1 ========================
2 ftrace - Function Tracer
3 ========================
5 Copyright 2008 Red Hat Inc.
7 :Author: Steven Rostedt <srostedt@redhat.com>
8 :License: The GNU Free Documentation License, Version 1.2
9 (dual licensed under the GPL v2)
10 :Original Reviewers: Elias Oltmanns, Randy Dunlap, Andrew Morton,
11 John Kacur, and David Teigland.
13 - Written for: 2.6.28-rc2
15 - Updated for: 4.13 - Copyright 2017 VMware Inc. Steven Rostedt
16 - Converted to rst format - Changbin Du <changbin.du@intel.com>
21 Ftrace is an internal tracer designed to help out developers and
22 designers of systems to find what is going on inside the kernel.
23 It can be used for debugging or analyzing latencies and
24 performance issues that take place outside of user-space.
26 Although ftrace is typically considered the function tracer, it
27 is really a framework of several assorted tracing utilities.
28 There's latency tracing to examine what occurs between interrupts
29 disabled and enabled, as well as for preemption and from a time
30 a task is woken to the task is actually scheduled in.
32 One of the most common uses of ftrace is the event tracing.
33 Throughout the kernel is hundreds of static event points that
34 can be enabled via the tracefs file system to see what is
35 going on in certain parts of the kernel.
37 See events.txt for more information.
40 Implementation Details
41 ----------------------
43 See :doc:`ftrace-design` for details for arch porters and such.
49 Ftrace uses the tracefs file system to hold the control files as
50 well as the files to display output.
52 When tracefs is configured into the kernel (which selecting any ftrace
53 option will do) the directory /sys/kernel/tracing will be created. To mount
54 this directory, you can add to your /etc/fstab file::
56 tracefs /sys/kernel/tracing tracefs defaults 0 0
58 Or you can mount it at run time with::
60 mount -t tracefs nodev /sys/kernel/tracing
62 For quicker access to that directory you may want to make a soft link to
65 ln -s /sys/kernel/tracing /tracing
69 Before 4.1, all ftrace tracing control files were within the debugfs
70 file system, which is typically located at /sys/kernel/debug/tracing.
71 For backward compatibility, when mounting the debugfs file system,
72 the tracefs file system will be automatically mounted at:
74 /sys/kernel/debug/tracing
76 All files located in the tracefs file system will be located in that
77 debugfs file system directory as well.
81 Any selected ftrace option will also create the tracefs file system.
82 The rest of the document will assume that you are in the ftrace directory
83 (cd /sys/kernel/tracing) and will only concentrate on the files within that
84 directory and not distract from the content with the extended
85 "/sys/kernel/tracing" path name.
87 That's it! (assuming that you have ftrace configured into your kernel)
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 files:
93 Note: all time values are in microseconds.
97 This is used to set or display the current tracer
102 This holds the different types of tracers that
103 have been compiled into the kernel. The
104 tracers listed here can be configured by
105 echoing their name into current_tracer.
109 This sets or displays whether writing to the trace
110 ring buffer is enabled. Echo 0 into this file to disable
111 the tracer or 1 to enable it. Note, this only disables
112 writing to the ring buffer, the tracing overhead may
115 The kernel function tracing_off() can be used within the
116 kernel to disable writing to the ring buffer, which will
117 set this file to "0". User space can re-enable tracing by
118 echoing "1" into the file.
120 Note, the function and event trigger "traceoff" will also
121 set this file to zero and stop tracing. Which can also
122 be re-enabled by user space using this file.
126 This file holds the output of the trace in a human
127 readable format (described below). Note, tracing is temporarily
128 disabled while this file is being read (opened).
132 The output is the same as the "trace" file but this
133 file is meant to be streamed with live tracing.
134 Reads from this file will block until new data is
135 retrieved. Unlike the "trace" file, this file is a
136 consumer. This means reading from this file causes
137 sequential reads to display more current data. Once
138 data is read from this file, it is consumed, and
139 will not be read again with a sequential read. The
140 "trace" file is static, and if the tracer is not
141 adding more data, it will display the same
142 information every time it is read. This file will not
143 disable tracing while being read.
147 This file lets the user control the amount of data
148 that is displayed in one of the above output
149 files. Options also exist to modify how a tracer
150 or events work (stack traces, timestamps, etc).
154 This is a directory that has a file for every available
155 trace option (also in trace_options). Options may also be set
156 or cleared by writing a "1" or "0" respectively into the
157 corresponding file with the option name.
161 Some of the tracers record the max latency.
162 For example, the maximum time that interrupts are disabled.
163 The maximum time is saved in this file. The max trace will also be
164 stored, and displayed by "trace". A new max trace will only be
165 recorded if the latency is greater than the value in this file
168 By echoing in a time into this file, no latency will be recorded
169 unless it is greater than the time in this file.
173 Some latency tracers will record a trace whenever the
174 latency is greater than the number in this file.
175 Only active when the file contains a number greater than 0.
180 This sets or displays the number of kilobytes each CPU
181 buffer holds. By default, the trace buffers are the same size
182 for each CPU. The displayed number is the size of the
183 CPU buffer and not total size of all buffers. The
184 trace buffers are allocated in pages (blocks of memory
185 that the kernel uses for allocation, usually 4 KB in size).
186 If the last page allocated has room for more bytes
187 than requested, the rest of the page will be used,
188 making the actual allocation bigger than requested or shown.
189 ( Note, the size may not be a multiple of the page size
190 due to buffer management meta-data. )
192 Buffer sizes for individual CPUs may vary
193 (see "per_cpu/cpu0/buffer_size_kb" below), and if they do
194 this file will show "X".
196 buffer_total_size_kb:
198 This displays the total combined size of all the trace buffers.
202 If a process is performing tracing, and the ring buffer should be
203 shrunk "freed" when the process is finished, even if it were to be
204 killed by a signal, this file can be used for that purpose. On close
205 of this file, the ring buffer will be resized to its minimum size.
206 Having a process that is tracing also open this file, when the process
207 exits its file descriptor for this file will be closed, and in doing so,
208 the ring buffer will be "freed".
210 It may also stop tracing if disable_on_free option is set.
214 This is a mask that lets the user only trace on specified CPUs.
215 The format is a hex string representing the CPUs.
219 When dynamic ftrace is configured in (see the
220 section below "dynamic ftrace"), the code is dynamically
221 modified (code text rewrite) to disable calling of the
222 function profiler (mcount). This lets tracing be configured
223 in with practically no overhead in performance. This also
224 has a side effect of enabling or disabling specific functions
225 to be traced. Echoing names of functions into this file
226 will limit the trace to only those functions.
227 This influences the tracers "function" and "function_graph"
228 and thus also function profiling (see "function_profile_enabled").
230 The functions listed in "available_filter_functions" are what
231 can be written into this file.
233 This interface also allows for commands to be used. See the
234 "Filter commands" section for more details.
236 As a speed up, since processing strings can't be quite expensive
237 and requires a check of all functions registered to tracing, instead
238 an index can be written into this file. A number (starting with "1")
239 written will instead select the same corresponding at the line position
240 of the "available_filter_functions" file.
244 This has an effect opposite to that of
245 set_ftrace_filter. Any function that is added here will not
246 be traced. If a function exists in both set_ftrace_filter
247 and set_ftrace_notrace, the function will _not_ be traced.
251 Have the function tracer only trace the threads whose PID are
254 If the "function-fork" option is set, then when a task whose
255 PID is listed in this file forks, the child's PID will
256 automatically be added to this file, and the child will be
257 traced by the function tracer as well. This option will also
258 cause PIDs of tasks that exit to be removed from the file.
262 Have the events only trace a task with a PID listed in this file.
263 Note, sched_switch and sched_wake_up will also trace events
266 To have the PIDs of children of tasks with their PID in this file
267 added on fork, enable the "event-fork" option. That option will also
268 cause the PIDs of tasks to be removed from this file when the task
273 Functions listed in this file will cause the function graph
274 tracer to only trace these functions and the functions that
275 they call. (See the section "dynamic ftrace" for more details).
276 Note, set_ftrace_filter and set_ftrace_notrace still affects
277 what functions are being traced.
281 Similar to set_graph_function, but will disable function graph
282 tracing when the function is hit until it exits the function.
283 This makes it possible to ignore tracing functions that are called
284 by a specific function.
286 available_filter_functions:
288 This lists the functions that ftrace has processed and can trace.
289 These are the function names that you can pass to
290 "set_ftrace_filter", "set_ftrace_notrace",
291 "set_graph_function", or "set_graph_notrace".
292 (See the section "dynamic ftrace" below for more details.)
294 dyn_ftrace_total_info:
296 This file is for debugging purposes. The number of functions that
297 have been converted to nops and are available to be traced.
301 This file is more for debugging ftrace, but can also be useful
302 in seeing if any function has a callback attached to it.
303 Not only does the trace infrastructure use ftrace function
304 trace utility, but other subsystems might too. This file
305 displays all functions that have a callback attached to them
306 as well as the number of callbacks that have been attached.
307 Note, a callback may also call multiple functions which will
308 not be listed in this count.
310 If the callback registered to be traced by a function with
311 the "save regs" attribute (thus even more overhead), a 'R'
312 will be displayed on the same line as the function that
313 is returning registers.
315 If the callback registered to be traced by a function with
316 the "ip modify" attribute (thus the regs->ip can be changed),
317 an 'I' will be displayed on the same line as the function that
320 If the architecture supports it, it will also show what callback
321 is being directly called by the function. If the count is greater
322 than 1 it most likely will be ftrace_ops_list_func().
324 If the callback of the function jumps to a trampoline that is
325 specific to a the callback and not the standard trampoline,
326 its address will be printed as well as the function that the
329 function_profile_enabled:
331 When set it will enable all functions with either the function
332 tracer, or if configured, the function graph tracer. It will
333 keep a histogram of the number of functions that were called
334 and if the function graph tracer was configured, it will also keep
335 track of the time spent in those functions. The histogram
336 content can be displayed in the files:
338 trace_stat/function<cpu> ( function0, function1, etc).
