trace.c 119 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
/*
 * ring buffer based function tracer
 *
 * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
 * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com>
 *
 * Originally taken from the RT patch by:
 *    Arnaldo Carvalho de Melo <acme@redhat.com>
 *
 * Based on code from the latency_tracer, that is:
 *  Copyright (C) 2004-2006 Ingo Molnar
12
 *  Copyright (C) 2004 Nadia Yvette Chambers
13
 */
14
#include <linux/ring_buffer.h>
15
#include <generated/utsrelease.h>
16
17
#include <linux/stacktrace.h>
#include <linux/writeback.h>
18
19
#include <linux/kallsyms.h>
#include <linux/seq_file.h>
20
#include <linux/notifier.h>
21
#include <linux/irqflags.h>
22
#include <linux/irq_work.h>
23
#include <linux/debugfs.h>
24
#include <linux/pagemap.h>
25
26
27
#include <linux/hardirq.h>
#include <linux/linkage.h>
#include <linux/uaccess.h>
28
#include <linux/kprobes.h>
29
30
31
#include <linux/ftrace.h>
#include <linux/module.h>
#include <linux/percpu.h>
32
#include <linux/splice.h>
33
#include <linux/kdebug.h>
34
#include <linux/string.h>
35
#include <linux/rwsem.h>
36
#include <linux/slab.h>
37
38
#include <linux/ctype.h>
#include <linux/init.h>
39
#include <linux/poll.h>
40
#include <linux/nmi.h>
41
#include <linux/fs.h>
Ingo Molnar's avatar
Ingo Molnar committed
42

43
#include "trace.h"
44
#include "trace_output.h"
45

46
47
48
49
/*
 * On boot up, the ring buffer is set to the minimum size, so that
 * we do not waste memory on systems that are not using tracing.
 */
50
int ring_buffer_expanded;
51

52
53
/*
 * We need to change this state when a selftest is running.
54
55
 * A selftest will lurk into the ring-buffer to count the
 * entries inserted during the selftest although some concurrent
56
 * insertions into the ring-buffer such as trace_printk could occurred
57
58
 * at the same time, giving false positive or negative results.
 */
59
static bool __read_mostly tracing_selftest_running;
60

61
62
63
/*
 * If a tracer is running, we do not want to run SELFTEST.
 */
64
bool __read_mostly tracing_selftest_disabled;
65

66
67
68
69
70
71
72
73
74
75
76
77
78
79
/* For tracers that don't implement custom flags */
static struct tracer_opt dummy_tracer_opt[] = {
	{ }
};

static struct tracer_flags dummy_tracer_flags = {
	.val = 0,
	.opts = dummy_tracer_opt
};

static int dummy_set_flag(u32 old_flags, u32 bit, int set)
{
	return 0;
}
80

81
82
83
84
85
86
87
/*
 * To prevent the comm cache from being overwritten when no
 * tracing is active, only save the comm when a trace event
 * occurred.
 */
static DEFINE_PER_CPU(bool, trace_cmdline_save);

88
89
90
91
92
93
94
95
/*
 * When a reader is waiting for data, then this variable is
 * set to true.
 */
static bool trace_wakeup_needed;

static struct irq_work trace_work_wakeup;

96
97
98
99
100
101
/*
 * Kill all tracing for good (never come back).
 * It is initialized to 1 but will turn to zero if the initialization
 * of the tracer is successful. But that is the only place that sets
 * this back to zero.
 */
102
static int tracing_disabled = 1;
103

104
DEFINE_PER_CPU(int, ftrace_cpu_disabled);
105

106
cpumask_var_t __read_mostly	tracing_buffer_mask;
107

108
109
110
111
112
113
114
115
116
117
118
/*
 * ftrace_dump_on_oops - variable to dump ftrace buffer on oops
 *
 * If there is an oops (or kernel panic) and the ftrace_dump_on_oops
 * is set, then ftrace_dump is called. This will output the contents
 * of the ftrace buffers to the console.  This is very useful for
 * capturing traces that lead to crashes and outputing it to a
 * serial console.
 *
 * It is default off, but you can enable it with either specifying
 * "ftrace_dump_on_oops" in the kernel command line, or setting
119
120
121
 * /proc/sys/kernel/ftrace_dump_on_oops
 * Set 1 if you want to dump buffers of all CPUs
 * Set 2 if you want to dump the buffer of the CPU that triggered oops
122
 */
123
124

enum ftrace_dump_mode ftrace_dump_on_oops;
125

126
127
static int tracing_set_tracer(const char *buf);

Li Zefan's avatar
Li Zefan committed
128
129
#define MAX_TRACER_SIZE		100
static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata;
130
static char *default_bootup_tracer;
131

132
static int __init set_cmdline_ftrace(char *str)
133
{
Li Zefan's avatar
Li Zefan committed
134
	strncpy(bootup_tracer_buf, str, MAX_TRACER_SIZE);
135
	default_bootup_tracer = bootup_tracer_buf;
136
137
	/* We are using ftrace early, expand it */
	ring_buffer_expanded = 1;
138
139
	return 1;
}
140
__setup("ftrace=", set_cmdline_ftrace);
141

142
143
static int __init set_ftrace_dump_on_oops(char *str)
{
144
145
146
147
148
149
150
151
152
153
154
	if (*str++ != '=' || !*str) {
		ftrace_dump_on_oops = DUMP_ALL;
		return 1;
	}

	if (!strcmp("orig_cpu", str)) {
		ftrace_dump_on_oops = DUMP_ORIG;
                return 1;
        }

        return 0;
155
156
}
__setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops);
Steven Rostedt's avatar
Steven Rostedt committed
157

