trace.c 142 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
bool 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
125
static bool allocate_snapshot;

126
static int __init set_cmdline_ftrace(char *str)
127
{
Li Zefan's avatar
Li Zefan committed
128
	strncpy(bootup_tracer_buf, str, MAX_TRACER_SIZE);
129
	default_bootup_tracer = bootup_tracer_buf;
130
	/* We are using ftrace early, expand it */
131
	ring_buffer_expanded = true;
132
133
	return 1;
}
134
__setup("ftrace=", set_cmdline_ftrace);
135

136
137
static int __init set_ftrace_dump_on_oops(char *str)
{
138
139
140
141
142
143
144
145
146
147
148
	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;
149
150
}
__setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops);
Steven Rostedt's avatar
Steven Rostedt committed
151

152
153
154
155
156
157
158
159
160
static int __init alloc_snapshot(char *str)
{
	allocate_snapshot = true;
	/* We also need the main ring buffer expanded */
	ring_buffer_expanded = true;
	return 1;
}
__setup("alloc_snapshot", alloc_snapshot);

161
162
163
164
165
166
167
168
169
170
171
172

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);

173
unsigned long long ns2usecs(cycle_t nsec)
174
175
176
177
178
179
{
	nsec += 500;
	do_div(nsec, 1000);
	return nsec;
}

Steven Rostedt's avatar
Steven Rostedt committed
180
181
182
183
184
185
186
187
188
189
190
191
/*
 * 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.
 */
192
193
static struct trace_array	global_trace;

194
195
LIST_HEAD(ftrace_trace_arrays);

196
197
int filter_current_check_discard(struct ring_buffer *buffer,
				 struct ftrace_event_call *call, void *rec,
198
199
				 struct ring_buffer_event *event)
{
200
	return filter_check_discard(call, rec, buffer, event);
201
}
202
EXPORT_SYMBOL_GPL(filter_current_check_discard);
203

204
205
206
207
208
cycle_t ftrace_now(int cpu)
{
	u64 ts;

	/* Early boot up does not have a buffer yet */
209
	if (!global_trace.trace_buffer.buffer)
210
211
		return trace_clock_local();

212
213
	ts = ring_buffer_time_stamp(global_trace.trace_buffer.buffer, cpu);
	ring_buffer_normalize_time_stamp(global_trace.trace_buffer.buffer, cpu, &ts);
214
215
216

	return ts;
}
217

218
219
int tracing_is_enabled(void)
{
220
	return tracing_is_on();
221
222
}

Steven Rostedt's avatar
Steven Rostedt committed
223
/*
224
225
226
 * 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.
227
228
229
230
231
 *
 * 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
232
 */
233
#define TRACE_BUF_SIZE_DEFAULT	1441792UL /* 16384 * 88 (sizeof(entry)) */
234

235
static unsigned long		trace_buf_size = TRACE_BUF_SIZE_DEFAULT;
236

Steven Rostedt's avatar
Steven Rostedt committed
237
/* trace_types holds a link list of available tracers. */
238
static struct tracer		*trace_types __read_mostly;
Steven Rostedt's avatar
Steven Rostedt committed
239
240
241
242

/*
 * trace_types_lock is used to protect the trace_types list.
 */
243
static DEFINE_MUTEX(trace_types_lock);
Steven Rostedt's avatar
Steven Rostedt committed
244

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
/*
 * 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)
{
273
	if (cpu == RING_BUFFER_ALL_CPUS) {
274
275
276
277
278
		/* gain it for accessing the whole ring buffer. */
		down_write(&all_cpu_access_lock);
	} else {
		/* gain it for accessing a cpu ring buffer. */

279
		/* Firstly block other trace_access_lock(RING_BUFFER_ALL_CPUS). */
280
281
282
283
284
285
286
287
288
		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)
{
289
	if (cpu == RING_BUFFER_ALL_CPUS) {
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
		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

327
/* trace_flags holds trace_options default values */
328
unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK |
329
	TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | TRACE_ITER_SLEEP_TIME |
330
	TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE |
331
	TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS;
332

