1 ======================== 1 ========================
2 ftrace - Function Tracer 2 ftrace - Function Tracer
3 ======================== 3 ========================
4 4
5 Copyright 2008 Red Hat Inc. 5 Copyright 2008 Red Hat Inc.
6 6
7 :Author: Steven Rostedt <srostedt@redhat.com> 7 :Author: Steven Rostedt <srostedt@redhat.com>
8 :License: The GNU Free Documentation License, 8 :License: The GNU Free Documentation License, Version 1.2
9 (dual licensed under the GPL v2) 9 (dual licensed under the GPL v2)
10 :Original Reviewers: Elias Oltmanns, Randy Du 10 :Original Reviewers: Elias Oltmanns, Randy Dunlap, Andrew Morton,
11 John Kacur, and David Te 11 John Kacur, and David Teigland.
12 12
13 - Written for: 2.6.28-rc2 13 - Written for: 2.6.28-rc2
14 - Updated for: 3.10 14 - Updated for: 3.10
15 - Updated for: 4.13 - Copyright 2017 VMware In 15 - Updated for: 4.13 - Copyright 2017 VMware Inc. Steven Rostedt
16 - Converted to rst format - Changbin Du <changb 16 - Converted to rst format - Changbin Du <changbin.du@intel.com>
17 17
18 Introduction 18 Introduction
19 ------------ 19 ------------
20 20
21 Ftrace is an internal tracer designed to help 21 Ftrace is an internal tracer designed to help out developers and
22 designers of systems to find what is going on 22 designers of systems to find what is going on inside the kernel.
23 It can be used for debugging or analyzing late 23 It can be used for debugging or analyzing latencies and
24 performance issues that take place outside of 24 performance issues that take place outside of user-space.
25 25
26 Although ftrace is typically considered the fu 26 Although ftrace is typically considered the function tracer, it
27 is really a framework of several assorted trac !! 27 is really a frame work of several assorted tracing utilities.
28 There's latency tracing to examine what occurs 28 There's latency tracing to examine what occurs between interrupts
29 disabled and enabled, as well as for preemptio 29 disabled and enabled, as well as for preemption and from a time
30 a task is woken to the task is actually schedu 30 a task is woken to the task is actually scheduled in.
31 31
32 One of the most common uses of ftrace is the e 32 One of the most common uses of ftrace is the event tracing.
33 Throughout the kernel is hundreds of static ev !! 33 Through out the kernel is hundreds of static event points that
34 can be enabled via the tracefs file system to 34 can be enabled via the tracefs file system to see what is
35 going on in certain parts of the kernel. 35 going on in certain parts of the kernel.
36 36
37 See events.rst for more information. !! 37 See events.txt for more information.
38 38
39 39
40 Implementation Details 40 Implementation Details
41 ---------------------- 41 ----------------------
42 42
43 See Documentation/trace/ftrace-design.rst for !! 43 See :doc:`ftrace-design` for details for arch porters and such.
44 44
45 45
46 The File System 46 The File System
47 --------------- 47 ---------------
48 48
49 Ftrace uses the tracefs file system to hold th 49 Ftrace uses the tracefs file system to hold the control files as
50 well as the files to display output. 50 well as the files to display output.
51 51
52 When tracefs is configured into the kernel (wh 52 When tracefs is configured into the kernel (which selecting any ftrace
53 option will do) the directory /sys/kernel/trac 53 option will do) the directory /sys/kernel/tracing will be created. To mount
54 this directory, you can add to your /etc/fstab 54 this directory, you can add to your /etc/fstab file::
55 55
56 tracefs /sys/kernel/tracing trace 56 tracefs /sys/kernel/tracing tracefs defaults 0 0
57 57
58 Or you can mount it at run time with:: 58 Or you can mount it at run time with::
59 59
60 mount -t tracefs nodev /sys/kernel/tracing 60 mount -t tracefs nodev /sys/kernel/tracing
61 61
62 For quicker access to that directory you may w 62 For quicker access to that directory you may want to make a soft link to
63 it:: 63 it::
64 64
65 ln -s /sys/kernel/tracing /tracing 65 ln -s /sys/kernel/tracing /tracing
66 66
67 .. attention:: 67 .. attention::
68 68
69 Before 4.1, all ftrace tracing control files 69 Before 4.1, all ftrace tracing control files were within the debugfs
70 file system, which is typically located at / 70 file system, which is typically located at /sys/kernel/debug/tracing.
71 For backward compatibility, when mounting th 71 For backward compatibility, when mounting the debugfs file system,
72 the tracefs file system will be automaticall 72 the tracefs file system will be automatically mounted at:
73 73
74 /sys/kernel/debug/tracing 74 /sys/kernel/debug/tracing
75 75
76 All files located in the tracefs file system 76 All files located in the tracefs file system will be located in that
77 debugfs file system directory as well. 77 debugfs file system directory as well.
78 78
79 .. attention:: 79 .. attention::
80 80
81 Any selected ftrace option will also create 81 Any selected ftrace option will also create the tracefs file system.
82 The rest of the document will assume that yo 82 The rest of the document will assume that you are in the ftrace directory
83 (cd /sys/kernel/tracing) and will only conce 83 (cd /sys/kernel/tracing) and will only concentrate on the files within that
84 directory and not distract from the content 84 directory and not distract from the content with the extended
85 "/sys/kernel/tracing" path name. 85 "/sys/kernel/tracing" path name.
86 86
87 That's it! (assuming that you have ftrace conf 87 That's it! (assuming that you have ftrace configured into your kernel)
88 88
89 After mounting tracefs you will have access to 89 After mounting tracefs you will have access to the control and output files
90 of ftrace. Here is a list of some of the key f 90 of ftrace. Here is a list of some of the key files:
91 91
92 92
93 Note: all time values are in microseconds. 93 Note: all time values are in microseconds.
94 94
95 current_tracer: 95 current_tracer:
96 96
97 This is used to set or display the cur 97 This is used to set or display the current tracer
98 that is configured. Changing the curre !! 98 that is configured.
99 the ring buffer content as well as the <<
100 99
101 available_tracers: 100 available_tracers:
102 101
103 This holds the different types of trac 102 This holds the different types of tracers that
104 have been compiled into the kernel. Th 103 have been compiled into the kernel. The
105 tracers listed here can be configured 104 tracers listed here can be configured by
106 echoing their name into current_tracer 105 echoing their name into current_tracer.
107 106
108 tracing_on: 107 tracing_on:
109 108
110 This sets or displays whether writing 109 This sets or displays whether writing to the trace
111 ring buffer is enabled. Echo 0 into th 110 ring buffer is enabled. Echo 0 into this file to disable
112 the tracer or 1 to enable it. Note, th 111 the tracer or 1 to enable it. Note, this only disables
113 writing to the ring buffer, the tracin 112 writing to the ring buffer, the tracing overhead may
114 still be occurring. 113 still be occurring.
115 114
116 The kernel function tracing_off() can 115 The kernel function tracing_off() can be used within the
117 kernel to disable writing to the ring 116 kernel to disable writing to the ring buffer, which will
118 set this file to "0". User space can r 117 set this file to "0". User space can re-enable tracing by
119 echoing "1" into the file. 118 echoing "1" into the file.
120 119
121 Note, the function and event trigger " 120 Note, the function and event trigger "traceoff" will also
122 set this file to zero and stop tracing 121 set this file to zero and stop tracing. Which can also
123 be re-enabled by user space using this 122 be re-enabled by user space using this file.
124 123
125 trace: 124 trace:
126 125
127 This file holds the output of the trac 126 This file holds the output of the trace in a human
128 readable format (described below). Ope !! 127 readable format (described below). Note, tracing is temporarily
129 writing with the O_TRUNC flag clears t !! 128 disabled while this file is being read (opened).
130 Note, this file is not a consumer. If <<
131 (no tracer running, or tracing_on is z <<
132 the same output each time it is read. <<
133 it may produce inconsistent results as <<
134 the entire buffer without consuming it <<
135 129
136 trace_pipe: 130 trace_pipe:
137 131
138 The output is the same as the "trace" 132 The output is the same as the "trace" file but this
139 file is meant to be streamed with live 133 file is meant to be streamed with live tracing.
140 Reads from this file will block until 134 Reads from this file will block until new data is
141 retrieved. Unlike the "trace" file, t 135 retrieved. Unlike the "trace" file, this file is a
142 consumer. This means reading from this 136 consumer. This means reading from this file causes
143 sequential reads to display more curre 137 sequential reads to display more current data. Once
144 data is read from this file, it is con 138 data is read from this file, it is consumed, and
145 will not be read again with a sequenti 139 will not be read again with a sequential read. The
146 "trace" file is static, and if the tra 140 "trace" file is static, and if the tracer is not
147 adding more data, it will display the 141 adding more data, it will display the same
148 information every time it is read. !! 142 information every time it is read. This file will not
>> 143 disable tracing while being read.
149 144
150 trace_options: 145 trace_options:
151 146
152 This file lets the user control the am 147 This file lets the user control the amount of data
153 that is displayed in one of the above 148 that is displayed in one of the above output
154 files. Options also exist to modify ho 149 files. Options also exist to modify how a tracer
155 or events work (stack traces, timestam 150 or events work (stack traces, timestamps, etc).
156 151
157 options: 152 options:
158 153
159 This is a directory that has a file fo 154 This is a directory that has a file for every available
160 trace option (also in trace_options). 155 trace option (also in trace_options). Options may also be set
161 or cleared by writing a "1" or "0" res 156 or cleared by writing a "1" or "0" respectively into the
162 corresponding file with the option nam 157 corresponding file with the option name.
163 158
164 tracing_max_latency: 159 tracing_max_latency:
165 160
166 Some of the tracers record the max lat 161 Some of the tracers record the max latency.
167 For example, the maximum time that int 162 For example, the maximum time that interrupts are disabled.
168 The maximum time is saved in this file 163 The maximum time is saved in this file. The max trace will also be
169 stored, and displayed by "trace". A ne 164 stored, and displayed by "trace". A new max trace will only be
170 recorded if the latency is greater tha 165 recorded if the latency is greater than the value in this file
171 (in microseconds). 166 (in microseconds).
172 167
173 By echoing in a time into this file, n 168 By echoing in a time into this file, no latency will be recorded
174 unless it is greater than the time in 169 unless it is greater than the time in this file.
175 170
176 tracing_thresh: 171 tracing_thresh:
177 172
178 Some latency tracers will record a tra 173 Some latency tracers will record a trace whenever the
179 latency is greater than the number in 174 latency is greater than the number in this file.
180 Only active when the file contains a n 175 Only active when the file contains a number greater than 0.
181 (in microseconds) 176 (in microseconds)
182 177
183 buffer_percent: <<
184 <<
185 This is the watermark for how much the <<
186 before a waiter is woken up. That is, <<
187 blocking read syscall on one of the pe <<
188 will block until the given amount of d <<
189 is in the ring buffer before it wakes <<
190 controls how the splice system calls a <<
191 <<
192 0 - means to wake up as soon as th <<
193 50 - means to wake up when roughly <<
194 are full. <<
195 100 - means to block until the ring <<
196 about to start overwriting the <<
197 <<
198 buffer_size_kb: 178 buffer_size_kb:
199 179
200 This sets or displays the number of ki 180 This sets or displays the number of kilobytes each CPU
201 buffer holds. By default, the trace bu 181 buffer holds. By default, the trace buffers are the same size
202 for each CPU. The displayed number is 182 for each CPU. The displayed number is the size of the
203 CPU buffer and not total size of all b 183 CPU buffer and not total size of all buffers. The
204 trace buffers are allocated in pages ( 184 trace buffers are allocated in pages (blocks of memory
205 that the kernel uses for allocation, u 185 that the kernel uses for allocation, usually 4 KB in size).
206 A few extra pages may be allocated to !! 186 If the last page allocated has room for more bytes
207 meta-data. If the last page allocated <<
208 than requested, the rest of the page w 187 than requested, the rest of the page will be used,
209 making the actual allocation bigger th 188 making the actual allocation bigger than requested or shown.
210 ( Note, the size may not be a multiple 189 ( Note, the size may not be a multiple of the page size
211 due to buffer management meta-data. ) 190 due to buffer management meta-data. )
212 191
213 Buffer sizes for individual CPUs may v 192 Buffer sizes for individual CPUs may vary
214 (see "per_cpu/cpu0/buffer_size_kb" bel 193 (see "per_cpu/cpu0/buffer_size_kb" below), and if they do
215 this file will show "X". 194 this file will show "X".
216 195
217 buffer_total_size_kb: 196 buffer_total_size_kb:
218 197
219 This displays the total combined size 198 This displays the total combined size of all the trace buffers.
220 199
221 buffer_subbuf_size_kb: <<
222 <<
223 This sets or displays the sub buffer s <<
224 into several same size "sub buffers". <<
225 the size of the sub buffer. Normally, <<
226 architecture's page (4K on x86). The s <<
227 at the start which also limits the siz <<
228 the sub buffer is a page size, no even <<
229 size minus the sub buffer meta data. <<
230 <<
231 Note, the buffer_subbuf_size_kb is a w <<
232 minimum size of the subbuffer. The ker <<
233 implementation details, or simply fail <<
234 not handle the request. <<
235 <<
236 Changing the sub buffer size allows fo <<
237 page size. <<
238 <<
239 Note: When changing the sub-buffer siz <<
240 data in the ring buffer and the snapsh <<
241 <<
242 free_buffer: 200 free_buffer:
243 201
244 If a process is performing tracing, an 202 If a process is performing tracing, and the ring buffer should be
245 shrunk "freed" when the process is fin 203 shrunk "freed" when the process is finished, even if it were to be
246 killed by a signal, this file can be u 204 killed by a signal, this file can be used for that purpose. On close
247 of this file, the ring buffer will be 205 of this file, the ring buffer will be resized to its minimum size.
248 Having a process that is tracing also 206 Having a process that is tracing also open this file, when the process
249 exits its file descriptor for this fil 207 exits its file descriptor for this file will be closed, and in doing so,
250 the ring buffer will be "freed". 208 the ring buffer will be "freed".
251 209
252 It may also stop tracing if disable_on 210 It may also stop tracing if disable_on_free option is set.
253 211
254 tracing_cpumask: 212 tracing_cpumask:
255 213
256 This is a mask that lets the user only 214 This is a mask that lets the user only trace on specified CPUs.
257 The format is a hex string representin 215 The format is a hex string representing the CPUs.
258 216
259 set_ftrace_filter: 217 set_ftrace_filter:
260 218
261 When dynamic ftrace is configured in ( 219 When dynamic ftrace is configured in (see the
262 section below "dynamic ftrace"), the c 220 section below "dynamic ftrace"), the code is dynamically
263 modified (code text rewrite) to disabl 221 modified (code text rewrite) to disable calling of the
264 function profiler (mcount). This lets 222 function profiler (mcount). This lets tracing be configured
265 in with practically no overhead in per 223 in with practically no overhead in performance. This also
266 has a side effect of enabling or disab 224 has a side effect of enabling or disabling specific functions
267 to be traced. Echoing names of functio 225 to be traced. Echoing names of functions into this file
268 will limit the trace to only those fun 226 will limit the trace to only those functions.
269 This influences the tracers "function" <<
270 and thus also function profiling (see <<
271 227
272 The functions listed in "available_fil 228 The functions listed in "available_filter_functions" are what
273 can be written into this file. 229 can be written into this file.
274 230
275 This interface also allows for command 231 This interface also allows for commands to be used. See the
276 "Filter commands" section for more det 232 "Filter commands" section for more details.
277 233
278 As a speed up, since processing string <<
279 and requires a check of all functions <<
280 an index can be written into this file <<
281 written will instead select the same c <<
282 of the "available_filter_functions" fi <<
283 <<
284 set_ftrace_notrace: 234 set_ftrace_notrace:
285 235
286 This has an effect opposite to that of 236 This has an effect opposite to that of
287 set_ftrace_filter. Any function that i 237 set_ftrace_filter. Any function that is added here will not
288 be traced. If a function exists in bot 238 be traced. If a function exists in both set_ftrace_filter
289 and set_ftrace_notrace, the function w 239 and set_ftrace_notrace, the function will _not_ be traced.
290 240
291 set_ftrace_pid: 241 set_ftrace_pid:
292 242
293 Have the function tracer only trace th 243 Have the function tracer only trace the threads whose PID are
294 listed in this file. 244 listed in this file.
295 245
296 If the "function-fork" option is set, 246 If the "function-fork" option is set, then when a task whose
297 PID is listed in this file forks, the 247 PID is listed in this file forks, the child's PID will
298 automatically be added to this file, a 248 automatically be added to this file, and the child will be
299 traced by the function tracer as well. 249 traced by the function tracer as well. This option will also
300 cause PIDs of tasks that exit to be re 250 cause PIDs of tasks that exit to be removed from the file.
301 251
302 set_ftrace_notrace_pid: <<
303 <<
304 Have the function tracer ignore thread <<
305 this file. <<
306 <<
307 If the "function-fork" option is set, <<
308 PID is listed in this file forks, the <<
309 automatically be added to this file, a <<
310 traced by the function tracer as well. <<
311 cause PIDs of tasks that exit to be re <<
312 <<
313 If a PID is in both this file and "set <<
314 file takes precedence, and the thread <<
315 <<
316 set_event_pid: 252 set_event_pid:
317 253
318 Have the events only trace a task with 254 Have the events only trace a task with a PID listed in this file.
319 Note, sched_switch and sched_wake_up w 255 Note, sched_switch and sched_wake_up will also trace events
320 listed in this file. 256 listed in this file.
321 257
322 To have the PIDs of children of tasks 258 To have the PIDs of children of tasks with their PID in this file
323 added on fork, enable the "event-fork" 259 added on fork, enable the "event-fork" option. That option will also
324 cause the PIDs of tasks to be removed 260 cause the PIDs of tasks to be removed from this file when the task
325 exits. 261 exits.
326 262
327 set_event_notrace_pid: <<
328 <<
329 Have the events not trace a task with <<
330 Note, sched_switch and sched_wakeup wi <<
331 in this file, even if a thread's PID i <<
332 sched_switch or sched_wakeup events al <<
333 be traced. <<
334 <<
335 To have the PIDs of children of tasks <<
336 added on fork, enable the "event-fork" <<
337 cause the PIDs of tasks to be removed <<
338 exits. <<
339 <<
340 set_graph_function: 263 set_graph_function:
341 264
342 Functions listed in this file will cau 265 Functions listed in this file will cause the function graph
343 tracer to only trace these functions a 266 tracer to only trace these functions and the functions that
344 they call. (See the section "dynamic f 267 they call. (See the section "dynamic ftrace" for more details).
345 Note, set_ftrace_filter and set_ftrace <<
346 what functions are being traced. <<
347 268
348 set_graph_notrace: 269 set_graph_notrace:
349 270
350 Similar to set_graph_function, but wil 271 Similar to set_graph_function, but will disable function graph
351 tracing when the function is hit until 272 tracing when the function is hit until it exits the function.
352 This makes it possible to ignore traci 273 This makes it possible to ignore tracing functions that are called
353 by a specific function. 274 by a specific function.
354 275
355 available_filter_functions: 276 available_filter_functions:
356 277
357 This lists the functions that ftrace h 278 This lists the functions that ftrace has processed and can trace.
358 These are the function names that you 279 These are the function names that you can pass to
359 "set_ftrace_filter", "set_ftrace_notra !! 280 "set_ftrace_filter" or "set_ftrace_notrace".
360 "set_graph_function", or "set_graph_no <<
361 (See the section "dynamic ftrace" belo 281 (See the section "dynamic ftrace" below for more details.)
362 282
363 available_filter_functions_addrs: <<
364 <<
365 Similar to available_filter_functions, <<
366 for each function. The displayed addre <<
367 and can differ from /proc/kallsyms add <<
368 <<
369 dyn_ftrace_total_info: 283 dyn_ftrace_total_info:
370 284
371 This file is for debugging purposes. T 285 This file is for debugging purposes. The number of functions that
372 have been converted to nops and are av 286 have been converted to nops and are available to be traced.
373 287
374 enabled_functions: 288 enabled_functions:
375 289
376 This file is more for debugging ftrace 290 This file is more for debugging ftrace, but can also be useful
377 in seeing if any function has a callba 291 in seeing if any function has a callback attached to it.
378 Not only does the trace infrastructure 292 Not only does the trace infrastructure use ftrace function
379 trace utility, but other subsystems mi 293 trace utility, but other subsystems might too. This file
380 displays all functions that have a cal 294 displays all functions that have a callback attached to them
381 as well as the number of callbacks tha 295 as well as the number of callbacks that have been attached.
382 Note, a callback may also call multipl 296 Note, a callback may also call multiple functions which will
383 not be listed in this count. 297 not be listed in this count.
384 298
385 If the callback registered to be trace 299 If the callback registered to be traced by a function with
386 the "save regs" attribute (thus even m 300 the "save regs" attribute (thus even more overhead), a 'R'
387 will be displayed on the same line as 301 will be displayed on the same line as the function that
388 is returning registers. 302 is returning registers.
389 303
390 If the callback registered to be trace 304 If the callback registered to be traced by a function with
391 the "ip modify" attribute (thus the re 305 the "ip modify" attribute (thus the regs->ip can be changed),
392 an 'I' will be displayed on the same l 306 an 'I' will be displayed on the same line as the function that
393 can be overridden. 307 can be overridden.
394 308
395 If a non ftrace trampoline is attached <<
396 Note, normal ftrace trampolines can al <<
397 "direct" trampoline can be attached to <<
398 <<
399 Some architectures can not call direct <<
400 the ftrace ops function located above <<
401 such cases an 'O' will be displayed. <<
402 <<
403 If a function had either the "ip modif <<
404 it in the past, a 'M' will be shown. T <<
405 used to know if a function was every m <<
406 and can be used for debugging. <<
407 <<
408 If the architecture supports it, it wi 309 If the architecture supports it, it will also show what callback
409 is being directly called by the functi 310 is being directly called by the function. If the count is greater
410 than 1 it most likely will be ftrace_o 311 than 1 it most likely will be ftrace_ops_list_func().
411 312
412 If the callback of a function jumps to !! 313 If the callback of the function jumps to a trampoline that is
413 specific to the callback and which is !! 314 specific to a the callback and not the standard trampoline,
414 its address will be printed as well as 315 its address will be printed as well as the function that the
415 trampoline calls. 316 trampoline calls.
416 317
417 touched_functions: <<
418 <<
419 This file contains all the functions t <<
420 to it via the ftrace infrastructure. I <<
421 enabled_functions but shows all functi <<
422 traced. <<
423 <<
424 To see any function that has every bee <<
425 direct trampoline, one can perform the <<
426 <<
427 grep ' M ' /sys/kernel/tracing/touched <<
428 <<
429 function_profile_enabled: 318 function_profile_enabled:
430 319
431 When set it will enable all functions 320 When set it will enable all functions with either the function
432 tracer, or if configured, the function 321 tracer, or if configured, the function graph tracer. It will
433 keep a histogram of the number of func 322 keep a histogram of the number of functions that were called
434 and if the function graph tracer was c 323 and if the function graph tracer was configured, it will also keep
435 track of the time spent in those funct 324 track of the time spent in those functions. The histogram
436 content can be displayed in the files: 325 content can be displayed in the files:
437 326
438 trace_stat/function<cpu> ( function0, !! 327 trace_stats/function<cpu> ( function0, function1, etc).
439 328
440 trace_stat: !! 329 trace_stats:
441 330
442 A directory that holds different traci 331 A directory that holds different tracing stats.
443 332
444 kprobe_events: 333 kprobe_events:
445 334
446 Enable dynamic trace points. See kprob !! 335 Enable dynamic trace points. See kprobetrace.txt.
447 336
448 kprobe_profile: 337 kprobe_profile:
449 338
450 Dynamic trace points stats. See kprobe !! 339 Dynamic trace points stats. See kprobetrace.txt.
451 340
452 max_graph_depth: 341 max_graph_depth:
453 342
454 Used with the function graph tracer. T 343 Used with the function graph tracer. This is the max depth
455 it will trace into a function. Setting 344 it will trace into a function. Setting this to a value of
456 one will show only the first kernel fu 345 one will show only the first kernel function that is called
457 from user space. 346 from user space.
458 347
459 printk_formats: 348 printk_formats:
460 349
461 This is for tools that read the raw fo 350 This is for tools that read the raw format files. If an event in
462 the ring buffer references a string, o 351 the ring buffer references a string, only a pointer to the string
463 is recorded into the buffer and not th 352 is recorded into the buffer and not the string itself. This prevents
464 tools from knowing what that string wa 353 tools from knowing what that string was. This file displays the string
465 and address for the string allowing to 354 and address for the string allowing tools to map the pointers to what
466 the strings were. 355 the strings were.
467 356
468 saved_cmdlines: 357 saved_cmdlines:
469 358
470 Only the pid of the task is recorded i 359 Only the pid of the task is recorded in a trace event unless
471 the event specifically saves the task 360 the event specifically saves the task comm as well. Ftrace
472 makes a cache of pid mappings to comms 361 makes a cache of pid mappings to comms to try to display
473 comms for events. If a pid for a comm 362 comms for events. If a pid for a comm is not listed, then
474 "<...>" is displayed in the output. 363 "<...>" is displayed in the output.
475 364
476 If the option "record-cmd" is set to " 365 If the option "record-cmd" is set to "0", then comms of tasks
477 will not be saved during recording. By 366 will not be saved during recording. By default, it is enabled.
478 367
479 saved_cmdlines_size: 368 saved_cmdlines_size:
480 369
481 By default, 128 comms are saved (see " 370 By default, 128 comms are saved (see "saved_cmdlines" above). To
482 increase or decrease the amount of com 371 increase or decrease the amount of comms that are cached, echo
483 the number of comms to cache into this !! 372 in a the number of comms to cache, into this file.
484 373
485 saved_tgids: 374 saved_tgids:
486 375
487 If the option "record-tgid" is set, on 376 If the option "record-tgid" is set, on each scheduling context switch
488 the Task Group ID of a task is saved i 377 the Task Group ID of a task is saved in a table mapping the PID of
489 the thread to its TGID. By default, th 378 the thread to its TGID. By default, the "record-tgid" option is
490 disabled. 379 disabled.
491 380
492 snapshot: 381 snapshot:
493 382
494 This displays the "snapshot" buffer an 383 This displays the "snapshot" buffer and also lets the user
495 take a snapshot of the current running 384 take a snapshot of the current running trace.
496 See the "Snapshot" section below for m 385 See the "Snapshot" section below for more details.
497 386
498 stack_max_size: 387 stack_max_size:
499 388
500 When the stack tracer is activated, th 389 When the stack tracer is activated, this will display the
501 maximum stack size it has encountered. 390 maximum stack size it has encountered.
502 See the "Stack Trace" section below. 391 See the "Stack Trace" section below.
503 392
504 stack_trace: 393 stack_trace:
505 394
506 This displays the stack back trace of 395 This displays the stack back trace of the largest stack
507 that was encountered when the stack tr 396 that was encountered when the stack tracer is activated.
508 See the "Stack Trace" section below. 397 See the "Stack Trace" section below.
509 398
510 stack_trace_filter: 399 stack_trace_filter:
511 400
512 This is similar to "set_ftrace_filter" 401 This is similar to "set_ftrace_filter" but it limits what
513 functions the stack tracer will check. 402 functions the stack tracer will check.
514 403
515 trace_clock: 404 trace_clock:
516 405
517 Whenever an event is recorded into the 406 Whenever an event is recorded into the ring buffer, a
518 "timestamp" is added. This stamp comes 407 "timestamp" is added. This stamp comes from a specified
519 clock. By default, ftrace uses the "lo 408 clock. By default, ftrace uses the "local" clock. This
520 clock is very fast and strictly per cp 409 clock is very fast and strictly per cpu, but on some
521 systems it may not be monotonic with r 410 systems it may not be monotonic with respect to other
522 CPUs. In other words, the local clocks 411 CPUs. In other words, the local clocks may not be in sync
523 with local clocks on other CPUs. 412 with local clocks on other CPUs.
524 413
525 Usual clocks for tracing:: 414 Usual clocks for tracing::
526 415
527 # cat trace_clock 416 # cat trace_clock
528 [local] global counter x86-tsc 417 [local] global counter x86-tsc
529 418
530 The clock with the square brackets aro 419 The clock with the square brackets around it is the one in effect.
531 420
532 local: 421 local:
533 Default clock, but may not be 422 Default clock, but may not be in sync across CPUs
534 423
535 global: 424 global:
536 This clock is in sync with all 425 This clock is in sync with all CPUs but may
537 be a bit slower than the local 426 be a bit slower than the local clock.
538 427
539 counter: 428 counter:
540 This is not a clock at all, bu 429 This is not a clock at all, but literally an atomic
541 counter. It counts up one by o 430 counter. It counts up one by one, but is in sync
542 with all CPUs. This is useful 431 with all CPUs. This is useful when you need to
543 know exactly the order events 432 know exactly the order events occurred with respect to
544 each other on different CPUs. 433 each other on different CPUs.
545 434
546 uptime: 435 uptime:
547 This uses the jiffies counter 436 This uses the jiffies counter and the time stamp
548 is relative to the time since 437 is relative to the time since boot up.
549 438
550 perf: 439 perf:
551 This makes ftrace use the same 440 This makes ftrace use the same clock that perf uses.
552 Eventually perf will be able t 441 Eventually perf will be able to read ftrace buffers
553 and this will help out in inte 442 and this will help out in interleaving the data.
554 443
555 x86-tsc: 444 x86-tsc:
556 Architectures may define their 445 Architectures may define their own clocks. For
557 example, x86 uses its own TSC 446 example, x86 uses its own TSC cycle clock here.
558 447
559 ppc-tb: 448 ppc-tb:
560 This uses the powerpc timebase 449 This uses the powerpc timebase register value.
561 This is in sync across CPUs an 450 This is in sync across CPUs and can also be used
562 to correlate events across hyp 451 to correlate events across hypervisor/guest if
563 tb_offset is known. 452 tb_offset is known.
564 453
565 mono: 454 mono:
566 This uses the fast monotonic c 455 This uses the fast monotonic clock (CLOCK_MONOTONIC)
567 which is monotonic and is subj 456 which is monotonic and is subject to NTP rate adjustments.
568 457
569 mono_raw: 458 mono_raw:
570 This is the raw monotonic cloc 459 This is the raw monotonic clock (CLOCK_MONOTONIC_RAW)
571 which is monotonic but is not !! 460 which is montonic but is not subject to any rate adjustments
572 and ticks at the same rate as 461 and ticks at the same rate as the hardware clocksource.
573 462
574 boot: 463 boot:
575 This is the boot clock (CLOCK_ 464 This is the boot clock (CLOCK_BOOTTIME) and is based on the
576 fast monotonic clock, but also 465 fast monotonic clock, but also accounts for time spent in
577 suspend. Since the clock acces 466 suspend. Since the clock access is designed for use in
578 tracing in the suspend path, s 467 tracing in the suspend path, some side effects are possible
579 if clock is accessed after the 468 if clock is accessed after the suspend time is accounted before
580 the fast mono clock is updated 469 the fast mono clock is updated. In this case, the clock update
581 appears to happen slightly soo 470 appears to happen slightly sooner than it normally would have.
582 Also on 32-bit systems, it's p 471 Also on 32-bit systems, it's possible that the 64-bit boot offset
583 sees a partial update. These e 472 sees a partial update. These effects are rare and post
584 processing should be able to h 473 processing should be able to handle them. See comments in the
585 ktime_get_boot_fast_ns() funct 474 ktime_get_boot_fast_ns() function for more information.
