aboutsummaryrefslogtreecommitdiff
path: root/kernel/trace/trace_sysprof.c
blob: 63528086337c0145b6e59e4a77c9e80b40686408 (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
/*
 * trace stack traces
 *
 * Copyright (C) 2004-2008, Soeren Sandmann
 * Copyright (C) 2007 Steven Rostedt <srostedt@redhat.com>
 * Copyright (C) 2008 Ingo Molnar <mingo@redhat.com>
 */
#include <linux/kallsyms.h>
#include <linux/debugfs.h>
#include <linux/hrtimer.h>
#include <linux/uaccess.h>
#include <linux/ftrace.h>
#include <linux/module.h>
#include <linux/irq.h>
#include <linux/fs.h>

#include <asm/stacktrace.h>

#include "trace.h"

static struct trace_array	*sysprof_trace;
static int __read_mostly	tracer_enabled;

/*
 * 1 msec sample interval by default:
 */
static unsigned long sample_period = 1000000;
static const unsigned int sample_max_depth = 512;

static DEFINE_MUTEX(sample_timer_lock);
/*
 * Per CPU hrtimers that do the profiling:
 */
static DEFINE_PER_CPU(struct hrtimer, stack_trace_hrtimer);

struct stack_frame {
	const void __user	*next_fp;
	unsigned long		return_address;
};

static int copy_stack_frame(const void __user *fp, struct stack_frame *frame)
{
	int ret;

	if (!access_ok(VERIFY_READ, fp, sizeof(*frame)))
		return 0;

	ret = 1;
	pagefault_disable();
	if (__copy_from_user_inatomic(frame, fp, sizeof(*frame)))
		ret = 0;
	pagefault_enable();

	return ret;
}

struct backtrace_info {
	struct trace_array_cpu	*data;
	struct trace_array	*tr;
	int			pos;
};

static void
backtrace_warning_symbol(void *data, char *msg, unsigned long symbol)
{
	/* Ignore warnings */
}

static void backtrace_warning(void *data, char *msg)
{
	/* Ignore warnings */
}

static int backtrace_stack(void *data, char *name)
{
	/* Don't bother with IRQ stacks for now */
	return -1;
}

static void backtrace_address(void *data, unsigned long addr, int reliable)
{
	struct backtrace_info *info = data;

	if (info->pos < sample_max_depth && reliable) {
		__trace_special(info->tr, info->data, 1, addr, 0);

		info->pos++;
	}
}

const static struct stacktrace_ops backtrace_ops = {
	.warning		= backtrace_warning,
	.warning_symbol		= backtrace_warning_symbol,
	.stack			= backtrace_stack,
	.address		= backtrace_address,
};

static int
trace_kernel(struct pt_regs *regs, struct trace_array *tr,
	     struct trace_array_cpu *data)
{
	struct backtrace_info info;
	unsigned long bp;
	char *stack;

	info.tr = tr;
	info.data = data;
	info.pos = 1;

	__trace_special(info.tr, info.data, 1, regs->ip, 0);

	stack = ((char *)regs + sizeof(struct pt_regs));
#ifdef CONFIG_FRAME_POINTER
	bp = regs->bp;
#else
	bp = 0;
#endif

	dump_trace(NULL, regs, (void *)stack, bp, &backtrace_ops, &info);

	return info.pos;
}

static void timer_notify(struct pt_regs *regs, int cpu)
{
	struct trace_array_cpu *data;
	struct stack_frame frame;
	struct trace_array *tr;
	const void __user *fp;
	int is_user;
	int i;

	if (!regs)
		return;

	tr = sysprof_trace;
	data = tr->data[cpu];
	is_user = user_mode(regs);

	if (!current || current->pid == 0)
		return;

	if (is_user && current->state != TASK_RUNNING)
		return;

	__trace_special(tr, data, 0, 0, current->pid);

	if (!is_user)
		i = trace_kernel(regs, tr, data);
	else
		i = 0;

	/*
	 * Trace user stack if we are not a kernel thread
	 */
	if (current->mm && i < sample_max_depth) {
		regs = (struct pt_regs *)current->thread.sp0 - 1;

		fp = (void __user *)regs->bp;

		__trace_special(tr, data, 2, regs->ip, 0);

		while (i < sample_max_depth) {
			frame.next_fp = 0;
			frame.return_address = 0;
			if (!copy_stack_frame(fp, &frame))
				break;
			if ((unsigned long)fp < regs->sp)
				break;

			__trace_special(tr, data, 2, frame.return_address,
					(unsigned long)fp);
			fp = frame.next_fp;

			i++;
		}

	}

	/*
	 * Special trace entry if we overflow the max depth:
	 */
	if (i == sample_max_depth)
		__trace_special(tr, data, -1, -1, -1);