158
159
160
161
162
163
164
165
166
167
168
169

static char trace_boot_options_buf[MAX_TRACER_SIZE] __initdata;
static char *trace_boot_options __initdata;

static int __init set_trace_boot_options(char *str)
{
	strncpy(trace_boot_options_buf, str, MAX_TRACER_SIZE);
	trace_boot_options = trace_boot_options_buf;
	return 0;
}
__setup("trace_options=", set_trace_boot_options);

170
unsigned long long ns2usecs(cycle_t nsec)
171
172
173
174
175
176
{
	nsec += 500;
	do_div(nsec, 1000);
	return nsec;
}

Steven Rostedt's avatar
Steven Rostedt committed
177
178
179
180
181
182
183
184
185
186
187
188
/*
 * The global_trace is the descriptor that holds the tracing
 * buffers for the live tracing. For each CPU, it contains
 * a link list of pages that will store trace entries. The
 * page descriptor of the pages in the memory is used to hold
 * the link list by linking the lru item in the page descriptor
 * to each of the pages in the buffer per CPU.
 *
 * For each active CPU there is a data field that holds the
 * pages for the buffer for that CPU. Each CPU has the same number
 * of pages allocated for its buffer.
 */
189
190
191
192
static struct trace_array	global_trace;

static DEFINE_PER_CPU(struct trace_array_cpu, global_trace_cpu);

193
194
int filter_current_check_discard(struct ring_buffer *buffer,
				 struct ftrace_event_call *call, void *rec,
195
196
				 struct ring_buffer_event *event)
{
197
	return filter_check_discard(call, rec, buffer, event);
198
}
199
EXPORT_SYMBOL_GPL(filter_current_check_discard);
200

201
202
203
204
205
206
207
208
209
210
211
212
213
cycle_t ftrace_now(int cpu)
{
	u64 ts;

	/* Early boot up does not have a buffer yet */
	if (!global_trace.buffer)
		return trace_clock_local();

	ts = ring_buffer_time_stamp(global_trace.buffer, cpu);
	ring_buffer_normalize_time_stamp(global_trace.buffer, cpu, &ts);

	return ts;
}
214

Steven Rostedt's avatar
Steven Rostedt committed
215
216
217
218
219
220
221
222
223
224
/*
 * The max_tr is used to snapshot the global_trace when a maximum
 * latency is reached. Some tracers will use this to store a maximum
 * trace while it continues examining live traces.
 *
 * The buffers for the max_tr are set up the same as the global_trace.
 * When a snapshot is taken, the link list of the max_tr is swapped
 * with the link list of the global_trace and the buffers are reset for
 * the global_trace so the tracing can continue.
 */
225
226
static struct trace_array	max_tr;

227
static DEFINE_PER_CPU(struct trace_array_cpu, max_tr_data);
228

229
230
int tracing_is_enabled(void)
{
231
	return tracing_is_on();
232
233
}

Steven Rostedt's avatar
Steven Rostedt committed
234
/*
235
236
237
 * trace_buf_size is the size in bytes that is allocated
 * for a buffer. Note, the number of bytes is always rounded
 * to page size.
238
239
240
241
242
 *
 * This number is purposely set to a low number of 16384.
 * If the dump on oops happens, it will be much appreciated
 * to not have to wait for all that output. Anyway this can be
 * boot time and run time configurable.
Steven Rostedt's avatar
Steven Rostedt committed
243
 */
244
#define TRACE_BUF_SIZE_DEFAULT	1441792UL /* 16384 * 88 (sizeof(entry)) */
245

246
static unsigned long		trace_buf_size = TRACE_BUF_SIZE_DEFAULT;
247

Steven Rostedt's avatar
Steven Rostedt committed
248
/* trace_types holds a link list of available tracers. */
249
static struct tracer		*trace_types __read_mostly;
Steven Rostedt's avatar
Steven Rostedt committed
250
251

/* current_trace points to the tracer that is currently active */
252
static struct tracer		*current_trace __read_mostly;
Steven Rostedt's avatar
Steven Rostedt committed
253
254
255
256

/*
 * trace_types_lock is used to protect the trace_types list.
 */
257
static DEFINE_MUTEX(trace_types_lock);
Steven Rostedt's avatar
Steven Rostedt committed
258

259
260
261
262
263
264
265
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
309
310
311
312
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
339
340
/*
 * serialize the access of the ring buffer
 *
 * ring buffer serializes readers, but it is low level protection.
 * The validity of the events (which returns by ring_buffer_peek() ..etc)
 * are not protected by ring buffer.
 *
 * The content of events may become garbage if we allow other process consumes
 * these events concurrently:
 *   A) the page of the consumed events may become a normal page
 *      (not reader page) in ring buffer, and this page will be rewrited
 *      by events producer.
 *   B) The page of the consumed events may become a page for splice_read,
 *      and this page will be returned to system.
 *
 * These primitives allow multi process access to different cpu ring buffer
 * concurrently.
 *
 * These primitives don't distinguish read-only and read-consume access.
 * Multi read-only access are also serialized.
 */

