ftrace.c 44.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
/*
 * Infrastructure for profiling code inserted by 'gcc -pg'.
 *
 * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
 * Copyright (C) 2004-2008 Ingo Molnar <mingo@redhat.com>
 *
 * Originally ported from the -rt patch by:
 *   Copyright (C) 2007 Arnaldo Carvalho de Melo <acme@redhat.com>
 *
 * Based on code in the latency_tracer, that is:
 *
 *  Copyright (C) 2004-2006 Ingo Molnar
 *  Copyright (C) 2004 William Lee Irwin III
 */

16 17 18
#include <linux/stop_machine.h>
#include <linux/clocksource.h>
#include <linux/kallsyms.h>
19
#include <linux/seq_file.h>
20
#include <linux/suspend.h>
21
#include <linux/debugfs.h>
22
#include <linux/hardirq.h>
I
Ingo Molnar 已提交
23
#include <linux/kthread.h>
24
#include <linux/uaccess.h>
25
#include <linux/kprobes.h>
I
Ingo Molnar 已提交
26
#include <linux/ftrace.h>
27
#include <linux/sysctl.h>
28
#include <linux/ctype.h>
29 30
#include <linux/list.h>

31 32
#include <asm/ftrace.h>

33
#include "trace.h"
34

35 36 37 38 39 40 41 42 43 44 45 46
#define FTRACE_WARN_ON(cond)			\
	do {					\
		if (WARN_ON(cond))		\
			ftrace_kill();		\
	} while (0)

#define FTRACE_WARN_ON_ONCE(cond)		\
	do {					\
		if (WARN_ON_ONCE(cond))		\
			ftrace_kill();		\
	} while (0)

47 48
/* ftrace_enabled is a method to turn ftrace on or off */
int ftrace_enabled __read_mostly;
49
static int last_ftrace_enabled;
50

51
/* set when tracing only a pid */
S
Steven Rostedt 已提交
52
struct pid *ftrace_pid_trace;
53
static struct pid * const ftrace_swapper_pid = &init_struct_pid;
54

55 56 57
/* Quick disabling of function tracer. */
int function_trace_stop;

58 59 60 61 62 63
/*
 * ftrace_disabled is set when an anomaly is discovered.
 * ftrace_disabled is much stronger than ftrace_enabled.
 */
static int ftrace_disabled __read_mostly;

64
static DEFINE_SPINLOCK(ftrace_lock);
65
static DEFINE_MUTEX(ftrace_sysctl_lock);
66
static DEFINE_MUTEX(ftrace_start_lock);
67

68 69 70 71 72 73 74
static struct ftrace_ops ftrace_list_end __read_mostly =
{
	.func = ftrace_stub,
};

static struct ftrace_ops *ftrace_list __read_mostly = &ftrace_list_end;
ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub;
75
ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub;
76
ftrace_func_t ftrace_pid_function __read_mostly = ftrace_stub;
77

I
Ingo Molnar 已提交
78
static void ftrace_list_func(unsigned long ip, unsigned long parent_ip)
79 80 81 82 83 84 85 86 87 88 89 90 91 92
{
	struct ftrace_ops *op = ftrace_list;

	/* in case someone actually ports this to alpha! */
	read_barrier_depends();

	while (op != &ftrace_list_end) {
		/* silly alpha */
		read_barrier_depends();
		op->func(ip, parent_ip);
		op = op->next;
	};
}

93 94
static void ftrace_pid_func(unsigned long ip, unsigned long parent_ip)
{
95
	if (!test_tsk_trace_trace(current))
96 97 98 99 100 101 102 103 104 105 106 107
		return;

	ftrace_pid_function(ip, parent_ip);
}

static void set_ftrace_pid_function(ftrace_func_t func)
{
	/* do not set ftrace_pid_function to itself! */
	if (func != ftrace_pid_func)
		ftrace_pid_function = func;
}

108
/**
109
 * clear_ftrace_function - reset the ftrace function
110
 *
111 112
 * This NULLs the ftrace function and in essence stops
 * tracing.  There may be lag
113
 */
114
void clear_ftrace_function(void)
115
{
116
	ftrace_trace_function = ftrace_stub;
117
	__ftrace_trace_function = ftrace_stub;
118
	ftrace_pid_function = ftrace_stub;
119 120
}

121 122 123 124 125 126 127 128 129 130 131 132 133 134
#ifndef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
/*
 * For those archs that do not test ftrace_trace_stop in their
 * mcount call site, we need to do it from C.
 */
static void ftrace_test_stop_func(unsigned long ip, unsigned long parent_ip)
{
	if (function_trace_stop)
		return;

	__ftrace_trace_function(ip, parent_ip);
}
#endif

I
Ingo Molnar 已提交
135
static int __register_ftrace_function(struct ftrace_ops *ops)
136
{
137
	/* should not be called from interrupt context */
138
	spin_lock(&ftrace_lock);
139 140 141 142 143 144 145 146 147 148

	ops->next = ftrace_list;
	/*
	 * We are entering ops into the ftrace_list but another
	 * CPU might be walking that list. We need to make sure
	 * the ops->next pointer is valid before another CPU sees
	 * the ops pointer included into the ftrace_list.
	 */
	smp_wmb();
	ftrace_list = ops;
149

150
	if (ftrace_enabled) {
151 152 153 154 155 156 157
		ftrace_func_t func;

		if (ops->next == &ftrace_list_end)
			func = ops->func;
		else
			func = ftrace_list_func;

S
Steven Rostedt 已提交
158
		if (ftrace_pid_trace) {
159 160 161 162
			set_ftrace_pid_function(func);
			func = ftrace_pid_func;
		}

163 164 165 166
		/*
		 * For one func, simply call it directly.
		 * For more than one func, call the chain.
		 */
167
#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
168
		ftrace_trace_function = func;
169
#else
170
		__ftrace_trace_function = func;
171 172
		ftrace_trace_function = ftrace_test_stop_func;
#endif
173
	}
174 175

	spin_unlock(&ftrace_lock);
176 177 178 179

	return 0;
}

I
Ingo Molnar 已提交
180
static int __unregister_ftrace_function(struct ftrace_ops *ops)
181 182 183 184
{
	struct ftrace_ops **p;
	int ret = 0;

185
	/* should not be called from interrupt context */
186
	spin_lock(&ftrace_lock);
187 188

	/*
189 190
	 * If we are removing the last function, then simply point
	 * to the ftrace_stub.
191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
	 */
	if (ftrace_list == ops && ops->next == &ftrace_list_end) {
		ftrace_trace_function = ftrace_stub;
		ftrace_list = &ftrace_list_end;
		goto out;
	}

	for (p = &ftrace_list; *p != &ftrace_list_end; p = &(*p)->next)
		if (*p == ops)
			break;

	if (*p != ops) {
		ret = -1;
		goto out;
	}

	*p = (*p)->next;

209 210
	if (ftrace_enabled) {
		/* If we only have one func left, then call that directly */
211 212 213
		if (ftrace_list->next == &ftrace_list_end) {
			ftrace_func_t func = ftrace_list->func;

S
Steven Rostedt 已提交
214
			if (ftrace_pid_trace) {
215 216 217 218 219 220 221 222 223
				set_ftrace_pid_function(func);
				func = ftrace_pid_func;
			}
#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
			ftrace_trace_function = func;
#else
			__ftrace_trace_function = func;
#endif
		}
224
	}
225 226

 out:
227 228 229 230 231
	spin_unlock(&ftrace_lock);

	return ret;
}

232 233 234 235 236 237 238 239 240 241 242 243
static void ftrace_update_pid_func(void)
{
	ftrace_func_t func;

	/* should not be called from interrupt context */
	spin_lock(&ftrace_lock);

	if (ftrace_trace_function == ftrace_stub)
		goto out;

	func = ftrace_trace_function;

S
Steven Rostedt 已提交
244
	if (ftrace_pid_trace) {
245 246 247
		set_ftrace_pid_function(func);
		func = ftrace_pid_func;
	} else {
248 249
		if (func == ftrace_pid_func)
			func = ftrace_pid_function;
250 251 252 253 254 255 256 257 258 259 260 261
	}

#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST
	ftrace_trace_function = func;
#else
	__ftrace_trace_function = func;
#endif

 out:
	spin_unlock(&ftrace_lock);
}

262
#ifdef CONFIG_DYNAMIC_FTRACE
263
#ifndef CONFIG_FTRACE_MCOUNT_RECORD
S
Steven Rostedt 已提交
264
# error Dynamic ftrace depends on MCOUNT_RECORD
265 266
#endif

