trace.c 132 KB
Newer Older
1
2
3
/*
 * ring buffer based function tracer
 *
4
 * Copyright (C) 2007-2012 Steven Rostedt <srostedt@redhat.com>
5
6
7
8
9
10
11
 * 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/debugfs.h>
23
#include <linux/pagemap.h>
24
25
26
#include <linux/hardirq.h>
#include <linux/linkage.h>
#include <linux/uaccess.h>
27
#include <linux/kprobes.h>
28
29
30
#include <linux/ftrace.h>
#include <linux/module.h>
#include <linux/percpu.h>
31
#include <linux/splice.h>
32
#include <linux/kdebug.h>
33
#include <linux/string.h>
34
#include <linux/rwsem.h>
35
#include <linux/slab.h>
36
37
#include <linux/ctype.h>
#include <linux/init.h>
38
#include <linux/poll.h>
39
#include <linux/nmi.h>
40
#include <linux/fs.h>
41
#include <linux/sched/rt.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
/*
 * 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.
 */
94
static int tracing_disabled = 1;
95

96
DEFINE_PER_CPU(int, ftrace_cpu_disabled);
97

98
cpumask_var_t __read_mostly	tracing_buffer_mask;
99

100
101
102
103
104
105
106
107
108
109
110
/*
 * 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
111
112
113
 * /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
114
 */
115
116

enum ftrace_dump_mode ftrace_dump_on_oops;
117

118
119
static int tracing_set_tracer(const char *buf);

Li Zefan's avatar
Li Zefan committed
120
121
#define MAX_TRACER_SIZE		100
static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata;
122
static char *default_bootup_tracer;
123

124
static int __init set_cmdline_ftrace(char *str)
125
{
Li Zefan's avatar
Li Zefan committed
126
	strncpy(bootup_tracer_buf, str, MAX_TRACER_SIZE);
127
	default_bootup_tracer = bootup_tracer_buf;
128
129
	/* We are using ftrace early, expand it */
	ring_buffer_expanded = 1;
130
131
	return 1;
}
132
__setup("ftrace=", set_cmdline_ftrace);
133

134
135
static int __init set_ftrace_dump_on_oops(char *str)
{
136
137
138
139
140
141
142
143
144
145
146
	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;
147
148
}
__setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops);
Steven Rostedt's avatar
Steven Rostedt committed
149

150
151
152
153
154
155
156
157
158
159
160
161

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);

162
unsigned long long ns2usecs(cycle_t nsec)
163
164
165
166
167
168
{
	nsec += 500;
	do_div(nsec, 1000);
	return nsec;
}

Steven Rostedt's avatar
Steven Rostedt committed
169
170
171
172
173
174
175
176
177
178
179
180
/*
 * 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.
 */
181
182
static struct trace_array	global_trace;

183
184
LIST_HEAD(ftrace_trace_arrays);

185
186
int filter_current_check_discard(struct ring_buffer *buffer,
				 struct ftrace_event_call *call, void *rec,
187
188
				 struct ring_buffer_event *event)
{
189
	return filter_check_discard(call, rec, buffer, event);
190
}
191
EXPORT_SYMBOL_GPL(filter_current_check_discard);
192

193
194
195
196
197
198
199
200
201
202
203
204
205
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;
}
206

Steven Rostedt's avatar
Steven Rostedt committed
207
208
209
210
211
212
213
214
215
216
/*
 * 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.
 */
217
218
static struct trace_array	max_tr;

219
220
int tracing_is_enabled(void)
{
221
	return tracing_is_on();
222
223
}

Steven Rostedt's avatar
Steven Rostedt committed
224
/*
225
226
227
 * 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.
228
229
230
231
232
 *
 * 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
233
 */
234
#define TRACE_BUF_SIZE_DEFAULT	1441792UL /* 16384 * 88 (sizeof(entry)) */
235

236
static unsigned long		trace_buf_size = TRACE_BUF_SIZE_DEFAULT;
237

Steven Rostedt's avatar
Steven Rostedt committed
238
/* trace_types holds a link list of available tracers. */
239
static struct tracer		*trace_types __read_mostly;
Steven Rostedt's avatar
Steven Rostedt committed
240
241
242
243

/*
 * trace_types_lock is used to protect the trace_types list.
 */
244
static DEFINE_MUTEX(trace_types_lock);
Steven Rostedt's avatar
Steven Rostedt committed
245