586 475
587 tai: <<
588 This is the tai clock (CLOCK_T <<
589 clock time. However, this cloc <<
590 discontinuities and backwards <<
591 seconds. Since the clock acces <<
592 side effects are possible. The <<
593 readouts in case the internal <<
594 by setting the system time or <<
595 These effects are rare and pos <<
596 handle them. See comments in t <<
597 function for more information. <<
598 <<
599 To set a clock, simply echo the clock 476 To set a clock, simply echo the clock name into this file::
600 477
601 # echo global > trace_clock 478 # echo global > trace_clock
602 479
603 Setting a clock clears the ring buffer <<
604 "snapshot" buffer. <<
605 <<
606 trace_marker: 480 trace_marker:
607 481
608 This is a very useful file for synchro 482 This is a very useful file for synchronizing user space
609 with events happening in the kernel. W 483 with events happening in the kernel. Writing strings into
610 this file will be written into the ftr 484 this file will be written into the ftrace buffer.
611 485
612 It is useful in applications to open t 486 It is useful in applications to open this file at the start
613 of the application and just reference 487 of the application and just reference the file descriptor
614 for the file:: 488 for the file::
615 489
616 void trace_write(const char *f 490 void trace_write(const char *fmt, ...)
617 { 491 {
618 va_list ap; 492 va_list ap;
619 char buf[256]; 493 char buf[256];
620 int n; 494 int n;
621 495
622 if (trace_fd < 0) 496 if (trace_fd < 0)
623 return; 497 return;
624 498
625 va_start(ap, fmt); 499 va_start(ap, fmt);
626 n = vsnprintf(buf, 256 500 n = vsnprintf(buf, 256, fmt, ap);
627 va_end(ap); 501 va_end(ap);
628 502
629 write(trace_fd, buf, n 503 write(trace_fd, buf, n);
630 } 504 }
631 505
632 start:: 506 start::
633 507
634 trace_fd = open("trace_marker" !! 508 trace_fd = open("trace_marker", WR_ONLY);
635 <<
636 Note: Writing into the trace_marker fi <<
637 that are written into /sys/kerne <<
638 See "Event triggers" in Document <<
639 example in Documentation/trace/h <<
640 509
641 trace_marker_raw: 510 trace_marker_raw:
642 511
643 This is similar to trace_marker above, !! 512 This is similar to trace_marker above, but is meant for for binary data
644 to be written to it, where a tool can 513 to be written to it, where a tool can be used to parse the data
645 from trace_pipe_raw. 514 from trace_pipe_raw.
646 515
647 uprobe_events: 516 uprobe_events:
648 517
649 Add dynamic tracepoints in programs. 518 Add dynamic tracepoints in programs.
650 See uprobetracer.rst !! 519 See uprobetracer.txt
651 520
652 uprobe_profile: 521 uprobe_profile:
653 522
654 Uprobe statistics. See uprobetrace.txt 523 Uprobe statistics. See uprobetrace.txt
655 524
656 instances: 525 instances:
657 526
658 This is a way to make multiple trace b 527 This is a way to make multiple trace buffers where different
659 events can be recorded in different bu 528 events can be recorded in different buffers.
660 See "Instances" section below. 529 See "Instances" section below.
661 530
662 events: 531 events:
663 532
664 This is the trace event directory. It 533 This is the trace event directory. It holds event tracepoints
665 (also known as static tracepoints) tha 534 (also known as static tracepoints) that have been compiled
666 into the kernel. It shows what event t 535 into the kernel. It shows what event tracepoints exist
667 and how they are grouped by system. Th 536 and how they are grouped by system. There are "enable"
668 files at various levels that can enabl 537 files at various levels that can enable the tracepoints
669 when a "1" is written to them. 538 when a "1" is written to them.
670 539
671 See events.rst for more information. !! 540 See events.txt for more information.
672 541
673 set_event: 542 set_event:
674 543
675 By echoing in the event into this file 544 By echoing in the event into this file, will enable that event.
676 545
677 See events.rst for more information. !! 546 See events.txt for more information.
678 547
679 available_events: 548 available_events:
680 549
681 A list of events that can be enabled i 550 A list of events that can be enabled in tracing.
682 551
683 See events.rst for more information. !! 552 See events.txt for more information.
684 553
685 timestamp_mode: 554 timestamp_mode:
686 555
687 Certain tracers may change the timesta 556 Certain tracers may change the timestamp mode used when
688 logging trace events into the event bu 557 logging trace events into the event buffer. Events with
689 different modes can coexist within a b 558 different modes can coexist within a buffer but the mode in
690 effect when an event is logged determi 559 effect when an event is logged determines which timestamp mode
691 is used for that event. The default t 560 is used for that event. The default timestamp mode is
692 'delta'. 561 'delta'.
693 562
694 Usual timestamp modes for tracing: 563 Usual timestamp modes for tracing:
695 564
696 # cat timestamp_mode 565 # cat timestamp_mode
697 [delta] absolute 566 [delta] absolute
698 567
699 The timestamp mode with the square b 568 The timestamp mode with the square brackets around it is the
700 one in effect. 569 one in effect.
701 570
702 delta: Default timestamp mode - time 571 delta: Default timestamp mode - timestamp is a delta against
703 a per-buffer timestamp. 572 a per-buffer timestamp.
704 573
705 absolute: The timestamp is a full ti 574 absolute: The timestamp is a full timestamp, not a delta
706 against some other value. As 575 against some other value. As such it takes up more
707 space and is less efficient. 576 space and is less efficient.
708 577
709 hwlat_detector: 578 hwlat_detector:
710 579
711 Directory for the Hardware Latency Det 580 Directory for the Hardware Latency Detector.
712 See "Hardware Latency Detector" sectio 581 See "Hardware Latency Detector" section below.
713 582
714 per_cpu: 583 per_cpu:
715 584
716 This is a directory that contains the 585 This is a directory that contains the trace per_cpu information.
717 586
718 per_cpu/cpu0/buffer_size_kb: 587 per_cpu/cpu0/buffer_size_kb:
719 588
720 The ftrace buffer is defined per_cpu. 589 The ftrace buffer is defined per_cpu. That is, there's a separate
721 buffer for each CPU to allow writes to 590 buffer for each CPU to allow writes to be done atomically,
722 and free from cache bouncing. These bu 591 and free from cache bouncing. These buffers may have different
723 size buffers. This file is similar to 592 size buffers. This file is similar to the buffer_size_kb
724 file, but it only displays or sets the 593 file, but it only displays or sets the buffer size for the
725 specific CPU. (here cpu0). 594 specific CPU. (here cpu0).
726 595
727 per_cpu/cpu0/trace: 596 per_cpu/cpu0/trace:
728 597
729 This is similar to the "trace" file, b 598 This is similar to the "trace" file, but it will only display
730 the data specific for the CPU. If writ 599 the data specific for the CPU. If written to, it only clears
731 the specific CPU buffer. 600 the specific CPU buffer.
732 601
733 per_cpu/cpu0/trace_pipe 602 per_cpu/cpu0/trace_pipe
734 603
735 This is similar to the "trace_pipe" fi 604 This is similar to the "trace_pipe" file, and is a consuming
736 read, but it will only display (and co 605 read, but it will only display (and consume) the data specific
737 for the CPU. 606 for the CPU.
738 607
739 per_cpu/cpu0/trace_pipe_raw 608 per_cpu/cpu0/trace_pipe_raw
740 609
741 For tools that can parse the ftrace ri 610 For tools that can parse the ftrace ring buffer binary format,
742 the trace_pipe_raw file can be used to 611 the trace_pipe_raw file can be used to extract the data
743 from the ring buffer directly. With th 612 from the ring buffer directly. With the use of the splice()
744 system call, the buffer data can be qu 613 system call, the buffer data can be quickly transferred to
745 a file or to the network where a serve 614 a file or to the network where a server is collecting the
746 data. 615 data.
747 616
748 Like trace_pipe, this is a consuming r 617 Like trace_pipe, this is a consuming reader, where multiple
749 reads will always produce different da 618 reads will always produce different data.
750 619
751 per_cpu/cpu0/snapshot: 620 per_cpu/cpu0/snapshot:
752 621
753 This is similar to the main "snapshot" 622 This is similar to the main "snapshot" file, but will only
754 snapshot the current CPU (if supported 623 snapshot the current CPU (if supported). It only displays
755 the content of the snapshot for a give 624 the content of the snapshot for a given CPU, and if
756 written to, only clears this CPU buffe 625 written to, only clears this CPU buffer.
757 626
758 per_cpu/cpu0/snapshot_raw: 627 per_cpu/cpu0/snapshot_raw:
759 628
760 Similar to the trace_pipe_raw, but wil 629 Similar to the trace_pipe_raw, but will read the binary format
761 from the snapshot buffer for the given 630 from the snapshot buffer for the given CPU.
762 631
763 per_cpu/cpu0/stats: 632 per_cpu/cpu0/stats:
764 633
765 This displays certain stats about the 634 This displays certain stats about the ring buffer:
766 635
767 entries: 636 entries:
768 The number of events that are 637 The number of events that are still in the buffer.
769 638
770 overrun: 639 overrun:
771 The number of lost events due 640 The number of lost events due to overwriting when
772 the buffer was full. 641 the buffer was full.
773 642
774 commit overrun: 643 commit overrun:
775 Should always be zero. 644 Should always be zero.
776 This gets set if so many event 645 This gets set if so many events happened within a nested
777 event (ring buffer is re-entra 646 event (ring buffer is re-entrant), that it fills the
778 buffer and starts dropping eve 647 buffer and starts dropping events.
779 648
780 bytes: 649 bytes:
781 Bytes actually read (not overw 650 Bytes actually read (not overwritten).
782 651
783 oldest event ts: 652 oldest event ts:
784 The oldest timestamp in the bu 653 The oldest timestamp in the buffer
785 654
786 now ts: 655 now ts:
787 The current timestamp 656 The current timestamp
788 657
789 dropped events: 658 dropped events:
790 Events lost due to overwrite o 659 Events lost due to overwrite option being off.
791 660
792 read events: 661 read events:
793 The number of events read. 662 The number of events read.
794 663
795 The Tracers 664 The Tracers
796 ----------- 665 -----------
797 666
798 Here is the list of current tracers that may b 667 Here is the list of current tracers that may be configured.
799 668
800 "function" 669 "function"
801 670
802 Function call tracer to trace all kern 671 Function call tracer to trace all kernel functions.
803 672
804 "function_graph" 673 "function_graph"
805 674
806 Similar to the function tracer except 675 Similar to the function tracer except that the
807 function tracer probes the functions o 676 function tracer probes the functions on their entry
808 whereas the function graph tracer trac 677 whereas the function graph tracer traces on both entry
809 and exit of the functions. It then pro 678 and exit of the functions. It then provides the ability
810 to draw a graph of function calls simi 679 to draw a graph of function calls similar to C code
811 source. 680 source.
812 681
813 "blk" 682 "blk"
814 683
815 The block tracer. The tracer used by t 684 The block tracer. The tracer used by the blktrace user
816 application. 685 application.
817 686
818 "hwlat" 687 "hwlat"
819 688
820 The Hardware Latency tracer is used to 689 The Hardware Latency tracer is used to detect if the hardware
821 produces any latency. See "Hardware La 690 produces any latency. See "Hardware Latency Detector" section
822 below. 691 below.
823 692
824 "irqsoff" 693 "irqsoff"
825 694
826 Traces the areas that disable interrup 695 Traces the areas that disable interrupts and saves
827 the trace with the longest max latency 696 the trace with the longest max latency.
828 See tracing_max_latency. When a new ma 697 See tracing_max_latency. When a new max is recorded,
829 it replaces the old trace. It is best 698 it replaces the old trace. It is best to view this
830 trace with the latency-format option e 699 trace with the latency-format option enabled, which
831 happens automatically when the tracer 700 happens automatically when the tracer is selected.
832 701
833 "preemptoff" 702 "preemptoff"
834 703
835 Similar to irqsoff but traces and reco 704 Similar to irqsoff but traces and records the amount of
836 time for which preemption is disabled. 705 time for which preemption is disabled.
837 706
838 "preemptirqsoff" 707 "preemptirqsoff"
839 708
840 Similar to irqsoff and preemptoff, but 709 Similar to irqsoff and preemptoff, but traces and
841 records the largest time for which irq 710 records the largest time for which irqs and/or preemption
842 is disabled. 711 is disabled.
843 712
844 "wakeup" 713 "wakeup"
845 714
846 Traces and records the max latency tha 715 Traces and records the max latency that it takes for
847 the highest priority task to get sched 716 the highest priority task to get scheduled after
848 it has been woken up. 717 it has been woken up.
849 Traces all tasks as an average develop 718 Traces all tasks as an average developer would expect.
850 719
851 "wakeup_rt" 720 "wakeup_rt"
852 721
853 Traces and records the max latency tha 722 Traces and records the max latency that it takes for just
854 RT tasks (as the current "wakeup" does 723 RT tasks (as the current "wakeup" does). This is useful
855 for those interested in wake up timing 724 for those interested in wake up timings of RT tasks.
856 725
857 "wakeup_dl" 726 "wakeup_dl"
858 727
859 Traces and records the max latency tha 728 Traces and records the max latency that it takes for
860 a SCHED_DEADLINE task to be woken (as 729 a SCHED_DEADLINE task to be woken (as the "wakeup" and
861 "wakeup_rt" does). 730 "wakeup_rt" does).
862 731
863 "mmiotrace" 732 "mmiotrace"
864 733
865 A special tracer that is used to trace 734 A special tracer that is used to trace binary module.
866 It will trace all the calls that a mod 735 It will trace all the calls that a module makes to the
867 hardware. Everything it writes and rea 736 hardware. Everything it writes and reads from the I/O
868 as well. 737 as well.
869 738
870 "branch" 739 "branch"
871 740
872 This tracer can be configured when tra 741 This tracer can be configured when tracing likely/unlikely
873 calls within the kernel. It will trace 742 calls within the kernel. It will trace when a likely and
874 unlikely branch is hit and if it was c 743 unlikely branch is hit and if it was correct in its prediction
875 of being correct. 744 of being correct.
876 745
877 "nop" 746 "nop"
878 747
879 This is the "trace nothing" tracer. To 748 This is the "trace nothing" tracer. To remove all
880 tracers from tracing simply echo "nop" 749 tracers from tracing simply echo "nop" into
881 current_tracer. 750 current_tracer.
882 751
883 Error conditions <<
884 ---------------- <<
885 <<
886 For most ftrace commands, failure modes are <<
887 using standard return codes. <<
888 <<
889 For other more involved commands, extended e <<
890 available via the tracing/error_log file. F <<
891 support it, reading the tracing/error_log fi <<
892 display more detailed information about what <<
893 information is available. The tracing/error <<
894 error log displaying a small number (current <<
895 for the last (8) failed commands. <<
896 <<
897 The extended error information and usage tak <<
898 this example:: <<
899 <<
900 # echo xxx > /sys/kernel/tracing/events/sc <<
901 echo: write error: Invalid argument <<
902 <<
903 # cat /sys/kernel/tracing/error_log <<
904 [ 5348.887237] location: error: Couldn't y <<
905 Command: xxx <<
906 ^ <<
907 [ 7517.023364] location: error: Bad rrr: s <<
908 Command: ppp qqq <<
909 ^ <<
910 <<
911 To clear the error log, echo the empty strin <<
912 <<
913 # echo > /sys/kernel/tracing/error_log <<
914 752
915 Examples of using the tracer 753 Examples of using the tracer
916 ---------------------------- 754 ----------------------------
917 755
918 Here are typical examples of using the tracers 756 Here are typical examples of using the tracers when controlling
919 them only with the tracefs interface (without 757 them only with the tracefs interface (without using any
920 user-land utilities). 758 user-land utilities).
921 759
922 Output format: 760 Output format:
923 -------------- 761 --------------
924 762
925 Here is an example of the output format of the 763 Here is an example of the output format of the file "trace"::
926 764
927 # tracer: function 765 # tracer: function
928 # 766 #
929 # entries-in-buffer/entries-written: 140080/ 767 # entries-in-buffer/entries-written: 140080/250280 #P:4
930 # 768 #
931 # _-----=> irqs 769 # _-----=> irqs-off
932 # / _----=> need 770 # / _----=> need-resched
933 # | / _---=> hard 771 # | / _---=> hardirq/softirq
934 # || / _--=> pree 772 # || / _--=> preempt-depth
935 # ||| / delay 773 # ||| / delay
936 # TASK-PID CPU# |||| TIMESTA 774 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
937 # | | | |||| | 775 # | | | |||| | |
938 bash-1977 [000] .... 17284.9936 776 bash-1977 [000] .... 17284.993652: sys_close <-system_call_fastpath
939 bash-1977 [000] .... 17284.9936 777 bash-1977 [000] .... 17284.993653: __close_fd <-sys_close
940 bash-1977 [000] .... 17284.9936 778 bash-1977 [000] .... 17284.993653: _raw_spin_lock <-__close_fd
941 sshd-1974 [003] .... 17284.9936 779 sshd-1974 [003] .... 17284.993653: __srcu_read_unlock <-fsnotify
942 bash-1977 [000] .... 17284.9936 780 bash-1977 [000] .... 17284.993654: add_preempt_count <-_raw_spin_lock
943 bash-1977 [000] ...1 17284.9936 781 bash-1977 [000] ...1 17284.993655: _raw_spin_unlock <-__close_fd
944 bash-1977 [000] ...1 17284.9936 782 bash-1977 [000] ...1 17284.993656: sub_preempt_count <-_raw_spin_unlock
945 bash-1977 [000] .... 17284.9936 783 bash-1977 [000] .... 17284.993657: filp_close <-__close_fd
946 bash-1977 [000] .... 17284.9936 784 bash-1977 [000] .... 17284.993657: dnotify_flush <-filp_close
947 sshd-1974 [003] .... 17284.9936 785 sshd-1974 [003] .... 17284.993658: sys_select <-system_call_fastpath
948 .... 786 ....
949 787
950 A header is printed with the tracer name that 788 A header is printed with the tracer name that is represented by
951 the trace. In this case the tracer is "functio 789 the trace. In this case the tracer is "function". Then it shows the
952 number of events in the buffer as well as the 790 number of events in the buffer as well as the total number of entries
953 that were written. The difference is the numbe 791 that were written. The difference is the number of entries that were
954 lost due to the buffer filling up (250280 - 14 792 lost due to the buffer filling up (250280 - 140080 = 110200 events
955 lost). 793 lost).
956 794
957 The header explains the content of the events. 795 The header explains the content of the events. Task name "bash", the task
958 PID "1977", the CPU that it was running on "00 796 PID "1977", the CPU that it was running on "000", the latency format
959 (explained below), the timestamp in <secs>.<us 797 (explained below), the timestamp in <secs>.<usecs> format, the
960 function name that was traced "sys_close" and 798 function name that was traced "sys_close" and the parent function that
961 called this function "system_call_fastpath". T 799 called this function "system_call_fastpath". The timestamp is the time
962 at which the function was entered. 800 at which the function was entered.
963 801
964 Latency trace format 802 Latency trace format
965 -------------------- 803 --------------------
966 804
967 When the latency-format option is enabled or w 805 When the latency-format option is enabled or when one of the latency
968 tracers is set, the trace file gives somewhat 806 tracers is set, the trace file gives somewhat more information to see
969 why a latency happened. Here is a typical trac 807 why a latency happened. Here is a typical trace::
970 808
971 # tracer: irqsoff 809 # tracer: irqsoff
972 # 810 #
973 # irqsoff latency trace v1.1.5 on 3.8.0-test 811 # irqsoff latency trace v1.1.5 on 3.8.0-test+
974 # ------------------------------------------ 812 # --------------------------------------------------------------------
975 # latency: 259 us, #4/4, CPU#2 | (M:preempt 813 # latency: 259 us, #4/4, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
976 # ----------------- 814 # -----------------
977 # | task: ps-6143 (uid:0 nice:0 policy:0 815 # | task: ps-6143 (uid:0 nice:0 policy:0 rt_prio:0)
978 # ----------------- 816 # -----------------
979 # => started at: __lock_task_sighand 817 # => started at: __lock_task_sighand
980 # => ended at: _raw_spin_unlock_irqrestor 818 # => ended at: _raw_spin_unlock_irqrestore
981 # 819 #
982 # 820 #
983 # _------=> CPU# 821 # _------=> CPU#
984 # / _-----=> irqs-off 822 # / _-----=> irqs-off
985 # | / _----=> need-resched 823 # | / _----=> need-resched
986 # || / _---=> hardirq/softirq 824 # || / _---=> hardirq/softirq
987 # ||| / _--=> preempt-depth 825 # ||| / _--=> preempt-depth
988 # |||| / delay 826 # |||| / delay
989 # cmd pid ||||| time | caller 827 # cmd pid ||||| time | caller
990 # \ / ||||| \ | / 828 # \ / ||||| \ | /
991 ps-6143 2d... 0us!: trace_hardir 829 ps-6143 2d... 0us!: trace_hardirqs_off <-__lock_task_sighand
992 ps-6143 2d..1 259us+: trace_hardir 830 ps-6143 2d..1 259us+: trace_hardirqs_on <-_raw_spin_unlock_irqrestore
993 ps-6143 2d..1 263us+: time_hardirq 831 ps-6143 2d..1 263us+: time_hardirqs_on <-_raw_spin_unlock_irqrestore
994 ps-6143 2d..1 306us : <stack trace 832 ps-6143 2d..1 306us : <stack trace>
995 => trace_hardirqs_on_caller 833 => trace_hardirqs_on_caller
996 => trace_hardirqs_on 834 => trace_hardirqs_on
997 => _raw_spin_unlock_irqrestore 835 => _raw_spin_unlock_irqrestore
998 => do_task_stat 836 => do_task_stat
999 => proc_tgid_stat 837 => proc_tgid_stat
1000 => proc_single_show 838 => proc_single_show
1001 => seq_read 839 => seq_read
1002 => vfs_read 840 => vfs_read
1003 => sys_read 841 => sys_read
1004 => system_call_fastpath 842 => system_call_fastpath
1005 843
1006 844
1007 This shows that the current tracer is "irqsof 845 This shows that the current tracer is "irqsoff" tracing the time
1008 for which interrupts were disabled. It gives 846 for which interrupts were disabled. It gives the trace version (which
1009 never changes) and the version of the kernel 847 never changes) and the version of the kernel upon which this was executed on
1010 (3.8). Then it displays the max latency in mi 848 (3.8). Then it displays the max latency in microseconds (259 us). The number
1011 of trace entries displayed and the total numb 849 of trace entries displayed and the total number (both are four: #4/4).
1012 VP, KP, SP, and HP are always zero and are re 850 VP, KP, SP, and HP are always zero and are reserved for later use.
1013 #P is the number of online CPUs (#P:4). 851 #P is the number of online CPUs (#P:4).
1014 852
1015 The task is the process that was running when 853 The task is the process that was running when the latency
1016 occurred. (ps pid: 6143). 854 occurred. (ps pid: 6143).
1017 855
1018 The start and stop (the functions in which th 856 The start and stop (the functions in which the interrupts were
1019 disabled and enabled respectively) that cause 857 disabled and enabled respectively) that caused the latencies:
1020 858
1021 - __lock_task_sighand is where the interrup 859 - __lock_task_sighand is where the interrupts were disabled.
1022 - _raw_spin_unlock_irqrestore is where they 860 - _raw_spin_unlock_irqrestore is where they were enabled again.
1023 861
1024 The next lines after the header are the trace 862 The next lines after the header are the trace itself. The header
1025 explains which is which. 863 explains which is which.
1026 864
1027 cmd: The name of the process in the trace. 865 cmd: The name of the process in the trace.
1028 866
1029 pid: The PID of that process. 867 pid: The PID of that process.
1030 868
1031 CPU#: The CPU which the process was running 869 CPU#: The CPU which the process was running on.
1032 870
1033 irqs-off: 'd' interrupts are disabled. '.' 871 irqs-off: 'd' interrupts are disabled. '.' otherwise.
1034 .. caution:: If the architecture does 872 .. caution:: If the architecture does not support a way to
1035 read the irq flags variable, 873 read the irq flags variable, an 'X' will always
1036 be printed here. 874 be printed here.
1037 875
1038 need-resched: 876 need-resched:
1039 - 'N' both TIF_NEED_RESCHED and PREEM 877 - 'N' both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set,
1040 - 'n' only TIF_NEED_RESCHED is set, 878 - 'n' only TIF_NEED_RESCHED is set,
1041 - 'p' only PREEMPT_NEED_RESCHED is se 879 - 'p' only PREEMPT_NEED_RESCHED is set,
1042 - '.' otherwise. 880 - '.' otherwise.
1043 881
1044 hardirq/softirq: 882 hardirq/softirq:
1045 - 'Z' - NMI occurred inside a hardirq 883 - 'Z' - NMI occurred inside a hardirq
1046 - 'z' - NMI is running 884 - 'z' - NMI is running
1047 - 'H' - hard irq occurred inside a so 885 - 'H' - hard irq occurred inside a softirq.
1048 - 'h' - hard irq is running 886 - 'h' - hard irq is running
1049 - 's' - soft irq is running 887 - 's' - soft irq is running
1050 - '.' - normal context. 888 - '.' - normal context.
1051 889
1052 preempt-depth: The level of preempt_disable 890 preempt-depth: The level of preempt_disabled
1053 891
1054 The above is mostly meaningful for kernel dev 892 The above is mostly meaningful for kernel developers.
1055 893
1056 time: 894 time:
1057 When the latency-format option is ena 895 When the latency-format option is enabled, the trace file
1058 output includes a timestamp relative 896 output includes a timestamp relative to the start of the
1059 trace. This differs from the output w 897 trace. This differs from the output when latency-format
1060 is disabled, which includes an absolu 898 is disabled, which includes an absolute timestamp.
1061 899
1062 delay: 900 delay:
1063 This is just to help catch your eye a 901 This is just to help catch your eye a bit better. And
1064 needs to be fixed to be only relative 902 needs to be fixed to be only relative to the same CPU.
1065 The marks are determined by the diffe 903 The marks are determined by the difference between this
1066 current trace and the next trace. 904 current trace and the next trace.
1067 905
1068 - '$' - greater than 1 second 906 - '$' - greater than 1 second
1069 - '@' - greater than 100 millisecon !! 907 - '@' - greater than 100 milisecond
1070 - '*' - greater than 10 millisecond !! 908 - '*' - greater than 10 milisecond
1071 - '#' - greater than 1000 microseco 909 - '#' - greater than 1000 microsecond
1072 - '!' - greater than 100 microsecon 910 - '!' - greater than 100 microsecond
1073 - '+' - greater than 10 microsecond 911 - '+' - greater than 10 microsecond
1074 - ' ' - less than or equal to 10 mi 912 - ' ' - less than or equal to 10 microsecond.
1075 913
1076 The rest is the same as the 'trace' file. 914 The rest is the same as the 'trace' file.
1077 915
1078 Note, the latency tracers will usually end 916 Note, the latency tracers will usually end with a back trace
1079 to easily find where the latency occurred. 917 to easily find where the latency occurred.
1080 918
1081 trace_options 919 trace_options
1082 ------------- 920 -------------
1083 921
1084 The trace_options file (or the options direct 922 The trace_options file (or the options directory) is used to control
1085 what gets printed in the trace output, or man 923 what gets printed in the trace output, or manipulate the tracers.
1086 To see what is available, simply cat the file 924 To see what is available, simply cat the file::
1087 925
1088 cat trace_options 926 cat trace_options
1089 print-parent 927 print-parent
1090 nosym-offset 928 nosym-offset
1091 nosym-addr 929 nosym-addr
1092 noverbose 930 noverbose
1093 noraw 931 noraw
1094 nohex 932 nohex
1095 nobin 933 nobin
1096 noblock 934 noblock
1097 nofields <<
1098 trace_printk 935 trace_printk
1099 annotate 936 annotate
1100 nouserstacktrace 937 nouserstacktrace
1101 nosym-userobj 938 nosym-userobj
1102 noprintk-msg-only 939 noprintk-msg-only
1103 context-info 940 context-info
1104 nolatency-format 941 nolatency-format
1105 record-cmd 942 record-cmd
1106 norecord-tgid 943 norecord-tgid
1107 overwrite 944 overwrite
1108 nodisable_on_free 945 nodisable_on_free
1109 irq-info 946 irq-info
1110 markers 947 markers
1111 noevent-fork 948 noevent-fork
1112 function-trace 949 function-trace
1113 nofunction-fork 950 nofunction-fork
1114 nodisplay-graph 951 nodisplay-graph
1115 nostacktrace 952 nostacktrace
1116 nobranch 953 nobranch
1117 954
1118 To disable one of the options, echo in the op 955 To disable one of the options, echo in the option prepended with
1119 "no":: 956 "no"::
1120 957
1121 echo noprint-parent > trace_options 958 echo noprint-parent > trace_options
1122 959
1123 To enable an option, leave off the "no":: 960 To enable an option, leave off the "no"::
1124 961
1125 echo sym-offset > trace_options 962 echo sym-offset > trace_options
1126 963
1127 Here are the available options: 964 Here are the available options:
1128 965
1129 print-parent 966 print-parent
1130 On function traces, display the calli 967 On function traces, display the calling (parent)
1131 function as well as the function bein 968 function as well as the function being traced.
1132 :: 969 ::
1133 970
1134 print-parent: 971 print-parent:
1135 bash-4000 [01] 1477.606694: simp 972 bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul
1136 973
1137 noprint-parent: 974 noprint-parent:
1138 bash-4000 [01] 1477.606694: simp 975 bash-4000 [01] 1477.606694: simple_strtoul
1139 976
1140 977
1141 sym-offset 978 sym-offset
1142 Display not only the function name, b 979 Display not only the function name, but also the
1143 offset in the function. For example, 980 offset in the function. For example, instead of
1144 seeing just "ktime_get", you will see 981 seeing just "ktime_get", you will see
1145 "ktime_get+0xb/0x20". 982 "ktime_get+0xb/0x20".
1146 :: 983 ::
1147 984
1148 sym-offset: 985 sym-offset:
1149 bash-4000 [01] 1477.606694: simp 986 bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0
1150 987
1151 sym-addr 988 sym-addr
1152 This will also display the function a 989 This will also display the function address as well
1153 as the function name. 990 as the function name.
1154 :: 991 ::
1155 992
1156 sym-addr: 993 sym-addr:
1157 bash-4000 [01] 1477.606694: simp 994 bash-4000 [01] 1477.606694: simple_strtoul <c0339346>
1158 995
1159 verbose 996 verbose
1160 This deals with the trace file when t 997 This deals with the trace file when the
1161 latency-format option is enabled. 998 latency-format option is enabled.
1162 :: 999 ::
1163 1000
1164 bash 4000 1 0 00000000 00010a95 1001 bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \
1165 (+0.000ms): simple_strtoul (kstrt 1002 (+0.000ms): simple_strtoul (kstrtoul)
1166 1003
1167 raw 1004 raw
1168 This will display raw numbers. This o 1005 This will display raw numbers. This option is best for
1169 use with user applications that can t 1006 use with user applications that can translate the raw
1170 numbers better than having it done in 1007 numbers better than having it done in the kernel.