333
334
335
336
337
338
339
340
/**
 * tracing_on - enable tracing buffers
 *
 * This function enables tracing buffers that may have been
 * disabled with tracing_off.
 */
void tracing_on(void)
{
341
342
	if (global_trace.trace_buffer.buffer)
		ring_buffer_record_on(global_trace.trace_buffer.buffer);
343
344
345
346
347
348
349
350
351
352
	/*
	 * 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);

353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
/**
 * __trace_puts - write a constant string into the trace buffer.
 * @ip:	   The address of the caller
 * @str:   The constant string to write
 * @size:  The size of the string.
 */
int __trace_puts(unsigned long ip, const char *str, int size)
{
	struct ring_buffer_event *event;
	struct ring_buffer *buffer;
	struct print_entry *entry;
	unsigned long irq_flags;
	int alloc;

	alloc = sizeof(*entry) + size + 2; /* possible \n added */

	local_save_flags(irq_flags);
	buffer = global_trace.trace_buffer.buffer;
	event = trace_buffer_lock_reserve(buffer, TRACE_PRINT, alloc, 
					  irq_flags, preempt_count());
	if (!event)
		return 0;

	entry = ring_buffer_event_data(event);
	entry->ip = ip;

	memcpy(&entry->buf, str, size);

	/* Add a newline if necessary */
	if (entry->buf[size - 1] != '\n') {
		entry->buf[size] = '\n';
		entry->buf[size + 1] = '\0';
	} else
		entry->buf[size] = '\0';

	__buffer_unlock_commit(buffer, event);

	return size;
}
EXPORT_SYMBOL_GPL(__trace_puts);

/**
 * __trace_bputs - write the pointer to a constant string into trace buffer
 * @ip:	   The address of the caller
 * @str:   The constant string to write to the buffer to
 */
int __trace_bputs(unsigned long ip, const char *str)
{
	struct ring_buffer_event *event;
	struct ring_buffer *buffer;
	struct bputs_entry *entry;
	unsigned long irq_flags;
	int size = sizeof(struct bputs_entry);

	local_save_flags(irq_flags);
	buffer = global_trace.trace_buffer.buffer;
	event = trace_buffer_lock_reserve(buffer, TRACE_BPUTS, size,
					  irq_flags, preempt_count());
	if (!event)
		return 0;

	entry = ring_buffer_event_data(event);
	entry->ip			= ip;
	entry->str			= str;

	__buffer_unlock_commit(buffer, event);

	return 1;
}
EXPORT_SYMBOL_GPL(__trace_bputs);

424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
#ifdef CONFIG_TRACER_SNAPSHOT
/**
 * trace_snapshot - take a snapshot of the current buffer.
 *
 * This causes a swap between the snapshot buffer and the current live
 * tracing buffer. You can use this to take snapshots of the live
 * trace when some condition is triggered, but continue to trace.
 *
 * Note, make sure to allocate the snapshot with either
 * a tracing_snapshot_alloc(), or by doing it manually
 * with: echo 1 > /sys/kernel/debug/tracing/snapshot
 *
 * If the snapshot buffer is not allocated, it will stop tracing.
 * Basically making a permanent snapshot.
 */
void tracing_snapshot(void)
{
	struct trace_array *tr = &global_trace;
	struct tracer *tracer = tr->current_trace;
	unsigned long flags;

	if (!tr->allocated_snapshot) {
446
447
		internal_trace_puts("*** SNAPSHOT NOT ALLOCATED ***\n");
		internal_trace_puts("*** stopping trace here!   ***\n");
448
449
450
451
452
453
		tracing_off();
		return;
	}