267 268 269 270 271 272 273 274
/*
 * Since MCOUNT_ADDR may point to mcount itself, we do not want
 * to get it confused by reading a reference in the code as we
 * are parsing on objcopy output of text. Use a variable for
 * it instead.
 */
static unsigned long mcount_addr = MCOUNT_ADDR;

275 276 277 278 279 280
enum {
	FTRACE_ENABLE_CALLS		= (1 << 0),
	FTRACE_DISABLE_CALLS		= (1 << 1),
	FTRACE_UPDATE_TRACE_FUNC	= (1 << 2),
	FTRACE_ENABLE_MCOUNT		= (1 << 3),
	FTRACE_DISABLE_MCOUNT		= (1 << 4),
281 282
	FTRACE_START_FUNC_RET		= (1 << 5),
	FTRACE_STOP_FUNC_RET		= (1 << 6),
283 284
};

285 286
static int ftrace_filtered;

S
Steven Rostedt 已提交
287
static LIST_HEAD(ftrace_new_addrs);
288

289
static DEFINE_MUTEX(ftrace_regex_lock);
290

291 292
struct ftrace_page {
	struct ftrace_page	*next;
293
	unsigned long		index;
294
	struct dyn_ftrace	records[];
295
};
296 297 298 299 300 301 302 303 304 305

#define ENTRIES_PER_PAGE \
  ((PAGE_SIZE - sizeof(struct ftrace_page)) / sizeof(struct dyn_ftrace))

/* estimate from running different kernels */
#define NR_TO_INIT		10000

static struct ftrace_page	*ftrace_pages_start;
static struct ftrace_page	*ftrace_pages;

306 307
static struct dyn_ftrace *ftrace_free_records;

A
Abhishek Sagar 已提交
308 309

#ifdef CONFIG_KPROBES
310 311 312

static int frozen_record_count;

A
Abhishek Sagar 已提交
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338
static inline void freeze_record(struct dyn_ftrace *rec)
{
	if (!(rec->flags & FTRACE_FL_FROZEN)) {
		rec->flags |= FTRACE_FL_FROZEN;
		frozen_record_count++;
	}
}

static inline void unfreeze_record(struct dyn_ftrace *rec)
{
	if (rec->flags & FTRACE_FL_FROZEN) {
		rec->flags &= ~FTRACE_FL_FROZEN;
		frozen_record_count--;
	}
}

static inline int record_frozen(struct dyn_ftrace *rec)
{
	return rec->flags & FTRACE_FL_FROZEN;
}
#else
# define freeze_record(rec)			({ 0; })
# define unfreeze_record(rec)			({ 0; })
# define record_frozen(rec)			({ 0; })
#endif /* CONFIG_KPROBES */

I
Ingo Molnar 已提交
339
static void ftrace_free_rec(struct dyn_ftrace *rec)
340 341 342 343 344 345
{
	rec->ip = (unsigned long)ftrace_free_records;
	ftrace_free_records = rec;
	rec->flags |= FTRACE_FL_FREE;
}

346 347 348 349 350 351 352 353
void ftrace_release(void *start, unsigned long size)
{
	struct dyn_ftrace *rec;
	struct ftrace_page *pg;
	unsigned long s = (unsigned long)start;
	unsigned long e = s + size;
	int i;

354
	if (ftrace_disabled || !start)
355 356
		return;

357
	/* should not be called from interrupt context */
358 359 360 361 362 363 364 365 366 367 368 369 370
	spin_lock(&ftrace_lock);

	for (pg = ftrace_pages_start; pg; pg = pg->next) {
		for (i = 0; i < pg->index; i++) {
			rec = &pg->records[i];

			if ((rec->ip >= s) && (rec->ip < e))
				ftrace_free_rec(rec);
		}
	}
	spin_unlock(&ftrace_lock);
}

I
Ingo Molnar 已提交
371
static struct dyn_ftrace *ftrace_alloc_dyn_node(unsigned long ip)
372
{
373 374 375 376 377 378 379
	struct dyn_ftrace *rec;

	/* First check for freed records */
	if (ftrace_free_records) {
		rec = ftrace_free_records;

		if (unlikely(!(rec->flags & FTRACE_FL_FREE))) {
380
			FTRACE_WARN_ON_ONCE(1);
381 382 383 384 385 386 387 388 389
			ftrace_free_records = NULL;
			return NULL;
		}

		ftrace_free_records = (void *)rec->ip;
		memset(rec, 0, sizeof(*rec));
		return rec;
	}

390
	if (ftrace_pages->index == ENTRIES_PER_PAGE) {
S
Steven Rostedt 已提交
391 392 393 394 395 396 397
		if (!ftrace_pages->next) {
			/* allocate another page */
			ftrace_pages->next =
				(void *)get_zeroed_page(GFP_KERNEL);
			if (!ftrace_pages->next)
				return NULL;
		}
398 399 400 401 402 403
		ftrace_pages = ftrace_pages->next;
	}

	return &ftrace_pages->records[ftrace_pages->index++];
}

S
Steven Rostedt 已提交
404
static struct dyn_ftrace *
405
ftrace_record_ip(unsigned long ip)
406
{
S
Steven Rostedt 已提交
407
	struct dyn_ftrace *rec;
408

409
	if (ftrace_disabled)
S
Steven Rostedt 已提交
410
		return NULL;
411

S
Steven Rostedt 已提交
412 413 414
	rec = ftrace_alloc_dyn_node(ip);
	if (!rec)
		return NULL;
415

S
Steven Rostedt 已提交
416
	rec->ip = ip;
417

S
Steven Rostedt 已提交
418
	list_add(&rec->list, &ftrace_new_addrs);
419

S
Steven Rostedt 已提交
420
	return rec;
421 422
}

423 424 425 426 427 428 429 430 431 432
static void print_ip_ins(const char *fmt, unsigned char *p)
{
	int i;

	printk(KERN_CONT "%s", fmt);

	for (i = 0; i < MCOUNT_INSN_SIZE; i++)
		printk(KERN_CONT "%s%02x", i ? ":" : "", p[i]);
}

433
static void ftrace_bug(int failed, unsigned long ip)
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
{
	switch (failed) {
	case -EFAULT:
		FTRACE_WARN_ON_ONCE(1);
		pr_info("ftrace faulted on modifying ");
		print_ip_sym(ip);
		break;
	case -EINVAL:
		FTRACE_WARN_ON_ONCE(1);
		pr_info("ftrace failed to modify ");
		print_ip_sym(ip);
		print_ip_ins(" actual: ", (unsigned char *)ip);
		printk(KERN_CONT "\n");
		break;
	case -EPERM:
		FTRACE_WARN_ON_ONCE(1);
		pr_info("ftrace faulted on writing ");
		print_ip_sym(ip);
		break;
	default:
		FTRACE_WARN_ON_ONCE(1);
		pr_info("ftrace faulted on unknown error ");
		print_ip_sym(ip);
	}
}

460

461
static int
462
__ftrace_replace_code(struct dyn_ftrace *rec, int enable)
463
{
464
	unsigned long ip, fl;
465 466 467
	unsigned long ftrace_addr;

	ftrace_addr = (unsigned long)ftrace_caller;
468 469 470

	ip = rec->ip;

S
Steven Rostedt 已提交
471 472 473 474 475 476 477 478 479 480 481 482 483 484 485
	/*
	 * If this record is not to be traced and
	 * it is not enabled then do nothing.
	 *
	 * If this record is not to be traced and
	 * it is enabled then disabled it.
	 *
	 */
	if (rec->flags & FTRACE_FL_NOTRACE) {
		if (rec->flags & FTRACE_FL_ENABLED)
			rec->flags &= ~FTRACE_FL_ENABLED;
		else
			return 0;

	} else if (ftrace_filtered && enable) {
486
		/*
S
Steven Rostedt 已提交
487
		 * Filtering is on:
488
		 */
489

S
Steven Rostedt 已提交
490
		fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED);
491

S
Steven Rostedt 已提交
492 493
		/* Record is filtered and enabled, do nothing */
		if (fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED))
494
			return 0;
495

S
Steven Rostedt 已提交
496 497 498 499 500 501
		/* Record is not filtered and is not enabled do nothing */
		if (!fl)
			return 0;

		/* Record is not filtered but enabled, disable it */
		if (fl == FTRACE_FL_ENABLED)
502
			rec->flags &= ~FTRACE_FL_ENABLED;
S
Steven Rostedt 已提交
503 504
		else
		/* Otherwise record is filtered but not enabled, enable it */
505 506
			rec->flags |= FTRACE_FL_ENABLED;
	} else {
S
Steven Rostedt 已提交
507
		/* Disable or not filtered */
508