1171 1008
1172 hex 1009 hex
1173 Similar to raw, but the numbers will 1010 Similar to raw, but the numbers will be in a hexadecimal format.
1174 1011
1175 bin 1012 bin
1176 This will print out the formats in ra 1013 This will print out the formats in raw binary.
1177 1014
1178 block 1015 block
1179 When set, reading trace_pipe will not 1016 When set, reading trace_pipe will not block when polled.
1180 1017
1181 fields <<
1182 Print the fields as described by thei <<
1183 option than using hex, bin or raw, as <<
1184 of the content of the event. <<
1185 <<
1186 trace_printk 1018 trace_printk
1187 Can disable trace_printk() from writi 1019 Can disable trace_printk() from writing into the buffer.
1188 1020
1189 trace_printk_dest <<
1190 Set to have trace_printk() and simila <<
1191 write into this instance. Note, only <<
1192 this set. By setting this flag, it cl <<
1193 of the instance that had it set previ <<
1194 level trace has this set, and will ge <<
1195 instance has it set then clears it. <<
1196 <<
1197 This flag cannot be cleared by the to <<
1198 default instance. The only way the to <<
1199 cleared, is by it being set in anothe <<
1200 <<
1201 annotate 1021 annotate
1202 It is sometimes confusing when the CP 1022 It is sometimes confusing when the CPU buffers are full
1203 and one CPU buffer had a lot of event 1023 and one CPU buffer had a lot of events recently, thus
1204 a shorter time frame, were another CP 1024 a shorter time frame, were another CPU may have only had
1205 a few events, which lets it have olde 1025 a few events, which lets it have older events. When
1206 the trace is reported, it shows the o 1026 the trace is reported, it shows the oldest events first,
1207 and it may look like only one CPU ran 1027 and it may look like only one CPU ran (the one with the
1208 oldest events). When the annotate opt 1028 oldest events). When the annotate option is set, it will
1209 display when a new CPU buffer started 1029 display when a new CPU buffer started::
1210 1030
1211 <idle>-0 [001] 1031 <idle>-0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on
1212 <idle>-0 [001] 1032 <idle>-0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on
1213 <idle>-0 [001] 1033 <idle>-0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore
1214 ##### CPU 2 buffer started ## 1034 ##### CPU 2 buffer started ####
1215 <idle>-0 [002] 1035 <idle>-0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle
1216 <idle>-0 [001] 1036 <idle>-0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog
1217 <idle>-0 [001] 1037 <idle>-0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlock
1218 1038
1219 userstacktrace 1039 userstacktrace
1220 This option changes the trace. It rec 1040 This option changes the trace. It records a
1221 stacktrace of the current user space 1041 stacktrace of the current user space thread after
1222 each trace event. 1042 each trace event.
1223 1043
1224 sym-userobj 1044 sym-userobj
1225 when user stacktrace are enabled, loo 1045 when user stacktrace are enabled, look up which
1226 object the address belongs to, and pr 1046 object the address belongs to, and print a
1227 relative address. This is especially 1047 relative address. This is especially useful when
1228 ASLR is on, otherwise you don't get a 1048 ASLR is on, otherwise you don't get a chance to
1229 resolve the address to object/file/li 1049 resolve the address to object/file/line after
1230 the app is no longer running 1050 the app is no longer running
1231 1051
1232 The lookup is performed when you read 1052 The lookup is performed when you read
1233 trace,trace_pipe. Example:: 1053 trace,trace_pipe. Example::
1234 1054
1235 a.out-1623 [000] 40874.465 1055 a.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0
1236 x494] <- /root/a.out[+0x4a8 1056 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]
1237 1057
1238 1058
1239 printk-msg-only 1059 printk-msg-only
1240 When set, trace_printk()s will only s 1060 When set, trace_printk()s will only show the format
1241 and not their parameters (if trace_bp 1061 and not their parameters (if trace_bprintk() or
1242 trace_bputs() was used to save the tr 1062 trace_bputs() was used to save the trace_printk()).
1243 1063
1244 context-info 1064 context-info
1245 Show only the event data. Hides the c 1065 Show only the event data. Hides the comm, PID,
1246 timestamp, CPU, and other useful data 1066 timestamp, CPU, and other useful data.
1247 1067
1248 latency-format 1068 latency-format
1249 This option changes the trace output. 1069 This option changes the trace output. When it is enabled,
1250 the trace displays additional informa 1070 the trace displays additional information about the
1251 latency, as described in "Latency tra 1071 latency, as described in "Latency trace format".
1252 1072
1253 pause-on-trace <<
1254 When set, opening the trace file for <<
1255 writing to the ring buffer (as if tra <<
1256 This simulates the original behavior <<
1257 When the file is closed, tracing will <<
1258 <<
1259 hash-ptr <<
1260 When set, "%p" in the event printk fo <<
1261 hashed pointer value instead of real <<
1262 This will be useful if you want to fi <<
1263 value is corresponding to the real va <<
1264 <<
1265 record-cmd 1073 record-cmd
1266 When any event or tracer is enabled, 1074 When any event or tracer is enabled, a hook is enabled
1267 in the sched_switch trace point to fi 1075 in the sched_switch trace point to fill comm cache
1268 with mapped pids and comms. But this 1076 with mapped pids and comms. But this may cause some
1269 overhead, and if you only care about 1077 overhead, and if you only care about pids, and not the
1270 name of the task, disabling this opti 1078 name of the task, disabling this option can lower the
1271 impact of tracing. See "saved_cmdline 1079 impact of tracing. See "saved_cmdlines".
1272 1080
1273 record-tgid 1081 record-tgid
1274 When any event or tracer is enabled, 1082 When any event or tracer is enabled, a hook is enabled
1275 in the sched_switch trace point to fi 1083 in the sched_switch trace point to fill the cache of
1276 mapped Thread Group IDs (TGID) mappin 1084 mapped Thread Group IDs (TGID) mapping to pids. See
1277 "saved_tgids". 1085 "saved_tgids".
1278 1086
1279 overwrite 1087 overwrite
1280 This controls what happens when the t 1088 This controls what happens when the trace buffer is
1281 full. If "1" (default), the oldest ev 1089 full. If "1" (default), the oldest events are
1282 discarded and overwritten. If "0", th 1090 discarded and overwritten. If "0", then the newest
1283 events are discarded. 1091 events are discarded.
1284 (see per_cpu/cpu0/stats for overrun a 1092 (see per_cpu/cpu0/stats for overrun and dropped)
1285 1093
1286 disable_on_free 1094 disable_on_free
1287 When the free_buffer is closed, traci 1095 When the free_buffer is closed, tracing will
1288 stop (tracing_on set to 0). 1096 stop (tracing_on set to 0).
1289 1097
1290 irq-info 1098 irq-info
1291 Shows the interrupt, preempt count, n 1099 Shows the interrupt, preempt count, need resched data.
1292 When disabled, the trace looks like:: 1100 When disabled, the trace looks like::
1293 1101
1294 # tracer: function 1102 # tracer: function
1295 # 1103 #
1296 # entries-in-buffer/entries-w 1104 # entries-in-buffer/entries-written: 144405/9452052 #P:4
1297 # 1105 #
1298 # TASK-PID CPU# 1106 # TASK-PID CPU# TIMESTAMP FUNCTION
1299 # | | | 1107 # | | | | |
1300 <idle>-0 [002] 1108 <idle>-0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up
1301 <idle>-0 [002] 1109 <idle>-0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89
1302 <idle>-0 [002] 1110 <idle>-0 [002] 23636.756055: enqueue_task <-activate_task
1303 1111
1304 1112
1305 markers 1113 markers
1306 When set, the trace_marker is writabl 1114 When set, the trace_marker is writable (only by root).
1307 When disabled, the trace_marker will 1115 When disabled, the trace_marker will error with EINVAL
1308 on write. 1116 on write.
1309 1117
1310 event-fork 1118 event-fork
1311 When set, tasks with PIDs listed in s 1119 When set, tasks with PIDs listed in set_event_pid will have
1312 the PIDs of their children added to s 1120 the PIDs of their children added to set_event_pid when those
1313 tasks fork. Also, when tasks with PID 1121 tasks fork. Also, when tasks with PIDs in set_event_pid exit,
1314 their PIDs will be removed from the f 1122 their PIDs will be removed from the file.
1315 1123
1316 This affects PIDs listed in set_event <<
1317 <<
1318 function-trace 1124 function-trace
1319 The latency tracers will enable funct 1125 The latency tracers will enable function tracing
1320 if this option is enabled (default it 1126 if this option is enabled (default it is). When
1321 it is disabled, the latency tracers d 1127 it is disabled, the latency tracers do not trace
1322 functions. This keeps the overhead of 1128 functions. This keeps the overhead of the tracer down
1323 when performing latency tests. 1129 when performing latency tests.
1324 1130
1325 function-fork 1131 function-fork
1326 When set, tasks with PIDs listed in s 1132 When set, tasks with PIDs listed in set_ftrace_pid will
1327 have the PIDs of their children added 1133 have the PIDs of their children added to set_ftrace_pid
1328 when those tasks fork. Also, when tas 1134 when those tasks fork. Also, when tasks with PIDs in
1329 set_ftrace_pid exit, their PIDs will 1135 set_ftrace_pid exit, their PIDs will be removed from the
1330 file. 1136 file.
1331 1137
1332 This affects PIDs in set_ftrace_notra <<
1333 <<
1334 display-graph 1138 display-graph
1335 When set, the latency tracers (irqsof 1139 When set, the latency tracers (irqsoff, wakeup, etc) will
1336 use function graph tracing instead of 1140 use function graph tracing instead of function tracing.
1337 1141
1338 stacktrace 1142 stacktrace
1339 When set, a stack trace is recorded a 1143 When set, a stack trace is recorded after any trace event
1340 is recorded. 1144 is recorded.
1341 1145
1342 branch 1146 branch
1343 Enable branch tracing with the tracer 1147 Enable branch tracing with the tracer. This enables branch
1344 tracer along with the currently set t 1148 tracer along with the currently set tracer. Enabling this
1345 with the "nop" tracer is the same as 1149 with the "nop" tracer is the same as just enabling the
1346 "branch" tracer. 1150 "branch" tracer.
1347 1151
1348 .. tip:: Some tracers have their own options. 1152 .. tip:: Some tracers have their own options. They only appear in this
1349 file when the tracer is active. They a 1153 file when the tracer is active. They always appear in the
1350 options directory. 1154 options directory.
1351 1155
1352 1156
1353 Here are the per tracer options: 1157 Here are the per tracer options:
1354 1158
1355 Options for function tracer: 1159 Options for function tracer:
1356 1160
1357 func_stack_trace 1161 func_stack_trace
1358 When set, a stack trace is recorded a 1162 When set, a stack trace is recorded after every
1359 function that is recorded. NOTE! Limi 1163 function that is recorded. NOTE! Limit the functions
1360 that are recorded before enabling thi 1164 that are recorded before enabling this, with
1361 "set_ftrace_filter" otherwise the sys 1165 "set_ftrace_filter" otherwise the system performance
1362 will be critically degraded. Remember 1166 will be critically degraded. Remember to disable
1363 this option before clearing the funct 1167 this option before clearing the function filter.
1364 1168
1365 Options for function_graph tracer: 1169 Options for function_graph tracer:
1366 1170
1367 Since the function_graph tracer has a slight 1171 Since the function_graph tracer has a slightly different output
1368 it has its own options to control what is di 1172 it has its own options to control what is displayed.
1369 1173
1370 funcgraph-overrun 1174 funcgraph-overrun
1371 When set, the "overrun" of the graph 1175 When set, the "overrun" of the graph stack is
1372 displayed after each function traced. 1176 displayed after each function traced. The
1373 overrun, is when the stack depth of t 1177 overrun, is when the stack depth of the calls
1374 is greater than what is reserved for 1178 is greater than what is reserved for each task.
1375 Each task has a fixed array of functi 1179 Each task has a fixed array of functions to
1376 trace in the call graph. If the depth 1180 trace in the call graph. If the depth of the
1377 calls exceeds that, the function is n 1181 calls exceeds that, the function is not traced.
1378 The overrun is the number of function 1182 The overrun is the number of functions missed
1379 due to exceeding this array. 1183 due to exceeding this array.
1380 1184
1381 funcgraph-cpu 1185 funcgraph-cpu
1382 When set, the CPU number of the CPU w 1186 When set, the CPU number of the CPU where the trace
1383 occurred is displayed. 1187 occurred is displayed.
1384 1188
1385 funcgraph-overhead 1189 funcgraph-overhead
1386 When set, if the function takes longe 1190 When set, if the function takes longer than
1387 A certain amount, then a delay marker 1191 A certain amount, then a delay marker is
1388 displayed. See "delay" above, under t 1192 displayed. See "delay" above, under the
1389 header description. 1193 header description.
1390 1194
1391 funcgraph-proc 1195 funcgraph-proc
1392 Unlike other tracers, the process' co 1196 Unlike other tracers, the process' command line
1393 is not displayed by default, but inst 1197 is not displayed by default, but instead only
1394 when a task is traced in and out duri 1198 when a task is traced in and out during a context
1395 switch. Enabling this options has the 1199 switch. Enabling this options has the command
1396 of each process displayed at every li 1200 of each process displayed at every line.
1397 1201
1398 funcgraph-duration 1202 funcgraph-duration
1399 At the end of each function (the retu 1203 At the end of each function (the return)
1400 the duration of the amount of time in 1204 the duration of the amount of time in the
1401 function is displayed in microseconds 1205 function is displayed in microseconds.
1402 1206
1403 funcgraph-abstime 1207 funcgraph-abstime
1404 When set, the timestamp is displayed 1208 When set, the timestamp is displayed at each line.
1405 1209
1406 funcgraph-irqs 1210 funcgraph-irqs
1407 When disabled, functions that happen 1211 When disabled, functions that happen inside an
1408 interrupt will not be traced. 1212 interrupt will not be traced.
1409 1213
1410 funcgraph-tail 1214 funcgraph-tail
1411 When set, the return event will inclu 1215 When set, the return event will include the function
1412 that it represents. By default this i 1216 that it represents. By default this is off, and
1413 only a closing curly bracket "}" is d 1217 only a closing curly bracket "}" is displayed for
1414 the return of a function. 1218 the return of a function.
1415 1219
1416 funcgraph-retval <<
1417 When set, the return value of each tr <<
1418 will be printed after an equal sign " <<
1419 this is off. <<
1420 <<
1421 funcgraph-retval-hex <<
1422 When set, the return value will alway <<
1423 in hexadecimal format. If the option <<
1424 the return value is an error code, it <<
1425 in signed decimal format; otherwise i <<
1426 printed in hexadecimal format. By def <<
1427 is off. <<
1428 <<
1429 sleep-time 1220 sleep-time
1430 When running function graph tracer, t 1221 When running function graph tracer, to include
1431 the time a task schedules out in its 1222 the time a task schedules out in its function.
1432 When enabled, it will account time th 1223 When enabled, it will account time the task has been
1433 scheduled out as part of the function 1224 scheduled out as part of the function call.
1434 1225
1435 graph-time 1226 graph-time
1436 When running function profiler with f 1227 When running function profiler with function graph tracer,
1437 to include the time to call nested fu 1228 to include the time to call nested functions. When this is
1438 not set, the time reported for the fu 1229 not set, the time reported for the function will only
1439 include the time the function itself 1230 include the time the function itself executed for, not the
1440 time for functions that it called. 1231 time for functions that it called.
1441 1232
1442 Options for blk tracer: 1233 Options for blk tracer:
1443 1234
1444 blk_classic 1235 blk_classic
1445 Shows a more minimalistic output. 1236 Shows a more minimalistic output.
1446 1237
1447 1238
1448 irqsoff 1239 irqsoff
1449 ------- 1240 -------
1450 1241
1451 When interrupts are disabled, the CPU can not 1242 When interrupts are disabled, the CPU can not react to any other
1452 external event (besides NMIs and SMIs). This 1243 external event (besides NMIs and SMIs). This prevents the timer
1453 interrupt from triggering or the mouse interr 1244 interrupt from triggering or the mouse interrupt from letting
1454 the kernel know of a new mouse event. The res 1245 the kernel know of a new mouse event. The result is a latency
1455 with the reaction time. 1246 with the reaction time.
1456 1247
1457 The irqsoff tracer tracks the time for which 1248 The irqsoff tracer tracks the time for which interrupts are
1458 disabled. When a new maximum latency is hit, 1249 disabled. When a new maximum latency is hit, the tracer saves
1459 the trace leading up to that latency point so 1250 the trace leading up to that latency point so that every time a
1460 new maximum is reached, the old saved trace i 1251 new maximum is reached, the old saved trace is discarded and the
1461 new trace is saved. 1252 new trace is saved.
1462 1253
1463 To reset the maximum, echo 0 into tracing_max 1254 To reset the maximum, echo 0 into tracing_max_latency. Here is
1464 an example:: 1255 an example::
1465 1256
1466 # echo 0 > options/function-trace 1257 # echo 0 > options/function-trace
1467 # echo irqsoff > current_tracer 1258 # echo irqsoff > current_tracer
1468 # echo 1 > tracing_on 1259 # echo 1 > tracing_on
1469 # echo 0 > tracing_max_latency 1260 # echo 0 > tracing_max_latency
1470 # ls -ltr 1261 # ls -ltr
1471 [...] 1262 [...]
1472 # echo 0 > tracing_on 1263 # echo 0 > tracing_on
1473 # cat trace 1264 # cat trace
1474 # tracer: irqsoff 1265 # tracer: irqsoff
1475 # 1266 #
1476 # irqsoff latency trace v1.1.5 on 3.8.0-tes 1267 # irqsoff latency trace v1.1.5 on 3.8.0-test+
1477 # ----------------------------------------- 1268 # --------------------------------------------------------------------
1478 # latency: 16 us, #4/4, CPU#0 | (M:preempt 1269 # latency: 16 us, #4/4, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1479 # ----------------- 1270 # -----------------
1480 # | task: swapper/0-0 (uid:0 nice:0 poli 1271 # | task: swapper/0-0 (uid:0 nice:0 policy:0 rt_prio:0)
1481 # ----------------- 1272 # -----------------
1482 # => started at: run_timer_softirq 1273 # => started at: run_timer_softirq
1483 # => ended at: run_timer_softirq 1274 # => ended at: run_timer_softirq
1484 # 1275 #
1485 # 1276 #
1486 # _------=> CPU# 1277 # _------=> CPU#
1487 # / _-----=> irqs-off 1278 # / _-----=> irqs-off
1488 # | / _----=> need-resched 1279 # | / _----=> need-resched
1489 # || / _---=> hardirq/softir 1280 # || / _---=> hardirq/softirq
1490 # ||| / _--=> preempt-depth 1281 # ||| / _--=> preempt-depth
1491 # |||| / delay 1282 # |||| / delay
1492 # cmd pid ||||| time | caller 1283 # cmd pid ||||| time | caller
1493 # \ / ||||| \ | / 1284 # \ / ||||| \ | /
1494 <idle>-0 0d.s2 0us+: _raw_spin_l 1285 <idle>-0 0d.s2 0us+: _raw_spin_lock_irq <-run_timer_softirq
1495 <idle>-0 0dNs3 17us : _raw_spin_u 1286 <idle>-0 0dNs3 17us : _raw_spin_unlock_irq <-run_timer_softirq
1496 <idle>-0 0dNs3 17us+: trace_hardi 1287 <idle>-0 0dNs3 17us+: trace_hardirqs_on <-run_timer_softirq
1497 <idle>-0 0dNs3 25us : <stack trac 1288 <idle>-0 0dNs3 25us : <stack trace>
1498 => _raw_spin_unlock_irq 1289 => _raw_spin_unlock_irq
1499 => run_timer_softirq 1290 => run_timer_softirq
1500 => __do_softirq 1291 => __do_softirq
1501 => call_softirq 1292 => call_softirq
1502 => do_softirq 1293 => do_softirq
1503 => irq_exit 1294 => irq_exit
1504 => smp_apic_timer_interrupt 1295 => smp_apic_timer_interrupt
1505 => apic_timer_interrupt 1296 => apic_timer_interrupt
1506 => rcu_idle_exit 1297 => rcu_idle_exit
1507 => cpu_idle 1298 => cpu_idle
1508 => rest_init 1299 => rest_init
1509 => start_kernel 1300 => start_kernel
1510 => x86_64_start_reservations 1301 => x86_64_start_reservations
1511 => x86_64_start_kernel 1302 => x86_64_start_kernel
1512 1303
1513 Here we see that we had a latency of 16 micro !! 1304 Here we see that that we had a latency of 16 microseconds (which is
1514 very good). The _raw_spin_lock_irq in run_tim 1305 very good). The _raw_spin_lock_irq in run_timer_softirq disabled
1515 interrupts. The difference between the 16 and 1306 interrupts. The difference between the 16 and the displayed
1516 timestamp 25us occurred because the clock was 1307 timestamp 25us occurred because the clock was incremented
1517 between the time of recording the max latency 1308 between the time of recording the max latency and the time of
1518 recording the function that had that latency. 1309 recording the function that had that latency.
1519 1310
1520 Note the above example had function-trace not 1311 Note the above example had function-trace not set. If we set
1521 function-trace, we get a much larger output:: 1312 function-trace, we get a much larger output::
1522 1313
1523 with echo 1 > options/function-trace 1314 with echo 1 > options/function-trace
1524 1315
1525 # tracer: irqsoff 1316 # tracer: irqsoff
1526 # 1317 #
1527 # irqsoff latency trace v1.1.5 on 3.8.0-tes 1318 # irqsoff latency trace v1.1.5 on 3.8.0-test+
1528 # ----------------------------------------- 1319 # --------------------------------------------------------------------
1529 # latency: 71 us, #168/168, CPU#3 | (M:pree 1320 # latency: 71 us, #168/168, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1530 # ----------------- 1321 # -----------------
1531 # | task: bash-2042 (uid:0 nice:0 policy 1322 # | task: bash-2042 (uid:0 nice:0 policy:0 rt_prio:0)
1532 # ----------------- 1323 # -----------------
1533 # => started at: ata_scsi_queuecmd 1324 # => started at: ata_scsi_queuecmd
1534 # => ended at: ata_scsi_queuecmd 1325 # => ended at: ata_scsi_queuecmd
1535 # 1326 #
1536 # 1327 #
1537 # _------=> CPU# 1328 # _------=> CPU#
1538 # / _-----=> irqs-off 1329 # / _-----=> irqs-off
1539 # | / _----=> need-resched 1330 # | / _----=> need-resched
1540 # || / _---=> hardirq/softir 1331 # || / _---=> hardirq/softirq
1541 # ||| / _--=> preempt-depth 1332 # ||| / _--=> preempt-depth
1542 # |||| / delay 1333 # |||| / delay
1543 # cmd pid ||||| time | caller 1334 # cmd pid ||||| time | caller
1544 # \ / ||||| \ | / 1335 # \ / ||||| \ | /
1545 bash-2042 3d... 0us : _raw_spin_l 1336 bash-2042 3d... 0us : _raw_spin_lock_irqsave <-ata_scsi_queuecmd
1546 bash-2042 3d... 0us : add_preempt 1337 bash-2042 3d... 0us : add_preempt_count <-_raw_spin_lock_irqsave
1547 bash-2042 3d..1 1us : ata_scsi_fi 1338 bash-2042 3d..1 1us : ata_scsi_find_dev <-ata_scsi_queuecmd
1548 bash-2042 3d..1 1us : __ata_scsi_ 1339 bash-2042 3d..1 1us : __ata_scsi_find_dev <-ata_scsi_find_dev
1549 bash-2042 3d..1 2us : ata_find_de 1340 bash-2042 3d..1 2us : ata_find_dev.part.14 <-__ata_scsi_find_dev
1550 bash-2042 3d..1 2us : ata_qc_new_ 1341 bash-2042 3d..1 2us : ata_qc_new_init <-__ata_scsi_queuecmd
1551 bash-2042 3d..1 3us : ata_sg_init 1342 bash-2042 3d..1 3us : ata_sg_init <-__ata_scsi_queuecmd
1552 bash-2042 3d..1 4us : ata_scsi_rw 1343 bash-2042 3d..1 4us : ata_scsi_rw_xlat <-__ata_scsi_queuecmd
1553 bash-2042 3d..1 4us : ata_build_r 1344 bash-2042 3d..1 4us : ata_build_rw_tf <-ata_scsi_rw_xlat
1554 [...] 1345 [...]
1555 bash-2042 3d..1 67us : delay_tsc < 1346 bash-2042 3d..1 67us : delay_tsc <-__delay
1556 bash-2042 3d..1 67us : add_preempt 1347 bash-2042 3d..1 67us : add_preempt_count <-delay_tsc
1557 bash-2042 3d..2 67us : sub_preempt 1348 bash-2042 3d..2 67us : sub_preempt_count <-delay_tsc
1558 bash-2042 3d..1 67us : add_preempt 1349 bash-2042 3d..1 67us : add_preempt_count <-delay_tsc
1559 bash-2042 3d..2 68us : sub_preempt 1350 bash-2042 3d..2 68us : sub_preempt_count <-delay_tsc
1560 bash-2042 3d..1 68us+: ata_bmdma_s 1351 bash-2042 3d..1 68us+: ata_bmdma_start <-ata_bmdma_qc_issue
1561 bash-2042 3d..1 71us : _raw_spin_u 1352 bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1562 bash-2042 3d..1 71us : _raw_spin_u 1353 bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1563 bash-2042 3d..1 72us+: trace_hardi 1354 bash-2042 3d..1 72us+: trace_hardirqs_on <-ata_scsi_queuecmd
1564 bash-2042 3d..1 120us : <stack trac 1355 bash-2042 3d..1 120us : <stack trace>
1565 => _raw_spin_unlock_irqrestore 1356 => _raw_spin_unlock_irqrestore
1566 => ata_scsi_queuecmd 1357 => ata_scsi_queuecmd
1567 => scsi_dispatch_cmd 1358 => scsi_dispatch_cmd
1568 => scsi_request_fn 1359 => scsi_request_fn
1569 => __blk_run_queue_uncond 1360 => __blk_run_queue_uncond
1570 => __blk_run_queue 1361 => __blk_run_queue
1571 => blk_queue_bio 1362 => blk_queue_bio
1572 => submit_bio_noacct !! 1363 => generic_make_request
1573 => submit_bio 1364 => submit_bio
1574 => submit_bh 1365 => submit_bh
1575 => __ext3_get_inode_loc 1366 => __ext3_get_inode_loc
1576 => ext3_iget 1367 => ext3_iget
1577 => ext3_lookup 1368 => ext3_lookup
1578 => lookup_real 1369 => lookup_real
1579 => __lookup_hash 1370 => __lookup_hash
1580 => walk_component 1371 => walk_component
1581 => lookup_last 1372 => lookup_last
1582 => path_lookupat 1373 => path_lookupat
1583 => filename_lookup 1374 => filename_lookup
1584 => user_path_at_empty 1375 => user_path_at_empty
1585 => user_path_at 1376 => user_path_at
1586 => vfs_fstatat 1377 => vfs_fstatat
1587 => vfs_stat 1378 => vfs_stat
1588 => sys_newstat 1379 => sys_newstat
1589 => system_call_fastpath 1380 => system_call_fastpath
1590 1381
1591 1382
1592 Here we traced a 71 microsecond latency. But 1383 Here we traced a 71 microsecond latency. But we also see all the
1593 functions that were called during that time. 1384 functions that were called during that time. Note that by
1594 enabling function tracing, we incur an added 1385 enabling function tracing, we incur an added overhead. This
1595 overhead may extend the latency times. But ne 1386 overhead may extend the latency times. But nevertheless, this
1596 trace has provided some very helpful debuggin 1387 trace has provided some very helpful debugging information.
1597 1388
1598 If we prefer function graph output instead of <<
1599 display-graph option:: <<
1600 <<
1601 with echo 1 > options/display-graph <<
1602 <<
1603 # tracer: irqsoff <<
1604 # <<
1605 # irqsoff latency trace v1.1.5 on 4.20.0-rc <<
1606 # ----------------------------------------- <<
1607 # latency: 3751 us, #274/274, CPU#0 | (M:de <<
1608 # ----------------- <<
1609 # | task: bash-1507 (uid:0 nice:0 policy <<
1610 # ----------------- <<
1611 # => started at: free_debug_processing <<
1612 # => ended at: return_to_handler <<
1613 # <<
1614 # <<
1615 # _-- <<
1616 # / _- <<
1617 # | / _ <<
1618 # || / <<
1619 # ||| / <<
1620 # REL TIME CPU TASK/PID |||| <<
1621 # | | | | |||| <<
1622 0 us | 0) bash-1507 | d... <<
1623 0 us | 0) bash-1507 | d..1 <<
1624 1 us | 0) bash-1507 | d..2 <<
1625 2 us | 0) bash-1507 | d..2 <<
1626 2 us | 0) bash-1507 | d..2 <<
1627 3 us | 0) bash-1507 | d..2 <<
1628 3 us | 0) bash-1507 | d..2 <<
1629 3 us | 0) bash-1507 | d..2 <<
1630 4 us | 0) bash-1507 | d..2 <<
1631 [...] <<
1632 3750 us | 0) bash-1507 | d..1 <<
1633 3750 us | 0) bash-1507 | d..1 <<
1634 3764 us | 0) bash-1507 | d..1 <<
1635 bash-1507 0d..1 3792us : <stack trac <<
1636 => free_debug_processing <<
1637 => __slab_free <<
1638 => kmem_cache_free <<
1639 => vm_area_free <<
1640 => remove_vma <<
1641 => exit_mmap <<
1642 => mmput <<
1643 => begin_new_exec <<
1644 => load_elf_binary <<
1645 => search_binary_handler <<
1646 => __do_execve_file.isra.32 <<
1647 => __x64_sys_execve <<
1648 => do_syscall_64 <<
1649 => entry_SYSCALL_64_after_hwframe <<
1650 1389
1651 preemptoff 1390 preemptoff
1652 ---------- 1391 ----------
1653 1392
1654 When preemption is disabled, we may be able t 1393 When preemption is disabled, we may be able to receive
1655 interrupts but the task cannot be preempted a 1394 interrupts but the task cannot be preempted and a higher
1656 priority task must wait for preemption to be 1395 priority task must wait for preemption to be enabled again
1657 before it can preempt a lower priority task. 1396 before it can preempt a lower priority task.
1658 1397
1659 The preemptoff tracer traces the places that 1398 The preemptoff tracer traces the places that disable preemption.
1660 Like the irqsoff tracer, it records the maxim 1399 Like the irqsoff tracer, it records the maximum latency for
1661 which preemption was disabled. The control of 1400 which preemption was disabled. The control of preemptoff tracer
1662 is much like the irqsoff tracer. 1401 is much like the irqsoff tracer.
1663 :: 1402 ::
1664 1403
1665 # echo 0 > options/function-trace 1404 # echo 0 > options/function-trace
1666 # echo preemptoff > current_tracer 1405 # echo preemptoff > current_tracer
1667 # echo 1 > tracing_on 1406 # echo 1 > tracing_on
1668 # echo 0 > tracing_max_latency 1407 # echo 0 > tracing_max_latency
1669 # ls -ltr 1408 # ls -ltr
1670 [...] 1409 [...]
1671 # echo 0 > tracing_on 1410 # echo 0 > tracing_on
1672 # cat trace 1411 # cat trace
1673 # tracer: preemptoff 1412 # tracer: preemptoff
1674 # 1413 #
1675 # preemptoff latency trace v1.1.5 on 3.8.0- 1414 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1676 # ----------------------------------------- 1415 # --------------------------------------------------------------------
1677 # latency: 46 us, #4/4, CPU#1 | (M:preempt 1416 # latency: 46 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1678 # ----------------- 1417 # -----------------
1679 # | task: sshd-1991 (uid:0 nice:0 policy 1418 # | task: sshd-1991 (uid:0 nice:0 policy:0 rt_prio:0)
1680 # ----------------- 1419 # -----------------
1681 # => started at: do_IRQ 1420 # => started at: do_IRQ
1682 # => ended at: do_IRQ 1421 # => ended at: do_IRQ
1683 # 1422 #
1684 # 1423 #
1685 # _------=> CPU# 1424 # _------=> CPU#
1686 # / _-----=> irqs-off 1425 # / _-----=> irqs-off
1687 # | / _----=> need-resched 1426 # | / _----=> need-resched
1688 # || / _---=> hardirq/softir 1427 # || / _---=> hardirq/softirq
1689 # ||| / _--=> preempt-depth 1428 # ||| / _--=> preempt-depth
1690 # |||| / delay 1429 # |||| / delay
1691 # cmd pid ||||| time | caller 1430 # cmd pid ||||| time | caller
1692 # \ / ||||| \ | / 1431 # \ / ||||| \ | /
1693 sshd-1991 1d.h. 0us+: irq_enter < 1432 sshd-1991 1d.h. 0us+: irq_enter <-do_IRQ
1694 sshd-1991 1d..1 46us : irq_exit <- 1433 sshd-1991 1d..1 46us : irq_exit <-do_IRQ
1695 sshd-1991 1d..1 47us+: trace_preem 1434 sshd-1991 1d..1 47us+: trace_preempt_on <-do_IRQ
1696 sshd-1991 1d..1 52us : <stack trac 1435 sshd-1991 1d..1 52us : <stack trace>
1697 => sub_preempt_count 1436 => sub_preempt_count
1698 => irq_exit 1437 => irq_exit
1699 => do_IRQ 1438 => do_IRQ
1700 => ret_from_intr 1439 => ret_from_intr
1701 1440
1702 1441
1703 This has some more changes. Preemption was di 1442 This has some more changes. Preemption was disabled when an
1704 interrupt came in (notice the 'h'), and was e 1443 interrupt came in (notice the 'h'), and was enabled on exit.