	__trace_special(tr, data, 3, current->pid, i);
}

static enum hrtimer_restart stack_trace_timer_fn(struct hrtimer *hrtimer)
{
	/* trace here */
	timer_notify(get_irq_regs(), smp_processor_id());

	hrtimer_forward_now(hrtimer, ns_to_ktime(sample_period));

	return HRTIMER_RESTART;
}

static void start_stack_timer(int cpu)
{
	struct hrtimer *hrtimer = &per_cpu(stack_trace_hrtimer, cpu);

	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
	hrtimer->function = stack_trace_timer_fn;
	hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;

	hrtimer_start(hrtimer, ns_to_ktime(sample_period), HRTIMER_MODE_REL);
}

static void start_stack_timers(void)
{
	cpumask_t saved_mask = current->cpus_allowed;
	int cpu;

	for_each_online_cpu(cpu) {
		cpumask_of_cpu_ptr(new_mask, cpu);

		set_cpus_allowed_ptr(current, new_mask);
		start_stack_timer(cpu);
	}
	set_cpus_allowed_ptr(current, &saved_mask);
}

static void stop_stack_timer(int cpu)
{
	struct hrtimer *hrtimer = &per_cpu(stack_trace_hrtimer, cpu);

	hrtimer_cancel(hrtimer);
}

static void stop_stack_timers(void)
{
	int cpu;

	for_each_online_cpu(cpu)
		stop_stack_timer(cpu);
}

static void stack_reset(struct trace_array *tr)
{
	int cpu;

	tr->time_start = ftrace_now(tr->cpu);

	for_each_online_cpu(cpu)
		tracing_reset(tr->data[cpu]);
}

static void start_stack_trace(struct trace_array *tr)
{
	mutex_lock(&sample_timer_lock);
	stack_reset(tr);
	start_stack_timers();
	tracer_enabled = 1;
	mutex_unlock(&sample_timer_lock);
}

static void stop_stack_trace(struct trace_array *tr)
{
	mutex_lock(&sample_timer_lock);
	stop_stack_timers();
	tracer_enabled = 0;
	mutex_unlock(&sample_timer_lock);
}

static void stack_trace_init(struct trace_array *tr)
{
	sysprof_trace = tr;

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

static void stack_trace_reset(struct trace_array *tr)
{
	if (tr->ctrl)
		stop_stack_trace(tr);
}

static void stack_trace_ctrl_update(struct trace_array *tr)
{
	/* When starting a new trace, reset the buffers */
	if (tr->ctrl)
		start_stack_trace(tr);
	else
		stop_stack_trace(tr);
}

static struct tracer stack_trace __read_mostly =
{
	.name		= "sysprof",
	.init		= stack_trace_init,
	.reset		= stack_trace_reset,
	.ctrl_update	= stack_trace_ctrl_update,
#ifdef CONFIG_FTRACE_SELFTEST
	.selftest    = trace_selftest_startup_sysprof,
#endif
};

__init static int init_stack_trace(void)
{
	return register_tracer(&stack_trace);
}
device_initcall(init_stack_trace);

#define MAX_LONG_DIGITS 22

static ssize_t
sysprof_sample_read(struct file *filp, char __user *ubuf,
		    size_t cnt, loff_t *ppos)
{
	char buf[MAX_LONG_DIGITS];
	int r;

	r = sprintf(buf, "%ld\n", nsecs_to_usecs(sample_period));

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

static ssize_t
sysprof_sample_write(struct file *filp, const char __user *ubuf,
		     size_t cnt, loff_t *ppos)
{
	char buf[MAX_LONG_DIGITS];
	unsigned long val;

	if (cnt > MAX_LONG_DIGITS-1)
		cnt = MAX_LONG_DIGITS-1;

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

	buf[cnt] = 0;

	val = simple_strtoul(buf, NULL, 10);
	/*
	 * Enforce a minimum sample period of 100 usecs:
	 */
	if (val < 100)
		val = 100;

	mutex_lock(&sample_timer_lock);
	stop_stack_timers();
	sample_period = val * 1000;
	start_stack_timers();
	mutex_unlock(&sample_timer_lock);

	return cnt;
}

static struct file_operations sysprof_sample_fops = {
	.read		= sysprof_sample_read,
	.write		= sysprof_sample_write,
};

void init_tracer_sysprof_debugfs(struct dentry *d_tracer)
{
	struct dentry *entry;

	entry = debugfs_create_file("sysprof_sample_period", 0644,
			d_tracer, NULL, &sysprof_sample_fops);
	if (entry)
		return;
	pr_warning("Could not create debugfs 'dyn_ftrace_total_info' entry\n");
}