strtoul(arg + 5, 10, &param);
921 922 923
			if (ret || param > PARAM_MAX_STACK)
				ret = -EINVAL;
			else {
924
				f->fn = t->fetch[FETCH_MTD_stack];
925
				f->data = (void *)param;
926
			}
927 928 929
		} else
			ret = -EINVAL;
	} else
930 931 932 933
		ret = -EINVAL;
	return ret;
}

934
/* Recursive argument parser */
935 936
static int __parse_probe_arg(char *arg, const struct fetch_type *t,
			     struct fetch_param *f, int is_return)
937 938 939 940 941 942 943 944
{
	int ret = 0;
	unsigned long param;
	long offset;
	char *tmp;

	switch (arg[0]) {
	case '$':
945
		ret = parse_probe_vars(arg + 1, t, f, is_return);
946 947 948 949
		break;
	case '%':	/* named register */
		ret = regs_query_register_offset(arg + 1);
		if (ret >= 0) {
950
			f->fn = t->fetch[FETCH_MTD_reg];
951
			f->data = (void *)(unsigned long)ret;
952 953 954
			ret = 0;
		}
		break;
955 956 957 958 959
	case '@':	/* memory or symbol */
		if (isdigit(arg[1])) {
			ret = strict_strtoul(arg + 1, 0, &param);
			if (ret)
				break;
960
			f->fn = t->fetch[FETCH_MTD_memory];
961
			f->data = (void *)param;
962 963 964 965
		} else {
			ret = split_symbol_offset(arg + 1, &offset);
			if (ret)
				break;
966 967
			f->data = alloc_symbol_cache(arg + 1, offset);
			if (f->data)
968
				f->fn = t->fetch[FETCH_MTD_symbol];
969 970
		}
		break;
971
	case '+':	/* deref memory */
972
		arg++;	/* Skip '+', because strict_strtol() rejects it. */
973 974
	case '-':
		tmp = strchr(arg, '(');
975
		if (!tmp)
976 977
			break;
		*tmp = '\0';
978
		ret = strict_strtol(arg, 0, &offset);
979 980 981 982 983
		if (ret)
			break;
		arg = tmp + 1;
		tmp = strrchr(arg, ')');
		if (tmp) {
984 985
			struct deref_fetch_param *dprm;
			const struct fetch_type *t2 = find_fetch_type(NULL);
986
			*tmp = '\0';
987 988 989
			dprm = kzalloc(sizeof(struct deref_fetch_param),
				       GFP_KERNEL);
			if (!dprm)
990
				return -ENOMEM;
991 992 993
			dprm->offset = offset;
			ret = __parse_probe_arg(arg, t2, &dprm->orig,
						is_return);
994
			if (ret)
995
				kfree(dprm);
996
			else {
997
				f->fn = t->fetch[FETCH_MTD_deref];
998
				f->data = (void *)dprm;
999
			}
1000
		}
1001 1002
		break;
	}
1003 1004 1005
	if (!ret && !f->fn) {	/* Parsed, but do not find fetch method */
		pr_info("%s type has no corresponding fetch method.\n",
			t->name);
1006
		ret = -EINVAL;
1007
	}
1008 1009 1010
	return ret;
}

1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045
#define BYTES_TO_BITS(nb)	((BITS_PER_LONG * (nb)) / sizeof(long))

/* Bitfield type needs to be parsed into a fetch function */
static int __parse_bitfield_probe_arg(const char *bf,
				      const struct fetch_type *t,
				      struct fetch_param *f)
{
	struct bitfield_fetch_param *bprm;
	unsigned long bw, bo;
	char *tail;

	if (*bf != 'b')
		return 0;

	bprm = kzalloc(sizeof(*bprm), GFP_KERNEL);
	if (!bprm)
		return -ENOMEM;
	bprm->orig = *f;
	f->fn = t->fetch[FETCH_MTD_bitfield];
	f->data = (void *)bprm;

	bw = simple_strtoul(bf + 1, &tail, 0);	/* Use simple one */
	if (bw == 0 || *tail != '@')
		return -EINVAL;

	bf = tail + 1;
	bo = simple_strtoul(bf, &tail, 0);
	if (tail == bf || *tail != '/')
		return -EINVAL;

	bprm->hi_shift = BYTES_TO_BITS(t->size) - (bw + bo);
	bprm->low_shift = bprm->hi_shift + bo;
	return (BYTES_TO_BITS(t->size) < (bw + bo)) ? -EINVAL : 0;
}

1046
/* String length checking wrapper */
1047 1048
static int parse_probe_arg(char *arg, struct trace_probe *tp,
			   struct probe_arg *parg, int is_return)
1049
{
1050
	const char *t;
1051
	int ret;
1052