1705 But we also see that interrupts have been dis 1444 But we also see that interrupts have been disabled when entering
1706 the preempt off section and leaving it (the ' 1445 the preempt off section and leaving it (the 'd'). We do not know if
1707 interrupts were enabled in the mean time or s 1446 interrupts were enabled in the mean time or shortly after this
1708 was over. 1447 was over.
1709 :: 1448 ::
1710 1449
1711 # tracer: preemptoff 1450 # tracer: preemptoff
1712 # 1451 #
1713 # preemptoff latency trace v1.1.5 on 3.8.0- 1452 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1714 # ----------------------------------------- 1453 # --------------------------------------------------------------------
1715 # latency: 83 us, #241/241, CPU#1 | (M:pree 1454 # latency: 83 us, #241/241, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1716 # ----------------- 1455 # -----------------
1717 # | task: bash-1994 (uid:0 nice:0 policy 1456 # | task: bash-1994 (uid:0 nice:0 policy:0 rt_prio:0)
1718 # ----------------- 1457 # -----------------
1719 # => started at: wake_up_new_task 1458 # => started at: wake_up_new_task
1720 # => ended at: task_rq_unlock 1459 # => ended at: task_rq_unlock
1721 # 1460 #
1722 # 1461 #
1723 # _------=> CPU# 1462 # _------=> CPU#
1724 # / _-----=> irqs-off 1463 # / _-----=> irqs-off
1725 # | / _----=> need-resched 1464 # | / _----=> need-resched
1726 # || / _---=> hardirq/softir 1465 # || / _---=> hardirq/softirq
1727 # ||| / _--=> preempt-depth 1466 # ||| / _--=> preempt-depth
1728 # |||| / delay 1467 # |||| / delay
1729 # cmd pid ||||| time | caller 1468 # cmd pid ||||| time | caller
1730 # \ / ||||| \ | / 1469 # \ / ||||| \ | /
1731 bash-1994 1d..1 0us : _raw_spin_l 1470 bash-1994 1d..1 0us : _raw_spin_lock_irqsave <-wake_up_new_task
1732 bash-1994 1d..1 0us : select_task 1471 bash-1994 1d..1 0us : select_task_rq_fair <-select_task_rq
1733 bash-1994 1d..1 1us : __rcu_read_ 1472 bash-1994 1d..1 1us : __rcu_read_lock <-select_task_rq_fair
1734 bash-1994 1d..1 1us : source_load 1473 bash-1994 1d..1 1us : source_load <-select_task_rq_fair
1735 bash-1994 1d..1 1us : source_load 1474 bash-1994 1d..1 1us : source_load <-select_task_rq_fair
1736 [...] 1475 [...]
1737 bash-1994 1d..1 12us : irq_enter < 1476 bash-1994 1d..1 12us : irq_enter <-smp_apic_timer_interrupt
1738 bash-1994 1d..1 12us : rcu_irq_ent 1477 bash-1994 1d..1 12us : rcu_irq_enter <-irq_enter
1739 bash-1994 1d..1 13us : add_preempt 1478 bash-1994 1d..1 13us : add_preempt_count <-irq_enter
1740 bash-1994 1d.h1 13us : exit_idle < 1479 bash-1994 1d.h1 13us : exit_idle <-smp_apic_timer_interrupt
1741 bash-1994 1d.h1 13us : hrtimer_int 1480 bash-1994 1d.h1 13us : hrtimer_interrupt <-smp_apic_timer_interrupt
1742 bash-1994 1d.h1 13us : _raw_spin_l 1481 bash-1994 1d.h1 13us : _raw_spin_lock <-hrtimer_interrupt
1743 bash-1994 1d.h1 14us : add_preempt 1482 bash-1994 1d.h1 14us : add_preempt_count <-_raw_spin_lock
1744 bash-1994 1d.h2 14us : ktime_get_u 1483 bash-1994 1d.h2 14us : ktime_get_update_offsets <-hrtimer_interrupt
1745 [...] 1484 [...]
1746 bash-1994 1d.h1 35us : lapic_next_ 1485 bash-1994 1d.h1 35us : lapic_next_event <-clockevents_program_event
1747 bash-1994 1d.h1 35us : irq_exit <- 1486 bash-1994 1d.h1 35us : irq_exit <-smp_apic_timer_interrupt
1748 bash-1994 1d.h1 36us : sub_preempt 1487 bash-1994 1d.h1 36us : sub_preempt_count <-irq_exit
1749 bash-1994 1d..2 36us : do_softirq 1488 bash-1994 1d..2 36us : do_softirq <-irq_exit
1750 bash-1994 1d..2 36us : __do_softir 1489 bash-1994 1d..2 36us : __do_softirq <-call_softirq
1751 bash-1994 1d..2 36us : __local_bh_ 1490 bash-1994 1d..2 36us : __local_bh_disable <-__do_softirq
1752 bash-1994 1d.s2 37us : add_preempt 1491 bash-1994 1d.s2 37us : add_preempt_count <-_raw_spin_lock_irq
1753 bash-1994 1d.s3 38us : _raw_spin_u 1492 bash-1994 1d.s3 38us : _raw_spin_unlock <-run_timer_softirq
1754 bash-1994 1d.s3 39us : sub_preempt 1493 bash-1994 1d.s3 39us : sub_preempt_count <-_raw_spin_unlock
1755 bash-1994 1d.s2 39us : call_timer_ 1494 bash-1994 1d.s2 39us : call_timer_fn <-run_timer_softirq
1756 [...] 1495 [...]
1757 bash-1994 1dNs2 81us : cpu_needs_a 1496 bash-1994 1dNs2 81us : cpu_needs_another_gp <-rcu_process_callbacks
1758 bash-1994 1dNs2 82us : __local_bh_ 1497 bash-1994 1dNs2 82us : __local_bh_enable <-__do_softirq
1759 bash-1994 1dNs2 82us : sub_preempt 1498 bash-1994 1dNs2 82us : sub_preempt_count <-__local_bh_enable
1760 bash-1994 1dN.2 82us : idle_cpu <- 1499 bash-1994 1dN.2 82us : idle_cpu <-irq_exit
1761 bash-1994 1dN.2 83us : rcu_irq_exi 1500 bash-1994 1dN.2 83us : rcu_irq_exit <-irq_exit
1762 bash-1994 1dN.2 83us : sub_preempt 1501 bash-1994 1dN.2 83us : sub_preempt_count <-irq_exit
1763 bash-1994 1.N.1 84us : _raw_spin_u 1502 bash-1994 1.N.1 84us : _raw_spin_unlock_irqrestore <-task_rq_unlock
1764 bash-1994 1.N.1 84us+: trace_preem 1503 bash-1994 1.N.1 84us+: trace_preempt_on <-task_rq_unlock
1765 bash-1994 1.N.1 104us : <stack trac 1504 bash-1994 1.N.1 104us : <stack trace>
1766 => sub_preempt_count 1505 => sub_preempt_count
1767 => _raw_spin_unlock_irqrestore 1506 => _raw_spin_unlock_irqrestore
1768 => task_rq_unlock 1507 => task_rq_unlock
1769 => wake_up_new_task 1508 => wake_up_new_task
1770 => do_fork 1509 => do_fork
1771 => sys_clone 1510 => sys_clone
1772 => stub_clone 1511 => stub_clone
1773 1512
1774 1513
1775 The above is an example of the preemptoff tra 1514 The above is an example of the preemptoff trace with
1776 function-trace set. Here we see that interrup 1515 function-trace set. Here we see that interrupts were not disabled
1777 the entire time. The irq_enter code lets us k 1516 the entire time. The irq_enter code lets us know that we entered
1778 an interrupt 'h'. Before that, the functions 1517 an interrupt 'h'. Before that, the functions being traced still
1779 show that it is not in an interrupt, but we c 1518 show that it is not in an interrupt, but we can see from the
1780 functions themselves that this is not the cas 1519 functions themselves that this is not the case.
1781 1520
1782 preemptirqsoff 1521 preemptirqsoff
1783 -------------- 1522 --------------
1784 1523
1785 Knowing the locations that have interrupts di 1524 Knowing the locations that have interrupts disabled or
1786 preemption disabled for the longest times is 1525 preemption disabled for the longest times is helpful. But
1787 sometimes we would like to know when either p 1526 sometimes we would like to know when either preemption and/or
1788 interrupts are disabled. 1527 interrupts are disabled.
1789 1528
1790 Consider the following code:: 1529 Consider the following code::
1791 1530
1792 local_irq_disable(); 1531 local_irq_disable();
1793 call_function_with_irqs_off(); 1532 call_function_with_irqs_off();
1794 preempt_disable(); 1533 preempt_disable();
1795 call_function_with_irqs_and_preemption_of 1534 call_function_with_irqs_and_preemption_off();
1796 local_irq_enable(); 1535 local_irq_enable();
1797 call_function_with_preemption_off(); 1536 call_function_with_preemption_off();
1798 preempt_enable(); 1537 preempt_enable();
1799 1538
1800 The irqsoff tracer will record the total leng 1539 The irqsoff tracer will record the total length of
1801 call_function_with_irqs_off() and 1540 call_function_with_irqs_off() and
1802 call_function_with_irqs_and_preemption_off(). 1541 call_function_with_irqs_and_preemption_off().
1803 1542
1804 The preemptoff tracer will record the total l 1543 The preemptoff tracer will record the total length of
1805 call_function_with_irqs_and_preemption_off() 1544 call_function_with_irqs_and_preemption_off() and
1806 call_function_with_preemption_off(). 1545 call_function_with_preemption_off().
1807 1546
1808 But neither will trace the time that interrup 1547 But neither will trace the time that interrupts and/or
1809 preemption is disabled. This total time is th 1548 preemption is disabled. This total time is the time that we can
1810 not schedule. To record this time, use the pr 1549 not schedule. To record this time, use the preemptirqsoff
1811 tracer. 1550 tracer.
1812 1551
1813 Again, using this trace is much like the irqs 1552 Again, using this trace is much like the irqsoff and preemptoff
1814 tracers. 1553 tracers.
1815 :: 1554 ::
1816 1555
1817 # echo 0 > options/function-trace 1556 # echo 0 > options/function-trace
1818 # echo preemptirqsoff > current_tracer 1557 # echo preemptirqsoff > current_tracer
1819 # echo 1 > tracing_on 1558 # echo 1 > tracing_on
1820 # echo 0 > tracing_max_latency 1559 # echo 0 > tracing_max_latency
1821 # ls -ltr 1560 # ls -ltr
1822 [...] 1561 [...]
1823 # echo 0 > tracing_on 1562 # echo 0 > tracing_on
1824 # cat trace 1563 # cat trace
1825 # tracer: preemptirqsoff 1564 # tracer: preemptirqsoff
1826 # 1565 #
1827 # preemptirqsoff latency trace v1.1.5 on 3. 1566 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1828 # ----------------------------------------- 1567 # --------------------------------------------------------------------
1829 # latency: 100 us, #4/4, CPU#3 | (M:preempt 1568 # latency: 100 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1830 # ----------------- 1569 # -----------------
1831 # | task: ls-2230 (uid:0 nice:0 policy:0 1570 # | task: ls-2230 (uid:0 nice:0 policy:0 rt_prio:0)
1832 # ----------------- 1571 # -----------------
1833 # => started at: ata_scsi_queuecmd 1572 # => started at: ata_scsi_queuecmd
1834 # => ended at: ata_scsi_queuecmd 1573 # => ended at: ata_scsi_queuecmd
1835 # 1574 #
1836 # 1575 #
1837 # _------=> CPU# 1576 # _------=> CPU#
1838 # / _-----=> irqs-off 1577 # / _-----=> irqs-off
1839 # | / _----=> need-resched 1578 # | / _----=> need-resched
1840 # || / _---=> hardirq/softir 1579 # || / _---=> hardirq/softirq
1841 # ||| / _--=> preempt-depth 1580 # ||| / _--=> preempt-depth
1842 # |||| / delay 1581 # |||| / delay
1843 # cmd pid ||||| time | caller 1582 # cmd pid ||||| time | caller
1844 # \ / ||||| \ | / 1583 # \ / ||||| \ | /
1845 ls-2230 3d... 0us+: _raw_spin_l 1584 ls-2230 3d... 0us+: _raw_spin_lock_irqsave <-ata_scsi_queuecmd
1846 ls-2230 3...1 100us : _raw_spin_u 1585 ls-2230 3...1 100us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1847 ls-2230 3...1 101us+: trace_preem 1586 ls-2230 3...1 101us+: trace_preempt_on <-ata_scsi_queuecmd
1848 ls-2230 3...1 111us : <stack trac 1587 ls-2230 3...1 111us : <stack trace>
1849 => sub_preempt_count 1588 => sub_preempt_count
1850 => _raw_spin_unlock_irqrestore 1589 => _raw_spin_unlock_irqrestore
1851 => ata_scsi_queuecmd 1590 => ata_scsi_queuecmd
1852 => scsi_dispatch_cmd 1591 => scsi_dispatch_cmd
1853 => scsi_request_fn 1592 => scsi_request_fn
1854 => __blk_run_queue_uncond 1593 => __blk_run_queue_uncond
1855 => __blk_run_queue 1594 => __blk_run_queue
1856 => blk_queue_bio 1595 => blk_queue_bio
1857 => submit_bio_noacct !! 1596 => generic_make_request
1858 => submit_bio 1597 => submit_bio
1859 => submit_bh 1598 => submit_bh
1860 => ext3_bread 1599 => ext3_bread
1861 => ext3_dir_bread 1600 => ext3_dir_bread
1862 => htree_dirblock_to_tree 1601 => htree_dirblock_to_tree
1863 => ext3_htree_fill_tree 1602 => ext3_htree_fill_tree
1864 => ext3_readdir 1603 => ext3_readdir
1865 => vfs_readdir 1604 => vfs_readdir
1866 => sys_getdents 1605 => sys_getdents
1867 => system_call_fastpath 1606 => system_call_fastpath
1868 1607
1869 1608
1870 The trace_hardirqs_off_thunk is called from a 1609 The trace_hardirqs_off_thunk is called from assembly on x86 when
1871 interrupts are disabled in the assembly code. 1610 interrupts are disabled in the assembly code. Without the
1872 function tracing, we do not know if interrupt 1611 function tracing, we do not know if interrupts were enabled
1873 within the preemption points. We do see that 1612 within the preemption points. We do see that it started with
1874 preemption enabled. 1613 preemption enabled.
1875 1614
1876 Here is a trace with function-trace set:: 1615 Here is a trace with function-trace set::
1877 1616
1878 # tracer: preemptirqsoff 1617 # tracer: preemptirqsoff
1879 # 1618 #
1880 # preemptirqsoff latency trace v1.1.5 on 3. 1619 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1881 # ----------------------------------------- 1620 # --------------------------------------------------------------------
1882 # latency: 161 us, #339/339, CPU#3 | (M:pre 1621 # latency: 161 us, #339/339, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1883 # ----------------- 1622 # -----------------
1884 # | task: ls-2269 (uid:0 nice:0 policy:0 1623 # | task: ls-2269 (uid:0 nice:0 policy:0 rt_prio:0)
1885 # ----------------- 1624 # -----------------
1886 # => started at: schedule 1625 # => started at: schedule
1887 # => ended at: mutex_unlock 1626 # => ended at: mutex_unlock
1888 # 1627 #
1889 # 1628 #
1890 # _------=> CPU# 1629 # _------=> CPU#
1891 # / _-----=> irqs-off 1630 # / _-----=> irqs-off
1892 # | / _----=> need-resched 1631 # | / _----=> need-resched
1893 # || / _---=> hardirq/softir 1632 # || / _---=> hardirq/softirq
1894 # ||| / _--=> preempt-depth 1633 # ||| / _--=> preempt-depth
1895 # |||| / delay 1634 # |||| / delay
1896 # cmd pid ||||| time | caller 1635 # cmd pid ||||| time | caller
1897 # \ / ||||| \ | / 1636 # \ / ||||| \ | /
1898 kworker/-59 3...1 0us : __schedule 1637 kworker/-59 3...1 0us : __schedule <-schedule
1899 kworker/-59 3d..1 0us : rcu_preempt 1638 kworker/-59 3d..1 0us : rcu_preempt_qs <-rcu_note_context_switch
1900 kworker/-59 3d..1 1us : add_preempt 1639 kworker/-59 3d..1 1us : add_preempt_count <-_raw_spin_lock_irq
1901 kworker/-59 3d..2 1us : deactivate_ 1640 kworker/-59 3d..2 1us : deactivate_task <-__schedule
1902 kworker/-59 3d..2 1us : dequeue_tas 1641 kworker/-59 3d..2 1us : dequeue_task <-deactivate_task
1903 kworker/-59 3d..2 2us : update_rq_c 1642 kworker/-59 3d..2 2us : update_rq_clock <-dequeue_task
1904 kworker/-59 3d..2 2us : dequeue_tas 1643 kworker/-59 3d..2 2us : dequeue_task_fair <-dequeue_task
1905 kworker/-59 3d..2 2us : update_curr 1644 kworker/-59 3d..2 2us : update_curr <-dequeue_task_fair
1906 kworker/-59 3d..2 2us : update_min_ 1645 kworker/-59 3d..2 2us : update_min_vruntime <-update_curr
1907 kworker/-59 3d..2 3us : cpuacct_cha 1646 kworker/-59 3d..2 3us : cpuacct_charge <-update_curr
1908 kworker/-59 3d..2 3us : __rcu_read_ 1647 kworker/-59 3d..2 3us : __rcu_read_lock <-cpuacct_charge
1909 kworker/-59 3d..2 3us : __rcu_read_ 1648 kworker/-59 3d..2 3us : __rcu_read_unlock <-cpuacct_charge
1910 kworker/-59 3d..2 3us : update_cfs_ 1649 kworker/-59 3d..2 3us : update_cfs_rq_blocked_load <-dequeue_task_fair
1911 kworker/-59 3d..2 4us : clear_buddi 1650 kworker/-59 3d..2 4us : clear_buddies <-dequeue_task_fair
1912 kworker/-59 3d..2 4us : account_ent 1651 kworker/-59 3d..2 4us : account_entity_dequeue <-dequeue_task_fair
1913 kworker/-59 3d..2 4us : update_min_ 1652 kworker/-59 3d..2 4us : update_min_vruntime <-dequeue_task_fair
1914 kworker/-59 3d..2 4us : update_cfs_ 1653 kworker/-59 3d..2 4us : update_cfs_shares <-dequeue_task_fair
1915 kworker/-59 3d..2 5us : hrtick_upda 1654 kworker/-59 3d..2 5us : hrtick_update <-dequeue_task_fair
1916 kworker/-59 3d..2 5us : wq_worker_s 1655 kworker/-59 3d..2 5us : wq_worker_sleeping <-__schedule
1917 kworker/-59 3d..2 5us : kthread_dat 1656 kworker/-59 3d..2 5us : kthread_data <-wq_worker_sleeping
1918 kworker/-59 3d..2 5us : put_prev_ta 1657 kworker/-59 3d..2 5us : put_prev_task_fair <-__schedule
1919 kworker/-59 3d..2 6us : pick_next_t 1658 kworker/-59 3d..2 6us : pick_next_task_fair <-pick_next_task
1920 kworker/-59 3d..2 6us : clear_buddi 1659 kworker/-59 3d..2 6us : clear_buddies <-pick_next_task_fair
1921 kworker/-59 3d..2 6us : set_next_en 1660 kworker/-59 3d..2 6us : set_next_entity <-pick_next_task_fair
1922 kworker/-59 3d..2 6us : update_stat 1661 kworker/-59 3d..2 6us : update_stats_wait_end <-set_next_entity
1923 ls-2269 3d..2 7us : finish_task 1662 ls-2269 3d..2 7us : finish_task_switch <-__schedule
1924 ls-2269 3d..2 7us : _raw_spin_u 1663 ls-2269 3d..2 7us : _raw_spin_unlock_irq <-finish_task_switch
1925 ls-2269 3d..2 8us : do_IRQ <-re 1664 ls-2269 3d..2 8us : do_IRQ <-ret_from_intr
1926 ls-2269 3d..2 8us : irq_enter < 1665 ls-2269 3d..2 8us : irq_enter <-do_IRQ
1927 ls-2269 3d..2 8us : rcu_irq_ent 1666 ls-2269 3d..2 8us : rcu_irq_enter <-irq_enter
1928 ls-2269 3d..2 9us : add_preempt 1667 ls-2269 3d..2 9us : add_preempt_count <-irq_enter
1929 ls-2269 3d.h2 9us : exit_idle < 1668 ls-2269 3d.h2 9us : exit_idle <-do_IRQ
1930 [...] 1669 [...]
1931 ls-2269 3d.h3 20us : sub_preempt 1670 ls-2269 3d.h3 20us : sub_preempt_count <-_raw_spin_unlock
1932 ls-2269 3d.h2 20us : irq_exit <- 1671 ls-2269 3d.h2 20us : irq_exit <-do_IRQ
1933 ls-2269 3d.h2 21us : sub_preempt 1672 ls-2269 3d.h2 21us : sub_preempt_count <-irq_exit
1934 ls-2269 3d..3 21us : do_softirq 1673 ls-2269 3d..3 21us : do_softirq <-irq_exit
1935 ls-2269 3d..3 21us : __do_softir 1674 ls-2269 3d..3 21us : __do_softirq <-call_softirq
1936 ls-2269 3d..3 21us+: __local_bh_ 1675 ls-2269 3d..3 21us+: __local_bh_disable <-__do_softirq
1937 ls-2269 3d.s4 29us : sub_preempt 1676 ls-2269 3d.s4 29us : sub_preempt_count <-_local_bh_enable_ip
1938 ls-2269 3d.s5 29us : sub_preempt 1677 ls-2269 3d.s5 29us : sub_preempt_count <-_local_bh_enable_ip
1939 ls-2269 3d.s5 31us : do_IRQ <-re 1678 ls-2269 3d.s5 31us : do_IRQ <-ret_from_intr
1940 ls-2269 3d.s5 31us : irq_enter < 1679 ls-2269 3d.s5 31us : irq_enter <-do_IRQ
1941 ls-2269 3d.s5 31us : rcu_irq_ent 1680 ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter
1942 [...] 1681 [...]
1943 ls-2269 3d.s5 31us : rcu_irq_ent 1682 ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter
1944 ls-2269 3d.s5 32us : add_preempt 1683 ls-2269 3d.s5 32us : add_preempt_count <-irq_enter
1945 ls-2269 3d.H5 32us : exit_idle < 1684 ls-2269 3d.H5 32us : exit_idle <-do_IRQ
1946 ls-2269 3d.H5 32us : handle_irq 1685 ls-2269 3d.H5 32us : handle_irq <-do_IRQ
1947 ls-2269 3d.H5 32us : irq_to_desc 1686 ls-2269 3d.H5 32us : irq_to_desc <-handle_irq
1948 ls-2269 3d.H5 33us : handle_fast 1687 ls-2269 3d.H5 33us : handle_fasteoi_irq <-handle_irq
1949 [...] 1688 [...]
1950 ls-2269 3d.s5 158us : _raw_spin_u 1689 ls-2269 3d.s5 158us : _raw_spin_unlock_irqrestore <-rtl8139_poll
1951 ls-2269 3d.s3 158us : net_rps_act 1690 ls-2269 3d.s3 158us : net_rps_action_and_irq_enable.isra.65 <-net_rx_action
1952 ls-2269 3d.s3 159us : __local_bh_ 1691 ls-2269 3d.s3 159us : __local_bh_enable <-__do_softirq
1953 ls-2269 3d.s3 159us : sub_preempt 1692 ls-2269 3d.s3 159us : sub_preempt_count <-__local_bh_enable
1954 ls-2269 3d..3 159us : idle_cpu <- 1693 ls-2269 3d..3 159us : idle_cpu <-irq_exit
1955 ls-2269 3d..3 159us : rcu_irq_exi 1694 ls-2269 3d..3 159us : rcu_irq_exit <-irq_exit
1956 ls-2269 3d..3 160us : sub_preempt 1695 ls-2269 3d..3 160us : sub_preempt_count <-irq_exit
1957 ls-2269 3d... 161us : __mutex_unl 1696 ls-2269 3d... 161us : __mutex_unlock_slowpath <-mutex_unlock
1958 ls-2269 3d... 162us+: trace_hardi 1697 ls-2269 3d... 162us+: trace_hardirqs_on <-mutex_unlock
1959 ls-2269 3d... 186us : <stack trac 1698 ls-2269 3d... 186us : <stack trace>
1960 => __mutex_unlock_slowpath 1699 => __mutex_unlock_slowpath
1961 => mutex_unlock 1700 => mutex_unlock
1962 => process_output 1701 => process_output
1963 => n_tty_write 1702 => n_tty_write
1964 => tty_write 1703 => tty_write
1965 => vfs_write 1704 => vfs_write
1966 => sys_write 1705 => sys_write
1967 => system_call_fastpath 1706 => system_call_fastpath
1968 1707
1969 This is an interesting trace. It started with 1708 This is an interesting trace. It started with kworker running and
1970 scheduling out and ls taking over. But as soo 1709 scheduling out and ls taking over. But as soon as ls released the
1971 rq lock and enabled interrupts (but not preem 1710 rq lock and enabled interrupts (but not preemption) an interrupt
1972 triggered. When the interrupt finished, it st 1711 triggered. When the interrupt finished, it started running softirqs.
1973 But while the softirq was running, another in 1712 But while the softirq was running, another interrupt triggered.
1974 When an interrupt is running inside a softirq 1713 When an interrupt is running inside a softirq, the annotation is 'H'.
1975 1714
1976 1715
1977 wakeup 1716 wakeup
1978 ------ 1717 ------
1979 1718
1980 One common case that people are interested in 1719 One common case that people are interested in tracing is the
1981 time it takes for a task that is woken to act 1720 time it takes for a task that is woken to actually wake up.
1982 Now for non Real-Time tasks, this can be arbi 1721 Now for non Real-Time tasks, this can be arbitrary. But tracing
1983 it nonetheless can be interesting. !! 1722 it none the less can be interesting.
1984 1723
1985 Without function tracing:: 1724 Without function tracing::
1986 1725
1987 # echo 0 > options/function-trace 1726 # echo 0 > options/function-trace
1988 # echo wakeup > current_tracer 1727 # echo wakeup > current_tracer
1989 # echo 1 > tracing_on 1728 # echo 1 > tracing_on
1990 # echo 0 > tracing_max_latency 1729 # echo 0 > tracing_max_latency
1991 # chrt -f 5 sleep 1 1730 # chrt -f 5 sleep 1
1992 # echo 0 > tracing_on 1731 # echo 0 > tracing_on
1993 # cat trace 1732 # cat trace
1994 # tracer: wakeup 1733 # tracer: wakeup
1995 # 1734 #
1996 # wakeup latency trace v1.1.5 on 3.8.0-test 1735 # wakeup latency trace v1.1.5 on 3.8.0-test+
1997 # ----------------------------------------- 1736 # --------------------------------------------------------------------
1998 # latency: 15 us, #4/4, CPU#3 | (M:preempt 1737 # latency: 15 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1999 # ----------------- 1738 # -----------------
2000 # | task: kworker/3:1H-312 (uid:0 nice:- 1739 # | task: kworker/3:1H-312 (uid:0 nice:-20 policy:0 rt_prio:0)
2001 # ----------------- 1740 # -----------------
2002 # 1741 #
2003 # _------=> CPU# 1742 # _------=> CPU#
2004 # / _-----=> irqs-off 1743 # / _-----=> irqs-off
2005 # | / _----=> need-resched 1744 # | / _----=> need-resched
2006 # || / _---=> hardirq/softir 1745 # || / _---=> hardirq/softirq
2007 # ||| / _--=> preempt-depth 1746 # ||| / _--=> preempt-depth
2008 # |||| / delay 1747 # |||| / delay
2009 # cmd pid ||||| time | caller 1748 # cmd pid ||||| time | caller
2010 # \ / ||||| \ | / 1749 # \ / ||||| \ | /
2011 <idle>-0 3dNs7 0us : 0:120: 1750 <idle>-0 3dNs7 0us : 0:120:R + [003] 312:100:R kworker/3:1H
2012 <idle>-0 3dNs7 1us+: ttwu_do_act 1751 <idle>-0 3dNs7 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up
2013 <idle>-0 3d..3 15us : __schedule 1752 <idle>-0 3d..3 15us : __schedule <-schedule
2014 <idle>-0 3d..3 15us : 0:120: 1753 <idle>-0 3d..3 15us : 0:120:R ==> [003] 312:100:R kworker/3:1H
2015 1754
2016 The tracer only traces the highest priority t 1755 The tracer only traces the highest priority task in the system
2017 to avoid tracing the normal circumstances. He 1756 to avoid tracing the normal circumstances. Here we see that
2018 the kworker with a nice priority of -20 (not 1757 the kworker with a nice priority of -20 (not very nice), took
2019 just 15 microseconds from the time it woke up 1758 just 15 microseconds from the time it woke up, to the time it
2020 ran. 1759 ran.
2021 1760
2022 Non Real-Time tasks are not that interesting. 1761 Non Real-Time tasks are not that interesting. A more interesting
2023 trace is to concentrate only on Real-Time tas 1762 trace is to concentrate only on Real-Time tasks.
2024 1763
2025 wakeup_rt 1764 wakeup_rt
2026 --------- 1765 ---------
2027 1766
2028 In a Real-Time environment it is very importa 1767 In a Real-Time environment it is very important to know the
2029 wakeup time it takes for the highest priority 1768 wakeup time it takes for the highest priority task that is woken
2030 up to the time that it executes. This is also 1769 up to the time that it executes. This is also known as "schedule
2031 latency". I stress the point that this is abo 1770 latency". I stress the point that this is about RT tasks. It is
2032 also important to know the scheduling latency 1771 also important to know the scheduling latency of non-RT tasks,
2033 but the average schedule latency is better fo 1772 but the average schedule latency is better for non-RT tasks.
2034 Tools like LatencyTop are more appropriate fo 1773 Tools like LatencyTop are more appropriate for such
2035 measurements. 1774 measurements.
2036 1775
2037 Real-Time environments are interested in the 1776 Real-Time environments are interested in the worst case latency.
2038 That is the longest latency it takes for some 1777 That is the longest latency it takes for something to happen,
2039 and not the average. We can have a very fast 1778 and not the average. We can have a very fast scheduler that may
2040 only have a large latency once in a while, bu 1779 only have a large latency once in a while, but that would not
2041 work well with Real-Time tasks. The wakeup_r 1780 work well with Real-Time tasks. The wakeup_rt tracer was designed
2042 to record the worst case wakeups of RT tasks. 1781 to record the worst case wakeups of RT tasks. Non-RT tasks are
2043 not recorded because the tracer only records 1782 not recorded because the tracer only records one worst case and
2044 tracing non-RT tasks that are unpredictable w 1783 tracing non-RT tasks that are unpredictable will overwrite the
2045 worst case latency of RT tasks (just run the 1784 worst case latency of RT tasks (just run the normal wakeup
2046 tracer for a while to see that effect). 1785 tracer for a while to see that effect).
2047 1786
2048 Since this tracer only deals with RT tasks, w 1787 Since this tracer only deals with RT tasks, we will run this
2049 slightly differently than we did with the pre 1788 slightly differently than we did with the previous tracers.
2050 Instead of performing an 'ls', we will run 's 1789 Instead of performing an 'ls', we will run 'sleep 1' under
2051 'chrt' which changes the priority of the task 1790 'chrt' which changes the priority of the task.