#ifdef CONFIG_SMP
static DECLARE_RWSEM(all_cpu_access_lock);
static DEFINE_PER_CPU(struct mutex, cpu_access_lock);

static inline void trace_access_lock(int cpu)
{
	if (cpu == TRACE_PIPE_ALL_CPU) {
		/* gain it for accessing the whole ring buffer. */
		down_write(&all_cpu_access_lock);
	} else {
		/* gain it for accessing a cpu ring buffer. */

		/* Firstly block other trace_access_lock(TRACE_PIPE_ALL_CPU). */
		down_read(&all_cpu_access_lock);

		/* Secondly block other access to this @cpu ring buffer. */
		mutex_lock(&per_cpu(cpu_access_lock, cpu));
	}
}

static inline void trace_access_unlock(int cpu)
{
	if (cpu == TRACE_PIPE_ALL_CPU) {
		up_write(&all_cpu_access_lock);
	} else {
		mutex_unlock(&per_cpu(cpu_access_lock, cpu));
		up_read(&all_cpu_access_lock);
	}
}

static inline void trace_access_lock_init(void)
{
	int cpu;

	for_each_possible_cpu(cpu)
		mutex_init(&per_cpu(cpu_access_lock, cpu));
}

#else

static DEFINE_MUTEX(access_lock);

static inline void trace_access_lock(int cpu)
{
	(void)cpu;
	mutex_lock(&access_lock);
}

static inline void trace_access_unlock(int cpu)
{
	(void)cpu;
	mutex_unlock(&access_lock);
}

static inline void trace_access_lock_init(void)
{
}

#endif

Steven Rostedt's avatar
Steven Rostedt committed
341
/* trace_wait is a waitqueue for tasks blocked on trace_poll */
342
343
static DECLARE_WAIT_QUEUE_HEAD(trace_wait);

344
/* trace_flags holds trace_options default values */
345
unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK |
346
	TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | TRACE_ITER_SLEEP_TIME |
347
	TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE |
348
	TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS;
349

350
static int trace_stop_count;
351
static DEFINE_RAW_SPINLOCK(tracing_start_lock);
352

353
354
355
356
357
358
359
360
/**
 * trace_wake_up - wake up tasks waiting for trace input
 *
 * Schedules a delayed work to wake up any task that is blocked on the
 * trace_wait queue. These is used with trace_poll for tasks polling the
 * trace.
 */
static void trace_wake_up(struct irq_work *work)
361
{
362
	wake_up_all(&trace_wait);
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
392
393
394
395
396
/**
 * tracing_on - enable tracing buffers
 *
 * This function enables tracing buffers that may have been
 * disabled with tracing_off.
 */
void tracing_on(void)
{
	if (global_trace.buffer)
		ring_buffer_record_on(global_trace.buffer);
	/*
	 * This flag is only looked at when buffers haven't been
	 * allocated yet. We don't really care about the race
	 * between setting this flag and actually turning
	 * on the buffer.
	 */
	global_trace.buffer_disabled = 0;
}
EXPORT_SYMBOL_GPL(tracing_on);

/**
 * tracing_off - turn off tracing buffers
 *
 * This function stops the tracing buffers from recording data.
 * It does not disable any overhead the tracers themselves may
 * be causing. This function simply causes all recording to
 * the ring buffers to fail.
 */
void tracing_off(void)
{
	if (global_trace.buffer)
397
		ring_buffer_record_off(global_trace.buffer);
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
	/*
	 * This flag is only looked at when buffers haven't been
	 * allocated yet. We don't really care about the race
	 * between setting this flag and actually turning
	 * on the buffer.
	 */
	global_trace.buffer_disabled = 1;
}
EXPORT_SYMBOL_GPL(tracing_off);

/**
 * tracing_is_on - show state of ring buffers enabled
 */
int tracing_is_on(void)
{
	if (global_trace.buffer)
		return ring_buffer_record_is_on(global_trace.buffer);
	return !global_trace.buffer_disabled;
}
EXPORT_SYMBOL_GPL(tracing_is_on);

419
static int __init set_buf_size(char *str)
420
{
421
	unsigned long buf_size;
422

423
424
	if (!str)
		return 0;
425
	buf_size = memparse(str, &str);
426
	/* nr_entries can not be zero */
427
	if (buf_size == 0)
428
		return 0;
429
	trace_buf_size = buf_size;
430
431
	return 1;
}
432
__setup("trace_buf_size=", set_buf_size);
433

434
435
static int __init set_tracing_thresh(char *str)
{
436
	unsigned long threshold;
437
438
439
440
	int ret;

	if (!str)
		return 0;
441
	ret = kstrtoul(str, 0, &threshold);
442
443
	if (ret < 0)
		return 0;
444
	tracing_thresh = threshold * 1000;
445
446
447
448
	return 1;
}
__setup("tracing_thresh=", set_tracing_thresh);

Steven Rostedt's avatar
Steven Rostedt committed
449
450
451
452
453
unsigned long nsecs_to_usecs(unsigned long nsecs)
{
	return nsecs / 1000;
}