509
		if (enable) {
S
Steven Rostedt 已提交
510
			/* if record is enabled, do nothing */
511
			if (rec->flags & FTRACE_FL_ENABLED)
512
				return 0;
S
Steven Rostedt 已提交
513

514
			rec->flags |= FTRACE_FL_ENABLED;
S
Steven Rostedt 已提交
515

516
		} else {
S
Steven Rostedt 已提交
517 518

			/* if record is not enabled do nothing */
519
			if (!(rec->flags & FTRACE_FL_ENABLED))
520
				return 0;
S
Steven Rostedt 已提交
521

522 523 524 525
			rec->flags &= ~FTRACE_FL_ENABLED;
		}
	}

S
Steven Rostedt 已提交
526
	if (rec->flags & FTRACE_FL_ENABLED)
527
		return ftrace_make_call(rec, ftrace_addr);
528
	else
529
		return ftrace_make_nop(NULL, rec, ftrace_addr);
530 531
}

I
Ingo Molnar 已提交
532
static void ftrace_replace_code(int enable)
533
{
534
	int i, failed;
535 536 537 538 539 540 541
	struct dyn_ftrace *rec;
	struct ftrace_page *pg;

	for (pg = ftrace_pages_start; pg; pg = pg->next) {
		for (i = 0; i < pg->index; i++) {
			rec = &pg->records[i];

542 543 544 545 546 547
			/*
			 * Skip over free records and records that have
			 * failed.
			 */
			if (rec->flags & FTRACE_FL_FREE ||
			    rec->flags & FTRACE_FL_FAILED)
548 549
				continue;

550
			/* ignore updates to this record's mcount site */
551 552
			if (get_kprobe((void *)rec->ip)) {
				freeze_record(rec);
553
				continue;
554 555 556
			} else {
				unfreeze_record(rec);
			}
557

558
			failed = __ftrace_replace_code(rec, enable);
559 560 561
			if (failed && (rec->flags & FTRACE_FL_CONVERTED)) {
				rec->flags |= FTRACE_FL_FAILED;
				if ((system_state == SYSTEM_BOOTING) ||
562
				    !core_kernel_text(rec->ip)) {
563
					ftrace_free_rec(rec);
564
				} else
565
					ftrace_bug(failed, rec->ip);
566
			}
567 568 569 570
		}
	}
}

571
static int
572
ftrace_code_disable(struct module *mod, struct dyn_ftrace *rec)
573 574
{
	unsigned long ip;
575
	int ret;
576 577 578

	ip = rec->ip;

579
	ret = ftrace_make_nop(mod, rec, mcount_addr);
580
	if (ret) {
581
		ftrace_bug(ret, ip);
582
		rec->flags |= FTRACE_FL_FAILED;
583
		return 0;
584
	}
585
	return 1;
586 587
}

I
Ingo Molnar 已提交
588
static int __ftrace_modify_code(void *data)
589
{
590 591
	int *command = data;

592
	if (*command & FTRACE_ENABLE_CALLS)
593
		ftrace_replace_code(1);
594
	else if (*command & FTRACE_DISABLE_CALLS)
595 596 597 598 599
		ftrace_replace_code(0);

	if (*command & FTRACE_UPDATE_TRACE_FUNC)
		ftrace_update_ftrace_func(ftrace_trace_function);

600 601 602 603 604
	if (*command & FTRACE_START_FUNC_RET)
		ftrace_enable_ftrace_graph_caller();
	else if (*command & FTRACE_STOP_FUNC_RET)
		ftrace_disable_ftrace_graph_caller();

605
	return 0;
606 607
}

I
Ingo Molnar 已提交
608
static void ftrace_run_update_code(int command)
609
{
R
Rusty Russell 已提交
610
	stop_machine(__ftrace_modify_code, &command, NULL);
611 612
}

613
static ftrace_func_t saved_ftrace_func;
614
static int ftrace_start_up;
615 616 617 618 619 620 621 622 623 624 625 626 627

static void ftrace_startup_enable(int command)
{
	if (saved_ftrace_func != ftrace_trace_function) {
		saved_ftrace_func = ftrace_trace_function;
		command |= FTRACE_UPDATE_TRACE_FUNC;
	}

	if (!command || !ftrace_enabled)
		return;

	ftrace_run_update_code(command);
}
628

629
static void ftrace_startup(int command)
630
{
631 632 633
	if (unlikely(ftrace_disabled))
		return;

S
Steven Rostedt 已提交
634
	mutex_lock(&ftrace_start_lock);
635
	ftrace_start_up++;
S
Steven Rostedt 已提交
636
	command |= FTRACE_ENABLE_CALLS;
637

638
	ftrace_startup_enable(command);
639

S
Steven Rostedt 已提交
640
	mutex_unlock(&ftrace_start_lock);
641 642
}

643
static void ftrace_shutdown(int command)
644
{
645 646 647
	if (unlikely(ftrace_disabled))
		return;

S
Steven Rostedt 已提交
648
	mutex_lock(&ftrace_start_lock);
649 650
	ftrace_start_up--;
	if (!ftrace_start_up)
651
		command |= FTRACE_DISABLE_CALLS;
652

653 654 655 656
	if (saved_ftrace_func != ftrace_trace_function) {
		saved_ftrace_func = ftrace_trace_function;
		command |= FTRACE_UPDATE_TRACE_FUNC;
	}
657

658 659 660 661
	if (!command || !ftrace_enabled)
		goto out;

	ftrace_run_update_code(command);
662
 out:
S
Steven Rostedt 已提交
663
	mutex_unlock(&ftrace_start_lock);
664 665
}

I
Ingo Molnar 已提交
666
static void ftrace_startup_sysctl(void)
667
{
668 669
	int command = FTRACE_ENABLE_MCOUNT;

670 671 672
	if (unlikely(ftrace_disabled))
		return;

S
Steven Rostedt 已提交
673
	mutex_lock(&ftrace_start_lock);
674 675
	/* Force update next time */
	saved_ftrace_func = NULL;
676 677
	/* ftrace_start_up is true if we want ftrace running */
	if (ftrace_start_up)
678 679 680
		command |= FTRACE_ENABLE_CALLS;

	ftrace_run_update_code(command);
S
Steven Rostedt 已提交
681
	mutex_unlock(&ftrace_start_lock);
682 683
}

I
Ingo Molnar 已提交
684
static void ftrace_shutdown_sysctl(void)
685
{
686 687
	int command = FTRACE_DISABLE_MCOUNT;

688 689 690
	if (unlikely(ftrace_disabled))
		return;

S
Steven Rostedt 已提交
691
	mutex_lock(&ftrace_start_lock);
692 693
	/* ftrace_start_up is true if ftrace is running */
	if (ftrace_start_up)
694 695 696
		command |= FTRACE_DISABLE_CALLS;

	ftrace_run_update_code(command);
S
Steven Rostedt 已提交
697
	mutex_unlock(&ftrace_start_lock);
698 699
}

700 701 702 703
static cycle_t		ftrace_update_time;
static unsigned long	ftrace_update_cnt;
unsigned long		ftrace_update_tot_cnt;

704
static int ftrace_update_code(struct module *mod)
705
{
S
Steven Rostedt 已提交
706
	struct dyn_ftrace *p, *t;
707
	cycle_t start, stop;
708

I
Ingo Molnar 已提交
709
	start = ftrace_now(raw_smp_processor_id());
710 711
	ftrace_update_cnt = 0;

S
Steven Rostedt 已提交
712
	list_for_each_entry_safe(p, t, &ftrace_new_addrs, list) {
713

S
Steven Rostedt 已提交
714 715 716
		/* If something went wrong, bail without enabling anything */
		if (unlikely(ftrace_disabled))
			return -1;
717

S
Steven Rostedt 已提交
718
		list_del_init(&p->list);
719

S
Steven Rostedt 已提交
720
		/* convert record (i.e, patch mcount-call with NOP) */
721
		if (ftrace_code_disable(mod, p)) {
S
Steven Rostedt 已提交
722 723 724 725
			p->flags |= FTRACE_FL_CONVERTED;
			ftrace_update_cnt++;
		} else
			ftrace_free_rec(p);
726 727
	}

I
Ingo Molnar 已提交
728
	stop = ftrace_now(raw_smp_processor_id());
729 730 731
	ftrace_update_time = stop - start;
	ftrace_update_tot_cnt += ftrace_update_cnt;

732 733 734
	return 0;
}

735
static int __init ftrace_dyn_table_alloc(unsigned long num_to_init)
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761
{
	struct ftrace_page *pg;
	int cnt;
	int i;