2052 :: 1791 ::
2053 1792
2054 # echo 0 > options/function-trace 1793 # echo 0 > options/function-trace
2055 # echo wakeup_rt > current_tracer 1794 # echo wakeup_rt > current_tracer
2056 # echo 1 > tracing_on 1795 # echo 1 > tracing_on
2057 # echo 0 > tracing_max_latency 1796 # echo 0 > tracing_max_latency
2058 # chrt -f 5 sleep 1 1797 # chrt -f 5 sleep 1
2059 # echo 0 > tracing_on 1798 # echo 0 > tracing_on
2060 # cat trace 1799 # cat trace
2061 # tracer: wakeup 1800 # tracer: wakeup
2062 # 1801 #
2063 # tracer: wakeup_rt 1802 # tracer: wakeup_rt
2064 # 1803 #
2065 # wakeup_rt latency trace v1.1.5 on 3.8.0-t 1804 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
2066 # ----------------------------------------- 1805 # --------------------------------------------------------------------
2067 # latency: 5 us, #4/4, CPU#3 | (M:preempt V 1806 # latency: 5 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
2068 # ----------------- 1807 # -----------------
2069 # | task: sleep-2389 (uid:0 nice:0 polic 1808 # | task: sleep-2389 (uid:0 nice:0 policy:1 rt_prio:5)
2070 # ----------------- 1809 # -----------------
2071 # 1810 #
2072 # _------=> CPU# 1811 # _------=> CPU#
2073 # / _-----=> irqs-off 1812 # / _-----=> irqs-off
2074 # | / _----=> need-resched 1813 # | / _----=> need-resched
2075 # || / _---=> hardirq/softir 1814 # || / _---=> hardirq/softirq
2076 # ||| / _--=> preempt-depth 1815 # ||| / _--=> preempt-depth
2077 # |||| / delay 1816 # |||| / delay
2078 # cmd pid ||||| time | caller 1817 # cmd pid ||||| time | caller
2079 # \ / ||||| \ | / 1818 # \ / ||||| \ | /
2080 <idle>-0 3d.h4 0us : 0:120: 1819 <idle>-0 3d.h4 0us : 0:120:R + [003] 2389: 94:R sleep
2081 <idle>-0 3d.h4 1us+: ttwu_do_act 1820 <idle>-0 3d.h4 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up
2082 <idle>-0 3d..3 5us : __schedule 1821 <idle>-0 3d..3 5us : __schedule <-schedule
2083 <idle>-0 3d..3 5us : 0:120: 1822 <idle>-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep
2084 1823
2085 1824
2086 Running this on an idle system, we see that i 1825 Running this on an idle system, we see that it only took 5 microseconds
2087 to perform the task switch. Note, since the 1826 to perform the task switch. Note, since the trace point in the schedule
2088 is before the actual "switch", we stop the tr 1827 is before the actual "switch", we stop the tracing when the recorded task
2089 is about to schedule in. This may change if w 1828 is about to schedule in. This may change if we add a new marker at the
2090 end of the scheduler. 1829 end of the scheduler.
2091 1830
2092 Notice that the recorded task is 'sleep' with 1831 Notice that the recorded task is 'sleep' with the PID of 2389
2093 and it has an rt_prio of 5. This priority is 1832 and it has an rt_prio of 5. This priority is user-space priority
2094 and not the internal kernel priority. The pol 1833 and not the internal kernel priority. The policy is 1 for
2095 SCHED_FIFO and 2 for SCHED_RR. 1834 SCHED_FIFO and 2 for SCHED_RR.
2096 1835
2097 Note, that the trace data shows the internal 1836 Note, that the trace data shows the internal priority (99 - rtprio).
2098 :: 1837 ::
2099 1838
2100 <idle>-0 3d..3 5us : 0:120:R 1839 <idle>-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep
2101 1840
2102 The 0:120:R means idle was running with a nic 1841 The 0:120:R means idle was running with a nice priority of 0 (120 - 120)
2103 and in the running state 'R'. The sleep task 1842 and in the running state 'R'. The sleep task was scheduled in with
2104 2389: 94:R. That is the priority is the kerne 1843 2389: 94:R. That is the priority is the kernel rtprio (99 - 5 = 94)
2105 and it too is in the running state. 1844 and it too is in the running state.
2106 1845
2107 Doing the same with chrt -r 5 and function-tr 1846 Doing the same with chrt -r 5 and function-trace set.
2108 :: 1847 ::
2109 1848
2110 echo 1 > options/function-trace 1849 echo 1 > options/function-trace
2111 1850
2112 # tracer: wakeup_rt 1851 # tracer: wakeup_rt
2113 # 1852 #
2114 # wakeup_rt latency trace v1.1.5 on 3.8.0-t 1853 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
2115 # ----------------------------------------- 1854 # --------------------------------------------------------------------
2116 # latency: 29 us, #85/85, CPU#3 | (M:preemp 1855 # latency: 29 us, #85/85, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
2117 # ----------------- 1856 # -----------------
2118 # | task: sleep-2448 (uid:0 nice:0 polic 1857 # | task: sleep-2448 (uid:0 nice:0 policy:1 rt_prio:5)
2119 # ----------------- 1858 # -----------------
2120 # 1859 #
2121 # _------=> CPU# 1860 # _------=> CPU#
2122 # / _-----=> irqs-off 1861 # / _-----=> irqs-off
2123 # | / _----=> need-resched 1862 # | / _----=> need-resched
2124 # || / _---=> hardirq/softir 1863 # || / _---=> hardirq/softirq
2125 # ||| / _--=> preempt-depth 1864 # ||| / _--=> preempt-depth
2126 # |||| / delay 1865 # |||| / delay
2127 # cmd pid ||||| time | caller 1866 # cmd pid ||||| time | caller
2128 # \ / ||||| \ | / 1867 # \ / ||||| \ | /
2129 <idle>-0 3d.h4 1us+: 0:120: 1868 <idle>-0 3d.h4 1us+: 0:120:R + [003] 2448: 94:R sleep
2130 <idle>-0 3d.h4 2us : ttwu_do_act 1869 <idle>-0 3d.h4 2us : ttwu_do_activate.constprop.87 <-try_to_wake_up
2131 <idle>-0 3d.h3 3us : check_preem 1870 <idle>-0 3d.h3 3us : check_preempt_curr <-ttwu_do_wakeup
2132 <idle>-0 3d.h3 3us : resched_cur 1871 <idle>-0 3d.h3 3us : resched_curr <-check_preempt_curr
2133 <idle>-0 3dNh3 4us : task_woken_ 1872 <idle>-0 3dNh3 4us : task_woken_rt <-ttwu_do_wakeup
2134 <idle>-0 3dNh3 4us : _raw_spin_u 1873 <idle>-0 3dNh3 4us : _raw_spin_unlock <-try_to_wake_up
2135 <idle>-0 3dNh3 4us : sub_preempt 1874 <idle>-0 3dNh3 4us : sub_preempt_count <-_raw_spin_unlock
2136 <idle>-0 3dNh2 5us : ttwu_stat < 1875 <idle>-0 3dNh2 5us : ttwu_stat <-try_to_wake_up
2137 <idle>-0 3dNh2 5us : _raw_spin_u 1876 <idle>-0 3dNh2 5us : _raw_spin_unlock_irqrestore <-try_to_wake_up
2138 <idle>-0 3dNh2 6us : sub_preempt 1877 <idle>-0 3dNh2 6us : sub_preempt_count <-_raw_spin_unlock_irqrestore
2139 <idle>-0 3dNh1 6us : _raw_spin_l 1878 <idle>-0 3dNh1 6us : _raw_spin_lock <-__run_hrtimer
2140 <idle>-0 3dNh1 6us : add_preempt 1879 <idle>-0 3dNh1 6us : add_preempt_count <-_raw_spin_lock
2141 <idle>-0 3dNh2 7us : _raw_spin_u 1880 <idle>-0 3dNh2 7us : _raw_spin_unlock <-hrtimer_interrupt
2142 <idle>-0 3dNh2 7us : sub_preempt 1881 <idle>-0 3dNh2 7us : sub_preempt_count <-_raw_spin_unlock
2143 <idle>-0 3dNh1 7us : tick_progra 1882 <idle>-0 3dNh1 7us : tick_program_event <-hrtimer_interrupt
2144 <idle>-0 3dNh1 7us : clockevents 1883 <idle>-0 3dNh1 7us : clockevents_program_event <-tick_program_event
2145 <idle>-0 3dNh1 8us : ktime_get < 1884 <idle>-0 3dNh1 8us : ktime_get <-clockevents_program_event
2146 <idle>-0 3dNh1 8us : lapic_next_ 1885 <idle>-0 3dNh1 8us : lapic_next_event <-clockevents_program_event
2147 <idle>-0 3dNh1 8us : irq_exit <- 1886 <idle>-0 3dNh1 8us : irq_exit <-smp_apic_timer_interrupt
2148 <idle>-0 3dNh1 9us : sub_preempt 1887 <idle>-0 3dNh1 9us : sub_preempt_count <-irq_exit
2149 <idle>-0 3dN.2 9us : idle_cpu <- 1888 <idle>-0 3dN.2 9us : idle_cpu <-irq_exit
2150 <idle>-0 3dN.2 9us : rcu_irq_exi 1889 <idle>-0 3dN.2 9us : rcu_irq_exit <-irq_exit
2151 <idle>-0 3dN.2 10us : rcu_eqs_ent 1890 <idle>-0 3dN.2 10us : rcu_eqs_enter_common.isra.45 <-rcu_irq_exit
2152 <idle>-0 3dN.2 10us : sub_preempt 1891 <idle>-0 3dN.2 10us : sub_preempt_count <-irq_exit
2153 <idle>-0 3.N.1 11us : rcu_idle_ex 1892 <idle>-0 3.N.1 11us : rcu_idle_exit <-cpu_idle
2154 <idle>-0 3dN.1 11us : rcu_eqs_exi 1893 <idle>-0 3dN.1 11us : rcu_eqs_exit_common.isra.43 <-rcu_idle_exit
2155 <idle>-0 3.N.1 11us : tick_nohz_i 1894 <idle>-0 3.N.1 11us : tick_nohz_idle_exit <-cpu_idle
2156 <idle>-0 3dN.1 12us : menu_hrtime 1895 <idle>-0 3dN.1 12us : menu_hrtimer_cancel <-tick_nohz_idle_exit
2157 <idle>-0 3dN.1 12us : ktime_get < 1896 <idle>-0 3dN.1 12us : ktime_get <-tick_nohz_idle_exit
2158 <idle>-0 3dN.1 12us : tick_do_upd 1897 <idle>-0 3dN.1 12us : tick_do_update_jiffies64 <-tick_nohz_idle_exit
2159 <idle>-0 3dN.1 13us : cpu_load_up 1898 <idle>-0 3dN.1 13us : cpu_load_update_nohz <-tick_nohz_idle_exit
2160 <idle>-0 3dN.1 13us : _raw_spin_l 1899 <idle>-0 3dN.1 13us : _raw_spin_lock <-cpu_load_update_nohz
2161 <idle>-0 3dN.1 13us : add_preempt 1900 <idle>-0 3dN.1 13us : add_preempt_count <-_raw_spin_lock
2162 <idle>-0 3dN.2 13us : __cpu_load_ 1901 <idle>-0 3dN.2 13us : __cpu_load_update <-cpu_load_update_nohz
2163 <idle>-0 3dN.2 14us : sched_avg_u 1902 <idle>-0 3dN.2 14us : sched_avg_update <-__cpu_load_update
2164 <idle>-0 3dN.2 14us : _raw_spin_u 1903 <idle>-0 3dN.2 14us : _raw_spin_unlock <-cpu_load_update_nohz
2165 <idle>-0 3dN.2 14us : sub_preempt 1904 <idle>-0 3dN.2 14us : sub_preempt_count <-_raw_spin_unlock
2166 <idle>-0 3dN.1 15us : calc_load_n 1905 <idle>-0 3dN.1 15us : calc_load_nohz_stop <-tick_nohz_idle_exit
2167 <idle>-0 3dN.1 15us : touch_softl 1906 <idle>-0 3dN.1 15us : touch_softlockup_watchdog <-tick_nohz_idle_exit
2168 <idle>-0 3dN.1 15us : hrtimer_can 1907 <idle>-0 3dN.1 15us : hrtimer_cancel <-tick_nohz_idle_exit
2169 <idle>-0 3dN.1 15us : hrtimer_try 1908 <idle>-0 3dN.1 15us : hrtimer_try_to_cancel <-hrtimer_cancel
2170 <idle>-0 3dN.1 16us : lock_hrtime 1909 <idle>-0 3dN.1 16us : lock_hrtimer_base.isra.18 <-hrtimer_try_to_cancel
2171 <idle>-0 3dN.1 16us : _raw_spin_l 1910 <idle>-0 3dN.1 16us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18
2172 <idle>-0 3dN.1 16us : add_preempt 1911 <idle>-0 3dN.1 16us : add_preempt_count <-_raw_spin_lock_irqsave
2173 <idle>-0 3dN.2 17us : __remove_hr 1912 <idle>-0 3dN.2 17us : __remove_hrtimer <-remove_hrtimer.part.16
2174 <idle>-0 3dN.2 17us : hrtimer_for 1913 <idle>-0 3dN.2 17us : hrtimer_force_reprogram <-__remove_hrtimer
2175 <idle>-0 3dN.2 17us : tick_progra 1914 <idle>-0 3dN.2 17us : tick_program_event <-hrtimer_force_reprogram
2176 <idle>-0 3dN.2 18us : clockevents 1915 <idle>-0 3dN.2 18us : clockevents_program_event <-tick_program_event
2177 <idle>-0 3dN.2 18us : ktime_get < 1916 <idle>-0 3dN.2 18us : ktime_get <-clockevents_program_event
2178 <idle>-0 3dN.2 18us : lapic_next_ 1917 <idle>-0 3dN.2 18us : lapic_next_event <-clockevents_program_event
2179 <idle>-0 3dN.2 19us : _raw_spin_u 1918 <idle>-0 3dN.2 19us : _raw_spin_unlock_irqrestore <-hrtimer_try_to_cancel
2180 <idle>-0 3dN.2 19us : sub_preempt 1919 <idle>-0 3dN.2 19us : sub_preempt_count <-_raw_spin_unlock_irqrestore
2181 <idle>-0 3dN.1 19us : hrtimer_for 1920 <idle>-0 3dN.1 19us : hrtimer_forward <-tick_nohz_idle_exit
2182 <idle>-0 3dN.1 20us : ktime_add_s 1921 <idle>-0 3dN.1 20us : ktime_add_safe <-hrtimer_forward
2183 <idle>-0 3dN.1 20us : ktime_add_s 1922 <idle>-0 3dN.1 20us : ktime_add_safe <-hrtimer_forward
2184 <idle>-0 3dN.1 20us : hrtimer_sta 1923 <idle>-0 3dN.1 20us : hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11
2185 <idle>-0 3dN.1 20us : __hrtimer_s 1924 <idle>-0 3dN.1 20us : __hrtimer_start_range_ns <-hrtimer_start_range_ns
2186 <idle>-0 3dN.1 21us : lock_hrtime 1925 <idle>-0 3dN.1 21us : lock_hrtimer_base.isra.18 <-__hrtimer_start_range_ns
2187 <idle>-0 3dN.1 21us : _raw_spin_l 1926 <idle>-0 3dN.1 21us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18
2188 <idle>-0 3dN.1 21us : add_preempt 1927 <idle>-0 3dN.1 21us : add_preempt_count <-_raw_spin_lock_irqsave
2189 <idle>-0 3dN.2 22us : ktime_add_s 1928 <idle>-0 3dN.2 22us : ktime_add_safe <-__hrtimer_start_range_ns
2190 <idle>-0 3dN.2 22us : enqueue_hrt 1929 <idle>-0 3dN.2 22us : enqueue_hrtimer <-__hrtimer_start_range_ns
2191 <idle>-0 3dN.2 22us : tick_progra 1930 <idle>-0 3dN.2 22us : tick_program_event <-__hrtimer_start_range_ns
2192 <idle>-0 3dN.2 23us : clockevents 1931 <idle>-0 3dN.2 23us : clockevents_program_event <-tick_program_event
2193 <idle>-0 3dN.2 23us : ktime_get < 1932 <idle>-0 3dN.2 23us : ktime_get <-clockevents_program_event
2194 <idle>-0 3dN.2 23us : lapic_next_ 1933 <idle>-0 3dN.2 23us : lapic_next_event <-clockevents_program_event
2195 <idle>-0 3dN.2 24us : _raw_spin_u 1934 <idle>-0 3dN.2 24us : _raw_spin_unlock_irqrestore <-__hrtimer_start_range_ns
2196 <idle>-0 3dN.2 24us : sub_preempt 1935 <idle>-0 3dN.2 24us : sub_preempt_count <-_raw_spin_unlock_irqrestore
2197 <idle>-0 3dN.1 24us : account_idl 1936 <idle>-0 3dN.1 24us : account_idle_ticks <-tick_nohz_idle_exit
2198 <idle>-0 3dN.1 24us : account_idl 1937 <idle>-0 3dN.1 24us : account_idle_time <-account_idle_ticks
2199 <idle>-0 3.N.1 25us : sub_preempt 1938 <idle>-0 3.N.1 25us : sub_preempt_count <-cpu_idle
2200 <idle>-0 3.N.. 25us : schedule <- 1939 <idle>-0 3.N.. 25us : schedule <-cpu_idle
2201 <idle>-0 3.N.. 25us : __schedule 1940 <idle>-0 3.N.. 25us : __schedule <-preempt_schedule
2202 <idle>-0 3.N.. 26us : add_preempt 1941 <idle>-0 3.N.. 26us : add_preempt_count <-__schedule
2203 <idle>-0 3.N.1 26us : rcu_note_co 1942 <idle>-0 3.N.1 26us : rcu_note_context_switch <-__schedule
2204 <idle>-0 3.N.1 26us : rcu_sched_q 1943 <idle>-0 3.N.1 26us : rcu_sched_qs <-rcu_note_context_switch
2205 <idle>-0 3dN.1 27us : rcu_preempt 1944 <idle>-0 3dN.1 27us : rcu_preempt_qs <-rcu_note_context_switch
2206 <idle>-0 3.N.1 27us : _raw_spin_l 1945 <idle>-0 3.N.1 27us : _raw_spin_lock_irq <-__schedule
2207 <idle>-0 3dN.1 27us : add_preempt 1946 <idle>-0 3dN.1 27us : add_preempt_count <-_raw_spin_lock_irq
2208 <idle>-0 3dN.2 28us : put_prev_ta 1947 <idle>-0 3dN.2 28us : put_prev_task_idle <-__schedule
2209 <idle>-0 3dN.2 28us : pick_next_t 1948 <idle>-0 3dN.2 28us : pick_next_task_stop <-pick_next_task
2210 <idle>-0 3dN.2 28us : pick_next_t 1949 <idle>-0 3dN.2 28us : pick_next_task_rt <-pick_next_task
2211 <idle>-0 3dN.2 29us : dequeue_pus 1950 <idle>-0 3dN.2 29us : dequeue_pushable_task <-pick_next_task_rt
2212 <idle>-0 3d..3 29us : __schedule 1951 <idle>-0 3d..3 29us : __schedule <-preempt_schedule
2213 <idle>-0 3d..3 30us : 0:120: 1952 <idle>-0 3d..3 30us : 0:120:R ==> [003] 2448: 94:R sleep
2214 1953
2215 This isn't that big of a trace, even with fun 1954 This isn't that big of a trace, even with function tracing enabled,
2216 so I included the entire trace. 1955 so I included the entire trace.
2217 1956
2218 The interrupt went off while when the system 1957 The interrupt went off while when the system was idle. Somewhere
2219 before task_woken_rt() was called, the NEED_R 1958 before task_woken_rt() was called, the NEED_RESCHED flag was set,
2220 this is indicated by the first occurrence of 1959 this is indicated by the first occurrence of the 'N' flag.
2221 1960
2222 Latency tracing and events 1961 Latency tracing and events
2223 -------------------------- 1962 --------------------------
2224 As function tracing can induce a much larger 1963 As function tracing can induce a much larger latency, but without
2225 seeing what happens within the latency it is 1964 seeing what happens within the latency it is hard to know what
2226 caused it. There is a middle ground, and that 1965 caused it. There is a middle ground, and that is with enabling
2227 events. 1966 events.
2228 :: 1967 ::
2229 1968
2230 # echo 0 > options/function-trace 1969 # echo 0 > options/function-trace
2231 # echo wakeup_rt > current_tracer 1970 # echo wakeup_rt > current_tracer
2232 # echo 1 > events/enable 1971 # echo 1 > events/enable
2233 # echo 1 > tracing_on 1972 # echo 1 > tracing_on
2234 # echo 0 > tracing_max_latency 1973 # echo 0 > tracing_max_latency
2235 # chrt -f 5 sleep 1 1974 # chrt -f 5 sleep 1
2236 # echo 0 > tracing_on 1975 # echo 0 > tracing_on
2237 # cat trace 1976 # cat trace
2238 # tracer: wakeup_rt 1977 # tracer: wakeup_rt
2239 # 1978 #
2240 # wakeup_rt latency trace v1.1.5 on 3.8.0-t 1979 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
2241 # ----------------------------------------- 1980 # --------------------------------------------------------------------
2242 # latency: 6 us, #12/12, CPU#2 | (M:preempt 1981 # latency: 6 us, #12/12, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
2243 # ----------------- 1982 # -----------------
2244 # | task: sleep-5882 (uid:0 nice:0 polic 1983 # | task: sleep-5882 (uid:0 nice:0 policy:1 rt_prio:5)
2245 # ----------------- 1984 # -----------------
2246 # 1985 #
2247 # _------=> CPU# 1986 # _------=> CPU#
2248 # / _-----=> irqs-off 1987 # / _-----=> irqs-off
2249 # | / _----=> need-resched 1988 # | / _----=> need-resched
2250 # || / _---=> hardirq/softir 1989 # || / _---=> hardirq/softirq
2251 # ||| / _--=> preempt-depth 1990 # ||| / _--=> preempt-depth
2252 # |||| / delay 1991 # |||| / delay
2253 # cmd pid ||||| time | caller 1992 # cmd pid ||||| time | caller
2254 # \ / ||||| \ | / 1993 # \ / ||||| \ | /
2255 <idle>-0 2d.h4 0us : 0:120: 1994 <idle>-0 2d.h4 0us : 0:120:R + [002] 5882: 94:R sleep
2256 <idle>-0 2d.h4 0us : ttwu_do_act 1995 <idle>-0 2d.h4 0us : ttwu_do_activate.constprop.87 <-try_to_wake_up
2257 <idle>-0 2d.h4 1us : sched_wakeu 1996 <idle>-0 2d.h4 1us : sched_wakeup: comm=sleep pid=5882 prio=94 success=1 target_cpu=002
2258 <idle>-0 2dNh2 1us : hrtimer_exp 1997 <idle>-0 2dNh2 1us : hrtimer_expire_exit: hrtimer=ffff88007796feb8
2259 <idle>-0 2.N.2 2us : power_end: 1998 <idle>-0 2.N.2 2us : power_end: cpu_id=2
2260 <idle>-0 2.N.2 3us : cpu_idle: s 1999 <idle>-0 2.N.2 3us : cpu_idle: state=4294967295 cpu_id=2
2261 <idle>-0 2dN.3 4us : hrtimer_can 2000 <idle>-0 2dN.3 4us : hrtimer_cancel: hrtimer=ffff88007d50d5e0
2262 <idle>-0 2dN.3 4us : hrtimer_sta 2001 <idle>-0 2dN.3 4us : hrtimer_start: hrtimer=ffff88007d50d5e0 function=tick_sched_timer expires=34311211000000 softexpires=34311211000000
2263 <idle>-0 2.N.2 5us : rcu_utiliza 2002 <idle>-0 2.N.2 5us : rcu_utilization: Start context switch
2264 <idle>-0 2.N.2 5us : rcu_utiliza 2003 <idle>-0 2.N.2 5us : rcu_utilization: End context switch
2265 <idle>-0 2d..3 6us : __schedule 2004 <idle>-0 2d..3 6us : __schedule <-schedule
2266 <idle>-0 2d..3 6us : 0:120: 2005 <idle>-0 2d..3 6us : 0:120:R ==> [002] 5882: 94:R sleep
2267 2006
2268 2007
2269 Hardware Latency Detector 2008 Hardware Latency Detector
2270 ------------------------- 2009 -------------------------
2271 2010
2272 The hardware latency detector is executed by 2011 The hardware latency detector is executed by enabling the "hwlat" tracer.
2273 2012
2274 NOTE, this tracer will affect the performance 2013 NOTE, this tracer will affect the performance of the system as it will
2275 periodically make a CPU constantly busy with 2014 periodically make a CPU constantly busy with interrupts disabled.
2276 :: 2015 ::
2277 2016
2278 # echo hwlat > current_tracer 2017 # echo hwlat > current_tracer
2279 # sleep 100 2018 # sleep 100
2280 # cat trace 2019 # cat trace
2281 # tracer: hwlat 2020 # tracer: hwlat
2282 # 2021 #
2283 # entries-in-buffer/entries-written: 13/13 <<
2284 # <<
2285 # _-----=> irq 2022 # _-----=> irqs-off
2286 # / _----=> nee 2023 # / _----=> need-resched
2287 # | / _---=> har 2024 # | / _---=> hardirq/softirq
2288 # || / _--=> pre 2025 # || / _--=> preempt-depth
2289 # ||| / dela 2026 # ||| / delay
2290 # TASK-PID CPU# |||| TIMEST 2027 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2291 # | | | |||| | 2028 # | | | |||| | |
2292 <...>-1729 [001] d... 678.473 !! 2029 <...>-3638 [001] d... 19452.055471: #1 inner/outer(us): 12/14 ts:1499801089.066141940
2293 <...>-1729 [004] d... 689.556 !! 2030 <...>-3638 [003] d... 19454.071354: #2 inner/outer(us): 11/9 ts:1499801091.082164365
2294 <...>-1729 [005] d... 714.756 !! 2031 <...>-3638 [002] dn.. 19461.126852: #3 inner/outer(us): 12/9 ts:1499801098.138150062
2295 <...>-1729 [001] d... 718.788 !! 2032 <...>-3638 [001] d... 19488.340960: #4 inner/outer(us): 8/12 ts:1499801125.354139633
2296 <...>-1729 [002] d... 719.796 !! 2033 <...>-3638 [003] d... 19494.388553: #5 inner/outer(us): 8/12 ts:1499801131.402150961
2297 <...>-1729 [006] d... 844.787 !! 2034 <...>-3638 [003] d... 19501.283419: #6 inner/outer(us): 0/12 ts:1499801138.297435289 nmi-total:4 nmi-count:1
2298 <...>-1729 [003] d... 849.827 <<
2299 <...>-1729 [007] d... 853.859 <<
2300 <...>-1729 [001] d... 855.874 <<
2301 <...>-1729 [001] d... 863.938 <<
2302 <...>-1729 [007] d... 878.050 <<
2303 <...>-1729 [007] d... 886.114 <<
2304 2035
2305 2036
2306 The above output is somewhat the same in the 2037 The above output is somewhat the same in the header. All events will have
2307 interrupts disabled 'd'. Under the FUNCTION t 2038 interrupts disabled 'd'. Under the FUNCTION title there is:
2308 2039
2309 #1 2040 #1
2310 This is the count of events recorded 2041 This is the count of events recorded that were greater than the
2311 tracing_threshold (See below). 2042 tracing_threshold (See below).
2312 2043
2313 inner/outer(us): 11/11 !! 2044 inner/outer(us): 12/14
2314 2045
2315 This shows two numbers as "inner latenc 2046 This shows two numbers as "inner latency" and "outer latency". The test
2316 runs in a loop checking a timestamp twi 2047 runs in a loop checking a timestamp twice. The latency detected within
2317 the two timestamps is the "inner latenc 2048 the two timestamps is the "inner latency" and the latency detected
2318 after the previous timestamp and the ne 2049 after the previous timestamp and the next timestamp in the loop is
2319 the "outer latency". 2050 the "outer latency".
2320 2051
2321 ts:1581527483.343962693 !! 2052 ts:1499801089.066141940
2322 <<
2323 The absolute timestamp that the first l <<
2324 2053
2325 count:6 !! 2054 The absolute timestamp that the event happened.
2326 2055
2327 The number of times a latency was detec !! 2056 nmi-total:4 nmi-count:1
2328 <<
2329 nmi-total:7 nmi-count:1 <<
2330 2057
2331 On architectures that support it, if an 2058 On architectures that support it, if an NMI comes in during the
2332 test, the time spent in NMI is reported 2059 test, the time spent in NMI is reported in "nmi-total" (in
2333 microseconds). 2060 microseconds).
2334 2061
2335 All architectures that have NMIs will s 2062 All architectures that have NMIs will show the "nmi-count" if an
2336 NMI comes in during the test. 2063 NMI comes in during the test.
2337 2064
2338 hwlat files: 2065 hwlat files:
2339 2066
2340 tracing_threshold 2067 tracing_threshold
2341 This gets automatically set to "10" t 2068 This gets automatically set to "10" to represent 10
2342 microseconds. This is the threshold o 2069 microseconds. This is the threshold of latency that
2343 needs to be detected before the trace 2070 needs to be detected before the trace will be recorded.
2344 2071
2345 Note, when hwlat tracer is finished ( 2072 Note, when hwlat tracer is finished (another tracer is
2346 written into "current_tracer"), the o 2073 written into "current_tracer"), the original value for
2347 tracing_threshold is placed back into 2074 tracing_threshold is placed back into this file.
2348 2075
2349 hwlat_detector/width 2076 hwlat_detector/width
2350 The length of time the test runs with 2077 The length of time the test runs with interrupts disabled.
2351 2078
2352 hwlat_detector/window 2079 hwlat_detector/window
2353 The length of time of the window whic 2080 The length of time of the window which the test
2354 runs. That is, the test will run for 2081 runs. That is, the test will run for "width"
2355 microseconds per "window" microsecond 2082 microseconds per "window" microseconds
2356 2083
2357 tracing_cpumask 2084 tracing_cpumask
2358 When the test is started. A kernel th 2085 When the test is started. A kernel thread is created that
2359 runs the test. This thread will alter 2086 runs the test. This thread will alternate between CPUs
2360 listed in the tracing_cpumask between 2087 listed in the tracing_cpumask between each period
2361 (one "window"). To limit the test to 2088 (one "window"). To limit the test to specific CPUs
2362 set the mask in this file to only the 2089 set the mask in this file to only the CPUs that the test
2363 should run on. 2090 should run on.
2364 2091
2365 function 2092 function
2366 -------- 2093 --------
2367 2094
2368 This tracer is the function tracer. Enabling 2095 This tracer is the function tracer. Enabling the function tracer
2369 can be done from the debug file system. Make 2096 can be done from the debug file system. Make sure the
2370 ftrace_enabled is set; otherwise this tracer 2097 ftrace_enabled is set; otherwise this tracer is a nop.