1053 1054 1055 1056
	if (strlen(arg) > MAX_ARGSTR_LEN) {
		pr_info("Argument is too long.: %s\n",  arg);
		return -ENOSPC;
	}
1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073
	parg->comm = kstrdup(arg, GFP_KERNEL);
	if (!parg->comm) {
		pr_info("Failed to allocate memory for command '%s'.\n", arg);
		return -ENOMEM;
	}
	t = strchr(parg->comm, ':');
	if (t) {
		arg[t - parg->comm] = '\0';
		t++;
	}
	parg->type = find_fetch_type(t);
	if (!parg->type) {
		pr_info("Unsupported type: %s\n", t);
		return -EINVAL;
	}
	parg->offset = tp->size;
	tp->size += parg->type->size;
1074
	ret = __parse_probe_arg(arg, parg->type, &parg->fetch, is_return);
1075
	if (ret >= 0 && t != NULL)
1076
		ret = __parse_bitfield_probe_arg(t, parg->type, &parg->fetch);
1077 1078 1079 1080 1081 1082
	if (ret >= 0) {
		parg->fetch_size.fn = get_fetch_size_function(parg->type,
							      parg->fetch.fn);
		parg->fetch_size.data = parg->fetch.data;
	}
	return ret;
1083 1084
}

1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
/* Return 1 if name is reserved or already used by another argument */
static int conflict_field_name(const char *name,
			       struct probe_arg *args, int narg)
{
	int i;
	for (i = 0; i < ARRAY_SIZE(reserved_field_names); i++)
		if (strcmp(reserved_field_names[i], name) == 0)
			return 1;
	for (i = 0; i < narg; i++)
		if (strcmp(args[i].name, name) == 0)
			return 1;
	return 0;
}

1099 1100 1101 1102
static int create_trace_probe(int argc, char **argv)
{
	/*
	 * Argument syntax:
1103 1104
	 *  - Add kprobe: p[:[GRP/]EVENT] [MOD:]KSYM[+OFFS]|KADDR [FETCHARGS]
	 *  - Add kretprobe: r[:[GRP/]EVENT] [MOD:]KSYM[+0] [FETCHARGS]
1105
	 * Fetch args:
1106 1107 1108
	 *  $retval	: fetch return value
	 *  $stack	: fetch stack address
	 *  $stackN	: fetch Nth of stack (N:0-)
1109 1110 1111
	 *  @ADDR	: fetch memory at ADDR (ADDR should be in kernel)
	 *  @SYM[+|-offs] : fetch memory at SYM +|- offs (SYM is a data symbol)
	 *  %REG	: fetch register REG
1112
	 * Dereferencing memory fetch:
1113
	 *  +|-offs(ARG) : fetch memory at ARG +|- offs address.
1114 1115
	 * Alias name of args:
	 *  NAME=FETCHARG : set NAME as alias of FETCHARG.
1116 1117
	 * Type of args:
	 *  FETCHARG:TYPE : use TYPE instead of unsigned long.
1118 1119 1120
	 */
	struct trace_probe *tp;
	int i, ret = 0;
1121
	int is_return = 0, is_delete = 0;
1122
	char *symbol = NULL, *event = NULL, *group = NULL;
1123
	char *arg;
1124
	unsigned long offset = 0;
1125
	void *addr = NULL;
1126
	char buf[MAX_EVENT_NAME_LEN];
1127

1128
	/* argc must be >= 1 */
1129 1130 1131 1132
	if (argv[0][0] == 'p')
		is_return = 0;
	else if (argv[0][0] == 'r')
		is_return = 1;
1133 1134
	else if (argv[0][0] == '-')
		is_delete = 1;
1135
	else {
1136 1137
		pr_info("Probe definition must be started with 'p', 'r' or"
			" '-'.\n");
1138
		return -EINVAL;
1139
	}
1140 1141 1142

	if (argv[0][1] == ':') {
		event = &argv[0][2];
1143 1144 1145 1146 1147
		if (strchr(event, '/')) {
			group = event;
			event = strchr(group, '/') + 1;
			event[-1] = '\0';
			if (strlen(group) == 0) {
1148
				pr_info("Group name is not specified\n");
1149 1150 1151
				return -EINVAL;
			}
		}
1152
		if (strlen(event) == 0) {
1153
			pr_info("Event name is not specified\n");
1154 1155 1156
			return -EINVAL;
		}
	}
1157 1158
	if (!group)
		group = KPROBE_EVENT_SYSTEM;
1159

1160 1161 1162 1163 1164
	if (is_delete) {
		if (!event) {
			pr_info("Delete command needs an event name.\n");
			return -EINVAL;
		}
1165
		mutex_lock(&probe_lock);
1166
		tp = find_trace_probe(event, group);
1167
		if (!tp) {
1168
			mutex_unlock(&probe_lock);
1169 1170 1171 1172 1173 1174
			pr_info("Event %s/%s doesn't exist.\n", group, event);
			return -ENOENT;
		}
		/* delete an event */
		unregister_trace_probe(tp);
		free_trace_probe(tp);
1175
		mutex_unlock(&probe_lock);
1176 1177 1178 1179 1180 1181 1182
		return 0;
	}

	if (argc < 2) {
		pr_info("Probe point is not specified.\n");
		return -EINVAL;
	}
1183
	if (isdigit(argv[1][0])) {
1184 1185
		if (is_return) {
			pr_info("Return probe point must be a symbol.\n");
1186
			return -EINVAL;
1187
		}
1188
		/* an address specified */
1189
		ret = strict_strtoul(&argv[1][0], 0, (unsigned long *)&addr);
1190 1191
		if (ret) {
			pr_info("Failed to parse address.\n");
1192
			return ret;
1193
		}
1194 1195 1196 1197 1198
	} else {
		/* a symbol specified */
		symbol = argv[1];
		/* TODO: support .init module functions */
		ret = split_symbol_offset(symbol, &offset);
1199 1200
		if (ret) {
			pr_info("Failed to parse symbol.\n");
1201
			return ret;
1202 1203 1204
		}
		if (offset && is_return) {
			pr_info("Return probe must be used without offset.\n");
1205
			return -EINVAL;
1206
		}
1207
	}
1208
	argc -= 2; argv += 2;
1209 1210