246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
/*
 * 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)
{
274
	if (cpu == RING_BUFFER_ALL_CPUS) {
275
276
277
278
279
		/* gain it for accessing the whole ring buffer. */
		down_write(&all_cpu_access_lock);
	} else {
		/* gain it for accessing a cpu ring buffer. */

280
		/* Firstly block other trace_access_lock(RING_BUFFER_ALL_CPUS). */
281
282
283
284
285
286
287
288
289
		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)
{
290
	if (cpu == RING_BUFFER_ALL_CPUS) {
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
		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

328
/* trace_flags holds trace_options default values */
329
unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK |
330
	TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | TRACE_ITER_SLEEP_TIME |
331
	TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE |
332
	TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS;
333

334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
/**
 * 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)
365
		ring_buffer_record_off(global_trace.buffer);
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
	/*
	 * 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);

387
static int __init set_buf_size(char *str)
388
{
389
	unsigned long buf_size;
390

391
392
	if (!str)
		return 0;
393
	buf_size = memparse(str, &str);
394
	/* nr_entries can not be zero */
395
	if (buf_size == 0)
396
		return 0;
397
	trace_buf_size = buf_size;
398
399
	return 1;
}
400
__setup("trace_buf_size=", set_buf_size);
401

402
403
static int __init set_tracing_thresh(char *str)
{
404
	unsigned long threshold;
405
406
407
408
	int ret;

	if (!str)
		return 0;
409
	ret = kstrtoul(str, 0, &threshold);
410
411
	if (ret < 0)
		return 0;
412
	tracing_thresh = threshold * 1000;
413
414
415
416
	return 1;
}
__setup("tracing_thresh=", set_tracing_thresh);

Steven Rostedt's avatar
Steven Rostedt committed
417
418
419
420
421
unsigned long nsecs_to_usecs(unsigned long nsecs)
{
	return nsecs / 1000;
}

Steven Rostedt's avatar
Steven Rostedt committed
422
/* These must match the bit postions in trace_iterator_flags */
423
424
425
426
427
static const char *trace_options[] = {
	"print-parent",
	"sym-offset",
	"sym-addr",
	"verbose",
Ingo Molnar's avatar
Ingo Molnar committed
428
	"raw",
429
	"hex",
Ingo Molnar's avatar
Ingo Molnar committed
430
	"bin",
431
	"block",
Ingo Molnar's avatar
Ingo Molnar committed
432
	"stacktrace",
433
	"trace_printk",
434
	"ftrace_preempt",
435
	"branch",
436
	"annotate",
437
	"userstacktrace",
438
	"sym-userobj",
439
	"printk-msg-only",
440
	"context-info",
441
	"latency-format",
442
	"sleep-time",
443
	"graph-time",
444
	"record-cmd",
445
	"overwrite",
446
	"disable_on_free",
447
	"irq-info",
448
	"markers",
449
450
451
	NULL
};

452
453
454
static struct {
	u64 (*func)(void);
	const char *name;
455
	int in_ns;		/* is this clock in nanoseconds? */
456
} trace_clocks[] = {
457
458
459
	{ trace_clock_local,	"local",	1 },
	{ trace_clock_global,	"global",	1 },
	{ trace_clock_counter,	"counter",	0 },
460
	ARCH_TRACE_CLOCKS
461
462
463
464
};

int trace_clock_id;

465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
/*
 * 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)) {
542
		if (parser->idx < parser->size - 1)
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
			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;
}

571
572
573
574
575
ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt)
{
	int len;
	int ret;

576
577
578
	if (!cnt)
		return 0;

579
580
581
582
583
584
585
	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);
586
	if (ret == cnt)
587
588
		return -EFAULT;

589
590
	cnt -= ret;

591
	s->readpos += cnt;
592
	return cnt;
593
594
}

595
static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt)
596
597
598
599
600
601
602
603
604
{
	int len;

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

	len = s->len - s->readpos;
	if (cnt > len)
		cnt = len;
605
	memcpy(buf, s->buffer + s->readpos, cnt);
606

607
	s->readpos += cnt;
608
609
610
	return cnt;
}

611
612
613
614
615
616
/*
 * 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.
 *
617
 * This is defined as a arch_spinlock_t in order to help
618
619
620
621
622
623
 * 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.
 */
624
static arch_spinlock_t ftrace_max_lock =
625
	(arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
626