2371 See the "ftrace_enabled" section below. 2098 See the "ftrace_enabled" section below.
2372 :: 2099 ::
2373 2100
2374 # sysctl kernel.ftrace_enabled=1 2101 # sysctl kernel.ftrace_enabled=1
2375 # echo function > current_tracer 2102 # echo function > current_tracer
2376 # echo 1 > tracing_on 2103 # echo 1 > tracing_on
2377 # usleep 1 2104 # usleep 1
2378 # echo 0 > tracing_on 2105 # echo 0 > tracing_on
2379 # cat trace 2106 # cat trace
2380 # tracer: function 2107 # tracer: function
2381 # 2108 #
2382 # entries-in-buffer/entries-written: 24799/ 2109 # entries-in-buffer/entries-written: 24799/24799 #P:4
2383 # 2110 #
2384 # _-----=> irq 2111 # _-----=> irqs-off
2385 # / _----=> nee 2112 # / _----=> need-resched
2386 # | / _---=> har 2113 # | / _---=> hardirq/softirq
2387 # || / _--=> pre 2114 # || / _--=> preempt-depth
2388 # ||| / dela 2115 # ||| / delay
2389 # TASK-PID CPU# |||| TIMEST 2116 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2390 # | | | |||| | 2117 # | | | |||| | |
2391 bash-1994 [002] .... 3082.063 2118 bash-1994 [002] .... 3082.063030: mutex_unlock <-rb_simple_write
2392 bash-1994 [002] .... 3082.063 2119 bash-1994 [002] .... 3082.063031: __mutex_unlock_slowpath <-mutex_unlock
2393 bash-1994 [002] .... 3082.063 2120 bash-1994 [002] .... 3082.063031: __fsnotify_parent <-fsnotify_modify
2394 bash-1994 [002] .... 3082.063 2121 bash-1994 [002] .... 3082.063032: fsnotify <-fsnotify_modify
2395 bash-1994 [002] .... 3082.063 2122 bash-1994 [002] .... 3082.063032: __srcu_read_lock <-fsnotify
2396 bash-1994 [002] .... 3082.063 2123 bash-1994 [002] .... 3082.063032: add_preempt_count <-__srcu_read_lock
2397 bash-1994 [002] ...1 3082.063 2124 bash-1994 [002] ...1 3082.063032: sub_preempt_count <-__srcu_read_lock
2398 bash-1994 [002] .... 3082.063 2125 bash-1994 [002] .... 3082.063033: __srcu_read_unlock <-fsnotify
2399 [...] 2126 [...]
2400 2127
2401 2128
2402 Note: function tracer uses ring buffers to st 2129 Note: function tracer uses ring buffers to store the above
2403 entries. The newest data may overwrite the ol 2130 entries. The newest data may overwrite the oldest data.
2404 Sometimes using echo to stop the trace is not 2131 Sometimes using echo to stop the trace is not sufficient because
2405 the tracing could have overwritten the data t 2132 the tracing could have overwritten the data that you wanted to
2406 record. For this reason, it is sometimes bett 2133 record. For this reason, it is sometimes better to disable
2407 tracing directly from a program. This allows 2134 tracing directly from a program. This allows you to stop the
2408 tracing at the point that you hit the part th 2135 tracing at the point that you hit the part that you are
2409 interested in. To disable the tracing directl 2136 interested in. To disable the tracing directly from a C program,
2410 something like following code snippet can be 2137 something like following code snippet can be used::
2411 2138
2412 int trace_fd; 2139 int trace_fd;
2413 [...] 2140 [...]
2414 int main(int argc, char *argv[]) { 2141 int main(int argc, char *argv[]) {
2415 [...] 2142 [...]
2416 trace_fd = open(tracing_file( 2143 trace_fd = open(tracing_file("tracing_on"), O_WRONLY);
2417 [...] 2144 [...]
2418 if (condition_hit()) { 2145 if (condition_hit()) {
2419 write(trace_fd, "0", 2146 write(trace_fd, "0", 1);
2420 } 2147 }
2421 [...] 2148 [...]
2422 } 2149 }
2423 2150
2424 2151
2425 Single thread tracing 2152 Single thread tracing
2426 --------------------- 2153 ---------------------
2427 2154
2428 By writing into set_ftrace_pid you can trace 2155 By writing into set_ftrace_pid you can trace a
2429 single thread. For example:: 2156 single thread. For example::
2430 2157
2431 # cat set_ftrace_pid 2158 # cat set_ftrace_pid
2432 no pid 2159 no pid
2433 # echo 3111 > set_ftrace_pid 2160 # echo 3111 > set_ftrace_pid
2434 # cat set_ftrace_pid 2161 # cat set_ftrace_pid
2435 3111 2162 3111
2436 # echo function > current_tracer 2163 # echo function > current_tracer
2437 # cat trace | head 2164 # cat trace | head
2438 # tracer: function 2165 # tracer: function
2439 # 2166 #
2440 # TASK-PID CPU# TIMESTAMP 2167 # TASK-PID CPU# TIMESTAMP FUNCTION
2441 # | | | | 2168 # | | | | |
2442 yum-updatesd-3111 [003] 1637.254676: 2169 yum-updatesd-3111 [003] 1637.254676: finish_task_switch <-thread_return
2443 yum-updatesd-3111 [003] 1637.254681: 2170 yum-updatesd-3111 [003] 1637.254681: hrtimer_cancel <-schedule_hrtimeout_range
2444 yum-updatesd-3111 [003] 1637.254682: 2171 yum-updatesd-3111 [003] 1637.254682: hrtimer_try_to_cancel <-hrtimer_cancel
2445 yum-updatesd-3111 [003] 1637.254683: 2172 yum-updatesd-3111 [003] 1637.254683: lock_hrtimer_base <-hrtimer_try_to_cancel
2446 yum-updatesd-3111 [003] 1637.254685: 2173 yum-updatesd-3111 [003] 1637.254685: fget_light <-do_sys_poll
2447 yum-updatesd-3111 [003] 1637.254686: 2174 yum-updatesd-3111 [003] 1637.254686: pipe_poll <-do_sys_poll
2448 # echo > set_ftrace_pid 2175 # echo > set_ftrace_pid
2449 # cat trace |head 2176 # cat trace |head
2450 # tracer: function 2177 # tracer: function
2451 # 2178 #
2452 # TASK-PID CPU# TIMESTAMP 2179 # TASK-PID CPU# TIMESTAMP FUNCTION
2453 # | | | | 2180 # | | | | |
2454 ##### CPU 3 buffer started #### 2181 ##### CPU 3 buffer started ####
2455 yum-updatesd-3111 [003] 1701.957688: 2182 yum-updatesd-3111 [003] 1701.957688: free_poll_entry <-poll_freewait
2456 yum-updatesd-3111 [003] 1701.957689: 2183 yum-updatesd-3111 [003] 1701.957689: remove_wait_queue <-free_poll_entry
2457 yum-updatesd-3111 [003] 1701.957691: 2184 yum-updatesd-3111 [003] 1701.957691: fput <-free_poll_entry
2458 yum-updatesd-3111 [003] 1701.957692: 2185 yum-updatesd-3111 [003] 1701.957692: audit_syscall_exit <-sysret_audit
2459 yum-updatesd-3111 [003] 1701.957693: 2186 yum-updatesd-3111 [003] 1701.957693: path_put <-audit_syscall_exit
2460 2187
2461 If you want to trace a function when executin 2188 If you want to trace a function when executing, you could use
2462 something like this simple program. 2189 something like this simple program.
2463 :: 2190 ::
2464 2191
2465 #include <stdio.h> 2192 #include <stdio.h>
2466 #include <stdlib.h> 2193 #include <stdlib.h>
2467 #include <sys/types.h> 2194 #include <sys/types.h>
2468 #include <sys/stat.h> 2195 #include <sys/stat.h>
2469 #include <fcntl.h> 2196 #include <fcntl.h>
2470 #include <unistd.h> 2197 #include <unistd.h>
2471 #include <string.h> 2198 #include <string.h>
2472 2199
2473 #define _STR(x) #x 2200 #define _STR(x) #x
2474 #define STR(x) _STR(x) 2201 #define STR(x) _STR(x)
2475 #define MAX_PATH 256 2202 #define MAX_PATH 256
2476 2203
2477 const char *find_tracefs(void) 2204 const char *find_tracefs(void)
2478 { 2205 {
2479 static char tracefs[MAX_PATH+1 2206 static char tracefs[MAX_PATH+1];
2480 static int tracefs_found; 2207 static int tracefs_found;
2481 char type[100]; 2208 char type[100];
2482 FILE *fp; 2209 FILE *fp;
2483 2210
2484 if (tracefs_found) 2211 if (tracefs_found)
2485 return tracefs; 2212 return tracefs;
2486 2213
2487 if ((fp = fopen("/proc/mounts" 2214 if ((fp = fopen("/proc/mounts","r")) == NULL) {
2488 perror("/proc/mounts") 2215 perror("/proc/mounts");
2489 return NULL; 2216 return NULL;
2490 } 2217 }
2491 2218
2492 while (fscanf(fp, "%*s %" 2219 while (fscanf(fp, "%*s %"
2493 STR(MAX_PATH) 2220 STR(MAX_PATH)
2494 "s %99s %*s %*d 2221 "s %99s %*s %*d %*d\n",
2495 tracefs, type) = 2222 tracefs, type) == 2) {
2496 if (strcmp(type, "trac 2223 if (strcmp(type, "tracefs") == 0)
2497 break; 2224 break;
2498 } 2225 }
2499 fclose(fp); 2226 fclose(fp);
2500 2227
2501 if (strcmp(type, "tracefs") != 2228 if (strcmp(type, "tracefs") != 0) {
2502 fprintf(stderr, "trace 2229 fprintf(stderr, "tracefs not mounted");
2503 return NULL; 2230 return NULL;
2504 } 2231 }
2505 2232
2506 strcat(tracefs, "/tracing/"); 2233 strcat(tracefs, "/tracing/");
2507 tracefs_found = 1; 2234 tracefs_found = 1;
2508 2235
2509 return tracefs; 2236 return tracefs;
2510 } 2237 }
2511 2238
2512 const char *tracing_file(const char * 2239 const char *tracing_file(const char *file_name)
2513 { 2240 {
2514 static char trace_file[MAX_PAT 2241 static char trace_file[MAX_PATH+1];
2515 snprintf(trace_file, MAX_PATH, 2242 snprintf(trace_file, MAX_PATH, "%s/%s", find_tracefs(), file_name);
2516 return trace_file; 2243 return trace_file;
2517 } 2244 }
2518 2245
2519 int main (int argc, char **argv) 2246 int main (int argc, char **argv)
2520 { 2247 {
2521 if (argc < 1) 2248 if (argc < 1)
2522 exit(-1); 2249 exit(-1);
2523 2250
2524 if (fork() > 0) { 2251 if (fork() > 0) {
2525 int fd, ffd; 2252 int fd, ffd;
2526 char line[64]; 2253 char line[64];
2527 int s; 2254 int s;
2528 2255
2529 ffd = open(tracing_fi 2256 ffd = open(tracing_file("current_tracer"), O_WRONLY);
2530 if (ffd < 0) 2257 if (ffd < 0)
2531 exit(-1); 2258 exit(-1);
2532 write(ffd, "nop", 3); 2259 write(ffd, "nop", 3);
2533 2260
2534 fd = open(tracing_fil 2261 fd = open(tracing_file("set_ftrace_pid"), O_WRONLY);
2535 s = sprintf(line, "%d 2262 s = sprintf(line, "%d\n", getpid());
2536 write(fd, line, s); 2263 write(fd, line, s);
2537 2264
2538 write(ffd, "function" 2265 write(ffd, "function", 8);
2539 2266
2540 close(fd); 2267 close(fd);
2541 close(ffd); 2268 close(ffd);
2542 2269
2543 execvp(argv[1], argv+ 2270 execvp(argv[1], argv+1);
2544 } 2271 }
2545 2272
2546 return 0; 2273 return 0;
2547 } 2274 }
2548 2275
2549 Or this simple script! 2276 Or this simple script!
2550 :: 2277 ::
2551 2278
2552 #!/bin/bash 2279 #!/bin/bash
2553 2280
2554 tracefs=`sed -ne 's/^tracefs \(.*\) tracefs 2281 tracefs=`sed -ne 's/^tracefs \(.*\) tracefs.*/\1/p' /proc/mounts`
2555 echo 0 > $tracefs/tracing_on !! 2282 echo nop > $tracefs/tracing/current_tracer
2556 echo $$ > $tracefs/set_ftrace_pid !! 2283 echo 0 > $tracefs/tracing/tracing_on
2557 echo function > $tracefs/current_tracer !! 2284 echo $$ > $tracefs/tracing/set_ftrace_pid
2558 echo 1 > $tracefs/tracing_on !! 2285 echo function > $tracefs/tracing/current_tracer
>> 2286 echo 1 > $tracefs/tracing/tracing_on
2559 exec "$@" 2287 exec "$@"
2560 2288
2561 2289
2562 function graph tracer 2290 function graph tracer
2563 --------------------------- 2291 ---------------------------
2564 2292
2565 This tracer is similar to the function tracer 2293 This tracer is similar to the function tracer except that it
2566 probes a function on its entry and its exit. 2294 probes a function on its entry and its exit. This is done by
2567 using a dynamically allocated stack of return 2295 using a dynamically allocated stack of return addresses in each
2568 task_struct. On function entry the tracer ove 2296 task_struct. On function entry the tracer overwrites the return
2569 address of each function traced to set a cust 2297 address of each function traced to set a custom probe. Thus the
2570 original return address is stored on the stac 2298 original return address is stored on the stack of return address
2571 in the task_struct. 2299 in the task_struct.
2572 2300
2573 Probing on both ends of a function leads to s 2301 Probing on both ends of a function leads to special features
2574 such as: 2302 such as:
2575 2303
2576 - measure of a function's time execution 2304 - measure of a function's time execution
2577 - having a reliable call stack to draw functi 2305 - having a reliable call stack to draw function calls graph
2578 2306
2579 This tracer is useful in several situations: 2307 This tracer is useful in several situations:
2580 2308
2581 - you want to find the reason of a strange ke 2309 - you want to find the reason of a strange kernel behavior and
2582 need to see what happens in detail on any a 2310 need to see what happens in detail on any areas (or specific
2583 ones). 2311 ones).
2584 2312
2585 - you are experiencing weird latencies but it 2313 - you are experiencing weird latencies but it's difficult to
2586 find its origin. 2314 find its origin.
2587 2315
2588 - you want to find quickly which path is take 2316 - you want to find quickly which path is taken by a specific
2589 function 2317 function
2590 2318
2591 - you just want to peek inside a working kern 2319 - you just want to peek inside a working kernel and want to see
2592 what happens there. 2320 what happens there.
2593 2321
2594 :: 2322 ::
2595 2323
2596 # tracer: function_graph 2324 # tracer: function_graph
2597 # 2325 #
2598 # CPU DURATION FUNCTION C 2326 # CPU DURATION FUNCTION CALLS
2599 # | | | | | | 2327 # | | | | | | |
2600 2328
2601 0) | sys_open() { 2329 0) | sys_open() {
2602 0) | do_sys_open() { 2330 0) | do_sys_open() {
2603 0) | getname() { 2331 0) | getname() {
2604 0) | kmem_cache_alloc 2332 0) | kmem_cache_alloc() {
2605 0) 1.382 us | __might_sleep( 2333 0) 1.382 us | __might_sleep();
2606 0) 2.478 us | } 2334 0) 2.478 us | }
2607 0) | strncpy_from_use 2335 0) | strncpy_from_user() {
2608 0) | might_fault() 2336 0) | might_fault() {
2609 0) 1.389 us | __might_slee 2337 0) 1.389 us | __might_sleep();
2610 0) 2.553 us | } 2338 0) 2.553 us | }
2611 0) 3.807 us | } 2339 0) 3.807 us | }
2612 0) 7.876 us | } 2340 0) 7.876 us | }
2613 0) | alloc_fd() { 2341 0) | alloc_fd() {
2614 0) 0.668 us | _spin_lock(); 2342 0) 0.668 us | _spin_lock();
2615 0) 0.570 us | expand_files(); 2343 0) 0.570 us | expand_files();
2616 0) 0.586 us | _spin_unlock(); 2344 0) 0.586 us | _spin_unlock();
2617 2345
2618 2346
2619 There are several columns that can be dynamic 2347 There are several columns that can be dynamically
2620 enabled/disabled. You can use every combinati 2348 enabled/disabled. You can use every combination of options you
2621 want, depending on your needs. 2349 want, depending on your needs.
2622 2350
2623 - The cpu number on which the function execut 2351 - The cpu number on which the function executed is default
2624 enabled. It is sometimes better to only tr 2352 enabled. It is sometimes better to only trace one cpu (see
2625 tracing_cpumask file) or you might sometime !! 2353 tracing_cpu_mask file) or you might sometimes see unordered
2626 function calls while cpu tracing switch. 2354 function calls while cpu tracing switch.
2627 2355
2628 - hide: echo nofuncgraph-cpu > trace_ 2356 - hide: echo nofuncgraph-cpu > trace_options
2629 - show: echo funcgraph-cpu > trace_op 2357 - show: echo funcgraph-cpu > trace_options
2630 2358
2631 - The duration (function's time of execution) 2359 - The duration (function's time of execution) is displayed on
2632 the closing bracket line of a function or o 2360 the closing bracket line of a function or on the same line
2633 than the current function in case of a leaf 2361 than the current function in case of a leaf one. It is default
2634 enabled. 2362 enabled.
2635 2363
2636 - hide: echo nofuncgraph-duration > t 2364 - hide: echo nofuncgraph-duration > trace_options
2637 - show: echo funcgraph-duration > tra 2365 - show: echo funcgraph-duration > trace_options
2638 2366
2639 - The overhead field precedes the duration fi 2367 - The overhead field precedes the duration field in case of
2640 reached duration thresholds. 2368 reached duration thresholds.
2641 2369
2642 - hide: echo nofuncgraph-overhead > t 2370 - hide: echo nofuncgraph-overhead > trace_options
2643 - show: echo funcgraph-overhead > tra 2371 - show: echo funcgraph-overhead > trace_options
2644 - depends on: funcgraph-duration 2372 - depends on: funcgraph-duration
2645 2373
2646 ie:: 2374 ie::
2647 2375
2648 3) # 1837.709 us | } /* __switch 2376 3) # 1837.709 us | } /* __switch_to */
2649 3) | finish_task_s 2377 3) | finish_task_switch() {
2650 3) 0.313 us | _raw_spin_u 2378 3) 0.313 us | _raw_spin_unlock_irq();
2651 3) 3.177 us | } 2379 3) 3.177 us | }
2652 3) # 1889.063 us | } /* __schedule 2380 3) # 1889.063 us | } /* __schedule */
2653 3) ! 140.417 us | } /* __schedule * 2381 3) ! 140.417 us | } /* __schedule */
2654 3) # 2034.948 us | } /* schedule */ 2382 3) # 2034.948 us | } /* schedule */
2655 3) * 33998.59 us | } /* schedule_preempt 2383 3) * 33998.59 us | } /* schedule_preempt_disabled */
2656 2384
2657 [...] 2385 [...]
2658 2386
2659 1) 0.260 us | msecs_to_ 2387 1) 0.260 us | msecs_to_jiffies();
2660 1) 0.313 us | __rcu_rea 2388 1) 0.313 us | __rcu_read_unlock();
2661 1) + 61.770 us | } 2389 1) + 61.770 us | }
2662 1) + 64.479 us | } 2390 1) + 64.479 us | }
2663 1) 0.313 us | rcu_bh_qs(); 2391 1) 0.313 us | rcu_bh_qs();
2664 1) 0.313 us | __local_bh_en 2392 1) 0.313 us | __local_bh_enable();
2665 1) ! 217.240 us | } 2393 1) ! 217.240 us | }
2666 1) 0.365 us | idle_cpu(); 2394 1) 0.365 us | idle_cpu();
2667 1) | rcu_irq_exit() 2395 1) | rcu_irq_exit() {
2668 1) 0.417 us | rcu_eqs_enter 2396 1) 0.417 us | rcu_eqs_enter_common.isra.47();
2669 1) 3.125 us | } 2397 1) 3.125 us | }
2670 1) ! 227.812 us | } 2398 1) ! 227.812 us | }
2671 1) ! 457.395 us | } 2399 1) ! 457.395 us | }
2672 1) @ 119760.2 us | } 2400 1) @ 119760.2 us | }
2673 2401
2674 [...] 2402 [...]
2675 2403
2676 2) | handle_IPI() { 2404 2) | handle_IPI() {
2677 1) 6.979 us | } 2405 1) 6.979 us | }
2678 2) 0.417 us | scheduler_ipi(); 2406 2) 0.417 us | scheduler_ipi();
2679 1) 9.791 us | } 2407 1) 9.791 us | }
2680 1) + 12.917 us | } 2408 1) + 12.917 us | }
2681 2) 3.490 us | } 2409 2) 3.490 us | }
2682 1) + 15.729 us | } 2410 1) + 15.729 us | }
2683 1) + 18.542 us | } 2411 1) + 18.542 us | }
2684 2) $ 3594274 us | } 2412 2) $ 3594274 us | }
2685 2413
2686 Flags:: 2414 Flags::
2687 2415
2688 + means that the function exceeded 10 usecs 2416 + means that the function exceeded 10 usecs.
2689 ! means that the function exceeded 100 usec 2417 ! means that the function exceeded 100 usecs.
2690 # means that the function exceeded 1000 use 2418 # means that the function exceeded 1000 usecs.
2691 * means that the function exceeded 10 msecs 2419 * means that the function exceeded 10 msecs.
2692 @ means that the function exceeded 100 msec 2420 @ means that the function exceeded 100 msecs.
2693 $ means that the function exceeded 1 sec. 2421 $ means that the function exceeded 1 sec.
2694 2422
2695 2423
2696 - The task/pid field displays the thread cmdl 2424 - The task/pid field displays the thread cmdline and pid which
2697 executed the function. It is default disabl 2425 executed the function. It is default disabled.
2698 2426
2699 - hide: echo nofuncgraph-proc > trace 2427 - hide: echo nofuncgraph-proc > trace_options
2700 - show: echo funcgraph-proc > trace_o 2428 - show: echo funcgraph-proc > trace_options
2701 2429
2702 ie:: 2430 ie::
2703 2431
2704 # tracer: function_graph 2432 # tracer: function_graph
2705 # 2433 #
2706 # CPU TASK/PID DURATION 2434 # CPU TASK/PID DURATION FUNCTION CALLS
2707 # | | | | | 2435 # | | | | | | | | |
2708 0) sh-4802 | | 2436 0) sh-4802 | | d_free() {
2709 0) sh-4802 | | 2437 0) sh-4802 | | call_rcu() {
2710 0) sh-4802 | | 2438 0) sh-4802 | | __call_rcu() {
2711 0) sh-4802 | 0.616 us | 2439 0) sh-4802 | 0.616 us | rcu_process_gp_end();
2712 0) sh-4802 | 0.586 us | 2440 0) sh-4802 | 0.586 us | check_for_new_grace_period();
2713 0) sh-4802 | 2.899 us | 2441 0) sh-4802 | 2.899 us | }
2714 0) sh-4802 | 4.040 us | 2442 0) sh-4802 | 4.040 us | }
2715 0) sh-4802 | 5.151 us | 2443 0) sh-4802 | 5.151 us | }
2716 0) sh-4802 | + 49.370 us | 2444 0) sh-4802 | + 49.370 us | }
2717 2445
2718 2446
2719 - The absolute time field is an absolute time 2447 - The absolute time field is an absolute timestamp given by the
2720 system clock since it started. A snapshot o 2448 system clock since it started. A snapshot of this time is
2721 given on each entry/exit of functions 2449 given on each entry/exit of functions
2722 2450
2723 - hide: echo nofuncgraph-abstime > tr 2451 - hide: echo nofuncgraph-abstime > trace_options
2724 - show: echo funcgraph-abstime > trac 2452 - show: echo funcgraph-abstime > trace_options
2725 2453
2726 ie:: 2454 ie::
2727 2455
2728 # 2456 #
2729 # TIME CPU DURATION 2457 # TIME CPU DURATION FUNCTION CALLS
2730 # | | | | 2458 # | | | | | | | |
2731 360.774522 | 1) 0.541 us | 2459 360.774522 | 1) 0.541 us | }
2732 360.774522 | 1) 4.663 us | 2460 360.774522 | 1) 4.663 us | }
2733 360.774523 | 1) 0.541 us | 2461 360.774523 | 1) 0.541 us | __wake_up_bit();
2734 360.774524 | 1) 6.796 us | 2462 360.774524 | 1) 6.796 us | }
2735 360.774524 | 1) 7.952 us | 2463 360.774524 | 1) 7.952 us | }
2736 360.774525 | 1) 9.063 us | 2464 360.774525 | 1) 9.063 us | }
2737 360.774525 | 1) 0.615 us | 2465 360.774525 | 1) 0.615 us | journal_mark_dirty();
2738 360.774527 | 1) 0.578 us | 2466 360.774527 | 1) 0.578 us | __brelse();
2739 360.774528 | 1) | 2467 360.774528 | 1) | reiserfs_prepare_for_journal() {
2740 360.774528 | 1) | 2468 360.774528 | 1) | unlock_buffer() {
2741 360.774529 | 1) | 2469 360.774529 | 1) | wake_up_bit() {
2742 360.774529 | 1) | 2470 360.774529 | 1) | bit_waitqueue() {
2743 360.774530 | 1) 0.594 us | 2471 360.774530 | 1) 0.594 us | __phys_addr();
2744 2472
2745 2473
2746 The function name is always displayed after t 2474 The function name is always displayed after the closing bracket
2747 for a function if the start of that function 2475 for a function if the start of that function is not in the
2748 trace buffer. 2476 trace buffer.
2749 2477
2750 Display of the function name after the closin 2478 Display of the function name after the closing bracket may be
2751 enabled for functions whose start is in the t 2479 enabled for functions whose start is in the trace buffer,
2752 allowing easier searching with grep for funct 2480 allowing easier searching with grep for function durations.
2753 It is default disabled. 2481 It is default disabled.
2754 2482
2755 - hide: echo nofuncgraph-tail > trace 2483 - hide: echo nofuncgraph-tail > trace_options
2756 - show: echo funcgraph-tail > trace_o 2484 - show: echo funcgraph-tail > trace_options
2757 2485
2758 Example with nofuncgraph-tail (default):: 2486 Example with nofuncgraph-tail (default)::
2759 2487
2760 0) | putname() { 2488 0) | putname() {
2761 0) | kmem_cache_free 2489 0) | kmem_cache_free() {
2762 0) 0.518 us | __phys_addr() 2490 0) 0.518 us | __phys_addr();
2763 0) 1.757 us | } 2491 0) 1.757 us | }
2764 0) 2.861 us | } 2492 0) 2.861 us | }
2765 2493
2766 Example with funcgraph-tail:: 2494 Example with funcgraph-tail::
2767 2495
2768 0) | putname() { 2496 0) | putname() {
2769 0) | kmem_cache_free 2497 0) | kmem_cache_free() {
2770 0) 0.518 us | __phys_addr() 2498 0) 0.518 us | __phys_addr();
2771 0) 1.757 us | } /* kmem_cache 2499 0) 1.757 us | } /* kmem_cache_free() */
2772 0) 2.861 us | } /* putname() */ 2500 0) 2.861 us | } /* putname() */
2773 2501
2774 The return value of each traced function can <<
2775 an equal sign "=". When encountering system c <<
2776 can be very helpful to quickly locate the fun <<
2777 returns an error code. <<
2778 <<
2779 - hide: echo nofuncgraph-retval > tra <<
2780 - show: echo funcgraph-retval > trace <<
2781 <<
2782 Example with funcgraph-retval:: <<
2783 <<
2784 1) | cgroup_migrate() { <<
2785 1) 0.651 us | cgroup_migrate_ad <<
2786 1) | cgroup_migrate_ex <<
2787 1) | cpu_cgroup_can_ <<
2788 1) | cgroup_taskse <<
2789 1) 0.732 us | cgroup_task <<
2790 1) 1.232 us | } /* cgroup_t <<
2791 1) 0.380 us | sched_rt_can_ <<
2792 1) 2.335 us | } /* cpu_cgroup <<
2793 1) 4.369 us | } /* cgroup_migra <<
2794 1) 7.143 us | } /* cgroup_migrate <<
2795 <<
2796 The above example shows that the function cpu <<
2797 returned the error code -22 firstly, then we <<
2798 of this function to get the root cause. <<
2799 <<
2800 When the option funcgraph-retval-hex is not s <<
2801 be displayed in a smart way. Specifically, if <<
2802 it will be printed in signed decimal format, <<
2803 printed in hexadecimal format. <<
2804 <<
2805 - smart: echo nofuncgraph-retval-hex <<
2806 - hexadecimal: echo funcgraph-retval- <<
2807 <<
2808 Example with funcgraph-retval-hex:: <<
2809 <<
2810 1) | cgroup_migrate() <<
2811 1) 0.651 us | cgroup_migrate_ <<
2812 1) | cgroup_migrate_ <<
2813 1) | cpu_cgroup_ca <<
2814 1) | cgroup_task <<
2815 1) 0.732 us | cgroup_ta <<
2816 1) 1.232 us | } /* cgroup <<
2817 1) 0.380 us | sched_rt_ca <<
2818 1) 2.335 us | } /* cpu_cgro <<
2819 1) 4.369 us | } /* cgroup_mig <<
2820 1) 7.143 us | } /* cgroup_migra <<
2821 <<
2822 At present, there are some limitations when u <<
2823 option, and these limitations will be elimina <<
2824 <<
2825 - Even if the function return type is void, a <<
2826 be printed, and you can just ignore it. <<
2827 <<
2828 - Even if return values are stored in multipl <<
2829 value contained in the first register will <<
2830 To illustrate, in the x86 architecture, eax <<
2831 a 64-bit return value, with the lower 32 bi <<
2832 upper 32 bits saved in edx. However, only t <<
2833 will be recorded and printed. <<
2834 <<
2835 - In certain procedure call standards, such a <<
2836 type is smaller than a GPR, it is the respo <<
2837 to perform the narrowing, and the upper bit <<
2838 Therefore, it is advisable to check the cod <<
2839 when using a u8 in a 64-bit GPR, bits [63:8 <<
2840 especially when larger types are truncated, <<
2841 Here are some specific cases to illustrate <<
2842 <<
2843 **Case One**: <<
2844 <<
2845 The function narrow_to_u8 is defined as fol <<
2846 <<
2847 u8 narrow_to_u8(u64 val) <<
2848 { <<
2849 // implicitly truncated <<
2850 return val; <<
2851 } <<
2852 <<
2853 It may be compiled to:: <<
2854 <<
2855 narrow_to_u8: <<
2856 < ... ftrace instrumentation <<
2857 RET <<
2858 <<
2859 If you pass 0x123456789abcdef to this funct <<
2860 it may be recorded as 0x123456789abcdef ins <<
2861 <<
2862 **Case Two**: <<
2863 <<
2864 The function error_if_not_4g_aligned is def <<
2865 <<
2866 int error_if_not_4g_aligned(u64 val) <<
2867 { <<
2868 if (val & GENMASK(31, 0)) <<
2869 return -EINVAL; <<
2870 <<
2871 return 0; <<
2872 } <<
2873 <<
2874 It could be compiled to:: <<
2875 <<
2876 error_if_not_4g_aligned: <<
2877 CBNZ w0, .Lnot_aligned <<
2878 RET // bi <<
2879 // [6 <<
2880 .Lnot_aligned: <<
2881 MOV x0, #-EINVAL <<
2882 RET <<
2883 <<
2884 When passing 0x2_0000_0000 to it, the retur <<
2885 0x2_0000_0000 instead of 0. <<
2886 <<
2887 You can put some comments on specific functio 2502 You can put some comments on specific functions by using
2888 trace_printk() For example, if you want to pu 2503 trace_printk() For example, if you want to put a comment inside
2889 the __might_sleep() function, you just have t 2504 the __might_sleep() function, you just have to include
2890 <linux/ftrace.h> and call trace_printk() insi 2505 <linux/ftrace.h> and call trace_printk() inside __might_sleep()::
2891 2506
2892 trace_printk("I'm a comment!\n") 2507 trace_printk("I'm a comment!\n")
2893 2508
2894 will produce:: 2509 will produce::
2895 2510
2896 1) | __might_sle 2511 1) | __might_sleep() {
2897 1) | /* I'm a 2512 1) | /* I'm a comment! */
2898 1) 1.449 us | } 2513 1) 1.449 us | }
2899 2514
2900 2515
2901 You might find other useful features for this 2516 You might find other useful features for this tracer in the
2902 following "dynamic ftrace" section such as tr 2517 following "dynamic ftrace" section such as tracing only specific
2903 functions or tasks. 2518 functions or tasks.