	/* Note, snapshot can not be used when the tracer uses it */
	if (tracer->use_max_tr) {
454
455
		internal_trace_puts("*** LATENCY TRACER ACTIVE ***\n");
		internal_trace_puts("*** Can not use snapshot (sorry) ***\n");
456
457
458
459
460
461
462
463
464
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
		return;
	}

	local_irq_save(flags);
	update_max_tr(tr, current, smp_processor_id());
	local_irq_restore(flags);
}

static int resize_buffer_duplicate_size(struct trace_buffer *trace_buf,
					struct trace_buffer *size_buf, int cpu_id);

/**
 * trace_snapshot_alloc - allocate and take a snapshot of the current buffer.
 *
 * This is similar to trace_snapshot(), but it will allocate the
 * snapshot buffer if it isn't already allocated. Use this only
 * where it is safe to sleep, as the allocation may sleep.
 *
 * This causes a swap between the snapshot buffer and the current live
 * tracing buffer. You can use this to take snapshots of the live
 * trace when some condition is triggered, but continue to trace.
 */
void tracing_snapshot_alloc(void)
{
	struct trace_array *tr = &global_trace;
	int ret;

	if (!tr->allocated_snapshot) {

		/* allocate spare buffer */
		ret = resize_buffer_duplicate_size(&tr->max_buffer,
				   &tr->trace_buffer, RING_BUFFER_ALL_CPUS);
		if (WARN_ON(ret < 0))
			return;

		tr->allocated_snapshot = true;
	}

	tracing_snapshot();
}
#else
void tracing_snapshot(void)
{
	WARN_ONCE(1, "Snapshot feature not enabled, but internal snapshot used");
}
void tracing_snapshot_alloc(void)
{
	/* Give warning */
	tracing_snapshot();
}
#endif /* CONFIG_TRACER_SNAPSHOT */

508
509
510
511
512
513
514
515
516
517
/**
 * 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)
{
518
519
	if (global_trace.trace_buffer.buffer)
		ring_buffer_record_off(global_trace.trace_buffer.buffer);
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
	/*
	 * 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)
{
535
536
	if (global_trace.trace_buffer.buffer)
		return ring_buffer_record_is_on(global_trace.trace_buffer.buffer);
537
538
539
540
	return !global_trace.buffer_disabled;
}
EXPORT_SYMBOL_GPL(tracing_is_on);

541
static int __init set_buf_size(char *str)
542
{
543
	unsigned long buf_size;
544

545
546
	if (!str)
		return 0;
547
	buf_size = memparse(str, &str);
548
	/* nr_entries can not be zero */
549
	if (buf_size == 0)
550
		return 0;
551
	trace_buf_size = buf_size;
552
553
	return 1;
}
554
__setup("trace_buf_size=", set_buf_size);
555

556
557
static int __init set_tracing_thresh(char *str)
{
558
	unsigned long threshold;
559
560
561
562
	int ret;

	if (!str)
		return 0;
563
	ret = kstrtoul(str, 0, &threshold);
564
565
	if (ret < 0)
		return 0;
566
	tracing_thresh = threshold * 1000;
567
568
569
570
	return 1;
}
__setup("tracing_thresh=", set_tracing_thresh);

Steven Rostedt's avatar
Steven Rostedt committed
571
572
573
574
575
unsigned long nsecs_to_usecs(unsigned long nsecs)
{
	return nsecs / 1000;
}

Steven Rostedt's avatar
Steven Rostedt committed
576
/* These must match the bit postions in trace_iterator_flags */
577
578
579
580
581
static const char *trace_options[] = {
	"print-parent",
	"sym-offset",
	"sym-addr",
	"verbose",
Ingo Molnar's avatar
Ingo Molnar committed
582
	"raw",
583
	"hex",
Ingo Molnar's avatar
Ingo Molnar committed
584
	"bin",
585
	"block",
Ingo Molnar's avatar
Ingo Molnar committed
586
	"stacktrace",
587
	"trace_printk",
588
	"ftrace_preempt",
589
	"branch",
590
	"annotate",
591
	"userstacktrace",
592
	"sym-userobj",
593
	"printk-msg-only",
594
	"context-info",
595
	"latency-format",
596
	"sleep-time",
597
	"graph-time",
598
	"record-cmd",
599
	"overwrite",
600
	"disable_on_free",
601
	"irq-info",
602
	"markers",
603
604
605
	NULL
};