627
628
unsigned long __read_mostly	tracing_thresh;

629
630
631
632
633
634
635
636
637
638
639
#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)
{
640
	struct trace_array_cpu *data = per_cpu_ptr(tr->data, cpu);
641
	struct trace_array_cpu *max_data;
642
643
644
645

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

646
	max_data = per_cpu_ptr(max_tr.data, cpu);
647
648
649
	max_data->saved_latency = tracing_max_latency;
	max_data->critical_start = data->critical_start;
	max_data->critical_end = data->critical_end;
650

651
	memcpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
652
653
654
655
656
	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;
657
658
659
660
661

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

Steven Rostedt's avatar
Steven Rostedt committed
662
663
664
665
666
667
668
669
670
/**
 * 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
671
void
672
673
update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
674
	struct ring_buffer *buf;
675

676
	if (tr->stop_count)
677
678
		return;

679
	WARN_ON_ONCE(!irqs_disabled());
680

681
	if (!tr->current_trace->allocated_snapshot) {
682
		/* Only the nop tracer should hit this when disabling */
683
		WARN_ON_ONCE(tr->current_trace != &nop_trace);
684
		return;
685
	}
686

687
	arch_spin_lock(&ftrace_max_lock);
688

689
	buf = tr->buffer;
690
691
692
	tr->buffer = max_tr.buffer;
	max_tr.buffer = buf;

693
	__update_max_tr(tr, tsk, cpu);
694
	arch_spin_unlock(&ftrace_max_lock);
695
696
697
698
699
700
701
}

/**
 * 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
702
703
 *
 * Flip the trace of a single CPU buffer between the @tr and the max_tr.
704
 */
Ingo Molnar's avatar
Ingo Molnar committed
705
void
706
707
update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
708
	int ret;
709

710
	if (tr->stop_count)
711
712
		return;

713
	WARN_ON_ONCE(!irqs_disabled());
714
	if (WARN_ON_ONCE(!tr->current_trace->allocated_snapshot))
715
716
		return;

717
	arch_spin_lock(&ftrace_max_lock);
718

719
720
	ret = ring_buffer_swap_cpu(max_tr.buffer, tr->buffer, cpu);

721
722
723
724
725
726
727
728
729
730
731
732
	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);
733
734

	__update_max_tr(tr, tsk, cpu);
735
	arch_spin_unlock(&ftrace_max_lock);
736
}
737
#endif /* CONFIG_TRACER_MAX_TRACE */
738

739
740
static void default_wait_pipe(struct trace_iterator *iter)
{
741
742
743
	/* Iterators are static, they should be filled or empty */
	if (trace_buffer_iter(iter, iter->cpu_file))
		return;
744

745
	ring_buffer_wait(iter->tr->buffer, iter->cpu_file);
746
747
}

Steven Rostedt's avatar
Steven Rostedt committed
748
749
750
751
752
753
/**
 * register_tracer - register a tracer with the ftrace system.
 * @type - the plugin for the tracer
 *
 * Register a new plugin tracer.
 */
754
755
756
757
758
759
760
761
762
763
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;
	}

764
	if (strlen(type->name) >= MAX_TRACER_SIZE) {
Li Zefan's avatar
Li Zefan committed
765
766
767
768
		pr_info("Tracer has a name longer than %d\n", MAX_TRACER_SIZE);
		return -1;
	}

769
	mutex_lock(&trace_types_lock);
Ingo Molnar's avatar
Ingo Molnar committed
770

771
772
	tracing_selftest_running = true;

773
774
775
	for (t = trace_types; t; t = t->next) {
		if (strcmp(type->name, t->name) == 0) {
			/* already found */
Li Zefan's avatar
Li Zefan committed
776
			pr_info("Tracer %s already registered\n",
777
778
779
780
781
782
				type->name);
			ret = -1;
			goto out;
		}
	}

783
784
785
786
787
788
789
	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;
790
791
792
	if (!type->wait_pipe)
		type->wait_pipe = default_wait_pipe;