	/* allocate a few pages */
	ftrace_pages_start = (void *)get_zeroed_page(GFP_KERNEL);
	if (!ftrace_pages_start)
		return -1;

	/*
	 * Allocate a few more pages.
	 *
	 * TODO: have some parser search vmlinux before
	 *   final linking to find all calls to ftrace.
	 *   Then we can:
	 *    a) know how many pages to allocate.
	 *     and/or
	 *    b) set up the table then.
	 *
	 *  The dynamic code is still necessary for
	 *  modules.
	 */

	pg = ftrace_pages = ftrace_pages_start;

762
	cnt = num_to_init / ENTRIES_PER_PAGE;
S
Steven Rostedt 已提交
763
	pr_info("ftrace: allocating %ld entries in %d pages\n",
764
		num_to_init, cnt + 1);
765 766 767 768 769 770 771 772 773 774 775 776 777 778

	for (i = 0; i < cnt; i++) {
		pg->next = (void *)get_zeroed_page(GFP_KERNEL);

		/* If we fail, we'll try later anyway */
		if (!pg->next)
			break;

		pg = pg->next;
	}

	return 0;
}

779 780 781
enum {
	FTRACE_ITER_FILTER	= (1 << 0),
	FTRACE_ITER_CONT	= (1 << 1),
782
	FTRACE_ITER_NOTRACE	= (1 << 2),
783
	FTRACE_ITER_FAILURES	= (1 << 3),
784 785 786 787 788 789 790 791 792 793 794 795 796
};

#define FTRACE_BUFF_MAX (KSYM_SYMBOL_LEN+4) /* room for wildcards */

struct ftrace_iterator {
	struct ftrace_page	*pg;
	unsigned		idx;
	unsigned		flags;
	unsigned char		buffer[FTRACE_BUFF_MAX+1];
	unsigned		buffer_idx;
	unsigned		filtered;
};

I
Ingo Molnar 已提交
797
static void *
798 799 800 801 802 803 804
t_next(struct seq_file *m, void *v, loff_t *pos)
{
	struct ftrace_iterator *iter = m->private;
	struct dyn_ftrace *rec = NULL;

	(*pos)++;

805 806
	/* should not be called from interrupt context */
	spin_lock(&ftrace_lock);
807 808 809 810 811 812
 retry:
	if (iter->idx >= iter->pg->index) {
		if (iter->pg->next) {
			iter->pg = iter->pg->next;
			iter->idx = 0;
			goto retry;
813 814
		} else {
			iter->idx = -1;
815 816 817
		}
	} else {
		rec = &iter->pg->records[iter->idx++];
818 819 820
		if ((rec->flags & FTRACE_FL_FREE) ||

		    (!(iter->flags & FTRACE_ITER_FAILURES) &&
821 822 823
		     (rec->flags & FTRACE_FL_FAILED)) ||

		    ((iter->flags & FTRACE_ITER_FAILURES) &&
824
		     !(rec->flags & FTRACE_FL_FAILED)) ||
825

S
Steven Rostedt 已提交
826 827 828
		    ((iter->flags & FTRACE_ITER_FILTER) &&
		     !(rec->flags & FTRACE_FL_FILTER)) ||

829 830
		    ((iter->flags & FTRACE_ITER_NOTRACE) &&
		     !(rec->flags & FTRACE_FL_NOTRACE))) {
831 832 833 834
			rec = NULL;
			goto retry;
		}
	}
835
	spin_unlock(&ftrace_lock);
836 837 838 839 840 841 842 843 844

	return rec;
}

static void *t_start(struct seq_file *m, loff_t *pos)
{
	struct ftrace_iterator *iter = m->private;
	void *p = NULL;

845 846 847 848 849 850
	if (*pos > 0) {
		if (iter->idx < 0)
			return p;
		(*pos)--;
		iter->idx--;
	}
851

852
	p = t_next(m, p, pos);
853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870

	return p;
}

static void t_stop(struct seq_file *m, void *p)
{
}

static int t_show(struct seq_file *m, void *v)
{
	struct dyn_ftrace *rec = v;
	char str[KSYM_SYMBOL_LEN];

	if (!rec)
		return 0;

	kallsyms_lookup(rec->ip, NULL, NULL, NULL, str);

871
	seq_printf(m, "%s\n", str);
872 873 874 875 876 877 878 879 880 881 882

	return 0;
}

static struct seq_operations show_ftrace_seq_ops = {
	.start = t_start,
	.next = t_next,
	.stop = t_stop,
	.show = t_show,
};

I
Ingo Molnar 已提交
883
static int
884 885 886 887 888
ftrace_avail_open(struct inode *inode, struct file *file)
{
	struct ftrace_iterator *iter;
	int ret;

889 890 891
	if (unlikely(ftrace_disabled))
		return -ENODEV;

892 893 894 895 896 897 898 899 900
	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
	if (!iter)
		return -ENOMEM;

	iter->pg = ftrace_pages_start;

	ret = seq_open(file, &show_ftrace_seq_ops);
	if (!ret) {
		struct seq_file *m = file->private_data;
I
Ingo Molnar 已提交
901

902
		m->private = iter;
I
Ingo Molnar 已提交
903
	} else {
904
		kfree(iter);
I
Ingo Molnar 已提交
905
	}
906 907 908 909 910 911 912 913 914 915 916

	return ret;
}

int ftrace_avail_release(struct inode *inode, struct file *file)
{
	struct seq_file *m = (struct seq_file *)file->private_data;
	struct ftrace_iterator *iter = m->private;

	seq_release(inode, file);
	kfree(iter);
I
Ingo Molnar 已提交
917

918 919 920
	return 0;
}

921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938
static int
ftrace_failures_open(struct inode *inode, struct file *file)
{
	int ret;
	struct seq_file *m;
	struct ftrace_iterator *iter;

	ret = ftrace_avail_open(inode, file);
	if (!ret) {
		m = (struct seq_file *)file->private_data;
		iter = (struct ftrace_iterator *)m->private;
		iter->flags = FTRACE_ITER_FAILURES;
	}

	return ret;
}


939
static void ftrace_filter_reset(int enable)
940 941 942
{
	struct ftrace_page *pg;
	struct dyn_ftrace *rec;
943
	unsigned long type = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE;
944 945
	unsigned i;

946 947
	/* should not be called from interrupt context */
	spin_lock(&ftrace_lock);
948 949
	if (enable)
		ftrace_filtered = 0;
950 951 952 953 954 955
	pg = ftrace_pages_start;
	while (pg) {
		for (i = 0; i < pg->index; i++) {
			rec = &pg->records[i];
			if (rec->flags & FTRACE_FL_FAILED)
				continue;
956
			rec->flags &= ~type;
957 958 959
		}
		pg = pg->next;
	}
960
	spin_unlock(&ftrace_lock);
961 962
}

I
Ingo Molnar 已提交
963
static int
964
ftrace_regex_open(struct inode *inode, struct file *file, int enable)
965 966 967 968
{
	struct ftrace_iterator *iter;
	int ret = 0;

969 970 971
	if (unlikely(ftrace_disabled))
		return -ENODEV;

972 973 974 975
	iter = kzalloc(sizeof(*iter), GFP_KERNEL);
	if (!iter)
		return -ENOMEM;

976
	mutex_lock(&ftrace_regex_lock);
977 978
	if ((file->f_mode & FMODE_WRITE) &&
	    !(file->f_flags & O_APPEND))
979
		ftrace_filter_reset(enable);
980 981 982

	if (file->f_mode & FMODE_READ) {
		iter->pg = ftrace_pages_start;
983 984
		iter->flags = enable ? FTRACE_ITER_FILTER :
			FTRACE_ITER_NOTRACE;
985 986 987 988 989 990 991 992 993

		ret = seq_open(file, &show_ftrace_seq_ops);
		if (!ret) {
			struct seq_file *m = file->private_data;
			m->private = iter;
		} else
			kfree(iter);
	} else
		file->private_data = iter;
994
	mutex_unlock(&ftrace_regex_lock);
995 996 997 998

	return ret;
}

999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010
static int
ftrace_filter_open(struct inode *inode, struct file *file)
{
	return ftrace_regex_open(inode, file, 1);
}

static int
ftrace_notrace_open(struct inode *inode, struct file *file)
{
	return ftrace_regex_open(inode, file, 0);
}

I
Ingo Molnar 已提交
1011
static ssize_t
1012
ftrace_regex_read(struct file *file, char __user *ubuf,
1013 1014 1015 1016 1017 1018 1019 1020
		       size_t cnt, loff_t *ppos)
{
	if (file->f_mode & FMODE_READ)
		return seq_read(file, ubuf, cnt, ppos);
	else
		return -EPERM;
}