	/* setup a probe */
1211 1212 1213
	if (!event) {
		/* Make a new event name */
		if (symbol)
1214
			snprintf(buf, MAX_EVENT_NAME_LEN, "%c_%s_%ld",
1215 1216
				 is_return ? 'r' : 'p', symbol, offset);
		else
1217
			snprintf(buf, MAX_EVENT_NAME_LEN, "%c_0x%p",
1218
				 is_return ? 'r' : 'p', addr);
1219 1220
		event = buf;
	}
1221 1222
	tp = alloc_trace_probe(group, event, addr, symbol, offset, argc,
			       is_return);
1223 1224 1225
	if (IS_ERR(tp)) {
		pr_info("Failed to allocate trace_probe.(%d)\n",
			(int)PTR_ERR(tp));
1226
		return PTR_ERR(tp);
1227
	}
1228 1229

	/* parse arguments */
1230 1231
	ret = 0;
	for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) {
1232 1233 1234
		/* Increment count for freeing args in error case */
		tp->nr_args++;

1235 1236
		/* Parse argument name */
		arg = strchr(argv[i], '=');
1237
		if (arg) {
1238
			*arg++ = '\0';
1239 1240
			tp->args[i].name = kstrdup(argv[i], GFP_KERNEL);
		} else {
1241
			arg = argv[i];
1242 1243 1244 1245
			/* If argument name is omitted, set "argN" */
			snprintf(buf, MAX_EVENT_NAME_LEN, "arg%d", i + 1);
			tp->args[i].name = kstrdup(buf, GFP_KERNEL);
		}
1246

1247
		if (!tp->args[i].name) {
1248
			pr_info("Failed to allocate argument[%d] name.\n", i);
1249
			ret = -ENOMEM;
1250 1251
			goto error;
		}
1252 1253 1254 1255 1256 1257 1258

		if (!is_good_name(tp->args[i].name)) {
			pr_info("Invalid argument[%d] name: %s\n",
				i, tp->args[i].name);
			ret = -EINVAL;
			goto error;
		}
1259 1260

		if (conflict_field_name(tp->args[i].name, tp->args, i)) {
1261
			pr_info("Argument[%d] name '%s' conflicts with "
1262 1263 1264 1265
				"another field.\n", i, argv[i]);
			ret = -EINVAL;
			goto error;
		}
1266 1267

		/* Parse fetch argument */
1268
		ret = parse_probe_arg(arg, tp, &tp->args[i], is_return);
1269
		if (ret) {
1270
			pr_info("Parse error at argument[%d]. (%d)\n", i, ret);
1271
			goto error;
1272
		}
1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284
	}

	ret = register_trace_probe(tp);
	if (ret)
		goto error;
	return 0;

error:
	free_trace_probe(tp);
	return ret;
}

1285
static void release_all_trace_probes(void)
1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318
{
	struct trace_probe *tp;

	mutex_lock(&probe_lock);
	/* TODO: Use batch unregistration */
	while (!list_empty(&probe_list)) {
		tp = list_entry(probe_list.next, struct trace_probe, list);
		unregister_trace_probe(tp);
		free_trace_probe(tp);
	}
	mutex_unlock(&probe_lock);
}

/* Probes listing interfaces */
static void *probes_seq_start(struct seq_file *m, loff_t *pos)
{
	mutex_lock(&probe_lock);
	return seq_list_start(&probe_list, *pos);
}

static void *probes_seq_next(struct seq_file *m, void *v, loff_t *pos)
{
	return seq_list_next(v, &probe_list, pos);
}

static void probes_seq_stop(struct seq_file *m, void *v)
{
	mutex_unlock(&probe_lock);
}

static int probes_seq_show(struct seq_file *m, void *v)
{
	struct trace_probe *tp = v;
1319
	int i;
1320

1321
	seq_printf(m, "%c", trace_probe_is_return(tp) ? 'r' : 'p');
1322
	seq_printf(m, ":%s/%s", tp->call.class->system, tp->call.name);
1323

1324 1325 1326
	if (!tp->symbol)
		seq_printf(m, " 0x%p", tp->rp.kp.addr);
	else if (tp->rp.kp.offset)
1327 1328
		seq_printf(m, " %s+%u", trace_probe_symbol(tp),
			   tp->rp.kp.offset);
1329
	else
1330
		seq_printf(m, " %s", trace_probe_symbol(tp));
1331

