Repository navigation
Expand file tree
/
Copy pathoffcputime.py
More file actions
executable file
·287 lines (258 loc) · 9.72 KB
/
Copy pathoffcputime.py
File metadata and controls
executable file
·287 lines (258 loc) · 9.72 KB
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
#!/usr/bin/python
#
# offcputime Summarize off-CPU time by stack trace
# For Linux, uses BCC, eBPF.
#
# USAGE: offcputime [-h] [-p PID | -u | -k] [-U | -K] [-f] [duration]
#
# Copyright 2016 Netflix, Inc.
# Licensed under the Apache License, Version 2.0 (the "License")
#
# 13-Jan-2016 Brendan Gregg Created this.
from __future__ import print_function
from bcc import BPF
from sys import stderr
from time import sleep, strftime
import argparse
import errno
import signal
# arg validation
def positive_int(val):
try:
ival = int(val)
except ValueError:
raise argparse.ArgumentTypeError("must be an integer")
if ival < 0:
raise argparse.ArgumentTypeError("must be positive")
return ival
def positive_nonzero_int(val):
ival = positive_int(val)
if ival == 0:
raise argparse.ArgumentTypeError("must be nonzero")
return ival
# arguments
examples = """examples:
./offcputime # trace off-CPU stack time until Ctrl-C
./offcputime 5 # trace for 5 seconds only
./offcputime -f 5 # 5 seconds, and output in folded format
./offcputime -m 1000 # trace only events that last more than 1000 usec
./offcputime -M 10000 # trace only events that last less than 10000 usec
./offcputime -p 185 # only trace threads for PID 185
./offcputime -t 188 # only trace thread 188
./offcputime -u # only trace user threads (no kernel)
./offcputime -k # only trace kernel threads (no user)
./offcputime -U # only show user space stacks (no kernel)
./offcputime -K # only show kernel space stacks (no user)
"""
parser = argparse.ArgumentParser(
description="Summarize off-CPU time by stack trace",
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog=examples)
thread_group = parser.add_mutually_exclusive_group()
# Note: this script provides --pid and --tid flags but their arguments are
# referred to internally using kernel nomenclature: TGID and PID.
thread_group.add_argument("-p", "--pid", metavar="PID", dest="tgid",
help="trace this PID only", type=positive_int)
thread_group.add_argument("-t", "--tid", metavar="TID", dest="pid",
help="trace this TID only", type=positive_int)
thread_group.add_argument("-u", "--user-threads-only", action="store_true",
help="user threads only (no kernel threads)")
thread_group.add_argument("-k", "--kernel-threads-only", action="store_true",
help="kernel threads only (no user threads)")
stack_group = parser.add_mutually_exclusive_group()
stack_group.add_argument("-U", "--user-stacks-only", action="store_tr
8000
ue",
help="show stacks from user space only (no kernel space stacks)")
stack_group.add_argument("-K", "--kernel-stacks-only", action="store_true",
help="show stacks from kernel space only (no user space stacks)")
parser.add_argument("-d", "--delimited", action="store_true",
help="insert delimiter between kernel/user stacks")
parser.add_argument("-f", "--folded", action="store_true",
help="output folded format")
parser.add_argument("--stack-storage-size", default=1024,
type=positive_nonzero_int,
help="the number of unique stack traces that can be stored and "
"displayed (default 1024)")
parser.add_argument("duration", nargs="?", default=99999999,
type=positive_nonzero_int,
help="duration of trace, in seconds")
parser.add_argument("-m", "--min-block-time", default=1,
type=positive_nonzero_int,
help="the amount of time in microseconds over which we " +
"store traces (default 1)")
parser.add_argument("-M", "--max-block-time", default=(1 << 64) - 1,
type=positive_nonzero_int,
help="the amount of time in microseconds under which we " +
"store traces (default U64_MAX)")
args = parser.parse_args()
if args.pid and args.tgid:
parser.error("specify only one of -p and -t")
folded = args.folded
duration = int(args.duration)
# signal handler
def signal_ignore(signal, frame):
print()
# define BPF program
bpf_text = """
#include <uapi/linux/ptrace.h>
#include <linux/sched.h>
#define MINBLOCK_US MINBLOCK_US_VALUEULL
#define MAXBLOCK_US MAXBLOCK_US_VALUEULL
struct key_t {
u32 pid;
u32 tgid;
int user_stack_id;
int kernel_stack_id;
char name[TASK_COMM_LEN];
};
BPF_HASH(counts, struct key_t);
BPF_HASH(start, u32);
BPF_STACK_TRACE(stack_traces, STACK_STORAGE_SIZE)
int oncpu(struct pt_regs *ctx, struct task_struct *prev) {
u32 pid = prev->pid;
u32 tgid = prev->tgid;
u64 ts, *tsp;
// record previous thread sleep time
if (THREAD_FILTER) {
ts = bpf_ktime_get_ns();
start.update(&pid, &ts);