793

Steven Rostedt's avatar
Steven Rostedt committed
794
#ifdef CONFIG_FTRACE_STARTUP_TEST
795
	if (type->selftest && !tracing_selftest_disabled) {
Steven Rostedt's avatar
Steven Rostedt committed
796
		struct trace_array *tr = &global_trace;
797
		struct tracer *saved_tracer = tr->current_trace;
798

Steven Rostedt's avatar
Steven Rostedt committed
799
800
801
802
803
804
805
		/*
		 * 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.
		 */
806
		tracing_reset_online_cpus(tr);
Ingo Molnar's avatar
Ingo Molnar committed
807

808
		tr->current_trace = type;
809

810
811
812
813
814
815
816
		if (type->use_max_tr) {
			/* If we expanded the buffers, make sure the max is expanded too */
			if (ring_buffer_expanded)
				ring_buffer_resize(max_tr.buffer, trace_buf_size,
						   RING_BUFFER_ALL_CPUS);
			type->allocated_snapshot = true;
		}
817

Steven Rostedt's avatar
Steven Rostedt committed
818
819
820
821
		/* 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 */
822
		tr->current_trace = saved_tracer;
Steven Rostedt's avatar
Steven Rostedt committed
823
824
		if (ret) {
			printk(KERN_CONT "FAILED!\n");
825
826
			/* Add the warning after printing 'FAILED' */
			WARN_ON(1);
Steven Rostedt's avatar
Steven Rostedt committed
827
828
			goto out;
		}
Steven Rostedt's avatar
Steven Rostedt committed
829
		/* Only reset on passing, to avoid touching corrupted buffers */
830
		tracing_reset_online_cpus(tr);
Ingo Molnar's avatar
Ingo Molnar committed
831

832
833
834
835
836
837
838
839
		if (type->use_max_tr) {
			type->allocated_snapshot = false;

			/* Shrink the max buffer again */
			if (ring_buffer_expanded)
				ring_buffer_resize(max_tr.buffer, 1,
						   RING_BUFFER_ALL_CPUS);
		}
840

Steven Rostedt's avatar
Steven Rostedt committed
841
842
843
844
		printk(KERN_CONT "PASSED\n");
	}
#endif

845
846
	type->next = trace_types;
	trace_types = type;
Steven Rostedt's avatar
Steven Rostedt committed
847

848
 out:
849
	tracing_selftest_running = false;
850
851
	mutex_unlock(&trace_types_lock);

Steven Rostedt's avatar
Steven Rostedt committed
852
853
854
	if (ret || !default_bootup_tracer)
		goto out_unlock;

Li Zefan's avatar
Li Zefan committed
855
	if (strncmp(default_bootup_tracer, type->name, MAX_TRACER_SIZE))
Steven Rostedt's avatar
Steven Rostedt committed
856
857
858
859
860
861
862
863
		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;
864
#ifdef CONFIG_FTRACE_STARTUP_TEST
Steven Rostedt's avatar
Steven Rostedt committed
865
866
	printk(KERN_INFO "Disabling FTRACE selftests due to running tracer '%s'\n",
	       type->name);
867
868
#endif

Steven Rostedt's avatar
Steven Rostedt committed
869
 out_unlock:
870
871
872
	return ret;
}

873
874
875
876
void tracing_reset(struct trace_array *tr, int cpu)
{
	struct ring_buffer *buffer = tr->buffer;

877
878
879
	if (!buffer)
		return;

880
881
882
883
	ring_buffer_record_disable(buffer);

	/* Make sure all commits have finished */
	synchronize_sched();
884
	ring_buffer_reset_cpu(buffer, cpu);
885
886
887
888

	ring_buffer_record_enable(buffer);
}

889
890
void tracing_reset_online_cpus(struct trace_array *tr)
{
891
	struct ring_buffer *buffer = tr->buffer;
892
893
	int cpu;

894
895
896
	if (!buffer)
		return;

897
898
899
900
901
	ring_buffer_record_disable(buffer);

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

902
903
904
	tr->time_start = ftrace_now(tr->cpu);

	for_each_online_cpu(cpu)
905
		ring_buffer_reset_cpu(buffer, cpu);
906
907

	ring_buffer_record_enable(buffer);
908
909
}

910
911
912
913
914
void tracing_reset_current(int cpu)
{
	tracing_reset(&global_trace, cpu);
}