1332 1333
	for (i = 0; i < tp->nr_args; i++)
		seq_printf(m, " %s=%s", tp->args[i].name, tp->args[i].comm);
1334
	seq_printf(m, "\n");
1335

1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349
	return 0;
}

static const struct seq_operations probes_seq_op = {
	.start  = probes_seq_start,
	.next   = probes_seq_next,
	.stop   = probes_seq_stop,
	.show   = probes_seq_show
};

static int probes_open(struct inode *inode, struct file *file)
{
	if ((file->f_mode & FMODE_WRITE) &&
	    (file->f_flags & O_TRUNC))
1350
		release_all_trace_probes();
1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370

	return seq_open(file, &probes_seq_op);
}

static int command_trace_probe(const char *buf)
{
	char **argv;
	int argc = 0, ret = 0;

	argv = argv_split(GFP_KERNEL, buf, &argc);
	if (!argv)
		return -ENOMEM;

	if (argc)
		ret = create_trace_probe(argc, argv);

	argv_free(argv);
	return ret;
}

1371
#define WRITE_BUFSIZE 4096
1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429

static ssize_t probes_write(struct file *file, const char __user *buffer,
			    size_t count, loff_t *ppos)
{
	char *kbuf, *tmp;
	int ret;
	size_t done;
	size_t size;

	kbuf = kmalloc(WRITE_BUFSIZE, GFP_KERNEL);
	if (!kbuf)
		return -ENOMEM;

	ret = done = 0;
	while (done < count) {
		size = count - done;
		if (size >= WRITE_BUFSIZE)
			size = WRITE_BUFSIZE - 1;
		if (copy_from_user(kbuf, buffer + done, size)) {
			ret = -EFAULT;
			goto out;
		}
		kbuf[size] = '\0';
		tmp = strchr(kbuf, '\n');
		if (tmp) {
			*tmp = '\0';
			size = tmp - kbuf + 1;
		} else if (done + size < count) {
			pr_warning("Line length is too long: "
				   "Should be less than %d.", WRITE_BUFSIZE);
			ret = -EINVAL;
			goto out;
		}
		done += size;
		/* Remove comments */
		tmp = strchr(kbuf, '#');
		if (tmp)
			*tmp = '\0';

		ret = command_trace_probe(kbuf);
		if (ret)
			goto out;
	}
	ret = done;
out:
	kfree(kbuf);
	return ret;
}

static const struct file_operations kprobe_events_ops = {
	.owner          = THIS_MODULE,
	.open           = probes_open,
	.read           = seq_read,
	.llseek         = seq_lseek,
	.release        = seq_release,
	.write		= probes_write,
};

1430 1431 1432 1433 1434 1435
/* Probes profiling interfaces */
static int probes_profile_seq_show(struct seq_file *m, void *v)
{
	struct trace_probe *tp = v;

	seq_printf(m, "  %-44s %15lu %15lu\n", tp->call.name, tp->nhit,
1436
		   tp->rp.kp.nmissed);
1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460

	return 0;
}

static const struct seq_operations profile_seq_op = {
	.start  = probes_seq_start,
	.next   = probes_seq_next,
	.stop   = probes_seq_stop,
	.show   = probes_profile_seq_show
};

static int profile_open(struct inode *inode, struct file *file)
{
	return seq_open(file, &profile_seq_op);
}

static const struct file_operations kprobe_profile_ops = {
	.owner          = THIS_MODULE,
	.open           = profile_open,
	.read           = seq_read,
	.llseek         = seq_lseek,
	.release        = seq_release,
};

1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
/* Sum up total data length for dynamic arraies (strings) */
static __kprobes int __get_data_size(struct trace_probe *tp,
				     struct pt_regs *regs)
{
	int i, ret = 0;
	u32 len;

	for (i = 0; i < tp->nr_args; i++)
		if (unlikely(tp->args[i].fetch_size.fn)) {
			call_fetch(&tp->args[i].fetch_size, regs, &len);
			ret += len;
		}

	return ret;
}

/* Store the value of each argument */
static __kprobes void store_trace_args(int ent_size, struct trace_probe *tp,
				       struct pt_regs *regs,
				       u8 *data, int maxlen)
{
	int i;
	u32 end = tp->size;
	u32 *dl;	/* Data (relative) location */

	for (i = 0; i < tp->nr_args; i++) {
		if (unlikely(tp->args[i].fetch_size.fn)) {
			/*
			 * First, we set the relative location and
			 * maximum data length to *dl
			 */
			dl = (u32 *)(data + tp->args[i].offset);
			*dl = make_data_rloc(maxlen, end - tp->args[i].offset);
			/* Then try to fetch string or dynamic array data */
			call_fetch(&tp->args[i].fetch, regs, dl);
			/* Reduce maximum length */
			end += get_rloc_len(*dl);
			maxlen -= get_rloc_len(*dl);
			/* Trick here, convert data_rloc to data_loc */
			*dl = convert_rloc_to_loc(*dl,
				 ent_size + tp->args[i].offset);
		} else
			/* Just fetching data normally */
			call_fetch(&tp->args[i].fetch, regs,
				   data + tp->args[i].offset);
	}
}