Steven Rostedt's avatar
Steven Rostedt committed
454
/* These must match the bit postions in trace_iterator_flags */
455
456
457
458
459
static const char *trace_options[] = {
	"print-parent",
	"sym-offset",
	"sym-addr",
	"verbose",
Ingo Molnar's avatar
Ingo Molnar committed
460
	"raw",
461
	"hex",
Ingo Molnar's avatar
Ingo Molnar committed
462
	"bin",
463
	"block",
Ingo Molnar's avatar
Ingo Molnar committed
464
	"stacktrace",
465
	"trace_printk",
466
	"ftrace_preempt",
467
	"branch",
468
	"annotate",
469
	"userstacktrace",
470
	"sym-userobj",
471
	"printk-msg-only",
472
	"context-info",
473
	"latency-format",
474
	"sleep-time",
475
	"graph-time",
476
	"record-cmd",
477
	"overwrite",
478
	"disable_on_free",
479
	"irq-info",
480
	"markers",
481
482
483
	NULL
};

484
485
486
static struct {
	u64 (*func)(void);
	const char *name;
487
	int in_ns;		/* is this clock in nanoseconds? */
488
} trace_clocks[] = {
489
490
491
	{ trace_clock_local,	"local",	1 },
	{ trace_clock_global,	"global",	1 },
	{ trace_clock_counter,	"counter",	0 },
492
	ARCH_TRACE_CLOCKS
493
494
495
496
};

int trace_clock_id;

497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
/*
 * trace_parser_get_init - gets the buffer for trace parser
 */
int trace_parser_get_init(struct trace_parser *parser, int size)
{
	memset(parser, 0, sizeof(*parser));

	parser->buffer = kmalloc(size, GFP_KERNEL);
	if (!parser->buffer)
		return 1;

	parser->size = size;
	return 0;
}

/*
 * trace_parser_put - frees the buffer for trace parser
 */
void trace_parser_put(struct trace_parser *parser)
{
	kfree(parser->buffer);
}

/*
 * trace_get_user - reads the user input string separated by  space
 * (matched by isspace(ch))
 *
 * For each string found the 'struct trace_parser' is updated,
 * and the function returns.
 *
 * Returns number of bytes read.
 *
 * See kernel/trace/trace.h for 'struct trace_parser' details.
 */
int trace_get_user(struct trace_parser *parser, const char __user *ubuf,
	size_t cnt, loff_t *ppos)
{
	char ch;
	size_t read = 0;
	ssize_t ret;

	if (!*ppos)
		trace_parser_clear(parser);

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

	read++;
	cnt--;

	/*
	 * The parser is not finished with the last write,
	 * continue reading the user input without skipping spaces.
	 */
	if (!parser->cont) {
		/* skip white space */
		while (cnt && isspace(ch)) {
			ret = get_user(ch, ubuf++);
			if (ret)
				goto out;
			read++;
			cnt--;
		}

		/* only spaces were written */
		if (isspace(ch)) {
			*ppos += read;
			ret = read;
			goto out;
		}

		parser->idx = 0;
	}

	/* read the non-space input */
	while (cnt && !isspace(ch)) {
574
		if (parser->idx < parser->size - 1)
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
			parser->buffer[parser->idx++] = ch;
		else {
			ret = -EINVAL;
			goto out;
		}
		ret = get_user(ch, ubuf++);
		if (ret)
			goto out;
		read++;
		cnt--;
	}

	/* We either got finished input or we have to wait for another call. */
	if (isspace(ch)) {
		parser->buffer[parser->idx] = 0;
		parser->cont = false;
	} else {
		parser->cont = true;
		parser->buffer[parser->idx++] = ch;
	}

	*ppos += read;
	ret = read;

out:
	return ret;
}

603
604
605
606
607
ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt)
{
	int len;
	int ret;

608
609
610
	if (!cnt)
		return 0;

611
612
613
614
615
616
617
	if (s->len <= s->readpos)
		return -EBUSY;

	len = s->len - s->readpos;
	if (cnt > len)
		cnt = len;
	ret = copy_to_user(ubuf, s->buffer + s->readpos, cnt);
618
	if (ret == cnt)
619
620
		return -EFAULT;

621
622
	cnt -= ret;

623
	s->readpos += cnt;
624
	return cnt;
625
626
}

627
static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt)
628
629
630
631
632
633
634
635
636
{
	int len;

	if (s->len <= s->readpos)
		return -EBUSY;

	len = s->len - s->readpos;
	if (cnt > len)
		cnt = len;
637
	memcpy(buf, s->buffer + s->readpos, cnt);
638

639
	s->readpos += cnt;
640
641
642
	return cnt;
}

643
644
645
646
647
648
/*
 * ftrace_max_lock is used to protect the swapping of buffers
 * when taking a max snapshot. The buffers themselves are
 * protected by per_cpu spinlocks. But the action of the swap
 * needs its own lock.
 *
649
 * This is defined as a arch_spinlock_t in order to help
650
651
652
653
654
655
 * with performance when lockdep debugging is enabled.
 *
 * It is also used in other places outside the update_max_tr
 * so it needs to be defined outside of the
 * CONFIG_TRACER_MAX_TRACE.
 */
656
static arch_spinlock_t ftrace_max_lock =
657
	(arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
658

659
660
unsigned long __read_mostly	tracing_thresh;

661
662
663
664
665
666
667
668
669
670
671
672
#ifdef CONFIG_TRACER_MAX_TRACE
unsigned long __read_mostly	tracing_max_latency;

/*
 * Copy the new maximum trace into the separate maximum-trace
 * structure. (this way the maximum trace is permanently saved,
 * for later retrieval via /sys/kernel/debug/tracing/latency_trace)
 */
static void
__update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
	struct trace_array_cpu *data = tr->data[cpu];
673
	struct trace_array_cpu *max_data;
674
675
676
677

	max_tr.cpu = cpu;
	max_tr.time_start = data->preempt_timestamp;