342 A directory that holds different tracing stats.
346 Enable dynamic trace points. See kprobetrace.txt.
350 Dynamic trace points stats. See kprobetrace.txt.
354 Used with the function graph tracer. This is the max depth
355 it will trace into a function. Setting this to a value of
356 one will show only the first kernel function that is called
361 This is for tools that read the raw format files. If an event in
362 the ring buffer references a string, only a pointer to the string
363 is recorded into the buffer and not the string itself. This prevents
364 tools from knowing what that string was. This file displays the string
365 and address for the string allowing tools to map the pointers to what
370 Only the pid of the task is recorded in a trace event unless
371 the event specifically saves the task comm as well. Ftrace
372 makes a cache of pid mappings to comms to try to display
373 comms for events. If a pid for a comm is not listed, then
374 "<...>" is displayed in the output.
376 If the option "record-cmd" is set to "0", then comms of tasks
377 will not be saved during recording. By default, it is enabled.
381 By default, 128 comms are saved (see "saved_cmdlines" above). To
382 increase or decrease the amount of comms that are cached, echo
383 in a the number of comms to cache, into this file.
387 If the option "record-tgid" is set, on each scheduling context switch
388 the Task Group ID of a task is saved in a table mapping the PID of
389 the thread to its TGID. By default, the "record-tgid" option is
394 This displays the "snapshot" buffer and also lets the user
395 take a snapshot of the current running trace.
396 See the "Snapshot" section below for more details.
400 When the stack tracer is activated, this will display the
401 maximum stack size it has encountered.
402 See the "Stack Trace" section below.
406 This displays the stack back trace of the largest stack
407 that was encountered when the stack tracer is activated.
408 See the "Stack Trace" section below.
412 This is similar to "set_ftrace_filter" but it limits what
413 functions the stack tracer will check.
417 Whenever an event is recorded into the ring buffer, a
418 "timestamp" is added. This stamp comes from a specified
419 clock. By default, ftrace uses the "local" clock. This
420 clock is very fast and strictly per cpu, but on some
421 systems it may not be monotonic with respect to other
422 CPUs. In other words, the local clocks may not be in sync
423 with local clocks on other CPUs.
425 Usual clocks for tracing::
428 [local] global counter x86-tsc
430 The clock with the square brackets around it is the one in effect.
433 Default clock, but may not be in sync across CPUs
436 This clock is in sync with all CPUs but may
437 be a bit slower than the local clock.
440 This is not a clock at all, but literally an atomic
441 counter. It counts up one by one, but is in sync
442 with all CPUs. This is useful when you need to
443 know exactly the order events occurred with respect to
444 each other on different CPUs.
447 This uses the jiffies counter and the time stamp
448 is relative to the time since boot up.
451 This makes ftrace use the same clock that perf uses.
452 Eventually perf will be able to read ftrace buffers
453 and this will help out in interleaving the data.
456 Architectures may define their own clocks. For
457 example, x86 uses its own TSC cycle clock here.
460 This uses the powerpc timebase register value.
461 This is in sync across CPUs and can also be used
462 to correlate events across hypervisor/guest if
466 This uses the fast monotonic clock (CLOCK_MONOTONIC)
467 which is monotonic and is subject to NTP rate adjustments.
470 This is the raw monotonic clock (CLOCK_MONOTONIC_RAW)
471 which is monotonic but is not subject to any rate adjustments
472 and ticks at the same rate as the hardware clocksource.
475 This is the boot clock (CLOCK_BOOTTIME) and is based on the
476 fast monotonic clock, but also accounts for time spent in
477 suspend. Since the clock access is designed for use in
478 tracing in the suspend path, some side effects are possible
479 if clock is accessed after the suspend time is accounted before
480 the fast mono clock is updated. In this case, the clock update
481 appears to happen slightly sooner than it normally would have.
482 Also on 32-bit systems, it's possible that the 64-bit boot offset
483 sees a partial update. These effects are rare and post
484 processing should be able to handle them. See comments in the
485 ktime_get_boot_fast_ns() function for more information.
487 To set a clock, simply echo the clock name into this file::
489 # echo global > trace_clock
493 This is a very useful file for synchronizing user space
494 with events happening in the kernel. Writing strings into
495 this file will be written into the ftrace buffer.
497 It is useful in applications to open this file at the start
498 of the application and just reference the file descriptor
501 void trace_write(const char *fmt, ...)
511 n = vsnprintf(buf, 256, fmt, ap);
514 write(trace_fd, buf, n);
519 trace_fd = open("trace_marker", WR_ONLY);
521 Note: Writing into the trace_marker file can also initiate triggers
522 that are written into /sys/kernel/tracing/events/ftrace/print/trigger
523 See "Event triggers" in Documentation/trace/events.rst and an
524 example in Documentation/trace/histogram.rst (Section 3.)
528 This is similar to trace_marker above, but is meant for for binary data
529 to be written to it, where a tool can be used to parse the data
534 Add dynamic tracepoints in programs.
539 Uprobe statistics. See uprobetrace.txt
543 This is a way to make multiple trace buffers where different
544 events can be recorded in different buffers.
545 See "Instances" section below.
549 This is the trace event directory. It holds event tracepoints
550 (also known as static tracepoints) that have been compiled
551 into the kernel. It shows what event tracepoints exist
552 and how they are grouped by system. There are "enable"
553 files at various levels that can enable the tracepoints
554 when a "1" is written to them.
556 See events.txt for more information.
560 By echoing in the event into this file, will enable that event.
562 See events.txt for more information.
566 A list of events that can be enabled in tracing.
568 See events.txt for more information.
572 Certain tracers may change the timestamp mode used when
573 logging trace events into the event buffer. Events with
574 different modes can coexist within a buffer but the mode in
575 effect when an event is logged determines which timestamp mode
576 is used for that event. The default timestamp mode is
579 Usual timestamp modes for tracing:
584 The timestamp mode with the square brackets around it is the
587 delta: Default timestamp mode - timestamp is a delta against
588 a per-buffer timestamp.
590 absolute: The timestamp is a full timestamp, not a delta
591 against some other value. As such it takes up more
592 space and is less efficient.
596 Directory for the Hardware Latency Detector.
597 See "Hardware Latency Detector" section below.
601 This is a directory that contains the trace per_cpu information.
603 per_cpu/cpu0/buffer_size_kb:
605 The ftrace buffer is defined per_cpu. That is, there's a separate
606 buffer for each CPU to allow writes to be done atomically,
607 and free from cache bouncing. These buffers may have different
608 size buffers. This file is similar to the buffer_size_kb
609 file, but it only displays or sets the buffer size for the
610 specific CPU. (here cpu0).
614 This is similar to the "trace" file, but it will only display
615 the data specific for the CPU. If written to, it only clears
616 the specific CPU buffer.
618 per_cpu/cpu0/trace_pipe
620 This is similar to the "trace_pipe" file, and is a consuming
621 read, but it will only display (and consume) the data specific
624 per_cpu/cpu0/trace_pipe_raw
626 For tools that can parse the ftrace ring buffer binary format,
627 the trace_pipe_raw file can be used to extract the data
628 from the ring buffer directly. With the use of the splice()
629 system call, the buffer data can be quickly transferred to
630 a file or to the network where a server is collecting the
633 Like trace_pipe, this is a consuming reader, where multiple
634 reads will always produce different data.
636 per_cpu/cpu0/snapshot:
638 This is similar to the main "snapshot" file, but will only
639 snapshot the current CPU (if supported). It only displays
640 the content of the snapshot for a given CPU, and if
641 written to, only clears this CPU buffer.
643 per_cpu/cpu0/snapshot_raw:
645 Similar to the trace_pipe_raw, but will read the binary format
646 from the snapshot buffer for the given CPU.
650 This displays certain stats about the ring buffer:
653 The number of events that are still in the buffer.
656 The number of lost events due to overwriting when
660 Should always be zero.
661 This gets set if so many events happened within a nested
662 event (ring buffer is re-entrant), that it fills the
663 buffer and starts dropping events.
666 Bytes actually read (not overwritten).
669 The oldest timestamp in the buffer
672 The current timestamp
675 Events lost due to overwrite option being off.
678 The number of events read.
683 Here is the list of current tracers that may be configured.
687 Function call tracer to trace all kernel functions.
691 Similar to the function tracer except that the
692 function tracer probes the functions on their entry
693 whereas the function graph tracer traces on both entry
694 and exit of the functions. It then provides the ability
695 to draw a graph of function calls similar to C code
700 The block tracer. The tracer used by the blktrace user
705 The Hardware Latency tracer is used to detect if the hardware
706 produces any latency. See "Hardware Latency Detector" section
711 Traces the areas that disable interrupts and saves
712 the trace with the longest max latency.
713 See tracing_max_latency. When a new max is recorded,
714 it replaces the old trace. It is best to view this
715 trace with the latency-format option enabled, which
716 happens automatically when the tracer is selected.
720 Similar to irqsoff but traces and records the amount of
721 time for which preemption is disabled.
725 Similar to irqsoff and preemptoff, but traces and
726 records the largest time for which irqs and/or preemption
731 Traces and records the max latency that it takes for
732 the highest priority task to get scheduled after
733 it has been woken up.
734 Traces all tasks as an average developer would expect.
738 Traces and records the max latency that it takes for just
739 RT tasks (as the current "wakeup" does). This is useful
740 for those interested in wake up timings of RT tasks.
744 Traces and records the max latency that it takes for
745 a SCHED_DEADLINE task to be woken (as the "wakeup" and
750 A special tracer that is used to trace binary module.
751 It will trace all the calls that a module makes to the
752 hardware. Everything it writes and reads from the I/O
757 This tracer can be configured when tracing likely/unlikely
758 calls within the kernel. It will trace when a likely and
759 unlikely branch is hit and if it was correct in its prediction
764 This is the "trace nothing" tracer. To remove all
765 tracers from tracing simply echo "nop" into
769 Examples of using the tracer
770 ----------------------------
772 Here are typical examples of using the tracers when controlling
773 them only with the tracefs interface (without using any
774 user-land utilities).