1509
/* Kprobe handler */
1510
static __kprobes void kprobe_trace_func(struct kprobe *kp, struct pt_regs *regs)
1511
{
1512
	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
1513
	struct kprobe_trace_entry_head *entry;
1514
	struct ring_buffer_event *event;
1515
	struct ring_buffer *buffer;
1516
	int size, dsize, pc;
1517
	unsigned long irq_flags;
1518
	struct ftrace_event_call *call = &tp->call;
1519

1520 1521
	tp->nhit++;

1522 1523 1524
	local_save_flags(irq_flags);
	pc = preempt_count();

1525 1526
	dsize = __get_data_size(tp, regs);
	size = sizeof(*entry) + tp->size + dsize;
1527

1528 1529
	event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
						  size, irq_flags, pc);
1530
	if (!event)
1531
		return;
1532 1533 1534

	entry = ring_buffer_event_data(event);
	entry->ip = (unsigned long)kp->addr;
1535
	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
1536

1537
	if (!filter_current_check_discard(buffer, call, entry, event))
1538 1539
		trace_nowake_buffer_unlock_commit_regs(buffer, event,
						       irq_flags, pc, regs);
1540 1541 1542
}

/* Kretprobe handler */
1543
static __kprobes void kretprobe_trace_func(struct kretprobe_instance *ri,
1544 1545 1546
					  struct pt_regs *regs)
{
	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
1547
	struct kretprobe_trace_entry_head *entry;
1548
	struct ring_buffer_event *event;
1549
	struct ring_buffer *buffer;
1550
	int size, pc, dsize;
1551
	unsigned long irq_flags;
1552
	struct ftrace_event_call *call = &tp->call;
1553 1554 1555 1556

	local_save_flags(irq_flags);
	pc = preempt_count();

1557 1558
	dsize = __get_data_size(tp, regs);
	size = sizeof(*entry) + tp->size + dsize;
1559

1560 1561
	event = trace_current_buffer_lock_reserve(&buffer, call->event.type,
						  size, irq_flags, pc);
1562
	if (!event)
1563
		return;
1564 1565

	entry = ring_buffer_event_data(event);
1566
	entry->func = (unsigned long)tp->rp.kp.addr;
1567
	entry->ret_ip = (unsigned long)ri->ret_addr;
1568
	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
1569

1570
	if (!filter_current_check_discard(buffer, call, entry, event))
1571 1572
		trace_nowake_buffer_unlock_commit_regs(buffer, event,
						       irq_flags, pc, regs);
1573 1574 1575 1576
}

/* Event entry printers */
enum print_line_t
1577 1578
print_kprobe_event(struct trace_iterator *iter, int flags,
		   struct trace_event *event)
1579
{
1580
	struct kprobe_trace_entry_head *field;
1581
	struct trace_seq *s = &iter->seq;
1582
	struct trace_probe *tp;
1583
	u8 *data;
1584 1585
	int i;

1586
	field = (struct kprobe_trace_entry_head *)iter->ent;
1587
	tp = container_of(event, struct trace_probe, call.event);
1588

1589 1590 1591
	if (!trace_seq_printf(s, "%s: (", tp->call.name))
		goto partial;

1592 1593 1594
	if (!seq_print_ip_sym(s, field->ip, flags | TRACE_ITER_SYM_OFFSET))
		goto partial;

1595
	if (!trace_seq_puts(s, ")"))
1596 1597
		goto partial;

1598 1599 1600
	data = (u8 *)&field[1];
	for (i = 0; i < tp->nr_args; i++)
		if (!tp->args[i].type->print(s, tp->args[i].name,
1601
					     data + tp->args[i].offset, field))
1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612
			goto partial;

	if (!trace_seq_puts(s, "\n"))
		goto partial;

	return TRACE_TYPE_HANDLED;
partial:
	return TRACE_TYPE_PARTIAL_LINE;
}

enum print_line_t
1613 1614
print_kretprobe_event(struct trace_iterator *iter, int flags,
		      struct trace_event *event)
1615
{
1616
	struct kretprobe_trace_entry_head *field;
1617
	struct trace_seq *s = &iter->seq;
1618
	struct trace_probe *tp;
1619
	u8 *data;
1620 1621
	int i;

1622
	field = (struct kretprobe_trace_entry_head *)iter->ent;
1623
	tp = container_of(event, struct trace_probe, call.event);
1624

1625 1626 1627
	if (!trace_seq_printf(s, "%s: (", tp->call.name))
		goto partial;

1628 1629 1630 1631 1632 1633 1634 1635 1636
	if (!seq_print_ip_sym(s, field->ret_ip, flags | TRACE_ITER_SYM_OFFSET))
		goto partial;

	if (!trace_seq_puts(s, " <- "))
		goto partial;

	if (!seq_print_ip_sym(s, field->func, flags & ~TRACE_ITER_SYM_OFFSET))
		goto partial;

1637
	if (!trace_seq_puts(s, ")"))
1638 1639
		goto partial;