678
679
680
681
	max_data = max_tr.data[cpu];
	max_data->saved_latency = tracing_max_latency;
	max_data->critical_start = data->critical_start;
	max_data->critical_end = data->critical_end;
682

683
	memcpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
684
685
686
687
688
	max_data->pid = tsk->pid;
	max_data->uid = task_uid(tsk);
	max_data->nice = tsk->static_prio - 20 - MAX_RT_PRIO;
	max_data->policy = tsk->policy;
	max_data->rt_priority = tsk->rt_priority;
689
690
691
692
693

	/* record this tasks comm */
	tracing_record_cmdline(tsk);
}

Steven Rostedt's avatar
Steven Rostedt committed
694
695
696
697
698
699
700
701
702
/**
 * update_max_tr - snapshot all trace buffers from global_trace to max_tr
 * @tr: tracer
 * @tsk: the task with the latency
 * @cpu: The cpu that initiated the trace.
 *
 * Flip the buffers between the @tr and the max_tr and record information
 * about which task was the cause of this latency.
 */
Ingo Molnar's avatar
Ingo Molnar committed
703
void
704
705
update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
706
	struct ring_buffer *buf = tr->buffer;
707

708
709
710
	if (trace_stop_count)
		return;

711
	WARN_ON_ONCE(!irqs_disabled());
712
713
714
715
	if (!current_trace->use_max_tr) {
		WARN_ON_ONCE(1);
		return;
	}
716
	arch_spin_lock(&ftrace_max_lock);
717
718
719
720

	tr->buffer = max_tr.buffer;
	max_tr.buffer = buf;

721
	__update_max_tr(tr, tsk, cpu);
722
	arch_spin_unlock(&ftrace_max_lock);
723
724
725
726
727
728
729
}

/**
 * update_max_tr_single - only copy one trace over, and reset the rest
 * @tr - tracer
 * @tsk - task with the latency
 * @cpu - the cpu of the buffer to copy.
Steven Rostedt's avatar
Steven Rostedt committed
730
731
 *
 * Flip the trace of a single CPU buffer between the @tr and the max_tr.
732
 */
Ingo Molnar's avatar
Ingo Molnar committed
733
void
734
735
update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
736
	int ret;
737

738
739
740
	if (trace_stop_count)
		return;

741
	WARN_ON_ONCE(!irqs_disabled());
742
743
744
745
746
	if (!current_trace->use_max_tr) {
		WARN_ON_ONCE(1);
		return;
	}

747
	arch_spin_lock(&ftrace_max_lock);
748

749
750
	ret = ring_buffer_swap_cpu(max_tr.buffer, tr->buffer, cpu);

751
752
753
754
755
756
757
758
759
760
761
762
	if (ret == -EBUSY) {
		/*
		 * We failed to swap the buffer due to a commit taking
		 * place on this CPU. We fail to record, but we reset
		 * the max trace buffer (no one writes directly to it)
		 * and flag that it failed.
		 */
		trace_array_printk(&max_tr, _THIS_IP_,
			"Failed to swap buffers due to commit in progress\n");
	}

	WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY);
763
764

	__update_max_tr(tr, tsk, cpu);
765
	arch_spin_unlock(&ftrace_max_lock);
766
}
767
#endif /* CONFIG_TRACER_MAX_TRACE */
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
static void default_wait_pipe(struct trace_iterator *iter)
{
	DEFINE_WAIT(wait);

	prepare_to_wait(&trace_wait, &wait, TASK_INTERRUPTIBLE);

	/*
	 * The events can happen in critical sections where
	 * checking a work queue can cause deadlocks.
	 * After adding a task to the queue, this flag is set
	 * only to notify events to try to wake up the queue
	 * using irq_work.
	 *
	 * We don't clear it even if the buffer is no longer
	 * empty. The flag only causes the next event to run
	 * irq_work to do the work queue wake up. The worse
	 * that can happen if we race with !trace_empty() is that
	 * an event will cause an irq_work to try to wake up
	 * an empty queue.
	 *
	 * There's no reason to protect this flag either, as
	 * the work queue and irq_work logic will do the necessary
	 * synchronization for the wake ups. The only thing
	 * that is necessary is that the wake up happens after
	 * a task has been queued. It's OK for spurious wake ups.
	 */
	trace_wakeup_needed = true;

	if (trace_empty(iter))
		schedule();

	finish_wait(&trace_wait, &wait);
}

Steven Rostedt's avatar
Steven Rostedt committed
803
804
805
806
807
808
/**
 * register_tracer - register a tracer with the ftrace system.
 * @type - the plugin for the tracer
 *
 * Register a new plugin tracer.
 */
809
810
811
812
813
814
815
816
817
818
int register_tracer(struct tracer *type)
{
	struct tracer *t;
	int ret = 0;

	if (!type->name) {
		pr_info("Tracer must have a name\n");
		return -1;
	}

819
	if (strlen(type->name) >= MAX_TRACER_SIZE) {
Li Zefan's avatar
Li Zefan committed
820
821
822
823
		pr_info("Tracer has a name longer than %d\n", MAX_TRACER_SIZE);
		return -1;
	}

824
	mutex_lock(&trace_types_lock);
Ingo Molnar's avatar
Ingo Molnar committed
825

826
827
	tracing_selftest_running = true;