I
Ingo Molnar 已提交
1021
static loff_t
1022
ftrace_regex_lseek(struct file *file, loff_t offset, int origin)
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040
{
	loff_t ret;

	if (file->f_mode & FMODE_READ)
		ret = seq_lseek(file, offset, origin);
	else
		file->f_pos = ret = 1;

	return ret;
}

enum {
	MATCH_FULL,
	MATCH_FRONT_ONLY,
	MATCH_MIDDLE_ONLY,
	MATCH_END_ONLY,
};

I
Ingo Molnar 已提交
1041
static void
1042
ftrace_match(unsigned char *buff, int len, int enable)
1043 1044 1045 1046 1047 1048
{
	char str[KSYM_SYMBOL_LEN];
	char *search = NULL;
	struct ftrace_page *pg;
	struct dyn_ftrace *rec;
	int type = MATCH_FULL;
1049
	unsigned long flag = enable ? FTRACE_FL_FILTER : FTRACE_FL_NOTRACE;
1050
	unsigned i, match = 0, search_len = 0;
1051 1052 1053 1054 1055 1056 1057
	int not = 0;

	if (buff[0] == '!') {
		not = 1;
		buff++;
		len--;
	}
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077

	for (i = 0; i < len; i++) {
		if (buff[i] == '*') {
			if (!i) {
				search = buff + i + 1;
				type = MATCH_END_ONLY;
				search_len = len - (i + 1);
			} else {
				if (type == MATCH_END_ONLY) {
					type = MATCH_MIDDLE_ONLY;
				} else {
					match = i;
					type = MATCH_FRONT_ONLY;
				}
				buff[i] = 0;
				break;
			}
		}
	}

1078 1079
	/* should not be called from interrupt context */
	spin_lock(&ftrace_lock);
1080 1081
	if (enable)
		ftrace_filtered = 1;
1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
	pg = ftrace_pages_start;
	while (pg) {
		for (i = 0; i < pg->index; i++) {
			int matched = 0;
			char *ptr;

			rec = &pg->records[i];
			if (rec->flags & FTRACE_FL_FAILED)
				continue;
			kallsyms_lookup(rec->ip, NULL, NULL, NULL, str);
			switch (type) {
			case MATCH_FULL:
				if (strcmp(str, buff) == 0)
					matched = 1;
				break;
			case MATCH_FRONT_ONLY:
				if (memcmp(str, buff, match) == 0)
					matched = 1;
				break;
			case MATCH_MIDDLE_ONLY:
				if (strstr(str, search))
					matched = 1;
				break;
			case MATCH_END_ONLY:
				ptr = strstr(str, search);
				if (ptr && (ptr[search_len] == 0))
					matched = 1;
				break;
			}
1111 1112 1113 1114 1115 1116
			if (matched) {
				if (not)
					rec->flags &= ~flag;
				else
					rec->flags |= flag;
			}
1117 1118 1119
		}
		pg = pg->next;
	}
1120
	spin_unlock(&ftrace_lock);
1121 1122
}

I
Ingo Molnar 已提交
1123
static ssize_t
1124 1125
ftrace_regex_write(struct file *file, const char __user *ubuf,
		   size_t cnt, loff_t *ppos, int enable)
1126 1127 1128 1129 1130 1131 1132 1133 1134
{
	struct ftrace_iterator *iter;
	char ch;
	size_t read = 0;
	ssize_t ret;

	if (!cnt || cnt < 0)
		return 0;

1135
	mutex_lock(&ftrace_regex_lock);
1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189

	if (file->f_mode & FMODE_READ) {
		struct seq_file *m = file->private_data;
		iter = m->private;
	} else
		iter = file->private_data;

	if (!*ppos) {
		iter->flags &= ~FTRACE_ITER_CONT;
		iter->buffer_idx = 0;
	}

	ret = get_user(ch, ubuf++);
	if (ret)
		goto out;
	read++;
	cnt--;

	if (!(iter->flags & ~FTRACE_ITER_CONT)) {
		/* skip white space */
		while (cnt && isspace(ch)) {
			ret = get_user(ch, ubuf++);
			if (ret)
				goto out;
			read++;
			cnt--;
		}

		if (isspace(ch)) {
			file->f_pos += read;
			ret = read;
			goto out;
		}

		iter->buffer_idx = 0;
	}

	while (cnt && !isspace(ch)) {
		if (iter->buffer_idx < FTRACE_BUFF_MAX)
			iter->buffer[iter->buffer_idx++] = ch;
		else {
			ret = -EINVAL;
			goto out;
		}
		ret = get_user(ch, ubuf++);
		if (ret)
			goto out;
		read++;
		cnt--;
	}

	if (isspace(ch)) {
		iter->filtered++;
		iter->buffer[iter->buffer_idx] = 0;
1190
		ftrace_match(iter->buffer, iter->buffer_idx, enable);
1191 1192 1193 1194 1195 1196 1197 1198 1199
		iter->buffer_idx = 0;
	} else
		iter->flags |= FTRACE_ITER_CONT;


	file->f_pos += read;

	ret = read;
 out:
1200
	mutex_unlock(&ftrace_regex_lock);
1201 1202 1203 1204

	return ret;
}

1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232
static ssize_t
ftrace_filter_write(struct file *file, const char __user *ubuf,
		    size_t cnt, loff_t *ppos)
{
	return ftrace_regex_write(file, ubuf, cnt, ppos, 1);
}

static ssize_t
ftrace_notrace_write(struct file *file, const char __user *ubuf,
		     size_t cnt, loff_t *ppos)
{
	return ftrace_regex_write(file, ubuf, cnt, ppos, 0);
}

static void
ftrace_set_regex(unsigned char *buf, int len, int reset, int enable)
{
	if (unlikely(ftrace_disabled))
		return;

	mutex_lock(&ftrace_regex_lock);
	if (reset)
		ftrace_filter_reset(enable);
	if (buf)
		ftrace_match(buf, len, enable);
	mutex_unlock(&ftrace_regex_lock);
}

1233 1234 1235 1236 1237 1238 1239 1240 1241
/**
 * ftrace_set_filter - set a function to filter on in ftrace
 * @buf - the string that holds the function filter text.
 * @len - the length of the string.
 * @reset - non zero to reset all filters before applying this filter.
 *
 * Filters denote which functions should be enabled when tracing is enabled.
 * If @buf is NULL and reset is set, all functions will be enabled for tracing.
 */
I
Ingo Molnar 已提交
1242
void ftrace_set_filter(unsigned char *buf, int len, int reset)
1243
{
1244 1245
	ftrace_set_regex(buf, len, reset, 1);
}
1246

1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259
/**
 * ftrace_set_notrace - set a function to not trace in ftrace
 * @buf - the string that holds the function notrace text.
 * @len - the length of the string.
 * @reset - non zero to reset all filters before applying this filter.
 *
 * Notrace Filters denote which functions should not be enabled when tracing
 * is enabled. If @buf is NULL and reset is set, all functions will be enabled
 * for tracing.
 */
void ftrace_set_notrace(unsigned char *buf, int len, int reset)
{
	ftrace_set_regex(buf, len, reset, 0);
1260 1261
}

I
Ingo Molnar 已提交
1262
static int
1263
ftrace_regex_release(struct inode *inode, struct file *file, int enable)
1264 1265 1266 1267
{
	struct seq_file *m = (struct seq_file *)file->private_data;
	struct ftrace_iterator *iter;

1268
	mutex_lock(&ftrace_regex_lock);
1269 1270 1271 1272 1273 1274 1275 1276 1277 1278
	if (file->f_mode & FMODE_READ) {
		iter = m->private;

		seq_release(inode, file);
	} else
		iter = file->private_data;

	if (iter->buffer_idx) {
		iter->filtered++;
		iter->buffer[iter->buffer_idx] = 0;
1279
		ftrace_match(iter->buffer, iter->buffer_idx, enable);
1280 1281 1282
	}

	mutex_lock(&ftrace_sysctl_lock);
S
Steven Rostedt 已提交
1283
	mutex_lock(&ftrace_start_lock);
1284
	if (ftrace_start_up && ftrace_enabled)
1285
		ftrace_run_update_code(FTRACE_ENABLE_CALLS);
S
Steven Rostedt 已提交
1286
	mutex_unlock(&ftrace_start_lock);
1287 1288 1289
	mutex_unlock(&ftrace_sysctl_lock);

	kfree(iter);
1290
	mutex_unlock(&ftrace_regex_lock);
1291 1292 1293
	return 0;
}

1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
static int
ftrace_filter_release(struct inode *inode, struct file *file)
{
	return ftrace_regex_release(inode, file, 1);
}

static int
ftrace_notrace_release(struct inode *inode, struct file *file)
{
	return ftrace_regex_release(inode, file, 0);
}