606
607
608
static struct {
	u64 (*func)(void);
	const char *name;
609
	int in_ns;		/* is this clock in nanoseconds? */
610
} trace_clocks[] = {
611
612
613
	{ trace_clock_local,	"local",	1 },
	{ trace_clock_global,	"global",	1 },
	{ trace_clock_counter,	"counter",	0 },
614
	ARCH_TRACE_CLOCKS
615
616
617
618
};

int trace_clock_id;

619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
/*
 * 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)) {
696
		if (parser->idx < parser->size - 1)
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
			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;
}

725
726
727
728
729
ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt)
{
	int len;
	int ret;

730
731
732
	if (!cnt)
		return 0;

733
734
735
736
737
738
739
	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);
740
	if (ret == cnt)
741
742
		return -EFAULT;

743
744
	cnt -= ret;

745
	s->readpos += cnt;
746
	return cnt;
747
748
}

749
static ssize_t trace_seq_to_buffer(struct trace_seq *s, void *buf, size_t cnt)
750
751
752
753
754
755
756
757
758
{
	int len;

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

	len = s->len - s->readpos;
	if (cnt > len)
		cnt = len;
759
	memcpy(buf, s->buffer + s->readpos, cnt);
760

761
	s->readpos += cnt;
762
763
764
	return cnt;
}

765
766
767
768
769
770
/*
 * 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.
 *
771
 * This is defined as a arch_spinlock_t in order to help
772
773
774
775
776
777
 * 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.
 */
778
static arch_spinlock_t ftrace_max_lock =
779
	(arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
780

781
782
unsigned long __read_mostly	tracing_thresh;

783
784
785
786
787
788
789
790
791
792
793
#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)
{
794
795
796
797
	struct trace_buffer *trace_buf = &tr->trace_buffer;
	struct trace_buffer *max_buf = &tr->max_buffer;
	struct trace_array_cpu *data = per_cpu_ptr(trace_buf->data, cpu);
	struct trace_array_cpu *max_data = per_cpu_ptr(max_buf->data, cpu);
798

799
800
	max_buf->cpu = cpu;
	max_buf->time_start = data->preempt_timestamp;
801

802
803
804
	max_data->saved_latency = tracing_max_latency;
	max_data->critical_start = data->critical_start;
	max_data->critical_end = data->critical_end;
805

806
	memcpy(max_data->comm, tsk->comm, TASK_COMM_LEN);
807
808
809
810
811
	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;
812
813
814
815
816

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

Steven Rostedt's avatar
Steven Rostedt committed
817
818
819
820
821
822
823
824
825
/**
 * 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
826
void
827
828
update_max_tr(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
829
	struct ring_buffer *buf;
830

831
	if (tr->stop_count)
832
833
		return;

834
	WARN_ON_ONCE(!irqs_disabled());
835

836
	if (!tr->allocated_snapshot) {
837
		/* Only the nop tracer should hit this when disabling */
838
		WARN_ON_ONCE(tr->current_trace != &nop_trace);
839
		return;
840
	}
841

842
	arch_spin_lock(&ftrace_max_lock);
843

844
845
846
	buf = tr->trace_buffer.buffer;
	tr->trace_buffer.buffer = tr->max_buffer.buffer;
	tr->max_buffer.buffer = buf;
847

848
	__update_max_tr(tr, tsk, cpu);
849
	arch_spin_unlock(&ftrace_max_lock);
850
851
852
853
854
855
856
}

/**
 * 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
857
858
 *
 * Flip the trace of a single CPU buffer between the @tr and the max_tr.
859
 */
Ingo Molnar's avatar
Ingo Molnar committed
860
void
861
862
update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
{
863
	int ret;
864

865
	if (tr->stop_count)
866
867
		return;