779 Here is an example of the output format of the file "trace"::
783 # entries-in-buffer/entries-written: 140080/250280 #P:4
786 # / _----=> need-resched
787 # | / _---=> hardirq/softirq
788 # || / _--=> preempt-depth
790 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
792 bash-1977 [000] .... 17284.993652: sys_close <-system_call_fastpath
793 bash-1977 [000] .... 17284.993653: __close_fd <-sys_close
794 bash-1977 [000] .... 17284.993653: _raw_spin_lock <-__close_fd
795 sshd-1974 [003] .... 17284.993653: __srcu_read_unlock <-fsnotify
796 bash-1977 [000] .... 17284.993654: add_preempt_count <-_raw_spin_lock
797 bash-1977 [000] ...1 17284.993655: _raw_spin_unlock <-__close_fd
798 bash-1977 [000] ...1 17284.993656: sub_preempt_count <-_raw_spin_unlock
799 bash-1977 [000] .... 17284.993657: filp_close <-__close_fd
800 bash-1977 [000] .... 17284.993657: dnotify_flush <-filp_close
801 sshd-1974 [003] .... 17284.993658: sys_select <-system_call_fastpath
804 A header is printed with the tracer name that is represented by
805 the trace. In this case the tracer is "function". Then it shows the
806 number of events in the buffer as well as the total number of entries
807 that were written. The difference is the number of entries that were
808 lost due to the buffer filling up (250280 - 140080 = 110200 events
811 The header explains the content of the events. Task name "bash", the task
812 PID "1977", the CPU that it was running on "000", the latency format
813 (explained below), the timestamp in <secs>.<usecs> format, the
814 function name that was traced "sys_close" and the parent function that
815 called this function "system_call_fastpath". The timestamp is the time
816 at which the function was entered.
821 When the latency-format option is enabled or when one of the latency
822 tracers is set, the trace file gives somewhat more information to see
823 why a latency happened. Here is a typical trace::
827 # irqsoff latency trace v1.1.5 on 3.8.0-test+
828 # --------------------------------------------------------------------
829 # latency: 259 us, #4/4, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
831 # | task: ps-6143 (uid:0 nice:0 policy:0 rt_prio:0)
833 # => started at: __lock_task_sighand
834 # => ended at: _raw_spin_unlock_irqrestore
838 # / _-----=> irqs-off
839 # | / _----=> need-resched
840 # || / _---=> hardirq/softirq
841 # ||| / _--=> preempt-depth
843 # cmd pid ||||| time | caller
845 ps-6143 2d... 0us!: trace_hardirqs_off <-__lock_task_sighand
846 ps-6143 2d..1 259us+: trace_hardirqs_on <-_raw_spin_unlock_irqrestore
847 ps-6143 2d..1 263us+: time_hardirqs_on <-_raw_spin_unlock_irqrestore
848 ps-6143 2d..1 306us : <stack trace>
849 => trace_hardirqs_on_caller
851 => _raw_spin_unlock_irqrestore
858 => system_call_fastpath
861 This shows that the current tracer is "irqsoff" tracing the time
862 for which interrupts were disabled. It gives the trace version (which
863 never changes) and the version of the kernel upon which this was executed on
864 (3.8). Then it displays the max latency in microseconds (259 us). The number
865 of trace entries displayed and the total number (both are four: #4/4).
866 VP, KP, SP, and HP are always zero and are reserved for later use.
867 #P is the number of online CPUs (#P:4).
869 The task is the process that was running when the latency
870 occurred. (ps pid: 6143).
872 The start and stop (the functions in which the interrupts were
873 disabled and enabled respectively) that caused the latencies:
875 - __lock_task_sighand is where the interrupts were disabled.
876 - _raw_spin_unlock_irqrestore is where they were enabled again.
878 The next lines after the header are the trace itself. The header
879 explains which is which.
881 cmd: The name of the process in the trace.
883 pid: The PID of that process.
885 CPU#: The CPU which the process was running on.
887 irqs-off: 'd' interrupts are disabled. '.' otherwise.
888 .. caution:: If the architecture does not support a way to
889 read the irq flags variable, an 'X' will always
893 - 'N' both TIF_NEED_RESCHED and PREEMPT_NEED_RESCHED is set,
894 - 'n' only TIF_NEED_RESCHED is set,
895 - 'p' only PREEMPT_NEED_RESCHED is set,
899 - 'Z' - NMI occurred inside a hardirq
900 - 'z' - NMI is running
901 - 'H' - hard irq occurred inside a softirq.
902 - 'h' - hard irq is running
903 - 's' - soft irq is running
904 - '.' - normal context.
906 preempt-depth: The level of preempt_disabled
908 The above is mostly meaningful for kernel developers.
911 When the latency-format option is enabled, the trace file
912 output includes a timestamp relative to the start of the
913 trace. This differs from the output when latency-format
914 is disabled, which includes an absolute timestamp.
917 This is just to help catch your eye a bit better. And
918 needs to be fixed to be only relative to the same CPU.
919 The marks are determined by the difference between this
920 current trace and the next trace.
922 - '$' - greater than 1 second
923 - '@' - greater than 100 millisecond
924 - '*' - greater than 10 millisecond
925 - '#' - greater than 1000 microsecond
926 - '!' - greater than 100 microsecond
927 - '+' - greater than 10 microsecond
928 - ' ' - less than or equal to 10 microsecond.
930 The rest is the same as the 'trace' file.
932 Note, the latency tracers will usually end with a back trace
933 to easily find where the latency occurred.
938 The trace_options file (or the options directory) is used to control
939 what gets printed in the trace output, or manipulate the tracers.
940 To see what is available, simply cat the file::
971 To disable one of the options, echo in the option prepended with
974 echo noprint-parent > trace_options
976 To enable an option, leave off the "no"::
978 echo sym-offset > trace_options
980 Here are the available options:
983 On function traces, display the calling (parent)
984 function as well as the function being traced.
988 bash-4000 [01] 1477.606694: simple_strtoul <-kstrtoul
991 bash-4000 [01] 1477.606694: simple_strtoul
995 Display not only the function name, but also the
996 offset in the function. For example, instead of
997 seeing just "ktime_get", you will see
998 "ktime_get+0xb/0x20".
1002 bash-4000 [01] 1477.606694: simple_strtoul+0x6/0xa0
1005 This will also display the function address as well
1006 as the function name.
1010 bash-4000 [01] 1477.606694: simple_strtoul <c0339346>
1013 This deals with the trace file when the
1014 latency-format option is enabled.
1017 bash 4000 1 0 00000000 00010a95 [58127d26] 1720.415ms \
1018 (+0.000ms): simple_strtoul (kstrtoul)
1021 This will display raw numbers. This option is best for
1022 use with user applications that can translate the raw
1023 numbers better than having it done in the kernel.
1026 Similar to raw, but the numbers will be in a hexadecimal format.
1029 This will print out the formats in raw binary.
1032 When set, reading trace_pipe will not block when polled.
1035 Can disable trace_printk() from writing into the buffer.
1038 It is sometimes confusing when the CPU buffers are full
1039 and one CPU buffer had a lot of events recently, thus
1040 a shorter time frame, were another CPU may have only had
1041 a few events, which lets it have older events. When
1042 the trace is reported, it shows the oldest events first,
1043 and it may look like only one CPU ran (the one with the
1044 oldest events). When the annotate option is set, it will
1045 display when a new CPU buffer started::
1047 <idle>-0 [001] dNs4 21169.031481: wake_up_idle_cpu <-add_timer_on
1048 <idle>-0 [001] dNs4 21169.031482: _raw_spin_unlock_irqrestore <-add_timer_on
1049 <idle>-0 [001] .Ns4 21169.031484: sub_preempt_count <-_raw_spin_unlock_irqrestore
1050 ##### CPU 2 buffer started ####
1051 <idle>-0 [002] .N.1 21169.031484: rcu_idle_exit <-cpu_idle
1052 <idle>-0 [001] .Ns3 21169.031484: _raw_spin_unlock <-clocksource_watchdog
1053 <idle>-0 [001] .Ns3 21169.031485: sub_preempt_count <-_raw_spin_unlock
1056 This option changes the trace. It records a
1057 stacktrace of the current user space thread after
1061 when user stacktrace are enabled, look up which
1062 object the address belongs to, and print a
1063 relative address. This is especially useful when
1064 ASLR is on, otherwise you don't get a chance to
1065 resolve the address to object/file/line after
1066 the app is no longer running
1068 The lookup is performed when you read
1069 trace,trace_pipe. Example::
1071 a.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0
1072 x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6]
1076 When set, trace_printk()s will only show the format
1077 and not their parameters (if trace_bprintk() or
1078 trace_bputs() was used to save the trace_printk()).
1081 Show only the event data. Hides the comm, PID,
1082 timestamp, CPU, and other useful data.
1085 This option changes the trace output. When it is enabled,
1086 the trace displays additional information about the
1087 latency, as described in "Latency trace format".
1090 When any event or tracer is enabled, a hook is enabled
1091 in the sched_switch trace point to fill comm cache
1092 with mapped pids and comms. But this may cause some
1093 overhead, and if you only care about pids, and not the
1094 name of the task, disabling this option can lower the
1095 impact of tracing. See "saved_cmdlines".
1098 When any event or tracer is enabled, a hook is enabled
1099 in the sched_switch trace point to fill the cache of
1100 mapped Thread Group IDs (TGID) mapping to pids. See
1104 This controls what happens when the trace buffer is
1105 full. If "1" (default), the oldest events are
1106 discarded and overwritten. If "0", then the newest
1107 events are discarded.
1108 (see per_cpu/cpu0/stats for overrun and dropped)
1111 When the free_buffer is closed, tracing will
1112 stop (tracing_on set to 0).
