aboutsummaryrefslogtreecommitdiff
path: root/kernel/trace/trace_sched_wakeup.c
blob: 5d2fb48e47f8194576a5bf87c9fc0600242f89a9 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
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
341
342
343
344
345
346
347
348
349
350
351
352
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
/*
 * trace task wakeup timings
 *
 * Copyright (C) 2007-2008 Steven Rostedt <srostedt@redhat.com>
 * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com>
 *
 * Based on code from the latency_tracer, that is:
 *
 *  Copyright (C) 2004-2006 Ingo Molnar
 *  Copyright (C) 2004 William Lee Irwin III
 */
#include <linux/module.h>
#include <linux/fs.h>
#include <linux/debugfs.h>
#include <linux/kallsyms.h>
#include <linux/uaccess.h>
#include <linux/ftrace.h>
#include <linux/marker.h>

#include "trace.h"

static struct trace_array	*wakeup_trace;
static int __read_mostly	tracer_enabled;

static struct task_struct	*wakeup_task;
static int			wakeup_cpu;
static unsigned			wakeup_prio = -1;

static DEFINE_SPINLOCK(wakeup_lock);

static void __wakeup_reset(struct trace_array *tr);

/*
 * Should this new latency be reported/recorded?
 */
static int report_latency(cycle_t delta)
{
	if (tracing_thresh) {
		if (delta < tracing_thresh)
			return 0;
	} else {
		if (delta <= tracing_max_latency)
			return 0;
	}
	return 1;
}

static void notrace
wakeup_sched_switch(void *private, void *rq, struct task_struct *prev,
	struct task_struct *next)
{
	unsigned long latency = 0, t0 = 0, t1 = 0;
	struct trace_array **ptr = private;
	struct trace_array *tr = *ptr;
	struct trace_array_cpu *data;
	cycle_t T0, T1, delta;
	unsigned long flags;
	long disabled;
	int cpu;

	if (unlikely(!tracer_enabled))
		return;

	/*
	 * When we start a new trace, we set wakeup_task to NULL
	 * and then set tracer_enabled = 1. We want to make sure
	 * that another CPU does not see the tracer_enabled = 1
	 * and the wakeup_task with an older task, that might
	 * actually be the same as next.
	 */
	smp_rmb();

	if (next != wakeup_task)
		return;

	/* The task we are waitng for is waking up */
	data = tr->data[wakeup_cpu];

	/* disable local data, not wakeup_cpu data */
	cpu = raw_smp_processor_id();
	disabled = atomic_inc_return(&tr->data[cpu]->disabled);
	if (likely(disabled != 1))
		goto out;

	spin_lock_irqsave(&wakeup_lock, flags);

	/* We could race with grabbing wakeup_lock */
	if (unlikely(!tracer_enabled || next != wakeup_task))
		goto out_unlock;

	trace_function(tr, data, CALLER_ADDR1, CALLER_ADDR2, flags);

	/*
	 * usecs conversion is slow so we try to delay the conversion
	 * as long as possible:
	 */
	T0 = data->preempt_timestamp;
	T1 = ftrace_now(cpu);
	delta = T1-T0;

	if (!report_latency(delta))
		goto out_unlock;

	latency = nsecs_to_usecs(delta);

	tracing_max_latency = delta;
	t0 = nsecs_to_usecs(T0);
	t1 = nsecs_to_usecs(T1);

	update_max_tr(tr, wakeup_task, wakeup_cpu);

out_unlock:
	__wakeup_reset(tr);
	spin_unlock_irqrestore(&wakeup_lock, flags);
out:
	atomic_dec(&tr->data[cpu]->disabled);
}

static notrace void
sched_switch_callback(void *probe_data, void *call_data,
		      const char *format, va_list *args)
{
	struct task_struct *prev;
	struct task_struct *next;
	struct rq *__rq;

	/* skip prev_pid %d next_pid %d prev_state %ld */
	(void)va_arg(*args, int);
	(void)va_arg(*args, int);
	(void)va_arg(*args, long);
	__rq = va_arg(*args, typeof(__rq));
	prev = va_arg(*args, typeof(prev));
	next = va_arg(*args, typeof(next));

	tracing_record_cmdline(prev);

	/*
	 * If tracer_switch_func only points to the local
	 * switch func, it still needs the ptr passed to it.
	 */
	wakeup_sched_switch(probe_data, __rq, prev, next);
}

static void __wakeup_reset(struct trace_array *tr)
{
	struct trace_array_cpu *data;
	int cpu;

	assert_spin_locked(&wakeup_lock);

	for_each_possible_cpu(cpu) {
		data = tr->data[cpu];
		tracing_reset(data);
	}

	wakeup_cpu = -1;
	wakeup_prio = -1;

	if (wakeup_task)
		put_task_struct(wakeup_task);

	wakeup_task = NULL;
}

static void wakeup_reset(struct trace_array *tr)
{
	unsigned long flags;

	spin_lock_irqsave(&wakeup_lock, flags);
	__wakeup_reset(tr);
	spin_unlock_irqrestore(&wakeup_lock, flags);
}

static void
wakeup_check_start(struct trace_array *tr, struct task_struct *p,
		   struct task_struct *curr)
{
	int cpu = smp_processor_id();
	unsigned long flags;
	long disabled;

	if (likely(!rt_task(p)) ||
			p->prio >= wakeup_prio ||
			p->prio >= curr->prio)
		return;

	disabled = atomic_inc_return(&tr->data[cpu]->disabled);
	if (unlikely(disabled != 1))
		goto out;