1306 1307 1308 1309 1310 1311 1312
static struct file_operations ftrace_avail_fops = {
	.open = ftrace_avail_open,
	.read = seq_read,
	.llseek = seq_lseek,
	.release = ftrace_avail_release,
};

1313 1314 1315 1316 1317 1318 1319
static struct file_operations ftrace_failures_fops = {
	.open = ftrace_failures_open,
	.read = seq_read,
	.llseek = seq_lseek,
	.release = ftrace_avail_release,
};

1320 1321
static struct file_operations ftrace_filter_fops = {
	.open = ftrace_filter_open,
1322
	.read = ftrace_regex_read,
1323
	.write = ftrace_filter_write,
1324
	.llseek = ftrace_regex_lseek,
1325 1326 1327
	.release = ftrace_filter_release,
};

1328 1329 1330 1331 1332 1333 1334 1335
static struct file_operations ftrace_notrace_fops = {
	.open = ftrace_notrace_open,
	.read = ftrace_regex_read,
	.write = ftrace_notrace_write,
	.llseek = ftrace_regex_lseek,
	.release = ftrace_notrace_release,
};

1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 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 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439
#ifdef CONFIG_FUNCTION_GRAPH_TRACER

static DEFINE_MUTEX(graph_lock);

int ftrace_graph_count;
unsigned long ftrace_graph_funcs[FTRACE_GRAPH_MAX_FUNCS] __read_mostly;

static void *
g_next(struct seq_file *m, void *v, loff_t *pos)
{
	unsigned long *array = m->private;
	int index = *pos;

	(*pos)++;

	if (index >= ftrace_graph_count)
		return NULL;

	return &array[index];
}

static void *g_start(struct seq_file *m, loff_t *pos)
{
	void *p = NULL;

	mutex_lock(&graph_lock);

	p = g_next(m, p, pos);

	return p;
}

static void g_stop(struct seq_file *m, void *p)
{
	mutex_unlock(&graph_lock);
}

static int g_show(struct seq_file *m, void *v)
{
	unsigned long *ptr = v;
	char str[KSYM_SYMBOL_LEN];

	if (!ptr)
		return 0;

	kallsyms_lookup(*ptr, NULL, NULL, NULL, str);

	seq_printf(m, "%s\n", str);

	return 0;
}

static struct seq_operations ftrace_graph_seq_ops = {
	.start = g_start,
	.next = g_next,
	.stop = g_stop,
	.show = g_show,
};

static int
ftrace_graph_open(struct inode *inode, struct file *file)
{
	int ret = 0;

	if (unlikely(ftrace_disabled))
		return -ENODEV;

	mutex_lock(&graph_lock);
	if ((file->f_mode & FMODE_WRITE) &&
	    !(file->f_flags & O_APPEND)) {
		ftrace_graph_count = 0;
		memset(ftrace_graph_funcs, 0, sizeof(ftrace_graph_funcs));
	}

	if (file->f_mode & FMODE_READ) {
		ret = seq_open(file, &ftrace_graph_seq_ops);
		if (!ret) {
			struct seq_file *m = file->private_data;
			m->private = ftrace_graph_funcs;
		}
	} else
		file->private_data = ftrace_graph_funcs;
	mutex_unlock(&graph_lock);

	return ret;
}

static ssize_t
ftrace_graph_read(struct file *file, char __user *ubuf,
		       size_t cnt, loff_t *ppos)
{
	if (file->f_mode & FMODE_READ)
		return seq_read(file, ubuf, cnt, ppos);
	else
		return -EPERM;
}

static int
ftrace_set_func(unsigned long *array, int idx, char *buffer)
{
	char str[KSYM_SYMBOL_LEN];
	struct dyn_ftrace *rec;
	struct ftrace_page *pg;
	int found = 0;
1440
	int i, j;
1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457

	if (ftrace_disabled)
		return -ENODEV;

	/* should not be called from interrupt context */
	spin_lock(&ftrace_lock);

	for (pg = ftrace_pages_start; pg; pg = pg->next) {
		for (i = 0; i < pg->index; i++) {
			rec = &pg->records[i];

			if (rec->flags & (FTRACE_FL_FAILED | FTRACE_FL_FREE))
				continue;

			kallsyms_lookup(rec->ip, NULL, NULL, NULL, str);
			if (strcmp(str, buffer) == 0) {
				found = 1;
1458 1459 1460 1461 1462 1463 1464
				for (j = 0; j < idx; j++)
					if (array[j] == rec->ip) {
						found = 0;
						break;
					}
				if (found)
					array[idx] = rec->ip;
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 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559
				break;
			}
		}
	}
	spin_unlock(&ftrace_lock);

	return found ? 0 : -EINVAL;
}

static ssize_t
ftrace_graph_write(struct file *file, const char __user *ubuf,
		   size_t cnt, loff_t *ppos)
{
	unsigned char buffer[FTRACE_BUFF_MAX+1];
	unsigned long *array;
	size_t read = 0;
	ssize_t ret;
	int index = 0;
	char ch;

	if (!cnt || cnt < 0)
		return 0;

	mutex_lock(&graph_lock);

	if (ftrace_graph_count >= FTRACE_GRAPH_MAX_FUNCS) {
		ret = -EBUSY;
		goto out;
	}

	if (file->f_mode & FMODE_READ) {
		struct seq_file *m = file->private_data;
		array = m->private;
	} else
		array = file->private_data;

	ret = get_user(ch, ubuf++);
	if (ret)
		goto out;
	read++;
	cnt--;

	/* skip white space */
	while (cnt && isspace(ch)) {
		ret = get_user(ch, ubuf++);
		if (ret)
			goto out;
		read++;
		cnt--;
	}

	if (isspace(ch)) {
		*ppos += read;
		ret = read;
		goto out;
	}

	while (cnt && !isspace(ch)) {
		if (index < FTRACE_BUFF_MAX)
			buffer[index++] = ch;
		else {
			ret = -EINVAL;
			goto out;
		}
		ret = get_user(ch, ubuf++);
		if (ret)
			goto out;
		read++;
		cnt--;
	}
	buffer[index] = 0;

	/* we allow only one at a time */
	ret = ftrace_set_func(array, ftrace_graph_count, buffer);
	if (ret)
		goto out;

	ftrace_graph_count++;

	file->f_pos += read;

	ret = read;
 out:
	mutex_unlock(&graph_lock);

	return ret;
}

static const struct file_operations ftrace_graph_fops = {
	.open = ftrace_graph_open,
	.read = ftrace_graph_read,
	.write = ftrace_graph_write,
};
#endif /* CONFIG_FUNCTION_GRAPH_TRACER */

1560
static __init int ftrace_init_dyn_debugfs(struct dentry *d_tracer)
1561 1562 1563 1564 1565 1566 1567 1568 1569
{
	struct dentry *entry;

	entry = debugfs_create_file("available_filter_functions", 0444,
				    d_tracer, NULL, &ftrace_avail_fops);
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'available_filter_functions' entry\n");

1570 1571 1572 1573 1574
	entry = debugfs_create_file("failures", 0444,
				    d_tracer, NULL, &ftrace_failures_fops);
	if (!entry)
		pr_warning("Could not create debugfs 'failures' entry\n");

1575 1576 1577 1578 1579
	entry = debugfs_create_file("set_ftrace_filter", 0644, d_tracer,
				    NULL, &ftrace_filter_fops);
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'set_ftrace_filter' entry\n");
1580 1581 1582 1583 1584 1585

	entry = debugfs_create_file("set_ftrace_notrace", 0644, d_tracer,
				    NULL, &ftrace_notrace_fops);
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'set_ftrace_notrace' entry\n");
1586

1587 1588 1589 1590 1591 1592 1593 1594 1595
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
	entry = debugfs_create_file("set_graph_function", 0444, d_tracer,
				    NULL,
				    &ftrace_graph_fops);
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'set_graph_function' entry\n");
#endif /* CONFIG_FUNCTION_GRAPH_TRACER */

1596 1597 1598
	return 0;
}

1599 1600
static int ftrace_convert_nops(struct module *mod,
			       unsigned long *start,
1601 1602 1603 1604 1605 1606
			       unsigned long *end)
{
	unsigned long *p;
	unsigned long addr;
	unsigned long flags;