1115 Shows the interrupt, preempt count, need resched data.
1116 When disabled, the trace looks like::
1120 # entries-in-buffer/entries-written: 144405/9452052 #P:4
1122 # TASK-PID CPU# TIMESTAMP FUNCTION
1124 <idle>-0 [002] 23636.756054: ttwu_do_activate.constprop.89 <-try_to_wake_up
1125 <idle>-0 [002] 23636.756054: activate_task <-ttwu_do_activate.constprop.89
1126 <idle>-0 [002] 23636.756055: enqueue_task <-activate_task
1130 When set, the trace_marker is writable (only by root).
1131 When disabled, the trace_marker will error with EINVAL
1135 When set, tasks with PIDs listed in set_event_pid will have
1136 the PIDs of their children added to set_event_pid when those
1137 tasks fork. Also, when tasks with PIDs in set_event_pid exit,
1138 their PIDs will be removed from the file.
1141 The latency tracers will enable function tracing
1142 if this option is enabled (default it is). When
1143 it is disabled, the latency tracers do not trace
1144 functions. This keeps the overhead of the tracer down
1145 when performing latency tests.
1148 When set, tasks with PIDs listed in set_ftrace_pid will
1149 have the PIDs of their children added to set_ftrace_pid
1150 when those tasks fork. Also, when tasks with PIDs in
1151 set_ftrace_pid exit, their PIDs will be removed from the
1155 When set, the latency tracers (irqsoff, wakeup, etc) will
1156 use function graph tracing instead of function tracing.
1159 When set, a stack trace is recorded after any trace event
1163 Enable branch tracing with the tracer. This enables branch
1164 tracer along with the currently set tracer. Enabling this
1165 with the "nop" tracer is the same as just enabling the
1168 .. tip:: Some tracers have their own options. They only appear in this
1169 file when the tracer is active. They always appear in the
1173 Here are the per tracer options:
1175 Options for function tracer:
1178 When set, a stack trace is recorded after every
1179 function that is recorded. NOTE! Limit the functions
1180 that are recorded before enabling this, with
1181 "set_ftrace_filter" otherwise the system performance
1182 will be critically degraded. Remember to disable
1183 this option before clearing the function filter.
1185 Options for function_graph tracer:
1187 Since the function_graph tracer has a slightly different output
1188 it has its own options to control what is displayed.
1191 When set, the "overrun" of the graph stack is
1192 displayed after each function traced. The
1193 overrun, is when the stack depth of the calls
1194 is greater than what is reserved for each task.
1195 Each task has a fixed array of functions to
1196 trace in the call graph. If the depth of the
1197 calls exceeds that, the function is not traced.
1198 The overrun is the number of functions missed
1199 due to exceeding this array.
1202 When set, the CPU number of the CPU where the trace
1203 occurred is displayed.
1206 When set, if the function takes longer than
1207 A certain amount, then a delay marker is
1208 displayed. See "delay" above, under the
1212 Unlike other tracers, the process' command line
1213 is not displayed by default, but instead only
1214 when a task is traced in and out during a context
1215 switch. Enabling this options has the command
1216 of each process displayed at every line.
1219 At the end of each function (the return)
1220 the duration of the amount of time in the
1221 function is displayed in microseconds.
1224 When set, the timestamp is displayed at each line.
1227 When disabled, functions that happen inside an
1228 interrupt will not be traced.
1231 When set, the return event will include the function
1232 that it represents. By default this is off, and
1233 only a closing curly bracket "}" is displayed for
1234 the return of a function.
1237 When running function graph tracer, to include
1238 the time a task schedules out in its function.
1239 When enabled, it will account time the task has been
1240 scheduled out as part of the function call.
1243 When running function profiler with function graph tracer,
1244 to include the time to call nested functions. When this is
1245 not set, the time reported for the function will only
1246 include the time the function itself executed for, not the
1247 time for functions that it called.
1249 Options for blk tracer:
1252 Shows a more minimalistic output.
1258 When interrupts are disabled, the CPU can not react to any other
1259 external event (besides NMIs and SMIs). This prevents the timer
1260 interrupt from triggering or the mouse interrupt from letting
1261 the kernel know of a new mouse event. The result is a latency
1262 with the reaction time.
1264 The irqsoff tracer tracks the time for which interrupts are
1265 disabled. When a new maximum latency is hit, the tracer saves
1266 the trace leading up to that latency point so that every time a
1267 new maximum is reached, the old saved trace is discarded and the
1270 To reset the maximum, echo 0 into tracing_max_latency. Here is
1273 # echo 0 > options/function-trace
1274 # echo irqsoff > current_tracer
1275 # echo 1 > tracing_on
1276 # echo 0 > tracing_max_latency
1279 # echo 0 > tracing_on
1283 # irqsoff latency trace v1.1.5 on 3.8.0-test+
1284 # --------------------------------------------------------------------
1285 # latency: 16 us, #4/4, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1287 # | task: swapper/0-0 (uid:0 nice:0 policy:0 rt_prio:0)
1289 # => started at: run_timer_softirq
1290 # => ended at: run_timer_softirq
1294 # / _-----=> irqs-off
1295 # | / _----=> need-resched
1296 # || / _---=> hardirq/softirq
1297 # ||| / _--=> preempt-depth
1299 # cmd pid ||||| time | caller
1301 <idle>-0 0d.s2 0us+: _raw_spin_lock_irq <-run_timer_softirq
1302 <idle>-0 0dNs3 17us : _raw_spin_unlock_irq <-run_timer_softirq
1303 <idle>-0 0dNs3 17us+: trace_hardirqs_on <-run_timer_softirq
1304 <idle>-0 0dNs3 25us : <stack trace>
1305 => _raw_spin_unlock_irq
1306 => run_timer_softirq
1311 => smp_apic_timer_interrupt
1312 => apic_timer_interrupt
1317 => x86_64_start_reservations
1318 => x86_64_start_kernel
1320 Here we see that that we had a latency of 16 microseconds (which is
1321 very good). The _raw_spin_lock_irq in run_timer_softirq disabled
1322 interrupts. The difference between the 16 and the displayed
1323 timestamp 25us occurred because the clock was incremented
1324 between the time of recording the max latency and the time of
1325 recording the function that had that latency.
1327 Note the above example had function-trace not set. If we set
1328 function-trace, we get a much larger output::
1330 with echo 1 > options/function-trace
1334 # irqsoff latency trace v1.1.5 on 3.8.0-test+
1335 # --------------------------------------------------------------------
1336 # latency: 71 us, #168/168, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1338 # | task: bash-2042 (uid:0 nice:0 policy:0 rt_prio:0)
1340 # => started at: ata_scsi_queuecmd
1341 # => ended at: ata_scsi_queuecmd
1345 # / _-----=> irqs-off
1346 # | / _----=> need-resched
1347 # || / _---=> hardirq/softirq
1348 # ||| / _--=> preempt-depth
1350 # cmd pid ||||| time | caller
1352 bash-2042 3d... 0us : _raw_spin_lock_irqsave <-ata_scsi_queuecmd
1353 bash-2042 3d... 0us : add_preempt_count <-_raw_spin_lock_irqsave
1354 bash-2042 3d..1 1us : ata_scsi_find_dev <-ata_scsi_queuecmd
1355 bash-2042 3d..1 1us : __ata_scsi_find_dev <-ata_scsi_find_dev
1356 bash-2042 3d..1 2us : ata_find_dev.part.14 <-__ata_scsi_find_dev
1357 bash-2042 3d..1 2us : ata_qc_new_init <-__ata_scsi_queuecmd
1358 bash-2042 3d..1 3us : ata_sg_init <-__ata_scsi_queuecmd
1359 bash-2042 3d..1 4us : ata_scsi_rw_xlat <-__ata_scsi_queuecmd
1360 bash-2042 3d..1 4us : ata_build_rw_tf <-ata_scsi_rw_xlat
1362 bash-2042 3d..1 67us : delay_tsc <-__delay
1363 bash-2042 3d..1 67us : add_preempt_count <-delay_tsc
1364 bash-2042 3d..2 67us : sub_preempt_count <-delay_tsc
1365 bash-2042 3d..1 67us : add_preempt_count <-delay_tsc
1366 bash-2042 3d..2 68us : sub_preempt_count <-delay_tsc
1367 bash-2042 3d..1 68us+: ata_bmdma_start <-ata_bmdma_qc_issue
1368 bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1369 bash-2042 3d..1 71us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1370 bash-2042 3d..1 72us+: trace_hardirqs_on <-ata_scsi_queuecmd
1371 bash-2042 3d..1 120us : <stack trace>
1372 => _raw_spin_unlock_irqrestore
1373 => ata_scsi_queuecmd
1374 => scsi_dispatch_cmd
1376 => __blk_run_queue_uncond
1379 => generic_make_request
1382 => __ext3_get_inode_loc
1391 => user_path_at_empty
1396 => system_call_fastpath
1399 Here we traced a 71 microsecond latency. But we also see all the
1400 functions that were called during that time. Note that by
1401 enabling function tracing, we incur an added overhead. This
1402 overhead may extend the latency times. But nevertheless, this
1403 trace has provided some very helpful debugging information.