2904 2519
2905 dynamic ftrace 2520 dynamic ftrace
2906 -------------- 2521 --------------
2907 2522
2908 If CONFIG_DYNAMIC_FTRACE is set, the system w 2523 If CONFIG_DYNAMIC_FTRACE is set, the system will run with
2909 virtually no overhead when function tracing i 2524 virtually no overhead when function tracing is disabled. The way
2910 this works is the mcount function call (place 2525 this works is the mcount function call (placed at the start of
2911 every kernel function, produced by the -pg sw 2526 every kernel function, produced by the -pg switch in gcc),
2912 starts of pointing to a simple return. (Enabl 2527 starts of pointing to a simple return. (Enabling FTRACE will
2913 include the -pg switch in the compiling of th 2528 include the -pg switch in the compiling of the kernel.)
2914 2529
2915 At compile time every C file object is run th 2530 At compile time every C file object is run through the
2916 recordmcount program (located in the scripts 2531 recordmcount program (located in the scripts directory). This
2917 program will parse the ELF headers in the C o 2532 program will parse the ELF headers in the C object to find all
2918 the locations in the .text section that call 2533 the locations in the .text section that call mcount. Starting
2919 with gcc version 4.6, the -mfentry has been a !! 2534 with gcc verson 4.6, the -mfentry has been added for x86, which
2920 calls "__fentry__" instead of "mcount". Which 2535 calls "__fentry__" instead of "mcount". Which is called before
2921 the creation of the stack frame. 2536 the creation of the stack frame.
2922 2537
2923 Note, not all sections are traced. They may b 2538 Note, not all sections are traced. They may be prevented by either
2924 a notrace, or blocked another way and all inl 2539 a notrace, or blocked another way and all inline functions are not
2925 traced. Check the "available_filter_functions 2540 traced. Check the "available_filter_functions" file to see what functions
2926 can be traced. 2541 can be traced.
2927 2542
2928 A section called "__mcount_loc" is created th 2543 A section called "__mcount_loc" is created that holds
2929 references to all the mcount/fentry call site 2544 references to all the mcount/fentry call sites in the .text section.
2930 The recordmcount program re-links this sectio 2545 The recordmcount program re-links this section back into the
2931 original object. The final linking stage of t 2546 original object. The final linking stage of the kernel will add all these
2932 references into a single table. 2547 references into a single table.
2933 2548
2934 On boot up, before SMP is initialized, the dy 2549 On boot up, before SMP is initialized, the dynamic ftrace code
2935 scans this table and updates all the location 2550 scans this table and updates all the locations into nops. It
2936 also records the locations, which are added t 2551 also records the locations, which are added to the
2937 available_filter_functions list. Modules are 2552 available_filter_functions list. Modules are processed as they
2938 are loaded and before they are executed. Whe 2553 are loaded and before they are executed. When a module is
2939 unloaded, it also removes its functions from 2554 unloaded, it also removes its functions from the ftrace function
2940 list. This is automatic in the module unload 2555 list. This is automatic in the module unload code, and the
2941 module author does not need to worry about it 2556 module author does not need to worry about it.
2942 2557
2943 When tracing is enabled, the process of modif 2558 When tracing is enabled, the process of modifying the function
2944 tracepoints is dependent on architecture. The 2559 tracepoints is dependent on architecture. The old method is to use
2945 kstop_machine to prevent races with the CPUs 2560 kstop_machine to prevent races with the CPUs executing code being
2946 modified (which can cause the CPU to do undes 2561 modified (which can cause the CPU to do undesirable things, especially
2947 if the modified code crosses cache (or page) 2562 if the modified code crosses cache (or page) boundaries), and the nops are
2948 patched back to calls. But this time, they do 2563 patched back to calls. But this time, they do not call mcount
2949 (which is just a function stub). They now cal 2564 (which is just a function stub). They now call into the ftrace
2950 infrastructure. 2565 infrastructure.
2951 2566
2952 The new method of modifying the function trac 2567 The new method of modifying the function tracepoints is to place
2953 a breakpoint at the location to be modified, 2568 a breakpoint at the location to be modified, sync all CPUs, modify
2954 the rest of the instruction not covered by th 2569 the rest of the instruction not covered by the breakpoint. Sync
2955 all CPUs again, and then remove the breakpoin 2570 all CPUs again, and then remove the breakpoint with the finished
2956 version to the ftrace call site. 2571 version to the ftrace call site.
2957 2572
2958 Some archs do not even need to monkey around 2573 Some archs do not even need to monkey around with the synchronization,
2959 and can just slap the new code on top of the 2574 and can just slap the new code on top of the old without any
2960 problems with other CPUs executing it at the 2575 problems with other CPUs executing it at the same time.
2961 2576
2962 One special side-effect to the recording of t 2577 One special side-effect to the recording of the functions being
2963 traced is that we can now selectively choose 2578 traced is that we can now selectively choose which functions we
2964 wish to trace and which ones we want the mcou 2579 wish to trace and which ones we want the mcount calls to remain
2965 as nops. 2580 as nops.
2966 2581
2967 Two files are used, one for enabling and one 2582 Two files are used, one for enabling and one for disabling the
2968 tracing of specified functions. They are: 2583 tracing of specified functions. They are:
2969 2584
2970 set_ftrace_filter 2585 set_ftrace_filter
2971 2586
2972 and 2587 and
2973 2588
2974 set_ftrace_notrace 2589 set_ftrace_notrace
2975 2590
2976 A list of available functions that you can ad 2591 A list of available functions that you can add to these files is
2977 listed in: 2592 listed in:
2978 2593
2979 available_filter_functions 2594 available_filter_functions
2980 2595
2981 :: 2596 ::
2982 2597
2983 # cat available_filter_functions 2598 # cat available_filter_functions
2984 put_prev_task_idle 2599 put_prev_task_idle
2985 kmem_cache_create 2600 kmem_cache_create
2986 pick_next_task_rt 2601 pick_next_task_rt
2987 cpus_read_lock !! 2602 get_online_cpus
2988 pick_next_task_fair 2603 pick_next_task_fair
2989 mutex_lock 2604 mutex_lock
2990 [...] 2605 [...]
2991 2606
2992 If I am only interested in sys_nanosleep and 2607 If I am only interested in sys_nanosleep and hrtimer_interrupt::
2993 2608
2994 # echo sys_nanosleep hrtimer_interrupt > se 2609 # echo sys_nanosleep hrtimer_interrupt > set_ftrace_filter
2995 # echo function > current_tracer 2610 # echo function > current_tracer
2996 # echo 1 > tracing_on 2611 # echo 1 > tracing_on
2997 # usleep 1 2612 # usleep 1
2998 # echo 0 > tracing_on 2613 # echo 0 > tracing_on
2999 # cat trace 2614 # cat trace
3000 # tracer: function 2615 # tracer: function
3001 # 2616 #
3002 # entries-in-buffer/entries-written: 5/5 2617 # entries-in-buffer/entries-written: 5/5 #P:4
3003 # 2618 #
3004 # _-----=> irq 2619 # _-----=> irqs-off
3005 # / _----=> nee 2620 # / _----=> need-resched
3006 # | / _---=> har 2621 # | / _---=> hardirq/softirq
3007 # || / _--=> pre 2622 # || / _--=> preempt-depth
3008 # ||| / dela 2623 # ||| / delay
3009 # TASK-PID CPU# |||| TIMEST 2624 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3010 # | | | |||| | 2625 # | | | |||| | |
3011 usleep-2665 [001] .... 4186.475 2626 usleep-2665 [001] .... 4186.475355: sys_nanosleep <-system_call_fastpath
3012 <idle>-0 [001] d.h1 4186.475 2627 <idle>-0 [001] d.h1 4186.475409: hrtimer_interrupt <-smp_apic_timer_interrupt
3013 usleep-2665 [001] d.h1 4186.475 2628 usleep-2665 [001] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt
3014 <idle>-0 [003] d.h1 4186.475 2629 <idle>-0 [003] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt
3015 <idle>-0 [002] d.h1 4186.475 2630 <idle>-0 [002] d.h1 4186.475427: hrtimer_interrupt <-smp_apic_timer_interrupt
3016 2631
3017 To see which functions are being traced, you 2632 To see which functions are being traced, you can cat the file:
3018 :: 2633 ::
3019 2634
3020 # cat set_ftrace_filter 2635 # cat set_ftrace_filter
3021 hrtimer_interrupt 2636 hrtimer_interrupt
3022 sys_nanosleep 2637 sys_nanosleep
3023 2638
3024 2639
3025 Perhaps this is not enough. The filters also 2640 Perhaps this is not enough. The filters also allow glob(7) matching.
3026 2641
3027 ``<match>*`` 2642 ``<match>*``
3028 will match functions that begin with 2643 will match functions that begin with <match>
3029 ``*<match>`` 2644 ``*<match>``
3030 will match functions that end with <m 2645 will match functions that end with <match>
3031 ``*<match>*`` 2646 ``*<match>*``
3032 will match functions that have <match 2647 will match functions that have <match> in it
3033 ``<match1>*<match2>`` 2648 ``<match1>*<match2>``
3034 will match functions that begin with 2649 will match functions that begin with <match1> and end with <match2>
3035 2650
3036 .. note:: 2651 .. note::
3037 It is better to use quotes to enclose t 2652 It is better to use quotes to enclose the wild cards,
3038 otherwise the shell may expand the para 2653 otherwise the shell may expand the parameters into names
3039 of files in the local directory. 2654 of files in the local directory.
3040 2655
3041 :: 2656 ::
3042 2657
3043 # echo 'hrtimer_*' > set_ftrace_filter 2658 # echo 'hrtimer_*' > set_ftrace_filter
3044 2659
3045 Produces:: 2660 Produces::
3046 2661
3047 # tracer: function 2662 # tracer: function
3048 # 2663 #
3049 # entries-in-buffer/entries-written: 897/89 2664 # entries-in-buffer/entries-written: 897/897 #P:4
3050 # 2665 #
3051 # _-----=> irq 2666 # _-----=> irqs-off
3052 # / _----=> nee 2667 # / _----=> need-resched
3053 # | / _---=> har 2668 # | / _---=> hardirq/softirq
3054 # || / _--=> pre 2669 # || / _--=> preempt-depth
3055 # ||| / dela 2670 # ||| / delay
3056 # TASK-PID CPU# |||| TIMEST 2671 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3057 # | | | |||| | 2672 # | | | |||| | |
3058 <idle>-0 [003] dN.1 4228.547 2673 <idle>-0 [003] dN.1 4228.547803: hrtimer_cancel <-tick_nohz_idle_exit
3059 <idle>-0 [003] dN.1 4228.547 2674 <idle>-0 [003] dN.1 4228.547804: hrtimer_try_to_cancel <-hrtimer_cancel
3060 <idle>-0 [003] dN.2 4228.547 2675 <idle>-0 [003] dN.2 4228.547805: hrtimer_force_reprogram <-__remove_hrtimer
3061 <idle>-0 [003] dN.1 4228.547 2676 <idle>-0 [003] dN.1 4228.547805: hrtimer_forward <-tick_nohz_idle_exit
3062 <idle>-0 [003] dN.1 4228.547 2677 <idle>-0 [003] dN.1 4228.547805: hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11
3063 <idle>-0 [003] d..1 4228.547 2678 <idle>-0 [003] d..1 4228.547858: hrtimer_get_next_event <-get_next_timer_interrupt
3064 <idle>-0 [003] d..1 4228.547 2679 <idle>-0 [003] d..1 4228.547859: hrtimer_start <-__tick_nohz_idle_enter
3065 <idle>-0 [003] d..2 4228.547 2680 <idle>-0 [003] d..2 4228.547860: hrtimer_force_reprogram <-__rem
3066 2681
3067 Notice that we lost the sys_nanosleep. 2682 Notice that we lost the sys_nanosleep.
3068 :: 2683 ::
3069 2684
3070 # cat set_ftrace_filter 2685 # cat set_ftrace_filter
3071 hrtimer_run_queues 2686 hrtimer_run_queues
3072 hrtimer_run_pending 2687 hrtimer_run_pending
3073 hrtimer_init 2688 hrtimer_init
3074 hrtimer_cancel 2689 hrtimer_cancel
3075 hrtimer_try_to_cancel 2690 hrtimer_try_to_cancel
3076 hrtimer_forward 2691 hrtimer_forward
3077 hrtimer_start 2692 hrtimer_start
3078 hrtimer_reprogram 2693 hrtimer_reprogram
3079 hrtimer_force_reprogram 2694 hrtimer_force_reprogram
3080 hrtimer_get_next_event 2695 hrtimer_get_next_event
3081 hrtimer_interrupt 2696 hrtimer_interrupt
3082 hrtimer_nanosleep 2697 hrtimer_nanosleep
3083 hrtimer_wakeup 2698 hrtimer_wakeup
3084 hrtimer_get_remaining 2699 hrtimer_get_remaining
3085 hrtimer_get_res 2700 hrtimer_get_res
3086 hrtimer_init_sleeper 2701 hrtimer_init_sleeper
3087 2702
3088 2703
3089 This is because the '>' and '>>' act just lik 2704 This is because the '>' and '>>' act just like they do in bash.
3090 To rewrite the filters, use '>' 2705 To rewrite the filters, use '>'
3091 To append to the filters, use '>>' 2706 To append to the filters, use '>>'
3092 2707
3093 To clear out a filter so that all functions w 2708 To clear out a filter so that all functions will be recorded
3094 again:: 2709 again::
3095 2710
3096 # echo > set_ftrace_filter 2711 # echo > set_ftrace_filter
3097 # cat set_ftrace_filter 2712 # cat set_ftrace_filter
3098 # 2713 #
3099 2714
3100 Again, now we want to append. 2715 Again, now we want to append.
3101 2716
3102 :: 2717 ::
3103 2718
3104 # echo sys_nanosleep > set_ftrace_filter 2719 # echo sys_nanosleep > set_ftrace_filter
3105 # cat set_ftrace_filter 2720 # cat set_ftrace_filter
3106 sys_nanosleep 2721 sys_nanosleep
3107 # echo 'hrtimer_*' >> set_ftrace_filter 2722 # echo 'hrtimer_*' >> set_ftrace_filter
3108 # cat set_ftrace_filter 2723 # cat set_ftrace_filter
3109 hrtimer_run_queues 2724 hrtimer_run_queues
3110 hrtimer_run_pending 2725 hrtimer_run_pending
3111 hrtimer_init 2726 hrtimer_init
3112 hrtimer_cancel 2727 hrtimer_cancel
3113 hrtimer_try_to_cancel 2728 hrtimer_try_to_cancel
3114 hrtimer_forward 2729 hrtimer_forward
3115 hrtimer_start 2730 hrtimer_start
3116 hrtimer_reprogram 2731 hrtimer_reprogram
3117 hrtimer_force_reprogram 2732 hrtimer_force_reprogram
3118 hrtimer_get_next_event 2733 hrtimer_get_next_event
3119 hrtimer_interrupt 2734 hrtimer_interrupt
3120 sys_nanosleep 2735 sys_nanosleep
3121 hrtimer_nanosleep 2736 hrtimer_nanosleep
3122 hrtimer_wakeup 2737 hrtimer_wakeup
3123 hrtimer_get_remaining 2738 hrtimer_get_remaining
3124 hrtimer_get_res 2739 hrtimer_get_res
3125 hrtimer_init_sleeper 2740 hrtimer_init_sleeper
3126 2741
3127 2742
3128 The set_ftrace_notrace prevents those functio 2743 The set_ftrace_notrace prevents those functions from being
3129 traced. 2744 traced.
3130 :: 2745 ::
3131 2746
3132 # echo '*preempt*' '*lock*' > set_ftrace_no 2747 # echo '*preempt*' '*lock*' > set_ftrace_notrace
3133 2748
3134 Produces:: 2749 Produces::
3135 2750
3136 # tracer: function 2751 # tracer: function
3137 # 2752 #
3138 # entries-in-buffer/entries-written: 39608/ 2753 # entries-in-buffer/entries-written: 39608/39608 #P:4
3139 # 2754 #
3140 # _-----=> irq 2755 # _-----=> irqs-off
3141 # / _----=> nee 2756 # / _----=> need-resched
3142 # | / _---=> har 2757 # | / _---=> hardirq/softirq
3143 # || / _--=> pre 2758 # || / _--=> preempt-depth
3144 # ||| / dela 2759 # ||| / delay
3145 # TASK-PID CPU# |||| TIMEST 2760 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3146 # | | | |||| | 2761 # | | | |||| | |
3147 bash-1994 [000] .... 4342.324 2762 bash-1994 [000] .... 4342.324896: file_ra_state_init <-do_dentry_open
3148 bash-1994 [000] .... 4342.324 2763 bash-1994 [000] .... 4342.324897: open_check_o_direct <-do_last
3149 bash-1994 [000] .... 4342.324 2764 bash-1994 [000] .... 4342.324897: ima_file_check <-do_last
3150 bash-1994 [000] .... 4342.324 2765 bash-1994 [000] .... 4342.324898: process_measurement <-ima_file_check
3151 bash-1994 [000] .... 4342.324 2766 bash-1994 [000] .... 4342.324898: ima_get_action <-process_measurement
3152 bash-1994 [000] .... 4342.324 2767 bash-1994 [000] .... 4342.324898: ima_match_policy <-ima_get_action
3153 bash-1994 [000] .... 4342.324 2768 bash-1994 [000] .... 4342.324899: do_truncate <-do_last
3154 bash-1994 [000] .... 4342.324 !! 2769 bash-1994 [000] .... 4342.324899: should_remove_suid <-do_truncate
3155 bash-1994 [000] .... 4342.324 2770 bash-1994 [000] .... 4342.324899: notify_change <-do_truncate
3156 bash-1994 [000] .... 4342.324 2771 bash-1994 [000] .... 4342.324900: current_fs_time <-notify_change
3157 bash-1994 [000] .... 4342.324 2772 bash-1994 [000] .... 4342.324900: current_kernel_time <-current_fs_time
3158 bash-1994 [000] .... 4342.324 2773 bash-1994 [000] .... 4342.324900: timespec_trunc <-current_fs_time
3159 2774
3160 We can see that there's no more lock or preem 2775 We can see that there's no more lock or preempt tracing.
3161 2776
3162 Selecting function filters via index <<
3163 ------------------------------------ <<
3164 <<
3165 Because processing of strings is expensive (t <<
3166 needs to be looked up before comparing to the <<
3167 an index can be used as well to enable functi <<
3168 case of setting thousands of specific functio <<
3169 in a list of numbers, no string processing wi <<
3170 at the specific location in the internal arra <<
3171 functions in the "available_filter_functions" <<
3172 <<
3173 :: <<
3174 <<
3175 # echo 1 > set_ftrace_filter <<
3176 <<
3177 Will select the first function listed in "ava <<
3178 <<
3179 :: <<
3180 <<
3181 # head -1 available_filter_functions <<
3182 trace_initcall_finish_cb <<
3183 <<
3184 # cat set_ftrace_filter <<
3185 trace_initcall_finish_cb <<
3186 <<
3187 # head -50 available_filter_functions | tai <<
3188 x86_pmu_commit_txn <<
3189 <<
3190 # echo 1 50 > set_ftrace_filter <<
3191 # cat set_ftrace_filter <<
3192 trace_initcall_finish_cb <<
3193 x86_pmu_commit_txn <<
3194 2777
3195 Dynamic ftrace with the function graph tracer 2778 Dynamic ftrace with the function graph tracer
3196 --------------------------------------------- 2779 ---------------------------------------------
3197 2780
3198 Although what has been explained above concer 2781 Although what has been explained above concerns both the
3199 function tracer and the function-graph-tracer 2782 function tracer and the function-graph-tracer, there are some
3200 special features only available in the functi 2783 special features only available in the function-graph tracer.
3201 2784
3202 If you want to trace only one function and al 2785 If you want to trace only one function and all of its children,
3203 you just have to echo its name into set_graph 2786 you just have to echo its name into set_graph_function::
3204 2787
3205 echo __do_fault > set_graph_function 2788 echo __do_fault > set_graph_function
3206 2789
3207 will produce the following "expanded" trace o 2790 will produce the following "expanded" trace of the __do_fault()
3208 function:: 2791 function::
3209 2792
3210 0) | __do_fault() { 2793 0) | __do_fault() {
3211 0) | filemap_fault() { 2794 0) | filemap_fault() {
3212 0) | find_lock_page() { 2795 0) | find_lock_page() {
3213 0) 0.804 us | find_get_page(); 2796 0) 0.804 us | find_get_page();
3214 0) | __might_sleep() 2797 0) | __might_sleep() {
3215 0) 1.329 us | } 2798 0) 1.329 us | }
3216 0) 3.904 us | } 2799 0) 3.904 us | }
3217 0) 4.979 us | } 2800 0) 4.979 us | }
3218 0) 0.653 us | _spin_lock(); 2801 0) 0.653 us | _spin_lock();
3219 0) 0.578 us | page_add_file_rmap() 2802 0) 0.578 us | page_add_file_rmap();
3220 0) 0.525 us | native_set_pte_at(); 2803 0) 0.525 us | native_set_pte_at();
3221 0) 0.585 us | _spin_unlock(); 2804 0) 0.585 us | _spin_unlock();
3222 0) | unlock_page() { 2805 0) | unlock_page() {
3223 0) 0.541 us | page_waitqueue(); 2806 0) 0.541 us | page_waitqueue();
3224 0) 0.639 us | __wake_up_bit(); 2807 0) 0.639 us | __wake_up_bit();
3225 0) 2.786 us | } 2808 0) 2.786 us | }
3226 0) + 14.237 us | } 2809 0) + 14.237 us | }
3227 0) | __do_fault() { 2810 0) | __do_fault() {
3228 0) | filemap_fault() { 2811 0) | filemap_fault() {
3229 0) | find_lock_page() { 2812 0) | find_lock_page() {
3230 0) 0.698 us | find_get_page(); 2813 0) 0.698 us | find_get_page();
3231 0) | __might_sleep() 2814 0) | __might_sleep() {
3232 0) 1.412 us | } 2815 0) 1.412 us | }
3233 0) 3.950 us | } 2816 0) 3.950 us | }
3234 0) 5.098 us | } 2817 0) 5.098 us | }
3235 0) 0.631 us | _spin_lock(); 2818 0) 0.631 us | _spin_lock();
3236 0) 0.571 us | page_add_file_rmap() 2819 0) 0.571 us | page_add_file_rmap();
3237 0) 0.526 us | native_set_pte_at(); 2820 0) 0.526 us | native_set_pte_at();
3238 0) 0.586 us | _spin_unlock(); 2821 0) 0.586 us | _spin_unlock();
3239 0) | unlock_page() { 2822 0) | unlock_page() {
3240 0) 0.533 us | page_waitqueue(); 2823 0) 0.533 us | page_waitqueue();
3241 0) 0.638 us | __wake_up_bit(); 2824 0) 0.638 us | __wake_up_bit();
3242 0) 2.793 us | } 2825 0) 2.793 us | }
3243 0) + 14.012 us | } 2826 0) + 14.012 us | }
3244 2827
3245 You can also expand several functions at once 2828 You can also expand several functions at once::
3246 2829
3247 echo sys_open > set_graph_function 2830 echo sys_open > set_graph_function
3248 echo sys_close >> set_graph_function 2831 echo sys_close >> set_graph_function
3249 2832
3250 Now if you want to go back to trace all funct 2833 Now if you want to go back to trace all functions you can clear
3251 this special filter via:: 2834 this special filter via::
3252 2835
3253 echo > set_graph_function 2836 echo > set_graph_function
3254 2837
3255 2838
3256 ftrace_enabled 2839 ftrace_enabled
3257 -------------- 2840 --------------
3258 2841
3259 Note, the proc sysctl ftrace_enable is a big 2842 Note, the proc sysctl ftrace_enable is a big on/off switch for the
3260 function tracer. By default it is enabled (wh 2843 function tracer. By default it is enabled (when function tracing is
3261 enabled in the kernel). If it is disabled, al 2844 enabled in the kernel). If it is disabled, all function tracing is
3262 disabled. This includes not only the function 2845 disabled. This includes not only the function tracers for ftrace, but
3263 also for any other uses (perf, kprobes, stack !! 2846 also for any other uses (perf, kprobes, stack tracing, profiling, etc).
3264 cannot be disabled if there is a callback wit <<
3265 registered. <<
3266 2847
3267 Please disable this with care. 2848 Please disable this with care.
3268 2849
3269 This can be disable (and enabled) with:: 2850 This can be disable (and enabled) with::
3270 2851
3271 sysctl kernel.ftrace_enabled=0 2852 sysctl kernel.ftrace_enabled=0
3272 sysctl kernel.ftrace_enabled=1 2853 sysctl kernel.ftrace_enabled=1
3273 2854
3274 or 2855 or
3275 2856
3276 echo 0 > /proc/sys/kernel/ftrace_enabled 2857 echo 0 > /proc/sys/kernel/ftrace_enabled
3277 echo 1 > /proc/sys/kernel/ftrace_enabled 2858 echo 1 > /proc/sys/kernel/ftrace_enabled
3278 2859
3279 2860
3280 Filter commands 2861 Filter commands
3281 --------------- 2862 ---------------
3282 2863
3283 A few commands are supported by the set_ftrac 2864 A few commands are supported by the set_ftrace_filter interface.
3284 Trace commands have the following format:: 2865 Trace commands have the following format::
3285 2866
3286 <function>:<command>:<parameter> 2867 <function>:<command>:<parameter>
3287 2868
3288 The following commands are supported: 2869 The following commands are supported:
3289 2870
3290 - mod: 2871 - mod:
3291 This command enables function filtering per 2872 This command enables function filtering per module. The
3292 parameter defines the module. For example, 2873 parameter defines the module. For example, if only the write*
3293 functions in the ext3 module are desired, r 2874 functions in the ext3 module are desired, run:
3294 2875
3295 echo 'write*:mod:ext3' > set_ftrace_filter 2876 echo 'write*:mod:ext3' > set_ftrace_filter
3296 2877
3297 This command interacts with the filter in t 2878 This command interacts with the filter in the same way as
3298 filtering based on function names. Thus, ad 2879 filtering based on function names. Thus, adding more functions
3299 in a different module is accomplished by ap 2880 in a different module is accomplished by appending (>>) to the
3300 filter file. Remove specific module functio 2881 filter file. Remove specific module functions by prepending
3301 '!':: 2882 '!'::
3302 2883
3303 echo '!writeback*:mod:ext3' >> set_ftrace_ 2884 echo '!writeback*:mod:ext3' >> set_ftrace_filter
3304 2885
3305 Mod command supports module globbing. Disab 2886 Mod command supports module globbing. Disable tracing for all
3306 functions except a specific module:: 2887 functions except a specific module::
3307 2888
3308 echo '!*:mod:!ext3' >> set_ftrace_filter 2889 echo '!*:mod:!ext3' >> set_ftrace_filter
3309 2890
3310 Disable tracing for all modules, but still 2891 Disable tracing for all modules, but still trace kernel::
3311 2892
3312 echo '!*:mod:*' >> set_ftrace_filter 2893 echo '!*:mod:*' >> set_ftrace_filter
3313 2894
3314 Enable filter only for kernel:: 2895 Enable filter only for kernel::
3315 2896
3316 echo '*write*:mod:!*' >> set_ftrace_filter 2897 echo '*write*:mod:!*' >> set_ftrace_filter
3317 2898
3318 Enable filter for module globbing:: 2899 Enable filter for module globbing::
3319 2900
3320 echo '*write*:mod:*snd*' >> set_ftrace_fil 2901 echo '*write*:mod:*snd*' >> set_ftrace_filter
3321 2902
3322 - traceon/traceoff: 2903 - traceon/traceoff:
3323 These commands turn tracing on and off when 2904 These commands turn tracing on and off when the specified
3324 functions are hit. The parameter determines 2905 functions are hit. The parameter determines how many times the
3325 tracing system is turned on and off. If uns 2906 tracing system is turned on and off. If unspecified, there is
3326 no limit. For example, to disable tracing w 2907 no limit. For example, to disable tracing when a schedule bug
3327 is hit the first 5 times, run:: 2908 is hit the first 5 times, run::
3328 2909
3329 echo '__schedule_bug:traceoff:5' > set_ftr 2910 echo '__schedule_bug:traceoff:5' > set_ftrace_filter
3330 2911
3331 To always disable tracing when __schedule_b 2912 To always disable tracing when __schedule_bug is hit::
3332 2913
3333 echo '__schedule_bug:traceoff' > set_ftrac 2914 echo '__schedule_bug:traceoff' > set_ftrace_filter
3334 2915
3335 These commands are cumulative whether or no 2916 These commands are cumulative whether or not they are appended
3336 to set_ftrace_filter. To remove a command, 2917 to set_ftrace_filter. To remove a command, prepend it by '!'
3337 and drop the parameter:: 2918 and drop the parameter::
3338 2919
3339 echo '!__schedule_bug:traceoff:0' > set_ft 2920 echo '!__schedule_bug:traceoff:0' > set_ftrace_filter
3340 2921
3341 The above removes the traceoff command for 2922 The above removes the traceoff command for __schedule_bug
3342 that have a counter. To remove commands wit 2923 that have a counter. To remove commands without counters::
3343 2924
3344 echo '!__schedule_bug:traceoff' > set_ftra 2925 echo '!__schedule_bug:traceoff' > set_ftrace_filter
3345 2926
3346 - snapshot: 2927 - snapshot:
3347 Will cause a snapshot to be triggered when 2928 Will cause a snapshot to be triggered when the function is hit.
3348 :: 2929 ::
3349 2930
3350 echo 'native_flush_tlb_others:snapshot' > 2931 echo 'native_flush_tlb_others:snapshot' > set_ftrace_filter
3351 2932
3352 To only snapshot once: 2933 To only snapshot once:
3353 :: 2934 ::
3354 2935
3355 echo 'native_flush_tlb_others:snapshot:1' 2936 echo 'native_flush_tlb_others:snapshot:1' > set_ftrace_filter
3356 2937
3357 To remove the above commands:: 2938 To remove the above commands::
3358 2939
3359 echo '!native_flush_tlb_others:snapshot' > 2940 echo '!native_flush_tlb_others:snapshot' > set_ftrace_filter
3360 echo '!native_flush_tlb_others:snapshot:0' 2941 echo '!native_flush_tlb_others:snapshot:0' > set_ftrace_filter
3361 2942
3362 - enable_event/disable_event: 2943 - enable_event/disable_event:
3363 These commands can enable or disable a trac 2944 These commands can enable or disable a trace event. Note, because
3364 function tracing callbacks are very sensiti 2945 function tracing callbacks are very sensitive, when these commands
3365 are registered, the trace point is activate 2946 are registered, the trace point is activated, but disabled in
3366 a "soft" mode. That is, the tracepoint will 2947 a "soft" mode. That is, the tracepoint will be called, but
3367 just will not be traced. The event tracepoi 2948 just will not be traced. The event tracepoint stays in this mode
3368 as long as there's a command that triggers 2949 as long as there's a command that triggers it.
3369 :: 2950 ::
3370 2951
3371 echo 'try_to_wake_up:enable_event:sched:sc 2952 echo 'try_to_wake_up:enable_event:sched:sched_switch:2' > \
3372 set_ftrace_filter 2953 set_ftrace_filter
3373 2954
3374 The format is:: 2955 The format is::
3375 2956
3376 <function>:enable_event:<system>:<event>[ 2957 <function>:enable_event:<system>:<event>[:count]
3377 <function>:disable_event:<system>:<event> 2958 <function>:disable_event:<system>:<event>[:count]
3378 2959
3379 To remove the events commands:: 2960 To remove the events commands::
3380 2961
3381 echo '!try_to_wake_up:enable_event:sched:s 2962 echo '!try_to_wake_up:enable_event:sched:sched_switch:0' > \
3382 set_ftrace_filter 2963 set_ftrace_filter
3383 echo '!schedule:disable_event:sched:sched_ 2964 echo '!schedule:disable_event:sched:sched_switch' > \
3384 set_ftrace_filter 2965 set_ftrace_filter
3385 2966
3386 - dump: 2967 - dump:
3387 When the function is hit, it will dump the 2968 When the function is hit, it will dump the contents of the ftrace
3388 ring buffer to the console. This is useful 2969 ring buffer to the console. This is useful if you need to debug
3389 something, and want to dump the trace when 2970 something, and want to dump the trace when a certain function
3390 is hit. Perhaps it's a function that is cal !! 2971 is hit. Perhaps its a function that is called before a tripple
3391 fault happens and does not allow you to get 2972 fault happens and does not allow you to get a regular dump.