S
Steven Rostedt 已提交
1607
	mutex_lock(&ftrace_start_lock);
1608 1609 1610
	p = start;
	while (p < end) {
		addr = ftrace_call_adjust(*p++);
1611 1612 1613 1614 1615 1616 1617 1618
		/*
		 * Some architecture linkers will pad between
		 * the different mcount_loc sections of different
		 * object files to satisfy alignments.
		 * Skip any NULL pointers.
		 */
		if (!addr)
			continue;
1619 1620 1621
		ftrace_record_ip(addr);
	}

S
Steven Rostedt 已提交
1622
	/* disable interrupts to prevent kstop machine */
1623
	local_irq_save(flags);
1624
	ftrace_update_code(mod);
1625
	local_irq_restore(flags);
S
Steven Rostedt 已提交
1626
	mutex_unlock(&ftrace_start_lock);
1627 1628 1629 1630

	return 0;
}

1631 1632
void ftrace_init_module(struct module *mod,
			unsigned long *start, unsigned long *end)
1633
{
1634
	if (ftrace_disabled || start == end)
1635
		return;
1636
	ftrace_convert_nops(mod, start, end);
1637 1638
}

1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665
extern unsigned long __start_mcount_loc[];
extern unsigned long __stop_mcount_loc[];

void __init ftrace_init(void)
{
	unsigned long count, addr, flags;
	int ret;

	/* Keep the ftrace pointer to the stub */
	addr = (unsigned long)ftrace_stub;

	local_irq_save(flags);
	ftrace_dyn_arch_init(&addr);
	local_irq_restore(flags);

	/* ftrace_dyn_arch_init places the return code in addr */
	if (addr)
		goto failed;

	count = __stop_mcount_loc - __start_mcount_loc;

	ret = ftrace_dyn_table_alloc(count);
	if (ret)
		goto failed;

	last_ftrace_enabled = ftrace_enabled = 1;

1666 1667
	ret = ftrace_convert_nops(NULL,
				  __start_mcount_loc,
1668 1669 1670 1671 1672 1673 1674
				  __stop_mcount_loc);

	return;
 failed:
	ftrace_disabled = 1;
}

1675
#else
1676 1677 1678 1679 1680 1681 1682 1683

static int __init ftrace_nodyn_init(void)
{
	ftrace_enabled = 1;
	return 0;
}
device_initcall(ftrace_nodyn_init);

1684 1685
static inline int ftrace_init_dyn_debugfs(struct dentry *d_tracer) { return 0; }
static inline void ftrace_startup_enable(int command) { }
1686 1687 1688
/* Keep as macros so we do not need to define the commands */
# define ftrace_startup(command)	do { } while (0)
# define ftrace_shutdown(command)	do { } while (0)
I
Ingo Molnar 已提交
1689 1690
# define ftrace_startup_sysctl()	do { } while (0)
# define ftrace_shutdown_sysctl()	do { } while (0)
1691 1692
#endif /* CONFIG_DYNAMIC_FTRACE */

1693 1694 1695 1696 1697 1698 1699
static ssize_t
ftrace_pid_read(struct file *file, char __user *ubuf,
		       size_t cnt, loff_t *ppos)
{
	char buf[64];
	int r;

1700 1701 1702
	if (ftrace_pid_trace == ftrace_swapper_pid)
		r = sprintf(buf, "swapper tasks\n");
	else if (ftrace_pid_trace)
S
Steven Rostedt 已提交
1703
		r = sprintf(buf, "%u\n", pid_nr(ftrace_pid_trace));
1704 1705 1706 1707 1708 1709
	else
		r = sprintf(buf, "no pid\n");

	return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
}

1710
static void clear_ftrace_swapper(void)
S
Steven Rostedt 已提交
1711 1712
{
	struct task_struct *p;
1713
	int cpu;
S
Steven Rostedt 已提交
1714

1715 1716 1717
	get_online_cpus();
	for_each_online_cpu(cpu) {
		p = idle_task(cpu);
S
Steven Rostedt 已提交
1718
		clear_tsk_trace_trace(p);
1719 1720 1721
	}
	put_online_cpus();
}
S
Steven Rostedt 已提交
1722

1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733
static void set_ftrace_swapper(void)
{
	struct task_struct *p;
	int cpu;

	get_online_cpus();
	for_each_online_cpu(cpu) {
		p = idle_task(cpu);
		set_tsk_trace_trace(p);
	}
	put_online_cpus();
S
Steven Rostedt 已提交
1734 1735
}

1736 1737 1738 1739
static void clear_ftrace_pid(struct pid *pid)
{
	struct task_struct *p;

1740
	rcu_read_lock();
1741 1742 1743
	do_each_pid_task(pid, PIDTYPE_PID, p) {
		clear_tsk_trace_trace(p);
	} while_each_pid_task(pid, PIDTYPE_PID, p);
1744 1745
	rcu_read_unlock();

1746 1747 1748 1749
	put_pid(pid);
}

static void set_ftrace_pid(struct pid *pid)
S
Steven Rostedt 已提交
1750 1751 1752
{
	struct task_struct *p;

1753
	rcu_read_lock();
S
Steven Rostedt 已提交
1754 1755 1756
	do_each_pid_task(pid, PIDTYPE_PID, p) {
		set_tsk_trace_trace(p);
	} while_each_pid_task(pid, PIDTYPE_PID, p);
1757
	rcu_read_unlock();
S
Steven Rostedt 已提交
1758 1759
}

1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777
static void clear_ftrace_pid_task(struct pid **pid)
{
	if (*pid == ftrace_swapper_pid)
		clear_ftrace_swapper();
	else
		clear_ftrace_pid(*pid);

	*pid = NULL;
}

static void set_ftrace_pid_task(struct pid *pid)
{
	if (pid == ftrace_swapper_pid)
		set_ftrace_swapper();
	else
		set_ftrace_pid(pid);
}

1778 1779 1780 1781
static ssize_t
ftrace_pid_write(struct file *filp, const char __user *ubuf,
		   size_t cnt, loff_t *ppos)
{
S
Steven Rostedt 已提交
1782
	struct pid *pid;
1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799
	char buf[64];
	long val;
	int ret;

	if (cnt >= sizeof(buf))
		return -EINVAL;

	if (copy_from_user(&buf, ubuf, cnt))
		return -EFAULT;

	buf[cnt] = 0;

	ret = strict_strtol(buf, 10, &val);
	if (ret < 0)
		return ret;

	mutex_lock(&ftrace_start_lock);
S
Steven Rostedt 已提交
1800
	if (val < 0) {
1801
		/* disable pid tracing */
S
Steven Rostedt 已提交
1802
		if (!ftrace_pid_trace)
1803
			goto out;
S
Steven Rostedt 已提交
1804 1805

		clear_ftrace_pid_task(&ftrace_pid_trace);
1806 1807

	} else {
1808 1809 1810 1811 1812 1813 1814
		/* swapper task is special */
		if (!val) {
			pid = ftrace_swapper_pid;
			if (pid == ftrace_pid_trace)
				goto out;
		} else {
			pid = find_get_pid(val);
1815

1816 1817 1818 1819
			if (pid == ftrace_pid_trace) {
				put_pid(pid);
				goto out;
			}
1820 1821
		}

S
Steven Rostedt 已提交
1822 1823 1824 1825 1826 1827 1828 1829 1830
		if (ftrace_pid_trace)
			clear_ftrace_pid_task(&ftrace_pid_trace);

		if (!pid)
			goto out;

		ftrace_pid_trace = pid;

		set_ftrace_pid_task(ftrace_pid_trace);
1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868
	}

	/* update the function call */
	ftrace_update_pid_func();
	ftrace_startup_enable(0);

 out:
	mutex_unlock(&ftrace_start_lock);

	return cnt;
}

static struct file_operations ftrace_pid_fops = {
	.read = ftrace_pid_read,
	.write = ftrace_pid_write,
};

static __init int ftrace_init_debugfs(void)
{
	struct dentry *d_tracer;
	struct dentry *entry;

	d_tracer = tracing_init_dentry();
	if (!d_tracer)
		return 0;

	ftrace_init_dyn_debugfs(d_tracer);

	entry = debugfs_create_file("set_ftrace_pid", 0644, d_tracer,
				    NULL, &ftrace_pid_fops);
	if (!entry)
		pr_warning("Could not create debugfs "
			   "'set_ftrace_pid' entry\n");
	return 0;
}

fs_initcall(ftrace_init_debugfs);

S
Steven Rostedt 已提交
1869
/**
1870
 * ftrace_kill - kill ftrace
S
Steven Rostedt 已提交
1871 1872 1873 1874 1875
 *
 * This function should be used by panic code. It stops ftrace
 * but in a not so nice way. If you need to simply kill ftrace
 * from a non-atomic section, use ftrace_kill.
 */
1876
void ftrace_kill(void)
S
Steven Rostedt 已提交
1877 1878 1879 1880 1881 1882
{
	ftrace_disabled = 1;
	ftrace_enabled = 0;
	clear_ftrace_function();
}