1405 If we prefer function graph output instead of function, we can set
1406 display-graph option::
1407 with echo 1 > options/display-graph
1411 # irqsoff latency trace v1.1.5 on 4.20.0-rc6+
1412 # --------------------------------------------------------------------
1413 # latency: 3751 us, #274/274, CPU#0 | (M:desktop VP:0, KP:0, SP:0 HP:0 #P:4)
1415 # | task: bash-1507 (uid:0 nice:0 policy:0 rt_prio:0)
1417 # => started at: free_debug_processing
1418 # => ended at: return_to_handler
1422 # / _----=> need-resched
1423 # | / _---=> hardirq/softirq
1424 # || / _--=> preempt-depth
1426 # REL TIME CPU TASK/PID |||| DURATION FUNCTION CALLS
1427 # | | | | |||| | | | | | |
1428 0 us | 0) bash-1507 | d... | 0.000 us | _raw_spin_lock_irqsave();
1429 0 us | 0) bash-1507 | d..1 | 0.378 us | do_raw_spin_trylock();
1430 1 us | 0) bash-1507 | d..2 | | set_track() {
1431 2 us | 0) bash-1507 | d..2 | | save_stack_trace() {
1432 2 us | 0) bash-1507 | d..2 | | __save_stack_trace() {
1433 3 us | 0) bash-1507 | d..2 | | __unwind_start() {
1434 3 us | 0) bash-1507 | d..2 | | get_stack_info() {
1435 3 us | 0) bash-1507 | d..2 | 0.351 us | in_task_stack();
1436 4 us | 0) bash-1507 | d..2 | 1.107 us | }
1438 3750 us | 0) bash-1507 | d..1 | 0.516 us | do_raw_spin_unlock();
1439 3750 us | 0) bash-1507 | d..1 | 0.000 us | _raw_spin_unlock_irqrestore();
1440 3764 us | 0) bash-1507 | d..1 | 0.000 us | tracer_hardirqs_on();
1441 bash-1507 0d..1 3792us : <stack trace>
1442 => free_debug_processing
1451 => search_binary_handler
1452 => __do_execve_file.isra.32
1455 => entry_SYSCALL_64_after_hwframe
1460 When preemption is disabled, we may be able to receive
1461 interrupts but the task cannot be preempted and a higher
1462 priority task must wait for preemption to be enabled again
1463 before it can preempt a lower priority task.
1465 The preemptoff tracer traces the places that disable preemption.
1466 Like the irqsoff tracer, it records the maximum latency for
1467 which preemption was disabled. The control of preemptoff tracer
1468 is much like the irqsoff tracer.
1471 # echo 0 > options/function-trace
1472 # echo preemptoff > current_tracer
1473 # echo 1 > tracing_on
1474 # echo 0 > tracing_max_latency
1477 # echo 0 > tracing_on
1479 # tracer: preemptoff
1481 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1482 # --------------------------------------------------------------------
1483 # latency: 46 us, #4/4, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1485 # | task: sshd-1991 (uid:0 nice:0 policy:0 rt_prio:0)
1487 # => started at: do_IRQ
1488 # => ended at: do_IRQ
1492 # / _-----=> irqs-off
1493 # | / _----=> need-resched
1494 # || / _---=> hardirq/softirq
1495 # ||| / _--=> preempt-depth
1497 # cmd pid ||||| time | caller
1499 sshd-1991 1d.h. 0us+: irq_enter <-do_IRQ
1500 sshd-1991 1d..1 46us : irq_exit <-do_IRQ
1501 sshd-1991 1d..1 47us+: trace_preempt_on <-do_IRQ
1502 sshd-1991 1d..1 52us : <stack trace>
1503 => sub_preempt_count
1509 This has some more changes. Preemption was disabled when an
1510 interrupt came in (notice the 'h'), and was enabled on exit.
1511 But we also see that interrupts have been disabled when entering
1512 the preempt off section and leaving it (the 'd'). We do not know if
1513 interrupts were enabled in the mean time or shortly after this
1517 # tracer: preemptoff
1519 # preemptoff latency trace v1.1.5 on 3.8.0-test+
1520 # --------------------------------------------------------------------
1521 # latency: 83 us, #241/241, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1523 # | task: bash-1994 (uid:0 nice:0 policy:0 rt_prio:0)
1525 # => started at: wake_up_new_task
1526 # => ended at: task_rq_unlock
1530 # / _-----=> irqs-off
1531 # | / _----=> need-resched
1532 # || / _---=> hardirq/softirq
1533 # ||| / _--=> preempt-depth
1535 # cmd pid ||||| time | caller
1537 bash-1994 1d..1 0us : _raw_spin_lock_irqsave <-wake_up_new_task
1538 bash-1994 1d..1 0us : select_task_rq_fair <-select_task_rq
1539 bash-1994 1d..1 1us : __rcu_read_lock <-select_task_rq_fair
1540 bash-1994 1d..1 1us : source_load <-select_task_rq_fair
1541 bash-1994 1d..1 1us : source_load <-select_task_rq_fair
1543 bash-1994 1d..1 12us : irq_enter <-smp_apic_timer_interrupt
1544 bash-1994 1d..1 12us : rcu_irq_enter <-irq_enter
1545 bash-1994 1d..1 13us : add_preempt_count <-irq_enter
1546 bash-1994 1d.h1 13us : exit_idle <-smp_apic_timer_interrupt
1547 bash-1994 1d.h1 13us : hrtimer_interrupt <-smp_apic_timer_interrupt
1548 bash-1994 1d.h1 13us : _raw_spin_lock <-hrtimer_interrupt
1549 bash-1994 1d.h1 14us : add_preempt_count <-_raw_spin_lock
1550 bash-1994 1d.h2 14us : ktime_get_update_offsets <-hrtimer_interrupt
1552 bash-1994 1d.h1 35us : lapic_next_event <-clockevents_program_event
1553 bash-1994 1d.h1 35us : irq_exit <-smp_apic_timer_interrupt
1554 bash-1994 1d.h1 36us : sub_preempt_count <-irq_exit
1555 bash-1994 1d..2 36us : do_softirq <-irq_exit
1556 bash-1994 1d..2 36us : __do_softirq <-call_softirq
1557 bash-1994 1d..2 36us : __local_bh_disable <-__do_softirq
1558 bash-1994 1d.s2 37us : add_preempt_count <-_raw_spin_lock_irq
1559 bash-1994 1d.s3 38us : _raw_spin_unlock <-run_timer_softirq
1560 bash-1994 1d.s3 39us : sub_preempt_count <-_raw_spin_unlock
1561 bash-1994 1d.s2 39us : call_timer_fn <-run_timer_softirq
1563 bash-1994 1dNs2 81us : cpu_needs_another_gp <-rcu_process_callbacks
1564 bash-1994 1dNs2 82us : __local_bh_enable <-__do_softirq
1565 bash-1994 1dNs2 82us : sub_preempt_count <-__local_bh_enable
1566 bash-1994 1dN.2 82us : idle_cpu <-irq_exit
1567 bash-1994 1dN.2 83us : rcu_irq_exit <-irq_exit
1568 bash-1994 1dN.2 83us : sub_preempt_count <-irq_exit
1569 bash-1994 1.N.1 84us : _raw_spin_unlock_irqrestore <-task_rq_unlock
1570 bash-1994 1.N.1 84us+: trace_preempt_on <-task_rq_unlock
1571 bash-1994 1.N.1 104us : <stack trace>
1572 => sub_preempt_count
1573 => _raw_spin_unlock_irqrestore
1581 The above is an example of the preemptoff trace with
1582 function-trace set. Here we see that interrupts were not disabled
1583 the entire time. The irq_enter code lets us know that we entered
1584 an interrupt 'h'. Before that, the functions being traced still
1585 show that it is not in an interrupt, but we can see from the
1586 functions themselves that this is not the case.
1591 Knowing the locations that have interrupts disabled or
1592 preemption disabled for the longest times is helpful. But
1593 sometimes we would like to know when either preemption and/or
1594 interrupts are disabled.
1596 Consider the following code::
1598 local_irq_disable();
1599 call_function_with_irqs_off();
1601 call_function_with_irqs_and_preemption_off();
1603 call_function_with_preemption_off();
1606 The irqsoff tracer will record the total length of
1607 call_function_with_irqs_off() and
1608 call_function_with_irqs_and_preemption_off().
1610 The preemptoff tracer will record the total length of
1611 call_function_with_irqs_and_preemption_off() and
1612 call_function_with_preemption_off().