1640 1641 1642
	data = (u8 *)&field[1];
	for (i = 0; i < tp->nr_args; i++)
		if (!tp->args[i].type->print(s, tp->args[i].name,
1643
					     data + tp->args[i].offset, field))
1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667
			goto partial;

	if (!trace_seq_puts(s, "\n"))
		goto partial;

	return TRACE_TYPE_HANDLED;
partial:
	return TRACE_TYPE_PARTIAL_LINE;
}

#undef DEFINE_FIELD
#define DEFINE_FIELD(type, item, name, is_signed)			\
	do {								\
		ret = trace_define_field(event_call, #type, name,	\
					 offsetof(typeof(field), item),	\
					 sizeof(field.item), is_signed, \
					 FILTER_OTHER);			\
		if (ret)						\
			return ret;					\
	} while (0)

static int kprobe_event_define_fields(struct ftrace_event_call *event_call)
{
	int ret, i;
1668
	struct kprobe_trace_entry_head field;
1669 1670
	struct trace_probe *tp = (struct trace_probe *)event_call->data;

1671
	DEFINE_FIELD(unsigned long, ip, FIELD_STRING_IP, 0);
1672
	/* Set argument names as fields */
1673
	for (i = 0; i < tp->nr_args; i++) {
1674
		ret = trace_define_field(event_call, tp->args[i].type->fmttype,
1675 1676 1677 1678 1679 1680 1681 1682
					 tp->args[i].name,
					 sizeof(field) + tp->args[i].offset,
					 tp->args[i].type->size,
					 tp->args[i].type->is_signed,
					 FILTER_OTHER);
		if (ret)
			return ret;
	}
1683 1684 1685 1686 1687 1688
	return 0;
}

static int kretprobe_event_define_fields(struct ftrace_event_call *event_call)
{
	int ret, i;
1689
	struct kretprobe_trace_entry_head field;
1690 1691
	struct trace_probe *tp = (struct trace_probe *)event_call->data;

1692 1693
	DEFINE_FIELD(unsigned long, func, FIELD_STRING_FUNC, 0);
	DEFINE_FIELD(unsigned long, ret_ip, FIELD_STRING_RETIP, 0);
1694
	/* Set argument names as fields */
1695
	for (i = 0; i < tp->nr_args; i++) {
1696
		ret = trace_define_field(event_call, tp->args[i].type->fmttype,
1697 1698 1699 1700 1701 1702 1703 1704
					 tp->args[i].name,
					 sizeof(field) + tp->args[i].offset,
					 tp->args[i].type->size,
					 tp->args[i].type->is_signed,
					 FILTER_OTHER);
		if (ret)
			return ret;
	}
1705 1706 1707
	return 0;
}

1708 1709 1710 1711 1712 1713 1714
static int __set_print_fmt(struct trace_probe *tp, char *buf, int len)
{
	int i;
	int pos = 0;

	const char *fmt, *arg;

1715
	if (!trace_probe_is_return(tp)) {
1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728
		fmt = "(%lx)";
		arg = "REC->" FIELD_STRING_IP;
	} else {
		fmt = "(%lx <- %lx)";
		arg = "REC->" FIELD_STRING_FUNC ", REC->" FIELD_STRING_RETIP;
	}

	/* When len=0, we just calculate the needed length */
#define LEN_OR_ZERO (len ? len - pos : 0)

	pos += snprintf(buf + pos, LEN_OR_ZERO, "\"%s", fmt);

	for (i = 0; i < tp->nr_args; i++) {
1729 1730
		pos += snprintf(buf + pos, LEN_OR_ZERO, " %s=%s",
				tp->args[i].name, tp->args[i].type->fmt);
1731 1732 1733 1734 1735
	}

	pos += snprintf(buf + pos, LEN_OR_ZERO, "\", %s", arg);

	for (i = 0; i < tp->nr_args; i++) {
1736 1737 1738 1739 1740 1741 1742
		if (strcmp(tp->args[i].type->name, "string") == 0)
			pos += snprintf(buf + pos, LEN_OR_ZERO,
					", __get_str(%s)",
					tp->args[i].name);
		else
			pos += snprintf(buf + pos, LEN_OR_ZERO, ", REC->%s",
					tp->args[i].name);
1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768
	}

#undef LEN_OR_ZERO

	/* return the length of print_fmt */
	return pos;
}

static int set_print_fmt(struct trace_probe *tp)
{
	int len;
	char *print_fmt;

	/* First: called with 0 length to calculate the needed length */
	len = __set_print_fmt(tp, NULL, 0);
	print_fmt = kmalloc(len + 1, GFP_KERNEL);
	if (!print_fmt)
		return -ENOMEM;

	/* Second: actually write the @print_fmt */
	__set_print_fmt(tp, print_fmt, len + 1);
	tp->call.print_fmt = print_fmt;

	return 0;
}

1769
#ifdef CONFIG_PERF_EVENTS
1770 1771

/* Kprobe profile handler */
1772
static __kprobes void kprobe_perf_func(struct kprobe *kp,
1773 1774 1775 1776
					 struct pt_regs *regs)
{
	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
	struct ftrace_event_call *call = &tp->call;
1777
	struct kprobe_trace_entry_head *entry;
1778
	struct hlist_head *head;
1779
	int size, __size, dsize;
1780
	int rctx;
1781