	/* interrupts should be off from try_to_wake_up */
	spin_lock(&wakeup_lock);

	/* check for races. */
	if (!tracer_enabled || p->prio >= wakeup_prio)
		goto out_locked;

	/* reset the trace */
	__wakeup_reset(tr);

	wakeup_cpu = task_cpu(p);
	wakeup_prio = p->prio;

	wakeup_task = p;
	get_task_struct(wakeup_task);

	local_save_flags(flags);

	tr->data[wakeup_cpu]->preempt_timestamp = ftrace_now(cpu);
	trace_function(tr, tr->data[wakeup_cpu],
		       CALLER_ADDR1, CALLER_ADDR2, flags);

out_locked:
	spin_unlock(&wakeup_lock);
out:
	atomic_dec(&tr->data[cpu]->disabled);
}

static notrace void
wake_up_callback(void *probe_data, void *call_data,
		 const char *format, va_list *args)
{
	struct trace_array **ptr = probe_data;
	struct trace_array *tr = *ptr;
	struct task_struct *curr;
	struct task_struct *task;
	struct rq *__rq;

	if (likely(!tracer_enabled))
		return;

	/* Skip pid %d state %ld */
	(void)va_arg(*args, int);
	(void)va_arg(*args, long);
	/* now get the meat: "rq %p task %p rq->curr %p" */
	__rq = va_arg(*args, typeof(__rq));
	task = va_arg(*args, typeof(task));
	curr = va_arg(*args, typeof(curr));

	tracing_record_cmdline(task);
	tracing_record_cmdline(curr);

	wakeup_check_start(tr, task, curr);
}

static void start_wakeup_tracer(struct trace_array *tr)
{
	int ret;

	ret = marker_probe_register("kernel_sched_wakeup",
			"pid %d state %ld ## rq %p task %p rq->curr %p",
			wake_up_callback,
			&wakeup_trace);
	if (ret) {
		pr_info("wakeup trace: Couldn't add marker"
			" probe to kernel_sched_wakeup\n");
		return;
	}

	ret = marker_probe_register("kernel_sched_wakeup_new",
			"pid %d state %ld ## rq %p task %p rq->curr %p",
			wake_up_callback,
			&wakeup_trace);
	if (ret) {
		pr_info("wakeup trace: Couldn't add marker"
			" probe to kernel_sched_wakeup_new\n");
		goto fail_deprobe;
	}

	ret = marker_probe_register("kernel_sched_schedule",
		"prev_pid %d next_pid %d prev_state %ld "
		"## rq %p prev %p next %p",
		sched_switch_callback,
		&wakeup_trace);
	if (ret) {
		pr_info("sched trace: Couldn't add marker"
			" probe to kernel_sched_schedule\n");
		goto fail_deprobe_wake_new;
	}

	wakeup_reset(tr);

	/*
	 * Don't let the tracer_enabled = 1 show up before
	 * the wakeup_task is reset. This may be overkill since
	 * wakeup_reset does a spin_unlock after setting the
	 * wakeup_task to NULL, but I want to be safe.
	 * This is a slow path anyway.
	 */
	smp_wmb();

	tracer_enabled = 1;

	return;
fail_deprobe_wake_new:
	marker_probe_unregister("kernel_sched_wakeup_new",
				wake_up_callback,
				&wakeup_trace);
fail_deprobe:
	marker_probe_unregister("kernel_sched_wakeup",
				wake_up_callback,
				&wakeup_trace);
}

static void stop_wakeup_tracer(struct trace_array *tr)
{
	tracer_enabled = 0;
	marker_probe_unregister("kernel_sched_schedule",
				sched_switch_callback,
				&wakeup_trace);
	marker_probe_unregister("kernel_sched_wakeup_new",
				wake_up_callback,
				&wakeup_trace);
	marker_probe_unregister("kernel_sched_wakeup",
				wake_up_callback,
				&wakeup_trace);
}

static void wakeup_tracer_init(struct trace_array *tr)
{
	wakeup_trace = tr;

	if (tr->ctrl)
		start_wakeup_tracer(tr);
}

static void wakeup_tracer_reset(struct trace_array *tr)
{
	if (tr->ctrl) {
		stop_wakeup_tracer(tr);
		/* make sure we put back any tasks we are tracing */
		wakeup_reset(tr);
	}
}

static void wakeup_tracer_ctrl_update(struct trace_array *tr)
{
	if (tr->ctrl)
		start_wakeup_tracer(tr);
	else
		stop_wakeup_tracer(tr);
}

static void wakeup_tracer_open(struct trace_iterator *iter)
{
	/* stop the trace while dumping */
	if (iter->tr->ctrl)
		stop_wakeup_tracer(iter->tr);
}

static void wakeup_tracer_close(struct trace_iterator *iter)
{
	/* forget about any processes we were recording */
	if (iter->tr->ctrl)
		start_wakeup_tracer(iter->tr);
}

static struct tracer wakeup_tracer __read_mostly =
{
	.name		= "wakeup",
	.init		= wakeup_tracer_init,
	.reset		= wakeup_tracer_reset,
	.open		= wakeup_tracer_open,
	.close		= wakeup_tracer_close,
	.ctrl_update	= wakeup_tracer_ctrl_update,
	.print_max	= 1,
#ifdef CONFIG_FTRACE_SELFTEST
	.selftest    = trace_selftest_startup_wakeup,
#endif
};

__init static int init_wakeup_tracer(void)
{
	int ret;

	ret = register_tracer(&wakeup_tracer);
	if (ret)
		return ret;

	return 0;
}
device_initcall(init_wakeup_tracer);