1614 But neither will trace the time that interrupts and/or
1615 preemption is disabled. This total time is the time that we can
1616 not schedule. To record this time, use the preemptirqsoff
1619 Again, using this trace is much like the irqsoff and preemptoff
1623 # echo 0 > options/function-trace
1624 # echo preemptirqsoff > current_tracer
1625 # echo 1 > tracing_on
1626 # echo 0 > tracing_max_latency
1629 # echo 0 > tracing_on
1631 # tracer: preemptirqsoff
1633 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1634 # --------------------------------------------------------------------
1635 # latency: 100 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1637 # | task: ls-2230 (uid:0 nice:0 policy:0 rt_prio:0)
1639 # => started at: ata_scsi_queuecmd
1640 # => ended at: ata_scsi_queuecmd
1644 # / _-----=> irqs-off
1645 # | / _----=> need-resched
1646 # || / _---=> hardirq/softirq
1647 # ||| / _--=> preempt-depth
1649 # cmd pid ||||| time | caller
1651 ls-2230 3d... 0us+: _raw_spin_lock_irqsave <-ata_scsi_queuecmd
1652 ls-2230 3...1 100us : _raw_spin_unlock_irqrestore <-ata_scsi_queuecmd
1653 ls-2230 3...1 101us+: trace_preempt_on <-ata_scsi_queuecmd
1654 ls-2230 3...1 111us : <stack trace>
1655 => sub_preempt_count
1656 => _raw_spin_unlock_irqrestore
1657 => ata_scsi_queuecmd
1658 => scsi_dispatch_cmd
1660 => __blk_run_queue_uncond
1663 => generic_make_request
1668 => htree_dirblock_to_tree
1669 => ext3_htree_fill_tree
1673 => system_call_fastpath
1676 The trace_hardirqs_off_thunk is called from assembly on x86 when
1677 interrupts are disabled in the assembly code. Without the
1678 function tracing, we do not know if interrupts were enabled
1679 within the preemption points. We do see that it started with
1682 Here is a trace with function-trace set::
1684 # tracer: preemptirqsoff
1686 # preemptirqsoff latency trace v1.1.5 on 3.8.0-test+
1687 # --------------------------------------------------------------------
1688 # latency: 161 us, #339/339, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1690 # | task: ls-2269 (uid:0 nice:0 policy:0 rt_prio:0)
1692 # => started at: schedule
1693 # => ended at: mutex_unlock
1697 # / _-----=> irqs-off
1698 # | / _----=> need-resched
1699 # || / _---=> hardirq/softirq
1700 # ||| / _--=> preempt-depth
1702 # cmd pid ||||| time | caller
1704 kworker/-59 3...1 0us : __schedule <-schedule
1705 kworker/-59 3d..1 0us : rcu_preempt_qs <-rcu_note_context_switch
1706 kworker/-59 3d..1 1us : add_preempt_count <-_raw_spin_lock_irq
1707 kworker/-59 3d..2 1us : deactivate_task <-__schedule
1708 kworker/-59 3d..2 1us : dequeue_task <-deactivate_task
1709 kworker/-59 3d..2 2us : update_rq_clock <-dequeue_task
1710 kworker/-59 3d..2 2us : dequeue_task_fair <-dequeue_task
1711 kworker/-59 3d..2 2us : update_curr <-dequeue_task_fair
1712 kworker/-59 3d..2 2us : update_min_vruntime <-update_curr
1713 kworker/-59 3d..2 3us : cpuacct_charge <-update_curr
1714 kworker/-59 3d..2 3us : __rcu_read_lock <-cpuacct_charge
1715 kworker/-59 3d..2 3us : __rcu_read_unlock <-cpuacct_charge
1716 kworker/-59 3d..2 3us : update_cfs_rq_blocked_load <-dequeue_task_fair
1717 kworker/-59 3d..2 4us : clear_buddies <-dequeue_task_fair
1718 kworker/-59 3d..2 4us : account_entity_dequeue <-dequeue_task_fair
1719 kworker/-59 3d..2 4us : update_min_vruntime <-dequeue_task_fair
1720 kworker/-59 3d..2 4us : update_cfs_shares <-dequeue_task_fair
1721 kworker/-59 3d..2 5us : hrtick_update <-dequeue_task_fair
1722 kworker/-59 3d..2 5us : wq_worker_sleeping <-__schedule
1723 kworker/-59 3d..2 5us : kthread_data <-wq_worker_sleeping
1724 kworker/-59 3d..2 5us : put_prev_task_fair <-__schedule
1725 kworker/-59 3d..2 6us : pick_next_task_fair <-pick_next_task
1726 kworker/-59 3d..2 6us : clear_buddies <-pick_next_task_fair
1727 kworker/-59 3d..2 6us : set_next_entity <-pick_next_task_fair
1728 kworker/-59 3d..2 6us : update_stats_wait_end <-set_next_entity
1729 ls-2269 3d..2 7us : finish_task_switch <-__schedule
1730 ls-2269 3d..2 7us : _raw_spin_unlock_irq <-finish_task_switch
1731 ls-2269 3d..2 8us : do_IRQ <-ret_from_intr
1732 ls-2269 3d..2 8us : irq_enter <-do_IRQ
1733 ls-2269 3d..2 8us : rcu_irq_enter <-irq_enter
1734 ls-2269 3d..2 9us : add_preempt_count <-irq_enter
1735 ls-2269 3d.h2 9us : exit_idle <-do_IRQ
1737 ls-2269 3d.h3 20us : sub_preempt_count <-_raw_spin_unlock
1738 ls-2269 3d.h2 20us : irq_exit <-do_IRQ
1739 ls-2269 3d.h2 21us : sub_preempt_count <-irq_exit
1740 ls-2269 3d..3 21us : do_softirq <-irq_exit
1741 ls-2269 3d..3 21us : __do_softirq <-call_softirq
1742 ls-2269 3d..3 21us+: __local_bh_disable <-__do_softirq
1743 ls-2269 3d.s4 29us : sub_preempt_count <-_local_bh_enable_ip
1744 ls-2269 3d.s5 29us : sub_preempt_count <-_local_bh_enable_ip
1745 ls-2269 3d.s5 31us : do_IRQ <-ret_from_intr
1746 ls-2269 3d.s5 31us : irq_enter <-do_IRQ
1747 ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter
1749 ls-2269 3d.s5 31us : rcu_irq_enter <-irq_enter
1750 ls-2269 3d.s5 32us : add_preempt_count <-irq_enter
1751 ls-2269 3d.H5 32us : exit_idle <-do_IRQ
1752 ls-2269 3d.H5 32us : handle_irq <-do_IRQ
1753 ls-2269 3d.H5 32us : irq_to_desc <-handle_irq
1754 ls-2269 3d.H5 33us : handle_fasteoi_irq <-handle_irq
1756 ls-2269 3d.s5 158us : _raw_spin_unlock_irqrestore <-rtl8139_poll
1757 ls-2269 3d.s3 158us : net_rps_action_and_irq_enable.isra.65 <-net_rx_action
1758 ls-2269 3d.s3 159us : __local_bh_enable <-__do_softirq
1759 ls-2269 3d.s3 159us : sub_preempt_count <-__local_bh_enable
1760 ls-2269 3d..3 159us : idle_cpu <-irq_exit
1761 ls-2269 3d..3 159us : rcu_irq_exit <-irq_exit
1762 ls-2269 3d..3 160us : sub_preempt_count <-irq_exit
1763 ls-2269 3d... 161us : __mutex_unlock_slowpath <-mutex_unlock
1764 ls-2269 3d... 162us+: trace_hardirqs_on <-mutex_unlock
1765 ls-2269 3d... 186us : <stack trace>
1766 => __mutex_unlock_slowpath
1773 => system_call_fastpath
1775 This is an interesting trace. It started with kworker running and
1776 scheduling out and ls taking over. But as soon as ls released the
1777 rq lock and enabled interrupts (but not preemption) an interrupt
1778 triggered. When the interrupt finished, it started running softirqs.
1779 But while the softirq was running, another interrupt triggered.
1780 When an interrupt is running inside a softirq, the annotation is 'H'.
1786 One common case that people are interested in tracing is the
1787 time it takes for a task that is woken to actually wake up.
1788 Now for non Real-Time tasks, this can be arbitrary. But tracing
1789 it none the less can be interesting.
1791 Without function tracing::
1793 # echo 0 > options/function-trace
1794 # echo wakeup > current_tracer
1795 # echo 1 > tracing_on
1796 # echo 0 > tracing_max_latency
1798 # echo 0 > tracing_on
1802 # wakeup latency trace v1.1.5 on 3.8.0-test+
1803 # --------------------------------------------------------------------
1804 # latency: 15 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1806 # | task: kworker/3:1H-312 (uid:0 nice:-20 policy:0 rt_prio:0)
1810 # / _-----=> irqs-off
1811 # | / _----=> need-resched
1812 # || / _---=> hardirq/softirq
1813 # ||| / _--=> preempt-depth
1815 # cmd pid ||||| time | caller
1817 <idle>-0 3dNs7 0us : 0:120:R + [003] 312:100:R kworker/3:1H
1818 <idle>-0 3dNs7 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up
1819 <idle>-0 3d..3 15us : __schedule <-schedule
1820 <idle>-0 3d..3 15us : 0:120:R ==> [003] 312:100:R kworker/3:1H
1822 The tracer only traces the highest priority task in the system
1823 to avoid tracing the normal circumstances. Here we see that
1824 the kworker with a nice priority of -20 (not very nice), took
1825 just 15 microseconds from the time it woke up, to the time it
1828 Non Real-Time tasks are not that interesting. A more interesting
1829 trace is to concentrate only on Real-Time tasks.
1834 In a Real-Time environment it is very important to know the
1835 wakeup time it takes for the highest priority task that is woken
1836 up to the time that it executes. This is also known as "schedule
1837 latency". I stress the point that this is about RT tasks. It is
1838 also important to know the scheduling latency of non-RT tasks,
1839 but the average schedule latency is better for non-RT tasks.
1840 Tools like LatencyTop are more appropriate for such
1843 Real-Time environments are interested in the worst case latency.
1844 That is the longest latency it takes for something to happen,
1845 and not the average. We can have a very fast scheduler that may
1846 only have a large latency once in a while, but that would not
1847 work well with Real-Time tasks. The wakeup_rt tracer was designed
1848 to record the worst case wakeups of RT tasks. Non-RT tasks are
1849 not recorded because the tracer only records one worst case and
1850 tracing non-RT tasks that are unpredictable will overwrite the
1851 worst case latency of RT tasks (just run the normal wakeup
1852 tracer for a while to see that effect).
1854 Since this tracer only deals with RT tasks, we will run this
1855 slightly differently than we did with the previous tracers.
1856 Instead of performing an 'ls', we will run 'sleep 1' under
1857 'chrt' which changes the priority of the task.
1860 # echo 0 > options/function-trace
1861 # echo wakeup_rt > current_tracer
1862 # echo 1 > tracing_on
1863 # echo 0 > tracing_max_latency
1865 # echo 0 > tracing_on
1871 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
1872 # --------------------------------------------------------------------
1873 # latency: 5 us, #4/4, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1875 # | task: sleep-2389 (uid:0 nice:0 policy:1 rt_prio:5)
1879 # / _-----=> irqs-off
1880 # | / _----=> need-resched
1881 # || / _---=> hardirq/softirq
1882 # ||| / _--=> preempt-depth
1884 # cmd pid ||||| time | caller
1886 <idle>-0 3d.h4 0us : 0:120:R + [003] 2389: 94:R sleep
1887 <idle>-0 3d.h4 1us+: ttwu_do_activate.constprop.87 <-try_to_wake_up
1888 <idle>-0 3d..3 5us : __schedule <-schedule
1889 <idle>-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep
1892 Running this on an idle system, we see that it only took 5 microseconds
1893 to perform the task switch. Note, since the trace point in the schedule
1894 is before the actual "switch", we stop the tracing when the recorded task
1895 is about to schedule in. This may change if we add a new marker at the
1896 end of the scheduler.