915
void tracing_reset_all_online_cpus(void)
916
{
917
918
919
920
921
922
923
	struct trace_array *tr;

	mutex_lock(&trace_types_lock);
	list_for_each_entry(tr, &ftrace_trace_arrays, list) {
		tracing_reset_online_cpus(tr);
	}
	mutex_unlock(&trace_types_lock);
924
925
}

926
#define SAVED_CMDLINES 128
927
#define NO_CMDLINE_MAP UINT_MAX
928
929
930
931
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;
932
static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
933
934

/* temporary disable recording */
935
static atomic_t trace_record_cmdline_disabled __read_mostly;
936
937
938

static void trace_init_cmdlines(void)
{
939
940
	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));
941
942
943
	cmdline_idx = 0;
}

944
945
int is_tracing_stopped(void)
{
946
	return global_trace.stop_count;
947
948
}

949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
/**
 * 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();
}

964
965
966
967
968
969
970
971
972
973
974
975
976
977
/**
 * 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;

978
979
980
	raw_spin_lock_irqsave(&global_trace.start_lock, flags);
	if (--global_trace.stop_count) {
		if (global_trace.stop_count < 0) {
981
982
			/* Someone screwed up their debugging */
			WARN_ON_ONCE(1);
983
			global_trace.stop_count = 0;
984
		}
985
986
987
		goto out;
	}

988
989
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);
990
991
992
993
994
995
996
997
998

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

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

999
1000
	arch_spin_unlock(&ftrace_max_lock);

1001
1002
	ftrace_start();
 out:
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
	raw_spin_unlock_irqrestore(&global_trace.start_lock, flags);
}

static void tracing_start_tr(struct trace_array *tr)
{
	struct ring_buffer *buffer;
	unsigned long flags;

	if (tracing_disabled)
		return;

	/* If global, we need to also start the max tracer */
	if (tr->flags & TRACE_ARRAY_FL_GLOBAL)
		return tracing_start();

	raw_spin_lock_irqsave(&tr->start_lock, flags);

	if (--tr->stop_count) {
		if (tr->stop_count < 0) {
			/* Someone screwed up their debugging */
			WARN_ON_ONCE(1);
			tr->stop_count = 0;
		}
		goto out;
	}

	buffer = tr->buffer;
	if (buffer)
		ring_buffer_record_enable(buffer);

 out:
	raw_spin_unlock_irqrestore(&tr->start_lock, flags);
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
}

/**
 * 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();
1049
1050
	raw_spin_lock_irqsave(&global_trace.start_lock, flags);
	if (global_trace.stop_count++)
1051
1052
		goto out;

1053
1054
1055
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);

1056
1057
1058
1059
1060
1061
1062
1063
	buffer = global_trace.buffer;
	if (buffer)
		ring_buffer_record_disable(buffer);

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

1064
1065
	arch_spin_unlock(&ftrace_max_lock);

1066
 out:
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
	raw_spin_unlock_irqrestore(&global_trace.start_lock, flags);
}

static void tracing_stop_tr(struct trace_array *tr)
{
	struct ring_buffer *buffer;
	unsigned long flags;

	/* If global, we need to also stop the max tracer */
	if (tr->flags & TRACE_ARRAY_FL_GLOBAL)
		return tracing_stop();

	raw_spin_lock_irqsave(&tr->start_lock, flags);
	if (tr->stop_count++)
		goto out;

	buffer = tr->buffer;
	if (buffer)
		ring_buffer_record_disable(buffer);

 out:
	raw_spin_unlock_irqrestore(&tr->start_lock, flags);
1089
1090
}

Ingo Molnar's avatar
Ingo Molnar committed
1091
void trace_stop_cmdline_recording(void);
1092

Ingo Molnar's avatar
Ingo Molnar committed
1093
static void trace_save_cmdline(struct task_struct *tsk)
1094
{
1095
	unsigned pid, idx;
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105

	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.
	 */
1106
	if (!arch_spin_trylock(&trace_cmdline_lock))
1107
1108
1109
		return;

	idx = map_pid_to_cmdline[tsk->pid];
1110
	if (idx == NO_CMDLINE_MAP) {
1111
1112
		idx = (cmdline_idx + 1) % SAVED_CMDLINES;

1113
1114
1115
1116
1117
1118
1119
1120
1121
		/*
		 * 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;
1122