1883
/**
1884 1885
 * register_ftrace_function - register a function for profiling
 * @ops - ops structure that holds the function for profiling.
1886
 *
1887 1888 1889 1890 1891 1892
 * Register a function to be called by all functions in the
 * kernel.
 *
 * Note: @ops->func and all the functions it calls must be labeled
 *       with "notrace", otherwise it will go into a
 *       recursive loop.
1893
 */
1894
int register_ftrace_function(struct ftrace_ops *ops)
1895
{
1896 1897
	int ret;

1898 1899 1900
	if (unlikely(ftrace_disabled))
		return -1;

1901
	mutex_lock(&ftrace_sysctl_lock);
1902

1903
	ret = __register_ftrace_function(ops);
1904
	ftrace_startup(0);
1905

1906
	mutex_unlock(&ftrace_sysctl_lock);
1907
	return ret;
1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919
}

/**
 * unregister_ftrace_function - unresgister a function for profiling.
 * @ops - ops structure that holds the function to unregister
 *
 * Unregister a function that was added to be called by ftrace profiling.
 */
int unregister_ftrace_function(struct ftrace_ops *ops)
{
	int ret;

1920
	mutex_lock(&ftrace_sysctl_lock);
1921
	ret = __unregister_ftrace_function(ops);
1922
	ftrace_shutdown(0);
1923 1924 1925 1926 1927
	mutex_unlock(&ftrace_sysctl_lock);

	return ret;
}

I
Ingo Molnar 已提交
1928
int
1929
ftrace_enable_sysctl(struct ctl_table *table, int write,
1930
		     struct file *file, void __user *buffer, size_t *lenp,
1931 1932 1933 1934
		     loff_t *ppos)
{
	int ret;

1935 1936 1937
	if (unlikely(ftrace_disabled))
		return -ENODEV;

1938 1939
	mutex_lock(&ftrace_sysctl_lock);

1940
	ret  = proc_dointvec(table, write, file, buffer, lenp, ppos);
1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967

	if (ret || !write || (last_ftrace_enabled == ftrace_enabled))
		goto out;

	last_ftrace_enabled = ftrace_enabled;

	if (ftrace_enabled) {

		ftrace_startup_sysctl();

		/* we are starting ftrace again */
		if (ftrace_list != &ftrace_list_end) {
			if (ftrace_list->next == &ftrace_list_end)
				ftrace_trace_function = ftrace_list->func;
			else
				ftrace_trace_function = ftrace_list_func;
		}

	} else {
		/* stopping ftrace calls (just send to ftrace_stub) */
		ftrace_trace_function = ftrace_stub;

		ftrace_shutdown_sysctl();
	}

 out:
	mutex_unlock(&ftrace_sysctl_lock);
1968
	return ret;
1969
}
1970

1971
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
1972

1973
static atomic_t ftrace_graph_active;
1974
static struct notifier_block ftrace_suspend_notifier;
1975

1976 1977 1978 1979 1980
int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace)
{
	return 0;
}

1981 1982 1983
/* The callbacks that hook a function */
trace_func_graph_ret_t ftrace_graph_return =
			(trace_func_graph_ret_t)ftrace_stub;
1984
trace_func_graph_ent_t ftrace_graph_entry = ftrace_graph_entry_stub;
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

/* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */
static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list)
{
	int i;
	int ret = 0;
	unsigned long flags;
	int start = 0, end = FTRACE_RETSTACK_ALLOC_SIZE;
	struct task_struct *g, *t;

	for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) {
		ret_stack_list[i] = kmalloc(FTRACE_RETFUNC_DEPTH
					* sizeof(struct ftrace_ret_stack),
					GFP_KERNEL);
		if (!ret_stack_list[i]) {
			start = 0;
			end = i;
			ret = -ENOMEM;
			goto free;
		}
	}

	read_lock_irqsave(&tasklist_lock, flags);
	do_each_thread(g, t) {
		if (start == end) {
			ret = -EAGAIN;
			goto unlock;
		}

		if (t->ret_stack == NULL) {
			t->curr_ret_stack = -1;
2016 2017 2018
			/* Make sure IRQs see the -1 first: */
			barrier();
			t->ret_stack = ret_stack_list[start++];
2019
			atomic_set(&t->tracing_graph_pause, 0);
2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032
			atomic_set(&t->trace_overrun, 0);
		}
	} while_each_thread(g, t);

unlock:
	read_unlock_irqrestore(&tasklist_lock, flags);
free:
	for (i = start; i < end; i++)
		kfree(ret_stack_list[i]);
	return ret;
}

/* Allocate a return stack for each task */
2033
static int start_graph_tracing(void)
2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052
{
	struct ftrace_ret_stack **ret_stack_list;
	int ret;

	ret_stack_list = kmalloc(FTRACE_RETSTACK_ALLOC_SIZE *
				sizeof(struct ftrace_ret_stack *),
				GFP_KERNEL);

	if (!ret_stack_list)
		return -ENOMEM;

	do {
		ret = alloc_retstack_tasklist(ret_stack_list);
	} while (ret == -EAGAIN);

	kfree(ret_stack_list);
	return ret;
}

2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073
/*
 * Hibernation protection.
 * The state of the current task is too much unstable during
 * suspend/restore to disk. We want to protect against that.
 */
static int
ftrace_suspend_notifier_call(struct notifier_block *bl, unsigned long state,
							void *unused)
{
	switch (state) {
	case PM_HIBERNATION_PREPARE:
		pause_graph_tracing();
		break;

	case PM_POST_HIBERNATION:
		unpause_graph_tracing();
		break;
	}
	return NOTIFY_DONE;
}

2074 2075
int register_ftrace_graph(trace_func_graph_ret_t retfunc,
			trace_func_graph_ent_t entryfunc)
2076
{
2077 2078 2079 2080
	int ret = 0;

	mutex_lock(&ftrace_sysctl_lock);

2081 2082 2083
	ftrace_suspend_notifier.notifier_call = ftrace_suspend_notifier_call;
	register_pm_notifier(&ftrace_suspend_notifier);

2084
	atomic_inc(&ftrace_graph_active);
2085
	ret = start_graph_tracing();
2086
	if (ret) {
2087
		atomic_dec(&ftrace_graph_active);
2088 2089
		goto out;
	}
2090

2091 2092
	ftrace_graph_return = retfunc;
	ftrace_graph_entry = entryfunc;
2093

2094
	ftrace_startup(FTRACE_START_FUNC_RET);
2095 2096 2097 2098

out:
	mutex_unlock(&ftrace_sysctl_lock);
	return ret;
2099 2100
}

2101
void unregister_ftrace_graph(void)
2102
{
2103 2104
	mutex_lock(&ftrace_sysctl_lock);

2105 2106
	atomic_dec(&ftrace_graph_active);
	ftrace_graph_return = (trace_func_graph_ret_t)ftrace_stub;
2107
	ftrace_graph_entry = ftrace_graph_entry_stub;
2108
	ftrace_shutdown(FTRACE_STOP_FUNC_RET);
2109
	unregister_pm_notifier(&ftrace_suspend_notifier);
2110 2111

	mutex_unlock(&ftrace_sysctl_lock);
2112
}
2113 2114

/* Allocate a return stack for newly created task */
2115
void ftrace_graph_init_task(struct task_struct *t)
2116
{
2117
	if (atomic_read(&ftrace_graph_active)) {
2118 2119 2120 2121 2122 2123
		t->ret_stack = kmalloc(FTRACE_RETFUNC_DEPTH
				* sizeof(struct ftrace_ret_stack),
				GFP_KERNEL);
		if (!t->ret_stack)
			return;
		t->curr_ret_stack = -1;
2124
		atomic_set(&t->tracing_graph_pause, 0);
2125 2126 2127 2128 2129
		atomic_set(&t->trace_overrun, 0);
	} else
		t->ret_stack = NULL;
}

2130
void ftrace_graph_exit_task(struct task_struct *t)
2131
{
2132 2133
	struct ftrace_ret_stack	*ret_stack = t->ret_stack;

2134
	t->ret_stack = NULL;
2135 2136 2137 2138
	/* NULL must become visible to IRQs before we free it: */
	barrier();

	kfree(ret_stack);
2139
}
S
Steven Rostedt 已提交
2140 2141 2142 2143 2144

void ftrace_graph_stop(void)
{
	ftrace_stop();
}
2145 2146
#endif