1898 Notice that the recorded task is 'sleep' with the PID of 2389
1899 and it has an rt_prio of 5. This priority is user-space priority
1900 and not the internal kernel priority. The policy is 1 for
1901 SCHED_FIFO and 2 for SCHED_RR.
1903 Note, that the trace data shows the internal priority (99 - rtprio).
1906 <idle>-0 3d..3 5us : 0:120:R ==> [003] 2389: 94:R sleep
1908 The 0:120:R means idle was running with a nice priority of 0 (120 - 120)
1909 and in the running state 'R'. The sleep task was scheduled in with
1910 2389: 94:R. That is the priority is the kernel rtprio (99 - 5 = 94)
1911 and it too is in the running state.
1913 Doing the same with chrt -r 5 and function-trace set.
1916 echo 1 > options/function-trace
1920 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
1921 # --------------------------------------------------------------------
1922 # latency: 29 us, #85/85, CPU#3 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
1924 # | task: sleep-2448 (uid:0 nice:0 policy:1 rt_prio:5)
1928 # / _-----=> irqs-off
1929 # | / _----=> need-resched
1930 # || / _---=> hardirq/softirq
1931 # ||| / _--=> preempt-depth
1933 # cmd pid ||||| time | caller
1935 <idle>-0 3d.h4 1us+: 0:120:R + [003] 2448: 94:R sleep
1936 <idle>-0 3d.h4 2us : ttwu_do_activate.constprop.87 <-try_to_wake_up
1937 <idle>-0 3d.h3 3us : check_preempt_curr <-ttwu_do_wakeup
1938 <idle>-0 3d.h3 3us : resched_curr <-check_preempt_curr
1939 <idle>-0 3dNh3 4us : task_woken_rt <-ttwu_do_wakeup
1940 <idle>-0 3dNh3 4us : _raw_spin_unlock <-try_to_wake_up
1941 <idle>-0 3dNh3 4us : sub_preempt_count <-_raw_spin_unlock
1942 <idle>-0 3dNh2 5us : ttwu_stat <-try_to_wake_up
1943 <idle>-0 3dNh2 5us : _raw_spin_unlock_irqrestore <-try_to_wake_up
1944 <idle>-0 3dNh2 6us : sub_preempt_count <-_raw_spin_unlock_irqrestore
1945 <idle>-0 3dNh1 6us : _raw_spin_lock <-__run_hrtimer
1946 <idle>-0 3dNh1 6us : add_preempt_count <-_raw_spin_lock
1947 <idle>-0 3dNh2 7us : _raw_spin_unlock <-hrtimer_interrupt
1948 <idle>-0 3dNh2 7us : sub_preempt_count <-_raw_spin_unlock
1949 <idle>-0 3dNh1 7us : tick_program_event <-hrtimer_interrupt
1950 <idle>-0 3dNh1 7us : clockevents_program_event <-tick_program_event
1951 <idle>-0 3dNh1 8us : ktime_get <-clockevents_program_event
1952 <idle>-0 3dNh1 8us : lapic_next_event <-clockevents_program_event
1953 <idle>-0 3dNh1 8us : irq_exit <-smp_apic_timer_interrupt
1954 <idle>-0 3dNh1 9us : sub_preempt_count <-irq_exit
1955 <idle>-0 3dN.2 9us : idle_cpu <-irq_exit
1956 <idle>-0 3dN.2 9us : rcu_irq_exit <-irq_exit
1957 <idle>-0 3dN.2 10us : rcu_eqs_enter_common.isra.45 <-rcu_irq_exit
1958 <idle>-0 3dN.2 10us : sub_preempt_count <-irq_exit
1959 <idle>-0 3.N.1 11us : rcu_idle_exit <-cpu_idle
1960 <idle>-0 3dN.1 11us : rcu_eqs_exit_common.isra.43 <-rcu_idle_exit
1961 <idle>-0 3.N.1 11us : tick_nohz_idle_exit <-cpu_idle
1962 <idle>-0 3dN.1 12us : menu_hrtimer_cancel <-tick_nohz_idle_exit
1963 <idle>-0 3dN.1 12us : ktime_get <-tick_nohz_idle_exit
1964 <idle>-0 3dN.1 12us : tick_do_update_jiffies64 <-tick_nohz_idle_exit
1965 <idle>-0 3dN.1 13us : cpu_load_update_nohz <-tick_nohz_idle_exit
1966 <idle>-0 3dN.1 13us : _raw_spin_lock <-cpu_load_update_nohz
1967 <idle>-0 3dN.1 13us : add_preempt_count <-_raw_spin_lock
1968 <idle>-0 3dN.2 13us : __cpu_load_update <-cpu_load_update_nohz
1969 <idle>-0 3dN.2 14us : sched_avg_update <-__cpu_load_update
1970 <idle>-0 3dN.2 14us : _raw_spin_unlock <-cpu_load_update_nohz
1971 <idle>-0 3dN.2 14us : sub_preempt_count <-_raw_spin_unlock
1972 <idle>-0 3dN.1 15us : calc_load_nohz_stop <-tick_nohz_idle_exit
1973 <idle>-0 3dN.1 15us : touch_softlockup_watchdog <-tick_nohz_idle_exit
1974 <idle>-0 3dN.1 15us : hrtimer_cancel <-tick_nohz_idle_exit
1975 <idle>-0 3dN.1 15us : hrtimer_try_to_cancel <-hrtimer_cancel
1976 <idle>-0 3dN.1 16us : lock_hrtimer_base.isra.18 <-hrtimer_try_to_cancel
1977 <idle>-0 3dN.1 16us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18
1978 <idle>-0 3dN.1 16us : add_preempt_count <-_raw_spin_lock_irqsave
1979 <idle>-0 3dN.2 17us : __remove_hrtimer <-remove_hrtimer.part.16
1980 <idle>-0 3dN.2 17us : hrtimer_force_reprogram <-__remove_hrtimer
1981 <idle>-0 3dN.2 17us : tick_program_event <-hrtimer_force_reprogram
1982 <idle>-0 3dN.2 18us : clockevents_program_event <-tick_program_event
1983 <idle>-0 3dN.2 18us : ktime_get <-clockevents_program_event
1984 <idle>-0 3dN.2 18us : lapic_next_event <-clockevents_program_event
1985 <idle>-0 3dN.2 19us : _raw_spin_unlock_irqrestore <-hrtimer_try_to_cancel
1986 <idle>-0 3dN.2 19us : sub_preempt_count <-_raw_spin_unlock_irqrestore
1987 <idle>-0 3dN.1 19us : hrtimer_forward <-tick_nohz_idle_exit
1988 <idle>-0 3dN.1 20us : ktime_add_safe <-hrtimer_forward
1989 <idle>-0 3dN.1 20us : ktime_add_safe <-hrtimer_forward
1990 <idle>-0 3dN.1 20us : hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11
1991 <idle>-0 3dN.1 20us : __hrtimer_start_range_ns <-hrtimer_start_range_ns
1992 <idle>-0 3dN.1 21us : lock_hrtimer_base.isra.18 <-__hrtimer_start_range_ns
1993 <idle>-0 3dN.1 21us : _raw_spin_lock_irqsave <-lock_hrtimer_base.isra.18
1994 <idle>-0 3dN.1 21us : add_preempt_count <-_raw_spin_lock_irqsave
1995 <idle>-0 3dN.2 22us : ktime_add_safe <-__hrtimer_start_range_ns
1996 <idle>-0 3dN.2 22us : enqueue_hrtimer <-__hrtimer_start_range_ns
1997 <idle>-0 3dN.2 22us : tick_program_event <-__hrtimer_start_range_ns
1998 <idle>-0 3dN.2 23us : clockevents_program_event <-tick_program_event
1999 <idle>-0 3dN.2 23us : ktime_get <-clockevents_program_event
2000 <idle>-0 3dN.2 23us : lapic_next_event <-clockevents_program_event
2001 <idle>-0 3dN.2 24us : _raw_spin_unlock_irqrestore <-__hrtimer_start_range_ns
2002 <idle>-0 3dN.2 24us : sub_preempt_count <-_raw_spin_unlock_irqrestore
2003 <idle>-0 3dN.1 24us : account_idle_ticks <-tick_nohz_idle_exit
2004 <idle>-0 3dN.1 24us : account_idle_time <-account_idle_ticks
2005 <idle>-0 3.N.1 25us : sub_preempt_count <-cpu_idle
2006 <idle>-0 3.N.. 25us : schedule <-cpu_idle
2007 <idle>-0 3.N.. 25us : __schedule <-preempt_schedule
2008 <idle>-0 3.N.. 26us : add_preempt_count <-__schedule
2009 <idle>-0 3.N.1 26us : rcu_note_context_switch <-__schedule
2010 <idle>-0 3.N.1 26us : rcu_sched_qs <-rcu_note_context_switch
2011 <idle>-0 3dN.1 27us : rcu_preempt_qs <-rcu_note_context_switch
2012 <idle>-0 3.N.1 27us : _raw_spin_lock_irq <-__schedule
2013 <idle>-0 3dN.1 27us : add_preempt_count <-_raw_spin_lock_irq
2014 <idle>-0 3dN.2 28us : put_prev_task_idle <-__schedule
2015 <idle>-0 3dN.2 28us : pick_next_task_stop <-pick_next_task
2016 <idle>-0 3dN.2 28us : pick_next_task_rt <-pick_next_task
2017 <idle>-0 3dN.2 29us : dequeue_pushable_task <-pick_next_task_rt
2018 <idle>-0 3d..3 29us : __schedule <-preempt_schedule
2019 <idle>-0 3d..3 30us : 0:120:R ==> [003] 2448: 94:R sleep
2021 This isn't that big of a trace, even with function tracing enabled,
2022 so I included the entire trace.
2024 The interrupt went off while when the system was idle. Somewhere
2025 before task_woken_rt() was called, the NEED_RESCHED flag was set,
2026 this is indicated by the first occurrence of the 'N' flag.