3392 2973
3393 - cpudump: 2974 - cpudump:
3394 When the function is hit, it will dump the 2975 When the function is hit, it will dump the contents of the ftrace
3395 ring buffer for the current CPU to the cons 2976 ring buffer for the current CPU to the console. Unlike the "dump"
3396 command, it only prints out the contents of 2977 command, it only prints out the contents of the ring buffer for the
3397 CPU that executed the function that trigger 2978 CPU that executed the function that triggered the dump.
3398 2979
3399 - stacktrace: <<
3400 When the function is hit, a stack trace is <<
3401 <<
3402 trace_pipe 2980 trace_pipe
3403 ---------- 2981 ----------
3404 2982
3405 The trace_pipe outputs the same content as th 2983 The trace_pipe outputs the same content as the trace file, but
3406 the effect on the tracing is different. Every 2984 the effect on the tracing is different. Every read from
3407 trace_pipe is consumed. This means that subse 2985 trace_pipe is consumed. This means that subsequent reads will be
3408 different. The trace is live. 2986 different. The trace is live.
3409 :: 2987 ::
3410 2988
3411 # echo function > current_tracer 2989 # echo function > current_tracer
3412 # cat trace_pipe > /tmp/trace.out & 2990 # cat trace_pipe > /tmp/trace.out &
3413 [1] 4153 2991 [1] 4153
3414 # echo 1 > tracing_on 2992 # echo 1 > tracing_on
3415 # usleep 1 2993 # usleep 1
3416 # echo 0 > tracing_on 2994 # echo 0 > tracing_on
3417 # cat trace 2995 # cat trace
3418 # tracer: function 2996 # tracer: function
3419 # 2997 #
3420 # entries-in-buffer/entries-written: 0/0 2998 # entries-in-buffer/entries-written: 0/0 #P:4
3421 # 2999 #
3422 # _-----=> irq 3000 # _-----=> irqs-off
3423 # / _----=> nee 3001 # / _----=> need-resched
3424 # | / _---=> har 3002 # | / _---=> hardirq/softirq
3425 # || / _--=> pre 3003 # || / _--=> preempt-depth
3426 # ||| / dela 3004 # ||| / delay
3427 # TASK-PID CPU# |||| TIMEST 3005 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3428 # | | | |||| | 3006 # | | | |||| | |
3429 3007
3430 # 3008 #
3431 # cat /tmp/trace.out 3009 # cat /tmp/trace.out
3432 bash-1994 [000] .... 5281.5689 3010 bash-1994 [000] .... 5281.568961: mutex_unlock <-rb_simple_write
3433 bash-1994 [000] .... 5281.5689 3011 bash-1994 [000] .... 5281.568963: __mutex_unlock_slowpath <-mutex_unlock
3434 bash-1994 [000] .... 5281.5689 3012 bash-1994 [000] .... 5281.568963: __fsnotify_parent <-fsnotify_modify
3435 bash-1994 [000] .... 5281.5689 3013 bash-1994 [000] .... 5281.568964: fsnotify <-fsnotify_modify
3436 bash-1994 [000] .... 5281.5689 3014 bash-1994 [000] .... 5281.568964: __srcu_read_lock <-fsnotify
3437 bash-1994 [000] .... 5281.5689 3015 bash-1994 [000] .... 5281.568964: add_preempt_count <-__srcu_read_lock
3438 bash-1994 [000] ...1 5281.5689 3016 bash-1994 [000] ...1 5281.568965: sub_preempt_count <-__srcu_read_lock
3439 bash-1994 [000] .... 5281.5689 3017 bash-1994 [000] .... 5281.568965: __srcu_read_unlock <-fsnotify
3440 bash-1994 [000] .... 5281.5689 3018 bash-1994 [000] .... 5281.568967: sys_dup2 <-system_call_fastpath
3441 3019
3442 3020
3443 Note, reading the trace_pipe file will block 3021 Note, reading the trace_pipe file will block until more input is
3444 added. This is contrary to the trace file. If !! 3022 added.
3445 the trace file for reading, it will actually <<
3446 prevent new entries from being added. The tra <<
3447 not have this limitation. <<
3448 3023
3449 trace entries 3024 trace entries
3450 ------------- 3025 -------------
3451 3026
3452 Having too much or not enough data can be tro 3027 Having too much or not enough data can be troublesome in
3453 diagnosing an issue in the kernel. The file b 3028 diagnosing an issue in the kernel. The file buffer_size_kb is
3454 used to modify the size of the internal trace 3029 used to modify the size of the internal trace buffers. The
3455 number listed is the number of entries that c 3030 number listed is the number of entries that can be recorded per
3456 CPU. To know the full size, multiply the numb 3031 CPU. To know the full size, multiply the number of possible CPUs
3457 with the number of entries. 3032 with the number of entries.
3458 :: 3033 ::
3459 3034
3460 # cat buffer_size_kb 3035 # cat buffer_size_kb
3461 1408 (units kilobytes) 3036 1408 (units kilobytes)
3462 3037
3463 Or simply read buffer_total_size_kb 3038 Or simply read buffer_total_size_kb
3464 :: 3039 ::
3465 3040
3466 # cat buffer_total_size_kb 3041 # cat buffer_total_size_kb
3467 5632 3042 5632
3468 3043
3469 To modify the buffer, simple echo in a number 3044 To modify the buffer, simple echo in a number (in 1024 byte segments).
3470 :: 3045 ::
3471 3046
3472 # echo 10000 > buffer_size_kb 3047 # echo 10000 > buffer_size_kb
3473 # cat buffer_size_kb 3048 # cat buffer_size_kb
3474 10000 (units kilobytes) 3049 10000 (units kilobytes)
3475 3050
3476 It will try to allocate as much as possible. 3051 It will try to allocate as much as possible. If you allocate too
3477 much, it can cause Out-Of-Memory to trigger. 3052 much, it can cause Out-Of-Memory to trigger.
3478 :: 3053 ::
3479 3054
3480 # echo 1000000000000 > buffer_size_kb 3055 # echo 1000000000000 > buffer_size_kb
3481 -bash: echo: write error: Cannot allocate m 3056 -bash: echo: write error: Cannot allocate memory
3482 # cat buffer_size_kb 3057 # cat buffer_size_kb
3483 85 3058 85
3484 3059
3485 The per_cpu buffers can be changed individual 3060 The per_cpu buffers can be changed individually as well:
3486 :: 3061 ::
3487 3062
3488 # echo 10000 > per_cpu/cpu0/buffer_size_kb 3063 # echo 10000 > per_cpu/cpu0/buffer_size_kb
3489 # echo 100 > per_cpu/cpu1/buffer_size_kb 3064 # echo 100 > per_cpu/cpu1/buffer_size_kb
3490 3065
3491 When the per_cpu buffers are not the same, th 3066 When the per_cpu buffers are not the same, the buffer_size_kb
3492 at the top level will just show an X 3067 at the top level will just show an X
3493 :: 3068 ::
3494 3069
3495 # cat buffer_size_kb 3070 # cat buffer_size_kb
3496 X 3071 X
3497 3072
3498 This is where the buffer_total_size_kb is use 3073 This is where the buffer_total_size_kb is useful:
3499 :: 3074 ::
3500 3075
3501 # cat buffer_total_size_kb 3076 # cat buffer_total_size_kb
3502 12916 3077 12916
3503 3078
3504 Writing to the top level buffer_size_kb will 3079 Writing to the top level buffer_size_kb will reset all the buffers
3505 to be the same again. 3080 to be the same again.
3506 3081
3507 Snapshot 3082 Snapshot
3508 -------- 3083 --------
3509 CONFIG_TRACER_SNAPSHOT makes a generic snapsh 3084 CONFIG_TRACER_SNAPSHOT makes a generic snapshot feature
3510 available to all non latency tracers. (Latenc 3085 available to all non latency tracers. (Latency tracers which
3511 record max latency, such as "irqsoff" or "wak 3086 record max latency, such as "irqsoff" or "wakeup", can't use
3512 this feature, since those are already using t 3087 this feature, since those are already using the snapshot
3513 mechanism internally.) 3088 mechanism internally.)
3514 3089
3515 Snapshot preserves a current trace buffer at 3090 Snapshot preserves a current trace buffer at a particular point
3516 in time without stopping tracing. Ftrace swap 3091 in time without stopping tracing. Ftrace swaps the current
3517 buffer with a spare buffer, and tracing conti 3092 buffer with a spare buffer, and tracing continues in the new
3518 current (=previous spare) buffer. 3093 current (=previous spare) buffer.
3519 3094
3520 The following tracefs files in "tracing" are 3095 The following tracefs files in "tracing" are related to this
3521 feature: 3096 feature:
3522 3097
3523 snapshot: 3098 snapshot:
3524 3099
3525 This is used to take a snapshot and t 3100 This is used to take a snapshot and to read the output
3526 of the snapshot. Echo 1 into this fil 3101 of the snapshot. Echo 1 into this file to allocate a
3527 spare buffer and to take a snapshot ( 3102 spare buffer and to take a snapshot (swap), then read
3528 the snapshot from this file in the sa 3103 the snapshot from this file in the same format as
3529 "trace" (described above in the secti 3104 "trace" (described above in the section "The File
3530 System"). Both reads snapshot and tra 3105 System"). Both reads snapshot and tracing are executable
3531 in parallel. When the spare buffer is 3106 in parallel. When the spare buffer is allocated, echoing
3532 0 frees it, and echoing else (positiv 3107 0 frees it, and echoing else (positive) values clear the
3533 snapshot contents. 3108 snapshot contents.
3534 More details are shown in the table b 3109 More details are shown in the table below.
3535 3110
3536 +--------------+------------+-------- 3111 +--------------+------------+------------+------------+
3537 |status\\input | 0 | 1 3112 |status\\input | 0 | 1 | else |
3538 +==============+============+======== 3113 +==============+============+============+============+
3539 |not allocated |(do nothing)| alloc+s 3114 |not allocated |(do nothing)| alloc+swap |(do nothing)|
3540 +--------------+------------+-------- 3115 +--------------+------------+------------+------------+
3541 |allocated | free | swap 3116 |allocated | free | swap | clear |
3542 +--------------+------------+-------- 3117 +--------------+------------+------------+------------+
3543 3118
3544 Here is an example of using the snapshot feat 3119 Here is an example of using the snapshot feature.
3545 :: 3120 ::
3546 3121
3547 # echo 1 > events/sched/enable 3122 # echo 1 > events/sched/enable
3548 # echo 1 > snapshot 3123 # echo 1 > snapshot
3549 # cat snapshot 3124 # cat snapshot
3550 # tracer: nop 3125 # tracer: nop
3551 # 3126 #
3552 # entries-in-buffer/entries-written: 71/71 3127 # entries-in-buffer/entries-written: 71/71 #P:8
3553 # 3128 #
3554 # _-----=> irq 3129 # _-----=> irqs-off
3555 # / _----=> nee 3130 # / _----=> need-resched
3556 # | / _---=> har 3131 # | / _---=> hardirq/softirq
3557 # || / _--=> pre 3132 # || / _--=> preempt-depth
3558 # ||| / dela 3133 # ||| / delay
3559 # TASK-PID CPU# |||| TIMEST 3134 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3560 # | | | |||| | 3135 # | | | |||| | |
3561 <idle>-0 [005] d... 2440.603 3136 <idle>-0 [005] d... 2440.603828: sched_switch: prev_comm=swapper/5 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2242 next_prio=120
3562 sleep-2242 [005] d... 2440.603 3137 sleep-2242 [005] d... 2440.603846: sched_switch: prev_comm=snapshot-test-2 prev_pid=2242 prev_prio=120 prev_state=R ==> next_comm=kworker/5:1 next_pid=60 next_prio=120
3563 [...] 3138 [...]
3564 <idle>-0 [002] d... 2440.70723 3139 <idle>-0 [002] d... 2440.707230: sched_switch: prev_comm=swapper/2 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2229 next_prio=120
3565 3140
3566 # cat trace 3141 # cat trace
3567 # tracer: nop 3142 # tracer: nop
3568 # 3143 #
3569 # entries-in-buffer/entries-written: 77/77 3144 # entries-in-buffer/entries-written: 77/77 #P:8
3570 # 3145 #
3571 # _-----=> irq 3146 # _-----=> irqs-off
3572 # / _----=> nee 3147 # / _----=> need-resched
3573 # | / _---=> har 3148 # | / _---=> hardirq/softirq
3574 # || / _--=> pre 3149 # || / _--=> preempt-depth
3575 # ||| / dela 3150 # ||| / delay
3576 # TASK-PID CPU# |||| TIMEST 3151 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3577 # | | | |||| | 3152 # | | | |||| | |
3578 <idle>-0 [007] d... 2440.707 3153 <idle>-0 [007] d... 2440.707395: sched_switch: prev_comm=swapper/7 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=snapshot-test-2 next_pid=2243 next_prio=120
3579 snapshot-test-2-2229 [002] d... 2440.707 3154 snapshot-test-2-2229 [002] d... 2440.707438: sched_switch: prev_comm=snapshot-test-2 prev_pid=2229 prev_prio=120 prev_state=S ==> next_comm=swapper/2 next_pid=0 next_prio=120
3580 [...] 3155 [...]
3581 3156
3582 3157
3583 If you try to use this snapshot feature when 3158 If you try to use this snapshot feature when current tracer is
3584 one of the latency tracers, you will get the 3159 one of the latency tracers, you will get the following results.
3585 :: 3160 ::
3586 3161
3587 # echo wakeup > current_tracer 3162 # echo wakeup > current_tracer
3588 # echo 1 > snapshot 3163 # echo 1 > snapshot
3589 bash: echo: write error: Device or resource 3164 bash: echo: write error: Device or resource busy
3590 # cat snapshot 3165 # cat snapshot
3591 cat: snapshot: Device or resource busy 3166 cat: snapshot: Device or resource busy
3592 3167
3593 3168
3594 Instances 3169 Instances
3595 --------- 3170 ---------
3596 In the tracefs tracing directory, there is a !! 3171 In the tracefs tracing directory is a directory called "instances".
3597 This directory can have new directories creat 3172 This directory can have new directories created inside of it using
3598 mkdir, and removing directories with rmdir. T 3173 mkdir, and removing directories with rmdir. The directory created
3599 with mkdir in this directory will already con 3174 with mkdir in this directory will already contain files and other
3600 directories after it is created. 3175 directories after it is created.
3601 :: 3176 ::
3602 3177
3603 # mkdir instances/foo 3178 # mkdir instances/foo
3604 # ls instances/foo 3179 # ls instances/foo
3605 buffer_size_kb buffer_total_size_kb event 3180 buffer_size_kb buffer_total_size_kb events free_buffer per_cpu
3606 set_event snapshot trace trace_clock tr 3181 set_event snapshot trace trace_clock trace_marker trace_options
3607 trace_pipe tracing_on 3182 trace_pipe tracing_on
3608 3183
3609 As you can see, the new directory looks simil 3184 As you can see, the new directory looks similar to the tracing directory
3610 itself. In fact, it is very similar, except t 3185 itself. In fact, it is very similar, except that the buffer and
3611 events are agnostic from the main directory, !! 3186 events are agnostic from the main director, or from any other
3612 instances that are created. 3187 instances that are created.
3613 3188
3614 The files in the new directory work just like 3189 The files in the new directory work just like the files with the
3615 same name in the tracing directory except the 3190 same name in the tracing directory except the buffer that is used
3616 is a separate and new buffer. The files affec 3191 is a separate and new buffer. The files affect that buffer but do not
3617 affect the main buffer with the exception of 3192 affect the main buffer with the exception of trace_options. Currently,
3618 the trace_options affect all instances and th 3193 the trace_options affect all instances and the top level buffer
3619 the same, but this may change in future relea 3194 the same, but this may change in future releases. That is, options
3620 may become specific to the instance they resi 3195 may become specific to the instance they reside in.
3621 3196
3622 Notice that none of the function tracer files 3197 Notice that none of the function tracer files are there, nor is
3623 current_tracer and available_tracers. This is 3198 current_tracer and available_tracers. This is because the buffers
3624 can currently only have events enabled for th 3199 can currently only have events enabled for them.
3625 :: 3200 ::
3626 3201
3627 # mkdir instances/foo 3202 # mkdir instances/foo
3628 # mkdir instances/bar 3203 # mkdir instances/bar
3629 # mkdir instances/zoot 3204 # mkdir instances/zoot
3630 # echo 100000 > buffer_size_kb 3205 # echo 100000 > buffer_size_kb
3631 # echo 1000 > instances/foo/buffer_size_kb 3206 # echo 1000 > instances/foo/buffer_size_kb
3632 # echo 5000 > instances/bar/per_cpu/cpu1/bu 3207 # echo 5000 > instances/bar/per_cpu/cpu1/buffer_size_kb
3633 # echo function > current_trace 3208 # echo function > current_trace
3634 # echo 1 > instances/foo/events/sched/sched 3209 # echo 1 > instances/foo/events/sched/sched_wakeup/enable
3635 # echo 1 > instances/foo/events/sched/sched 3210 # echo 1 > instances/foo/events/sched/sched_wakeup_new/enable
3636 # echo 1 > instances/foo/events/sched/sched 3211 # echo 1 > instances/foo/events/sched/sched_switch/enable
3637 # echo 1 > instances/bar/events/irq/enable 3212 # echo 1 > instances/bar/events/irq/enable
3638 # echo 1 > instances/zoot/events/syscalls/e 3213 # echo 1 > instances/zoot/events/syscalls/enable
3639 # cat trace_pipe 3214 # cat trace_pipe
3640 CPU:2 [LOST 11745 EVENTS] 3215 CPU:2 [LOST 11745 EVENTS]
3641 bash-2044 [002] .... 10594.481 3216 bash-2044 [002] .... 10594.481032: _raw_spin_lock_irqsave <-get_page_from_freelist
3642 bash-2044 [002] d... 10594.481 3217 bash-2044 [002] d... 10594.481032: add_preempt_count <-_raw_spin_lock_irqsave
3643 bash-2044 [002] d..1 10594.481 3218 bash-2044 [002] d..1 10594.481032: __rmqueue <-get_page_from_freelist
3644 bash-2044 [002] d..1 10594.481 3219 bash-2044 [002] d..1 10594.481033: _raw_spin_unlock <-get_page_from_freelist
3645 bash-2044 [002] d..1 10594.481 3220 bash-2044 [002] d..1 10594.481033: sub_preempt_count <-_raw_spin_unlock
3646 bash-2044 [002] d... 10594.481 3221 bash-2044 [002] d... 10594.481033: get_pageblock_flags_group <-get_pageblock_migratetype
3647 bash-2044 [002] d... 10594.481 3222 bash-2044 [002] d... 10594.481034: __mod_zone_page_state <-get_page_from_freelist
3648 bash-2044 [002] d... 10594.481 3223 bash-2044 [002] d... 10594.481034: zone_statistics <-get_page_from_freelist
3649 bash-2044 [002] d... 10594.481 3224 bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics
3650 bash-2044 [002] d... 10594.481 3225 bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics
3651 bash-2044 [002] .... 10594.481 3226 bash-2044 [002] .... 10594.481035: arch_dup_task_struct <-copy_process
3652 [...] 3227 [...]
3653 3228
3654 # cat instances/foo/trace_pipe 3229 # cat instances/foo/trace_pipe
3655 bash-1998 [000] d..4 136.676 3230 bash-1998 [000] d..4 136.676759: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000
3656 bash-1998 [000] dN.4 136.676 3231 bash-1998 [000] dN.4 136.676760: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000
3657 <idle>-0 [003] d.h3 136.676 3232 <idle>-0 [003] d.h3 136.676906: sched_wakeup: comm=rcu_preempt pid=9 prio=120 success=1 target_cpu=003
3658 <idle>-0 [003] d..3 136.676 3233 <idle>-0 [003] d..3 136.676909: sched_switch: prev_comm=swapper/3 prev_pid=0 prev_prio=120 prev_state=R ==> next_comm=rcu_preempt next_pid=9 next_prio=120
3659 rcu_preempt-9 [003] d..3 136.676 3234 rcu_preempt-9 [003] d..3 136.676916: sched_switch: prev_comm=rcu_preempt prev_pid=9 prev_prio=120 prev_state=S ==> next_comm=swapper/3 next_pid=0 next_prio=120
3660 bash-1998 [000] d..4 136.677 3235 bash-1998 [000] d..4 136.677014: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000
3661 bash-1998 [000] dN.4 136.677 3236 bash-1998 [000] dN.4 136.677016: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000
3662 bash-1998 [000] d..3 136.677 3237 bash-1998 [000] d..3 136.677018: sched_switch: prev_comm=bash prev_pid=1998 prev_prio=120 prev_state=R+ ==> next_comm=kworker/0:1 next_pid=59 next_prio=120
3663 kworker/0:1-59 [000] d..4 136.677 3238 kworker/0:1-59 [000] d..4 136.677022: sched_wakeup: comm=sshd pid=1995 prio=120 success=1 target_cpu=001
3664 kworker/0:1-59 [000] d..3 136.677 3239 kworker/0:1-59 [000] d..3 136.677025: sched_switch: prev_comm=kworker/0:1 prev_pid=59 prev_prio=120 prev_state=S ==> next_comm=bash next_pid=1998 next_prio=120
3665 [...] 3240 [...]
3666 3241
3667 # cat instances/bar/trace_pipe 3242 # cat instances/bar/trace_pipe
3668 migration/1-14 [001] d.h3 138.732 3243 migration/1-14 [001] d.h3 138.732674: softirq_raise: vec=3 [action=NET_RX]
3669 <idle>-0 [001] dNh3 138.732 3244 <idle>-0 [001] dNh3 138.732725: softirq_raise: vec=3 [action=NET_RX]
3670 bash-1998 [000] d.h1 138.733 3245 bash-1998 [000] d.h1 138.733101: softirq_raise: vec=1 [action=TIMER]
3671 bash-1998 [000] d.h1 138.733 3246 bash-1998 [000] d.h1 138.733102: softirq_raise: vec=9 [action=RCU]
3672 bash-1998 [000] ..s2 138.733 3247 bash-1998 [000] ..s2 138.733105: softirq_entry: vec=1 [action=TIMER]
3673 bash-1998 [000] ..s2 138.733 3248 bash-1998 [000] ..s2 138.733106: softirq_exit: vec=1 [action=TIMER]
3674 bash-1998 [000] ..s2 138.733 3249 bash-1998 [000] ..s2 138.733106: softirq_entry: vec=9 [action=RCU]
3675 bash-1998 [000] ..s2 138.733 3250 bash-1998 [000] ..s2 138.733109: softirq_exit: vec=9 [action=RCU]
3676 sshd-1995 [001] d.h1 138.733 3251 sshd-1995 [001] d.h1 138.733278: irq_handler_entry: irq=21 name=uhci_hcd:usb4
3677 sshd-1995 [001] d.h1 138.733 3252 sshd-1995 [001] d.h1 138.733280: irq_handler_exit: irq=21 ret=unhandled
3678 sshd-1995 [001] d.h1 138.733 3253 sshd-1995 [001] d.h1 138.733281: irq_handler_entry: irq=21 name=eth0
3679 sshd-1995 [001] d.h1 138.733 3254 sshd-1995 [001] d.h1 138.733283: irq_handler_exit: irq=21 ret=handled
3680 [...] 3255 [...]
3681 3256
3682 # cat instances/zoot/trace 3257 # cat instances/zoot/trace
3683 # tracer: nop 3258 # tracer: nop
3684 # 3259 #
3685 # entries-in-buffer/entries-written: 18996/ 3260 # entries-in-buffer/entries-written: 18996/18996 #P:4
3686 # 3261 #
3687 # _-----=> irq 3262 # _-----=> irqs-off
3688 # / _----=> nee 3263 # / _----=> need-resched
3689 # | / _---=> har 3264 # | / _---=> hardirq/softirq
3690 # || / _--=> pre 3265 # || / _--=> preempt-depth
3691 # ||| / dela 3266 # ||| / delay
3692 # TASK-PID CPU# |||| TIMEST 3267 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3693 # | | | |||| | 3268 # | | | |||| | |
3694 bash-1998 [000] d... 140.733 3269 bash-1998 [000] d... 140.733501: sys_write -> 0x2
3695 bash-1998 [000] d... 140.733 3270 bash-1998 [000] d... 140.733504: sys_dup2(oldfd: a, newfd: 1)
3696 bash-1998 [000] d... 140.733 3271 bash-1998 [000] d... 140.733506: sys_dup2 -> 0x1
3697 bash-1998 [000] d... 140.733 3272 bash-1998 [000] d... 140.733508: sys_fcntl(fd: a, cmd: 1, arg: 0)
3698 bash-1998 [000] d... 140.733 3273 bash-1998 [000] d... 140.733509: sys_fcntl -> 0x1
3699 bash-1998 [000] d... 140.733 3274 bash-1998 [000] d... 140.733510: sys_close(fd: a)
3700 bash-1998 [000] d... 140.733 3275 bash-1998 [000] d... 140.733510: sys_close -> 0x0
3701 bash-1998 [000] d... 140.733 3276 bash-1998 [000] d... 140.733514: sys_rt_sigprocmask(how: 0, nset: 0, oset: 6e2768, sigsetsize: 8)
3702 bash-1998 [000] d... 140.733 3277 bash-1998 [000] d... 140.733515: sys_rt_sigprocmask -> 0x0
3703 bash-1998 [000] d... 140.733 3278 bash-1998 [000] d... 140.733516: sys_rt_sigaction(sig: 2, act: 7fff718846f0, oact: 7fff71884650, sigsetsize: 8)
3704 bash-1998 [000] d... 140.733 3279 bash-1998 [000] d... 140.733516: sys_rt_sigaction -> 0x0
3705 3280
3706 You can see that the trace of the top most tr 3281 You can see that the trace of the top most trace buffer shows only
3707 the function tracing. The foo instance displa 3282 the function tracing. The foo instance displays wakeups and task
3708 switches. 3283 switches.
3709 3284
3710 To remove the instances, simply delete their 3285 To remove the instances, simply delete their directories:
3711 :: 3286 ::
3712 3287
3713 # rmdir instances/foo 3288 # rmdir instances/foo
3714 # rmdir instances/bar 3289 # rmdir instances/bar
3715 # rmdir instances/zoot 3290 # rmdir instances/zoot
3716 3291
3717 Note, if a process has a trace file open in o 3292 Note, if a process has a trace file open in one of the instance
3718 directories, the rmdir will fail with EBUSY. 3293 directories, the rmdir will fail with EBUSY.
3719 3294
3720 3295
3721 Stack trace 3296 Stack trace
3722 ----------- 3297 -----------
3723 Since the kernel has a fixed sized stack, it 3298 Since the kernel has a fixed sized stack, it is important not to
3724 waste it in functions. A kernel developer mus !! 3299 waste it in functions. A kernel developer must be conscience of
3725 what they allocate on the stack. If they add 3300 what they allocate on the stack. If they add too much, the system
3726 can be in danger of a stack overflow, and cor 3301 can be in danger of a stack overflow, and corruption will occur,
3727 usually leading to a system panic. 3302 usually leading to a system panic.
3728 3303
3729 There are some tools that check this, usually 3304 There are some tools that check this, usually with interrupts
3730 periodically checking usage. But if you can p 3305 periodically checking usage. But if you can perform a check
3731 at every function call that will become very 3306 at every function call that will become very useful. As ftrace provides
3732 a function tracer, it makes it convenient to 3307 a function tracer, it makes it convenient to check the stack size
3733 at every function call. This is enabled via t 3308 at every function call. This is enabled via the stack tracer.
3734 3309
3735 CONFIG_STACK_TRACER enables the ftrace stack 3310 CONFIG_STACK_TRACER enables the ftrace stack tracing functionality.
3736 To enable it, write a '1' into /proc/sys/kern 3311 To enable it, write a '1' into /proc/sys/kernel/stack_tracer_enabled.
3737 :: 3312 ::
3738 3313
3739 # echo 1 > /proc/sys/kernel/stack_tracer_ena 3314 # echo 1 > /proc/sys/kernel/stack_tracer_enabled
3740 3315
3741 You can also enable it from the kernel comman 3316 You can also enable it from the kernel command line to trace
3742 the stack size of the kernel during boot up, 3317 the stack size of the kernel during boot up, by adding "stacktrace"
3743 to the kernel command line parameter. 3318 to the kernel command line parameter.
3744 3319
3745 After running it for a few minutes, the outpu 3320 After running it for a few minutes, the output looks like:
3746 :: 3321 ::
3747 3322
3748 # cat stack_max_size 3323 # cat stack_max_size
3749 2928 3324 2928
3750 3325
3751 # cat stack_trace 3326 # cat stack_trace
3752 Depth Size Location (18 ent 3327 Depth Size Location (18 entries)
3753 ----- ---- -------- 3328 ----- ---- --------
3754 0) 2928 224 update_sd_lb_stats+ 3329 0) 2928 224 update_sd_lb_stats+0xbc/0x4ac
3755 1) 2704 160 find_busiest_group+ 3330 1) 2704 160 find_busiest_group+0x31/0x1f1
3756 2) 2544 256 load_balance+0xd9/0 3331 2) 2544 256 load_balance+0xd9/0x662
3757 3) 2288 80 idle_balance+0xbb/0 3332 3) 2288 80 idle_balance+0xbb/0x130
3758 4) 2208 128 __schedule+0x26e/0x 3333 4) 2208 128 __schedule+0x26e/0x5b9
3759 5) 2080 16 schedule+0x64/0x66 3334 5) 2080 16 schedule+0x64/0x66
3760 6) 2064 128 schedule_timeout+0x 3335 6) 2064 128 schedule_timeout+0x34/0xe0
3761 7) 1936 112 wait_for_common+0x9 3336 7) 1936 112 wait_for_common+0x97/0xf1
3762 8) 1824 16 wait_for_completion 3337 8) 1824 16 wait_for_completion+0x1d/0x1f
3763 9) 1808 128 flush_work+0xfe/0x1 3338 9) 1808 128 flush_work+0xfe/0x119
3764 10) 1680 16 tty_flush_to_ldisc+ 3339 10) 1680 16 tty_flush_to_ldisc+0x1e/0x20
3765 11) 1664 48 input_available_p+0 3340 11) 1664 48 input_available_p+0x1d/0x5c
3766 12) 1616 48 n_tty_poll+0x6d/0x1 3341 12) 1616 48 n_tty_poll+0x6d/0x134
3767 13) 1568 64 tty_poll+0x64/0x7f 3342 13) 1568 64 tty_poll+0x64/0x7f
3768 14) 1504 880 do_select+0x31e/0x5 3343 14) 1504 880 do_select+0x31e/0x511
3769 15) 624 400 core_sys_select+0x1 3344 15) 624 400 core_sys_select+0x177/0x216
3770 16) 224 96 sys_select+0x91/0xb 3345 16) 224 96 sys_select+0x91/0xb9
3771 17) 128 128 system_call_fastpat 3346 17) 128 128 system_call_fastpath+0x16/0x1b
3772 3347
3773 Note, if -mfentry is being used by gcc, funct 3348 Note, if -mfentry is being used by gcc, functions get traced before
3774 they set up the stack frame. This means that 3349 they set up the stack frame. This means that leaf level functions
3775 are not tested by the stack tracer when -mfen 3350 are not tested by the stack tracer when -mfentry is used.
3776 3351
3777 Currently, -mfentry is used by gcc 4.6.0 and 3352 Currently, -mfentry is used by gcc 4.6.0 and above on x86 only.
3778 3353
3779 More 3354 More
3780 ---- 3355 ----
3781 More details can be found in the source code, 3356 More details can be found in the source code, in the `kernel/trace/*.c` files.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.