1123
		map_cmdline_to_pid[idx] = tsk->pid;
1124
1125
1126
1127
1128
1129
1130
		map_pid_to_cmdline[tsk->pid] = idx;

		cmdline_idx = idx;
	}

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

1131
	arch_spin_unlock(&trace_cmdline_lock);
1132
1133
}

1134
void trace_find_cmdline(int pid, char comm[])
1135
1136
1137
{
	unsigned map;

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

1143
1144
1145
1146
1147
	if (WARN_ON_ONCE(pid < 0)) {
		strcpy(comm, "<XXX>");
		return;
	}

1148
1149
1150
1151
	if (pid > PID_MAX_DEFAULT) {
		strcpy(comm, "<...>");
		return;
	}
1152

1153
	preempt_disable();
1154
	arch_spin_lock(&trace_cmdline_lock);
1155
	map = map_pid_to_cmdline[pid];
1156
1157
1158
1159
	if (map != NO_CMDLINE_MAP)
		strcpy(comm, saved_cmdlines[map]);
	else
		strcpy(comm, "<...>");
1160

1161
	arch_spin_unlock(&trace_cmdline_lock);
1162
	preempt_enable();
1163
1164
}

Ingo Molnar's avatar
Ingo Molnar committed
1165
void tracing_record_cmdline(struct task_struct *tsk)
1166
{
1167
	if (atomic_read(&trace_record_cmdline_disabled) || !tracing_is_on())
1168
1169
		return;

1170
1171
1172
1173
1174
	if (!__this_cpu_read(trace_cmdline_save))
		return;

	__this_cpu_write(trace_cmdline_save, false);

1175
1176
1177
	trace_save_cmdline(tsk);
}

1178
void
1179
1180
tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
			     int pc)
1181
1182
1183
{
	struct task_struct *tsk = current;

1184
1185
1186
	entry->preempt_count		= pc & 0xff;
	entry->pid			= (tsk) ? tsk->pid : 0;
	entry->flags =
1187
#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
1188
		(irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) |
1189
1190
1191
#else
		TRACE_FLAG_IRQS_NOSUPPORT |
#endif
1192
1193
1194
1195
		((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
		((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
		(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
}
1196
EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
1197

1198
1199
1200
1201
1202
struct ring_buffer_event *
trace_buffer_lock_reserve(struct ring_buffer *buffer,
			  int type,
			  unsigned long len,
			  unsigned long flags, int pc)
1203
1204
1205
{
	struct ring_buffer_event *event;

1206
	event = ring_buffer_lock_reserve(buffer, len);
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
	if (event != NULL) {
		struct trace_entry *ent = ring_buffer_event_data(event);

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

	return event;
}

1217
1218
1219
1220
1221
1222
1223
void
__buffer_unlock_commit(struct ring_buffer *buffer, struct ring_buffer_event *event)
{
	__this_cpu_write(trace_cmdline_save, true);
	ring_buffer_unlock_commit(buffer, event);
}

1224
1225
1226
static inline void
__trace_buffer_unlock_commit(struct ring_buffer *buffer,
			     struct ring_buffer_event *event,
1227
			     unsigned long flags, int pc)
1228
{
1229
	__buffer_unlock_commit(buffer, event);
1230

1231
1232
	ftrace_trace_stack(buffer, flags, 6, pc);
	ftrace_trace_userstack(buffer, flags, pc);
1233
1234
}

1235
1236
1237
void trace_buffer_unlock_commit(struct ring_buffer *buffer,
				struct ring_buffer_event *event,
				unsigned long flags, int pc)
1238
{
1239
	__trace_buffer_unlock_commit(buffer, event, flags, pc);
1240
}
1241
EXPORT_SYMBOL_GPL(trace_buffer_unlock_commit);
1242

1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
struct ring_buffer_event *
trace_event_buffer_lock_reserve(struct ring_buffer **current_rb,
			  struct ftrace_event_file *ftrace_file,
			  int type, unsigned long len,
			  unsigned long flags, int pc)
{
	*current_rb = ftrace_file->tr->buffer;
	return trace_buffer_lock_reserve(*current_rb,
					 type, len, flags, pc);
}
EXPORT_SYMBOL_GPL(trace_event_buffer_lock_reserve);

1255
struct ring_buffer_event *
1256
1257
trace_current_buffer_lock_reserve(struct ring_buffer **current_rb