828
829
830
	for (t = trace_types; t; t = t->next) {
		if (strcmp(type->name, t->name) == 0) {
			/* already found */
Li Zefan's avatar
Li Zefan committed
831
			pr_info("Tracer %s already registered\n",
832
833
834
835
836
837
				type->name);
			ret = -1;
			goto out;
		}
	}

838
839
840
841
842
843
844
	if (!type->set_flag)
		type->set_flag = &dummy_set_flag;
	if (!type->flags)
		type->flags = &dummy_tracer_flags;
	else
		if (!type->flags->opts)
			type->flags->opts = dummy_tracer_opt;
845
846
847
	if (!type->wait_pipe)
		type->wait_pipe = default_wait_pipe;

848

Steven Rostedt's avatar
Steven Rostedt committed
849
#ifdef CONFIG_FTRACE_STARTUP_TEST
850
	if (type->selftest && !tracing_selftest_disabled) {
Steven Rostedt's avatar
Steven Rostedt committed
851
852
		struct tracer *saved_tracer = current_trace;
		struct trace_array *tr = &global_trace;
853

Steven Rostedt's avatar
Steven Rostedt committed
854
855
856
857
858
859
860
		/*
		 * Run a selftest on this tracer.
		 * Here we reset the trace buffer, and set the current
		 * tracer to be this tracer. The tracer can then run some
		 * internal tracing to verify that everything is in order.
		 * If we fail, we do not register this tracer.
		 */
861
		tracing_reset_online_cpus(tr);
Ingo Molnar's avatar
Ingo Molnar committed
862

Steven Rostedt's avatar
Steven Rostedt committed
863
		current_trace = type;
864
865
866

		/* If we expanded the buffers, make sure the max is expanded too */
		if (ring_buffer_expanded && type->use_max_tr)
867
868
			ring_buffer_resize(max_tr.buffer, trace_buf_size,
						RING_BUFFER_ALL_CPUS);
869

Steven Rostedt's avatar
Steven Rostedt committed
870
871
872
873
874
875
876
		/* the test is responsible for initializing and enabling */
		pr_info("Testing tracer %s: ", type->name);
		ret = type->selftest(type, tr);
		/* the test is responsible for resetting too */
		current_trace = saved_tracer;
		if (ret) {
			printk(KERN_CONT "FAILED!\n");
877
878
			/* Add the warning after printing 'FAILED' */
			WARN_ON(1);
Steven Rostedt's avatar
Steven Rostedt committed
879
880
			goto out;
		}
Steven Rostedt's avatar
Steven Rostedt committed
881
		/* Only reset on passing, to avoid touching corrupted buffers */
882
		tracing_reset_online_cpus(tr);
Ingo Molnar's avatar
Ingo Molnar committed
883

884
885
		/* Shrink the max buffer again */
		if (ring_buffer_expanded && type->use_max_tr)
886
887
			ring_buffer_resize(max_tr.buffer, 1,
						RING_BUFFER_ALL_CPUS);
888

Steven Rostedt's avatar
Steven Rostedt committed
889
890
891
892
		printk(KERN_CONT "PASSED\n");
	}
#endif

893
894
	type->next = trace_types;
	trace_types = type;
Steven Rostedt's avatar
Steven Rostedt committed
895

896
 out:
897
	tracing_selftest_running = false;
898
899
	mutex_unlock(&trace_types_lock);

Steven Rostedt's avatar
Steven Rostedt committed
900
901
902
	if (ret || !default_bootup_tracer)
		goto out_unlock;

Li Zefan's avatar
Li Zefan committed
903
	if (strncmp(default_bootup_tracer, type->name, MAX_TRACER_SIZE))
Steven Rostedt's avatar
Steven Rostedt committed
904
905
906
907
908
909
910
911
		goto out_unlock;

	printk(KERN_INFO "Starting tracer '%s'\n", type->name);
	/* Do we want this tracer to start on bootup? */
	tracing_set_tracer(type->name);
	default_bootup_tracer = NULL;
	/* disable other selftests, since this will break it. */
	tracing_selftest_disabled = 1;
912
#ifdef CONFIG_FTRACE_STARTUP_TEST
Steven Rostedt's avatar
Steven Rostedt committed
913
914
	printk(KERN_INFO "Disabling FTRACE selftests due to running tracer '%s'\n",
	       type->name);
915
916
#endif

Steven Rostedt's avatar
Steven Rostedt committed
917
 out_unlock:
918
919
920
	return ret;
}

921
922
923
924
void tracing_reset(struct trace_array *tr, int cpu)
{
	struct ring_buffer *buffer = tr->buffer;

925
926
927
	if (!buffer)
		return;

928
929
930
931
	ring_buffer_record_disable(buffer);

	/* Make sure all commits have finished */
	synchronize_sched();
932
	ring_buffer_reset_cpu(buffer, cpu);
933
934
935
936

	ring_buffer_record_enable(buffer);
}

937
938
void tracing_reset_online_cpus(struct trace_array *tr)
{
939
	struct ring_buffer *buffer = tr->buffer;
940
941
	int cpu;

942
943
944
	if (!buffer)
		return;

945
946
947
948
949
	ring_buffer_record_disable(buffer);

	/* Make sure all commits have finished */
	synchronize_sched();