868
	WARN_ON_ONCE(!irqs_disabled());
869
	if (WARN_ON_ONCE(!tr->allocated_snapshot))
870
871
		return;

872
	arch_spin_lock(&ftrace_max_lock);
873

874
	ret = ring_buffer_swap_cpu(tr->max_buffer.buffer, tr->trace_buffer.buffer, cpu);
875

876
877
878
879
880
881
882
	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.
		 */
883
		trace_array_printk_buf(tr->max_buffer.buffer, _THIS_IP_,
884
885
886
887
			"Failed to swap buffers due to commit in progress\n");
	}

	WARN_ON_ONCE(ret && ret != -EAGAIN && ret != -EBUSY);
888
889

	__update_max_tr(tr, tsk, cpu);
890
	arch_spin_unlock(&ftrace_max_lock);
891
}
892
#endif /* CONFIG_TRACER_MAX_TRACE */
893

894
895
static void default_wait_pipe(struct trace_iterator *iter)
{
896
897
898
	/* Iterators are static, they should be filled or empty */
	if (trace_buffer_iter(iter, iter->cpu_file))
		return;
899

900
	ring_buffer_wait(iter->trace_buffer->buffer, iter->cpu_file);
901
902
}

903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
#ifdef CONFIG_FTRACE_STARTUP_TEST
static int run_tracer_selftest(struct tracer *type)
{
	struct trace_array *tr = &global_trace;
	struct tracer *saved_tracer = tr->current_trace;
	int ret;

	if (!type->selftest || tracing_selftest_disabled)
		return 0;

	/*
	 * 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.
	 */
	tracing_reset_online_cpus(&tr->trace_buffer);

	tr->current_trace = type;

#ifdef CONFIG_TRACER_MAX_TRACE
	if (type->use_max_tr) {
		/* If we expanded the buffers, make sure the max is expanded too */
		if (ring_buffer_expanded)
			ring_buffer_resize(tr->max_buffer.buffer, trace_buf_size,
					   RING_BUFFER_ALL_CPUS);
		tr->allocated_snapshot = true;
	}
#endif

	/* 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 */
	tr->current_trace = saved_tracer;
	if (ret) {
		printk(KERN_CONT "FAILED!\n");
		/* Add the warning after printing 'FAILED' */
		WARN_ON(1);
		return -1;
	}
	/* Only reset on passing, to avoid touching corrupted buffers */
	tracing_reset_online_cpus(&tr->trace_buffer);

#ifdef CONFIG_TRACER_MAX_TRACE
	if (type->use_max_tr) {
		tr->allocated_snapshot = false;

		/* Shrink the max buffer again */
		if (ring_buffer_expanded)
			ring_buffer_resize(tr->max_buffer.buffer, 1,
					   RING_BUFFER_ALL_CPUS);
	}
#endif

	printk(KERN_CONT "PASSED\n");
	return 0;
}
#else
static inline int run_tracer_selftest(struct tracer *type)
{
	return 0;
}
#endif /* CONFIG_FTRACE_STARTUP_TEST */

Steven Rostedt's avatar
Steven Rostedt committed
969
970
971
972
973
974
/**
 * register_tracer - register a tracer with the ftrace system.
 * @type - the plugin for the tracer
 *
 * Register a new plugin tracer.
 */
975
976
977
978
979
980
981
982
983
984
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;
	}

985
	if (strlen(type->name) >= MAX_TRACER_SIZE) {
Li Zefan's avatar
Li Zefan committed
986
987
988
989
		pr_info("Tracer has a name longer than %d\n", MAX_TRACER_SIZE);
		return -1;
	}

990
	mutex_lock(&trace_types_lock);
Ingo Molnar's avatar
Ingo Molnar committed
991

992
993
	tracing_selftest_running = true;

994
995
996
	for (t = trace_types; t; t = t->next) {
		if (strcmp(type->name, t->name) == 0) {
			/* already found */
Li Zefan's avatar
Li Zefan committed
997
			pr_info("Tracer %s already registered\n",
998
999
1000
1001
1002
1003
				type->name);
			ret = -1;
			goto out;
		}
	}