}
// get the current thread's start time
pid = bpf_get_current_pid_tgid();
tgid = bpf_get_current_pid_tgid() >> 32;
tsp = start.lookup(&pid);
if (tsp == 0) {
return 0; // missed start or filtered
}
// calculate current thread's delta time
u64 delta = bpf_ktime_get_ns() - *tsp;
start.delete(&pid);
delta = delta / 1000;
if ((delta < MINBLOCK_US) || (delta > MAXBLOCK_US)) {
return 0;
}
// create map key
u64 zero = 0, *val;
struct key_t key = {};
key.pid = pid;
key.tgid = tgid;
key.user_stack_id = USER_STACK_GET;
key.kernel_stack_id = KERNEL_STACK_GET;
bpf_get_current_comm(&key.name, sizeof(key.name));
val = counts.lookup_or_init(&key, &zero);
(*val) += delta;
return 0;
}
"""
# set thread filter
thread_context = ""
if args.tgid is not None:
thread_context = "PID %d" % args.tgid
thread_filter = 'tgid == %d' % args.tgid
elif args.pid is not None:
thread_context = "TID %d" % args.pid
thread_filter = 'pid == %d' % args.pid
elif args.user_threads_only:
thread_context = "user threads"
thread_filter = '!(prev->flags & PF_KTHREAD)'
elif args.kernel_threads_only:
thread_context = "kernel threads"
thread_filter = 'prev->flags & PF_KTHREAD'
else:
thread_context = "all threads"
thread_filter = '1'
bpf_text = bpf_text.replace('THREAD_FILTER', thread_filter)
# set stack storage size
bpf_text = bpf_text.replace('STACK_STORAGE_SIZE', str(args.stack_storage_size))
bpf_text = bpf_text.replace('MINBLOCK_US_VALUE', str(args.min_block_time))
bpf_text = bpf_text.replace('MAXBLOCK_US_VALUE', str(args.max_block_time))
# handle stack args
kernel_stack_get = "stack_traces.get_stackid(ctx, BPF_F_REUSE_STACKID)"
user_stack_get = \
"stack_traces.get_stackid(ctx, BPF_F_REUSE_STACKID | BPF_F_USER_STACK)"
stack_context = ""
if args.user_stacks_only:
stack_context = "user"
kernel_stack_get = "-1"
el
8BA1
if args.kernel_stacks_only:
stack_context = "kernel"
user_stack_get = "-1"
else:
stack_context = "user + kernel"
bpf_text = bpf_text.replace('USER_STACK_GET', user_stack_get)
bpf_text = bpf_text.replace('KERNEL_STACK_GET', kernel_stack_get)
need_delimiter = args.delimited and not (args.kernel_stacks_only or
args.user_stacks_only)
# check for an edge case; the code below will handle this case correctly
# but ultimately nothing will be displayed
if args.kernel_threads_only and args.user_stacks_only:
print("ERROR: Displaying user stacks for kernel threads " +
"doesn't make sense.", file=stderr)
exit(1)
# initialize BPF
b = BPF(text=bpf_text)
b.attach_kprobe(event="finish_task_switch", fn_name="oncpu")
matched = b.num_open_kprobes()
if matched == 0:
print("error: 0 functions traced. Exiting.", file=stderr)
exit(1)
# header
if not folded:
print("Tracing off-CPU time (us) of %s by %s stack" %
(thread_context, stack_context), end="")
if duration < 99999999:
print(" for %d secs." % duration)
else:
print("... Hit Ctrl-C to end.")
try:
sleep(duration)
except KeyboardInterrupt:
# as cleanup can take many seconds, trap Ctrl-C:
signal.signal(signal.SIGINT, signal_ignore)
if not folded:
print()
missing_stacks = 0
has_enomem = False
counts = b.get_table("counts")
stack_traces = b.get_table("stack_traces")
for k, v in sorted(counts.items(), key=lambda counts: counts[1].value):
# handle get_stackid erorrs
if (not args.user_stacks_only and k.kernel_stack_id < 0) or \
(not args.kernel_stacks_only and k.user_stack_id < 0 and
k.user_stack_id != -errno.EFAULT):
missing_stacks += 1
# check for an ENOMEM error
if k.kernel_stack_id == -errno.ENOMEM or \
k.user_stack_id == -errno.ENOMEM:
has_enomem = True
continue
# user stacks will be symbolized by tgid, not pid, to avoid the overhead
# of one symbol resolver per thread
user_stack = [] if k.user_stack_id < 0 else \
stack_traces.walk(k.user_stack_id)
kernel_stack = [] if k.kernel_stack_id < 0 else \
stack_traces.walk(k.kernel_stack_id)
if folded:
# print folded stack output
user_stack = list(user_stack)
kernel_stack = list(kernel_stack)
line = [k.name.decode()] + \
[b.sym(addr, k.tgid) for addr in reversed(user_stack)] + \
(need_delimiter and ["-"] or []) + \
[b.ksym(addr) for addr in reversed(kernel_stack)]
print("%s %d" % (";".join(line), v.value))
else:
# print default multi-line stack output
for addr in kernel_stack:
print(" %s" % b.ksym(addr))
if need_delimiter:
print(" --")
for addr in user_stack:
print(" %s" % b.sym(addr, k.tgid))
print(" %-16s %s (%d)" % ("-", k.name, k.pid))
print(" %d\n" % v.value)
if missing_stacks > 0:
enomem_str = "" if not has_enomem else \
" Consider increasing --stack-storage-size."
print("WARNING: %d stack traces could not be displayed.%s" %
(missing_stacks, enomem_str),
file=stderr)