950
951
952
	tr->time_start = ftrace_now(tr->cpu);

	for_each_online_cpu(cpu)
953
		ring_buffer_reset_cpu(buffer, cpu);
954
955

	ring_buffer_record_enable(buffer);
956
957
}

958
959
960
961
962
963
964
965
966
967
void tracing_reset_current(int cpu)
{
	tracing_reset(&global_trace, cpu);
}

void tracing_reset_current_online_cpus(void)
{
	tracing_reset_online_cpus(&global_trace);
}

968
#define SAVED_CMDLINES 128
969
#define NO_CMDLINE_MAP UINT_MAX
970
971
972
973
static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];
static unsigned map_cmdline_to_pid[SAVED_CMDLINES];
static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN];
static int cmdline_idx;
974
static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
975
976

/* temporary disable recording */
977
static atomic_t trace_record_cmdline_disabled __read_mostly;
978
979
980

static void trace_init_cmdlines(void)
{
981
982
	memset(&map_pid_to_cmdline, NO_CMDLINE_MAP, sizeof(map_pid_to_cmdline));
	memset(&map_cmdline_to_pid, NO_CMDLINE_MAP, sizeof(map_cmdline_to_pid));
983
984
985
	cmdline_idx = 0;
}

986
987
988
989
990
int is_tracing_stopped(void)
{
	return trace_stop_count;
}

991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
/**
 * ftrace_off_permanent - disable all ftrace code permanently
 *
 * This should only be called when a serious anomally has
 * been detected.  This will turn off the function tracing,
 * ring buffers, and other tracing utilites. It takes no
 * locks and can be called from any context.
 */
void ftrace_off_permanent(void)
{
	tracing_disabled = 1;
	ftrace_stop();
	tracing_off_permanent();
}

1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
/**
 * tracing_start - quick start of the tracer
 *
 * If tracing is enabled but was stopped by tracing_stop,
 * this will start the tracer back up.
 */
void tracing_start(void)
{
	struct ring_buffer *buffer;
	unsigned long flags;

	if (tracing_disabled)
		return;

1020
	raw_spin_lock_irqsave(&tracing_start_lock, flags);
1021
1022
1023
1024
1025
1026
	if (--trace_stop_count) {
		if (trace_stop_count < 0) {
			/* Someone screwed up their debugging */
			WARN_ON_ONCE(1);
			trace_stop_count = 0;
		}
1027
1028
1029
		goto out;
	}

1030
1031
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);
1032
1033
1034
1035
1036
1037
1038
1039
1040

	buffer = global_trace.buffer;
	if (buffer)
		ring_buffer_record_enable(buffer);

	buffer = max_tr.buffer;
	if (buffer)
		ring_buffer_record_enable(buffer);

1041
1042
	arch_spin_unlock(&ftrace_max_lock);

1043
1044
	ftrace_start();
 out:
1045
	raw_spin_unlock_irqrestore(&tracing_start_lock, flags);
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
}

/**
 * tracing_stop - quick stop of the tracer
 *
 * Light weight way to stop tracing. Use in conjunction with
 * tracing_start.
 */
void tracing_stop(void)
{
	struct ring_buffer *buffer;
	unsigned long flags;

	ftrace_stop();
1060
	raw_spin_lock_irqsave(&tracing_start_lock, flags);
1061
1062
1063
	if (trace_stop_count++)
		goto out;

1064
1065
1066
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);

1067
1068
1069
1070
1071
1072
1073
1074
	buffer = global_trace.buffer;
	if (buffer)
		ring_buffer_record_disable(buffer);

	buffer = max_tr.buffer;
	if (buffer)
		ring_buffer_record_disable(buffer);

1075
1076
	arch_spin_unlock(&ftrace_max_lock);

1077
 out:
1078
	raw_spin_unlock_irqrestore(&tracing_start_lock, flags);
1079
1080
}

Ingo Molnar's avatar
Ingo Molnar committed
1081
void trace_stop_cmdline_recording(void);
1082

Ingo Molnar's avatar
Ingo Molnar committed
1083
static void trace_save_cmdline(struct task_struct *tsk)
1084
{
1085
	unsigned pid, idx;
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095

	if (!tsk->pid || unlikely(tsk->pid > PID_MAX_DEFAULT))
		return;

	/*
	 * It's not the end of the world if we don't get
	 * the lock, but we also don't want to spin
	 * nor do we want to disable interrupts,
	 * so if we miss here, then better luck next time.
	 */
1096
	if (!arch_spin_trylock(&trace_cmdline_lock))
1097
1098
1099
		return;

	idx = map_pid_to_cmdline[tsk->pid];
1100
	if (idx == NO_CMDLINE_MAP) {
1101
1102
		idx = (cmdline_idx + 1) % SAVED_CMDLINES;

1103
1104
1105
1106
1107
1108
1109
1110
1111
		/*
		 * Check whether the cmdline buffer at idx has a pid
		 * mapped. We are going to overwrite that entry so we
		 * need to clear the map_pid_to_cmdline. Otherwise we
		 * would read the new comm for the old pid.
		 */
		pid = map_cmdline_to_pid[idx];
		if (pid != NO_CMDLINE_MAP)
			map_pid_to_cmdline[pid] = NO_CMDLINE_MAP;
1112

1113
		map_cmdline_to_pid[idx] = tsk->pid;
1114
1115
1116
1117
1118
1119
1120
		map_pid_to_cmdline[tsk->pid] = idx;

		cmdline_idx = idx;
	}

	memcpy(&saved_cmdlines[idx], tsk->comm, TASK_COMM_LEN);