1004
1005
1006
1007
1008
1009
1010
	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;
1011
1012
1013
	if (!type->wait_pipe)
		type->wait_pipe = default_wait_pipe;

1014
1015
1016
	ret = run_tracer_selftest(type);
	if (ret < 0)
		goto out;
Steven Rostedt's avatar
Steven Rostedt committed
1017

1018
1019
	type->next = trace_types;
	trace_types = type;
Steven Rostedt's avatar
Steven Rostedt committed
1020

1021
 out:
1022
	tracing_selftest_running = false;
1023
1024
	mutex_unlock(&trace_types_lock);

Steven Rostedt's avatar
Steven Rostedt committed
1025
1026
1027
	if (ret || !default_bootup_tracer)
		goto out_unlock;

Li Zefan's avatar
Li Zefan committed
1028
	if (strncmp(default_bootup_tracer, type->name, MAX_TRACER_SIZE))
Steven Rostedt's avatar
Steven Rostedt committed
1029
1030
1031
1032
1033
1034
1035
		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. */
1036
	tracing_selftest_disabled = true;
1037
#ifdef CONFIG_FTRACE_STARTUP_TEST
Steven Rostedt's avatar
Steven Rostedt committed
1038
1039
	printk(KERN_INFO "Disabling FTRACE selftests due to running tracer '%s'\n",
	       type->name);
1040
1041
#endif

Steven Rostedt's avatar
Steven Rostedt committed
1042
 out_unlock:
1043
1044
1045
	return ret;
}

1046
void tracing_reset(struct trace_buffer *buf, int cpu)
1047
{
1048
	struct ring_buffer *buffer = buf->buffer;
1049

1050
1051
1052
	if (!buffer)
		return;

1053
1054
1055
1056
	ring_buffer_record_disable(buffer);

	/* Make sure all commits have finished */
	synchronize_sched();
1057
	ring_buffer_reset_cpu(buffer, cpu);
1058
1059
1060
1061

	ring_buffer_record_enable(buffer);
}

1062
void tracing_reset_online_cpus(struct trace_buffer *buf)
1063
{
1064
	struct ring_buffer *buffer = buf->buffer;
1065
1066
	int cpu;

1067
1068
1069
	if (!buffer)
		return;

1070
1071
1072
1073
1074
	ring_buffer_record_disable(buffer);

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

1075
	buf->time_start = ftrace_now(buf->cpu);
1076
1077

	for_each_online_cpu(cpu)
1078
		ring_buffer_reset_cpu(buffer, cpu);
1079
1080

	ring_buffer_record_enable(buffer);
1081
1082
}

1083
1084
void tracing_reset_current(int cpu)
{
1085
	tracing_reset(&global_trace.trace_buffer, cpu);
1086
1087
}

1088
void tracing_reset_all_online_cpus(void)
1089
{
1090
1091
1092
1093
	struct trace_array *tr;

	mutex_lock(&trace_types_lock);
	list_for_each_entry(tr, &ftrace_trace_arrays, list) {
1094
1095
1096
1097
		tracing_reset_online_cpus(&tr->trace_buffer);
#ifdef CONFIG_TRACER_MAX_TRACE
		tracing_reset_online_cpus(&tr->max_buffer);
#endif
1098
1099
	}
	mutex_unlock(&trace_types_lock);
1100
1101
}

1102
#define SAVED_CMDLINES 128
1103
#define NO_CMDLINE_MAP UINT_MAX
1104
1105
1106
1107
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;
1108
static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
1109
1110

/* temporary disable recording */
1111
static atomic_t trace_record_cmdline_disabled __read_mostly;
1112
1113
1114

static void trace_init_cmdlines(void)
{
1115
1116
	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));
1117
1118
1119
	cmdline_idx = 0;
}

1120
1121
int is_tracing_stopped(void)
{
1122
	return global_trace.stop_count;
1123
1124
}

1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
/**
 * 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();
}

1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
/**
 * 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;