1782 1783
	dsize = __get_data_size(tp, regs);
	__size = sizeof(*entry) + tp->size + dsize;
1784 1785
	size = ALIGN(__size + sizeof(u32), sizeof(u64));
	size -= sizeof(u32);
1786
	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
1787
		     "profile buffer not large enough"))
1788
		return;
1789

S
Steven Rostedt 已提交
1790
	entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx);
1791
	if (!entry)
1792
		return;
1793 1794

	entry->ip = (unsigned long)kp->addr;
1795 1796
	memset(&entry[1], 0, dsize);
	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
1797

1798
	head = this_cpu_ptr(call->perf_events);
1799
	perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head);
1800 1801 1802
}

/* Kretprobe profile handler */
1803
static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri,
1804 1805 1806 1807
					    struct pt_regs *regs)
{
	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);
	struct ftrace_event_call *call = &tp->call;
1808
	struct kretprobe_trace_entry_head *entry;
1809
	struct hlist_head *head;
1810
	int size, __size, dsize;
1811
	int rctx;
1812

1813 1814
	dsize = __get_data_size(tp, regs);
	__size = sizeof(*entry) + tp->size + dsize;
1815 1816
	size = ALIGN(__size + sizeof(u32), sizeof(u64));
	size -= sizeof(u32);
1817
	if (WARN_ONCE(size > PERF_MAX_TRACE_SIZE,
1818
		     "profile buffer not large enough"))
1819
		return;
1820

S
Steven Rostedt 已提交
1821
	entry = perf_trace_buf_prepare(size, call->event.type, regs, &rctx);
1822
	if (!entry)
1823
		return;
1824

1825 1826
	entry->func = (unsigned long)tp->rp.kp.addr;
	entry->ret_ip = (unsigned long)ri->ret_addr;
1827
	store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize);
1828

1829
	head = this_cpu_ptr(call->perf_events);
1830
	perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, regs, head);
1831
}
1832
#endif	/* CONFIG_PERF_EVENTS */
1833

1834 1835 1836
static __kprobes
int kprobe_register(struct ftrace_event_call *event, enum trace_reg type)
{
1837 1838
	struct trace_probe *tp = (struct trace_probe *)event->data;

1839 1840
	switch (type) {
	case TRACE_REG_REGISTER:
1841
		return enable_trace_probe(tp, TP_FLAG_TRACE);
1842
	case TRACE_REG_UNREGISTER:
1843
		disable_trace_probe(tp, TP_FLAG_TRACE);
1844 1845 1846 1847
		return 0;

#ifdef CONFIG_PERF_EVENTS
	case TRACE_REG_PERF_REGISTER:
1848
		return enable_trace_probe(tp, TP_FLAG_PROFILE);
1849
	case TRACE_REG_PERF_UNREGISTER:
1850
		disable_trace_probe(tp, TP_FLAG_PROFILE);
1851 1852 1853 1854 1855
		return 0;
#endif
	}
	return 0;
}
1856 1857 1858 1859 1860

static __kprobes
int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs)
{
	struct trace_probe *tp = container_of(kp, struct trace_probe, rp.kp);
1861

1862 1863
	if (tp->flags & TP_FLAG_TRACE)
		kprobe_trace_func(kp, regs);
1864
#ifdef CONFIG_PERF_EVENTS
1865
	if (tp->flags & TP_FLAG_PROFILE)
1866
		kprobe_perf_func(kp, regs);
1867
#endif
1868 1869 1870 1871 1872 1873 1874 1875 1876 1877
	return 0;	/* We don't tweek kernel, so just return 0 */
}

static __kprobes
int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs)
{
	struct trace_probe *tp = container_of(ri->rp, struct trace_probe, rp);

	if (tp->flags & TP_FLAG_TRACE)
		kretprobe_trace_func(ri, regs);
1878
#ifdef CONFIG_PERF_EVENTS
1879
	if (tp->flags & TP_FLAG_PROFILE)
1880
		kretprobe_perf_func(ri, regs);
1881
#endif
1882 1883
	return 0;	/* We don't tweek kernel, so just return 0 */
}
1884

1885 1886 1887 1888 1889 1890 1891 1892
static struct trace_event_functions kretprobe_funcs = {
	.trace		= print_kretprobe_event
};

static struct trace_event_functions kprobe_funcs = {
	.trace		= print_kprobe_event
};

1893 1894 1895 1896 1897 1898
static int register_probe_event(struct trace_probe *tp)
{
	struct ftrace_event_call *call = &tp->call;
	int ret;