1121
	arch_spin_unlock(&trace_cmdline_lock);
1122
1123
}

1124
void trace_find_cmdline(int pid, char comm[])
1125
1126
1127
{
	unsigned map;

1128
1129
1130
1131
	if (!pid) {
		strcpy(comm, "<idle>");
		return;
	}
1132

1133
1134
1135
1136
1137
	if (WARN_ON_ONCE(pid < 0)) {
		strcpy(comm, "<XXX>");
		return;
	}

1138
1139
1140
1141
	if (pid > PID_MAX_DEFAULT) {
		strcpy(comm, "<...>");
		return;
	}
1142

1143
	preempt_disable();
1144
	arch_spin_lock(&trace_cmdline_lock);
1145
	map = map_pid_to_cmdline[pid];
1146
1147
1148
1149
	if (map != NO_CMDLINE_MAP)
		strcpy(comm, saved_cmdlines[map]);
	else
		strcpy(comm, "<...>");
1150

1151
	arch_spin_unlock(&trace_cmdline_lock);
1152
	preempt_enable();
1153
1154
}

Ingo Molnar's avatar
Ingo Molnar committed
1155
void tracing_record_cmdline(struct task_struct *tsk)
1156
{
1157
	if (atomic_read(&trace_record_cmdline_disabled) || !tracing_is_on())
1158
1159
		return;

1160
1161
1162
1163
1164
	if (!__this_cpu_read(trace_cmdline_save))
		return;

	__this_cpu_write(trace_cmdline_save, false);

1165
1166
1167
	trace_save_cmdline(tsk);
}

1168
void
1169
1170
tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
			     int pc)
1171
1172
1173
{
	struct task_struct *tsk = current;

1174
1175
1176
	entry->preempt_count		= pc & 0xff;
	entry->pid			= (tsk) ? tsk->pid : 0;
	entry->flags =
1177
#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
1178
		(irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) |
1179
1180
1181
#else
		TRACE_FLAG_IRQS_NOSUPPORT |
#endif
1182
1183
1184
1185
		((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
		((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
		(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
}
1186
EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
1187

1188
1189
1190
1191
1192
struct ring_buffer_event *
trace_buffer_lock_reserve(struct ring_buffer *buffer,
			  int type,
			  unsigned long len,
			  unsigned long flags, int pc)
1193
1194
1195
{
	struct ring_buffer_event *event;

1196
	event = ring_buffer_lock_reserve(buffer, len);
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
	if (event != NULL) {
		struct trace_entry *ent = ring_buffer_event_data(event);

		tracing_generic_entry_update(ent, flags, pc);
		ent->type = type;
	}

	return event;
}

1207
1208
1209
1210
void
__buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event)
{
	__this_cpu_write(trace_cmdline_save, true);
1211
1212
1213
1214
1215
	if (trace_wakeup_needed) {
		trace_wakeup_needed = false;
		/* irq_work_queue() supplies it's own memory barriers */
		irq_work_queue(&trace_work_wakeup);
	}
1216
1217
1218
	ring_buffer_unlock_commit(buffer, event);
}

1219
1220
1221
static inline void
__trace_buffer_unlock_commit(struct ring_buffer *buffer,
			     struct ring_buffer_event *event,
1222
			     unsigned long flags, int pc)
1223
{
1224
	__buffer_unlock_commit(buffer, event);
1225

1226
1227
	ftrace_trace_stack(buffer, flags, 6, pc);
	ftrace_trace_userstack(buffer, flags, pc);
1228
1229
}

1230
1231
1232
void trace_buffer_unlock_commit(struct ring_buffer *buffer,
				struct ring_buffer_event *event,
				unsigned long flags, int pc)
1233
{
1234
	__trace_buffer_unlock_commit(buffer, event, flags, pc);
1235
}
1236
EXPORT_SYMBOL_GPL(trace_buffer_unlock_commit);
1237

1238
struct ring_buffer_event *
1239
1240
trace_current_buffer_lock_reserve(struct ring_buffer **current_rb,
				  int type, unsigned long len,
1241
1242
				  unsigned long flags, int pc)
{
1243
1244
	*current_rb = global_trace.buffer;
	return trace_buffer_lock_reserve(*current_rb,
1245
1246
					 type, len, flags, pc);
}
1247
EXPORT_SYMBOL_GPL(trace_current_buffer_lock_reserve);
1248

1249
1250
void trace_current_buffer_unlock_commit(struct ring_buffer *buffer,
					struct ring_buffer_event *event,
1251
1252
					unsigned long flags, int pc)
{
1253
	__trace_buffer_unlock_commit(buffer, event, flags, pc);
1254
}
1255
EXPORT_SYMBOL_GPL(trace_current_buffer_unlock_commit);
1256

1257
1258
1259
1260
void trace_buffer_unlock_commit_regs(struct ring_buffer *buffer,
				     struct ring_buffer_event *event,
				     unsigned long flags, int pc,
				     struct pt_regs *regs)
1261
{
1262
	__buffer_unlock_commit(buffer, event);
1263
1264
1265
1266

	ftrace_trace_stack_regs(buffer, flags, 0, pc, regs);
	ftrace_trace_userstack(buffer, flags, pc);
}
1267
EXPORT_SYMBOL_GPL(trace_buffer_unlock_commit_regs);
1268

1269
1270
void trace_current_buffer_discard_commit(struct ring_buffer *buffer,
					 struct ring_buffer_event *event)