1154
1155
1156
	raw_spin_lock_irqsave(&global_trace.start_lock, flags);
	if (--global_trace.stop_count) {
		if (global_trace.stop_count < 0) {
1157
1158
			/* Someone screwed up their debugging */
			WARN_ON_ONCE(1);
1159
			global_trace.stop_count = 0;
1160
		}
1161
1162
1163
		goto out;
	}

1164
1165
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);
1166

1167
	buffer = global_trace.trace_buffer.buffer;
1168
1169
1170
	if (buffer)
		ring_buffer_record_enable(buffer);

1171
1172
#ifdef CONFIG_TRACER_MAX_TRACE
	buffer = global_trace.max_buffer.buffer;
1173
1174
	if (buffer)
		ring_buffer_record_enable(buffer);
1175
#endif
1176

1177
1178
	arch_spin_unlock(&ftrace_max_lock);

1179
1180
	ftrace_start();
 out:
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
	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;
	}

1207
	buffer = tr->trace_buffer.buffer;
1208
1209
1210
1211
1212
	if (buffer)
		ring_buffer_record_enable(buffer);

 out:
	raw_spin_unlock_irqrestore(&tr->start_lock, flags);
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
}

/**
 * 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();
1227
1228
	raw_spin_lock_irqsave(&global_trace.start_lock, flags);
	if (global_trace.stop_count++)
1229
1230
		goto out;

1231
1232
1233
	/* Prevent the buffers from switching */
	arch_spin_lock(&ftrace_max_lock);

1234
	buffer = global_trace.trace_buffer.buffer;
1235
1236
1237
	if (buffer)
		ring_buffer_record_disable(buffer);

1238
1239
#ifdef CONFIG_TRACER_MAX_TRACE
	buffer = global_trace.max_buffer.buffer;
1240
1241
	if (buffer)
		ring_buffer_record_disable(buffer);
1242
#endif
1243

1244
1245
	arch_spin_unlock(&ftrace_max_lock);

1246
 out:
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
	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;

1263
	buffer = tr->trace_buffer.buffer;
1264
1265
1266
1267
1268
	if (buffer)
		ring_buffer_record_disable(buffer);

 out:
	raw_spin_unlock_irqrestore(&tr->start_lock, flags);
1269
1270
}

Ingo Molnar's avatar
Ingo Molnar committed
1271
void trace_stop_cmdline_recording(void);
1272

Ingo Molnar's avatar
Ingo Molnar committed
1273
static void trace_save_cmdline(struct task_struct *tsk)
1274
{
1275
	unsigned pid, idx;
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285

	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.
	 */
1286
	if (!arch_spin_trylock(&trace_cmdline_lock))
1287
1288
1289
		return;

	idx = map_pid_to_cmdline[tsk->pid];
1290
	if (idx == NO_CMDLINE_MAP) {
1291
1292
		idx = (cmdline_idx + 1) % SAVED_CMDLINES;

1293
1294
1295
1296
1297
1298
1299
1300
1301
		/*
		 * 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;
1302

1303
		map_cmdline_to_pid[idx] = tsk->pid;
1304
1305
1306
1307
1308
1309
1310
		map_pid_to_cmdline[tsk->pid] = idx;

		cmdline_idx = idx;
	}

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

1311
	arch_spin_unlock(&trace_cmdline_lock);
1312
1313
}

1314
void trace_find_cmdline(int pid, char comm[])
1315
1316
1317
{
	unsigned map;

1318
1319
1320
1321
	if (!pid) {
		strcpy(comm, "<idle>");
		return;
	}
1322

1323
1324
1325
1326
1327
	if (WARN_ON_ONCE(pid < 0)) {
		strcpy(comm, "<XXX>");
		return;
	}

1328
1329
1330
1331
	if (pid > PID_MAX_DEFAULT) {
		strcpy(comm, "<...>");
		return;
	}
1332

1333
	preempt_disable();
1334
	arch_spin_lock(&trace_cmdline_lock);
1335
	map = map_pid_to_cmdline[pid];