	/* Initialize ftrace_event_call */
1899
	INIT_LIST_HEAD(&call->class->fields);
1900
	if (trace_probe_is_return(tp)) {
1901
		call->event.funcs = &kretprobe_funcs;
1902
		call->class->define_fields = kretprobe_event_define_fields;
1903
	} else {
1904
		call->event.funcs = &kprobe_funcs;
1905
		call->class->define_fields = kprobe_event_define_fields;
1906
	}
1907 1908
	if (set_print_fmt(tp) < 0)
		return -ENOMEM;
1909 1910
	ret = register_ftrace_event(&call->event);
	if (!ret) {
1911
		kfree(call->print_fmt);
1912
		return -ENODEV;
1913
	}
1914
	call->flags = 0;
1915
	call->class->reg = kprobe_register;
1916 1917
	call->data = tp;
	ret = trace_add_event_call(call);
1918
	if (ret) {
1919
		pr_info("Failed to register kprobe event: %s\n", call->name);
1920
		kfree(call->print_fmt);
1921
		unregister_ftrace_event(&call->event);
1922
	}
1923 1924 1925 1926 1927
	return ret;
}

static void unregister_probe_event(struct trace_probe *tp)
{
1928
	/* tp->event is unregistered in trace_remove_event_call() */
1929
	trace_remove_event_call(&tp->call);
1930
	kfree(tp->call.print_fmt);
1931 1932
}

L
Lucas De Marchi 已提交
1933
/* Make a debugfs interface for controlling probe points */
1934 1935 1936 1937 1938
static __init int init_kprobe_trace(void)
{
	struct dentry *d_tracer;
	struct dentry *entry;

1939 1940 1941
	if (register_module_notifier(&trace_probe_module_nb))
		return -EINVAL;

1942 1943 1944 1945 1946 1947 1948
	d_tracer = tracing_init_dentry();
	if (!d_tracer)
		return 0;

	entry = debugfs_create_file("kprobe_events", 0644, d_tracer,
				    NULL, &kprobe_events_ops);

1949
	/* Event list interface */
1950 1951 1952
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'kprobe_events' entry\n");
1953 1954 1955 1956 1957 1958 1959 1960

	/* Profile interface */
	entry = debugfs_create_file("kprobe_profile", 0444, d_tracer,
				    NULL, &kprobe_profile_ops);

	if (!entry)
		pr_warning("Could not create debugfs "
			   "'kprobe_profile' entry\n");
1961 1962 1963 1964 1965 1966 1967
	return 0;
}
fs_initcall(init_kprobe_trace);


#ifdef CONFIG_FTRACE_STARTUP_TEST

1968 1969 1970 1971 1972 1973
/*
 * The "__used" keeps gcc from removing the function symbol
 * from the kallsyms table.
 */
static __used int kprobe_trace_selftest_target(int a1, int a2, int a3,
					       int a4, int a5, int a6)
1974 1975 1976 1977 1978 1979
{
	return a1 + a2 + a3 + a4 + a5 + a6;
}

static __init int kprobe_trace_self_tests_init(void)
{
1980
	int ret, warn = 0;
1981
	int (*target)(int, int, int, int, int, int);
1982
	struct trace_probe *tp;
1983 1984 1985 1986 1987 1988

	target = kprobe_trace_selftest_target;

	pr_info("Testing kprobe tracing: ");

	ret = command_trace_probe("p:testprobe kprobe_trace_selftest_target "
1989 1990 1991 1992 1993 1994
				  "$stack $stack0 +0($stack)");
	if (WARN_ON_ONCE(ret)) {
		pr_warning("error on probing function entry.\n");
		warn++;
	} else {
		/* Enable trace point */
1995
		tp = find_trace_probe("testprobe", KPROBE_EVENT_SYSTEM);
1996 1997 1998 1999
		if (WARN_ON_ONCE(tp == NULL)) {
			pr_warning("error on getting new probe.\n");
			warn++;
		} else
2000
			enable_trace_probe(tp, TP_FLAG_TRACE);
2001
	}
2002 2003

	ret = command_trace_probe("r:testprobe2 kprobe_trace_selftest_target "
2004
				  "$retval");
2005 2006 2007 2008 2009
	if (WARN_ON_ONCE(ret)) {
		pr_warning("error on probing function return.\n");
		warn++;
	} else {
		/* Enable trace point */
2010
		tp = find_trace_probe("testprobe2", KPROBE_EVENT_SYSTEM);
2011 2012 2013 2014
		if (WARN_ON_ONCE(tp == NULL)) {
			pr_warning("error on getting new probe.\n");
			warn++;
		} else
2015
			enable_trace_probe(tp, TP_FLAG_TRACE);
2016 2017 2018 2019
	}

	if (warn)
		goto end;
2020 2021 2022

	ret = target(1, 2, 3, 4, 5, 6);

2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033
	ret = command_trace_probe("-:testprobe");
	if (WARN_ON_ONCE(ret)) {
		pr_warning("error on deleting a probe.\n");
		warn++;
	}

	ret = command_trace_probe("-:testprobe2");
	if (WARN_ON_ONCE(ret)) {
		pr_warning("error on deleting a probe.\n");
		warn++;
	}
2034

2035
end:
2036
	release_all_trace_probes();
2037 2038 2039 2040
	if (warn)
		pr_cont("NG: Some tests are failed. Please check them.\n");
	else
		pr_cont("OK\n");
2041 2042 2043 2044 2045 2046
	return 0;
}

late_initcall(kprobe_trace_self_tests_init);

#endif