2028 Latency tracing and events
2029 --------------------------
2030 As function tracing can induce a much larger latency, but without
2031 seeing what happens within the latency it is hard to know what
2032 caused it. There is a middle ground, and that is with enabling
2036 # echo 0 > options/function-trace
2037 # echo wakeup_rt > current_tracer
2038 # echo 1 > events/enable
2039 # echo 1 > tracing_on
2040 # echo 0 > tracing_max_latency
2042 # echo 0 > tracing_on
2046 # wakeup_rt latency trace v1.1.5 on 3.8.0-test+
2047 # --------------------------------------------------------------------
2048 # latency: 6 us, #12/12, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
2050 # | task: sleep-5882 (uid:0 nice:0 policy:1 rt_prio:5)
2054 # / _-----=> irqs-off
2055 # | / _----=> need-resched
2056 # || / _---=> hardirq/softirq
2057 # ||| / _--=> preempt-depth
2059 # cmd pid ||||| time | caller
2061 <idle>-0 2d.h4 0us : 0:120:R + [002] 5882: 94:R sleep
2062 <idle>-0 2d.h4 0us : ttwu_do_activate.constprop.87 <-try_to_wake_up
2063 <idle>-0 2d.h4 1us : sched_wakeup: comm=sleep pid=5882 prio=94 success=1 target_cpu=002
2064 <idle>-0 2dNh2 1us : hrtimer_expire_exit: hrtimer=ffff88007796feb8
2065 <idle>-0 2.N.2 2us : power_end: cpu_id=2
2066 <idle>-0 2.N.2 3us : cpu_idle: state=4294967295 cpu_id=2
2067 <idle>-0 2dN.3 4us : hrtimer_cancel: hrtimer=ffff88007d50d5e0
2068 <idle>-0 2dN.3 4us : hrtimer_start: hrtimer=ffff88007d50d5e0 function=tick_sched_timer expires=34311211000000 softexpires=34311211000000
2069 <idle>-0 2.N.2 5us : rcu_utilization: Start context switch
2070 <idle>-0 2.N.2 5us : rcu_utilization: End context switch
2071 <idle>-0 2d..3 6us : __schedule <-schedule
2072 <idle>-0 2d..3 6us : 0:120:R ==> [002] 5882: 94:R sleep
2075 Hardware Latency Detector
2076 -------------------------
2078 The hardware latency detector is executed by enabling the "hwlat" tracer.
2080 NOTE, this tracer will affect the performance of the system as it will
2081 periodically make a CPU constantly busy with interrupts disabled.
2084 # echo hwlat > current_tracer
2090 # / _----=> need-resched
2091 # | / _---=> hardirq/softirq
2092 # || / _--=> preempt-depth
2094 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2096 <...>-3638 [001] d... 19452.055471: #1 inner/outer(us): 12/14 ts:1499801089.066141940
2097 <...>-3638 [003] d... 19454.071354: #2 inner/outer(us): 11/9 ts:1499801091.082164365
2098 <...>-3638 [002] dn.. 19461.126852: #3 inner/outer(us): 12/9 ts:1499801098.138150062
2099 <...>-3638 [001] d... 19488.340960: #4 inner/outer(us): 8/12 ts:1499801125.354139633
2100 <...>-3638 [003] d... 19494.388553: #5 inner/outer(us): 8/12 ts:1499801131.402150961
2101 <...>-3638 [003] d... 19501.283419: #6 inner/outer(us): 0/12 ts:1499801138.297435289 nmi-total:4 nmi-count:1
2104 The above output is somewhat the same in the header. All events will have
2105 interrupts disabled 'd'. Under the FUNCTION title there is:
2108 This is the count of events recorded that were greater than the
2109 tracing_threshold (See below).
2111 inner/outer(us): 12/14
2113 This shows two numbers as "inner latency" and "outer latency". The test
2114 runs in a loop checking a timestamp twice. The latency detected within
2115 the two timestamps is the "inner latency" and the latency detected
2116 after the previous timestamp and the next timestamp in the loop is
2117 the "outer latency".
2119 ts:1499801089.066141940
2121 The absolute timestamp that the event happened.
2123 nmi-total:4 nmi-count:1
2125 On architectures that support it, if an NMI comes in during the
2126 test, the time spent in NMI is reported in "nmi-total" (in
2129 All architectures that have NMIs will show the "nmi-count" if an
2130 NMI comes in during the test.
2135 This gets automatically set to "10" to represent 10
2136 microseconds. This is the threshold of latency that
2137 needs to be detected before the trace will be recorded.
2139 Note, when hwlat tracer is finished (another tracer is
2140 written into "current_tracer"), the original value for
2141 tracing_threshold is placed back into this file.
2143 hwlat_detector/width
2144 The length of time the test runs with interrupts disabled.
2146 hwlat_detector/window
2147 The length of time of the window which the test
2148 runs. That is, the test will run for "width"
2149 microseconds per "window" microseconds
2152 When the test is started. A kernel thread is created that
2153 runs the test. This thread will alternate between CPUs
2154 listed in the tracing_cpumask between each period
2155 (one "window"). To limit the test to specific CPUs
2156 set the mask in this file to only the CPUs that the test
2162 This tracer is the function tracer. Enabling the function tracer
2163 can be done from the debug file system. Make sure the
2164 ftrace_enabled is set; otherwise this tracer is a nop.
2165 See the "ftrace_enabled" section below.
2168 # sysctl kernel.ftrace_enabled=1
2169 # echo function > current_tracer
2170 # echo 1 > tracing_on
2172 # echo 0 > tracing_on
2176 # entries-in-buffer/entries-written: 24799/24799 #P:4
2179 # / _----=> need-resched
2180 # | / _---=> hardirq/softirq
2181 # || / _--=> preempt-depth
2183 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2185 bash-1994 [002] .... 3082.063030: mutex_unlock <-rb_simple_write
2186 bash-1994 [002] .... 3082.063031: __mutex_unlock_slowpath <-mutex_unlock
2187 bash-1994 [002] .... 3082.063031: __fsnotify_parent <-fsnotify_modify
2188 bash-1994 [002] .... 3082.063032: fsnotify <-fsnotify_modify
2189 bash-1994 [002] .... 3082.063032: __srcu_read_lock <-fsnotify
2190 bash-1994 [002] .... 3082.063032: add_preempt_count <-__srcu_read_lock
2191 bash-1994 [002] ...1 3082.063032: sub_preempt_count <-__srcu_read_lock
2192 bash-1994 [002] .... 3082.063033: __srcu_read_unlock <-fsnotify
2196 Note: function tracer uses ring buffers to store the above
2197 entries. The newest data may overwrite the oldest data.
2198 Sometimes using echo to stop the trace is not sufficient because
2199 the tracing could have overwritten the data that you wanted to
2200 record. For this reason, it is sometimes better to disable
2201 tracing directly from a program. This allows you to stop the
2202 tracing at the point that you hit the part that you are
2203 interested in. To disable the tracing directly from a C program,
2204 something like following code snippet can be used::
2208 int main(int argc, char *argv[]) {
2210 trace_fd = open(tracing_file("tracing_on"), O_WRONLY);
2212 if (condition_hit()) {
2213 write(trace_fd, "0", 1);
2219 Single thread tracing
2220 ---------------------
2222 By writing into set_ftrace_pid you can trace a
2223 single thread. For example::
2225 # cat set_ftrace_pid
2227 # echo 3111 > set_ftrace_pid
2228 # cat set_ftrace_pid
2230 # echo function > current_tracer
2234 # TASK-PID CPU# TIMESTAMP FUNCTION
2236 yum-updatesd-3111 [003] 1637.254676: finish_task_switch <-thread_return
2237 yum-updatesd-3111 [003] 1637.254681: hrtimer_cancel <-schedule_hrtimeout_range
2238 yum-updatesd-3111 [003] 1637.254682: hrtimer_try_to_cancel <-hrtimer_cancel
2239 yum-updatesd-3111 [003] 1637.254683: lock_hrtimer_base <-hrtimer_try_to_cancel
2240 yum-updatesd-3111 [003] 1637.254685: fget_light <-do_sys_poll
2241 yum-updatesd-3111 [003] 1637.254686: pipe_poll <-do_sys_poll
2242 # echo > set_ftrace_pid
2246 # TASK-PID CPU# TIMESTAMP FUNCTION
2248 ##### CPU 3 buffer started ####
2249 yum-updatesd-3111 [003] 1701.957688: free_poll_entry <-poll_freewait
2250 yum-updatesd-3111 [003] 1701.957689: remove_wait_queue <-free_poll_entry
2251 yum-updatesd-3111 [003] 1701.957691: fput <-free_poll_entry
2252 yum-updatesd-3111 [003] 1701.957692: audit_syscall_exit <-sysret_audit
2253 yum-updatesd-3111 [003] 1701.957693: path_put <-audit_syscall_exit
2255 If you want to trace a function when executing, you could use
2256 something like this simple program.
2261 #include <sys/types.h>
2262 #include <sys/stat.h>
2268 #define STR(x) _STR(x)
2269 #define MAX_PATH 256
2271 const char *find_tracefs(void)
2273 static char tracefs[MAX_PATH+1];
2274 static int tracefs_found;
2281 if ((fp = fopen("/proc/mounts","r")) == NULL) {
2282 perror("/proc/mounts");
2286 while (fscanf(fp, "%*s %"
2288 "s %99s %*s %*d %*d\n",
2289 tracefs, type) == 2) {
2290 if (strcmp(type, "tracefs") == 0)
2295 if (strcmp(type, "tracefs") != 0) {
2296 fprintf(stderr, "tracefs not mounted");
2300 strcat(tracefs, "/tracing/");
2306 const char *tracing_file(const char *file_name)
2308 static char trace_file[MAX_PATH+1];
2309 snprintf(trace_file, MAX_PATH, "%s/%s", find_tracefs(), file_name);
2313 int main (int argc, char **argv)
2323 ffd = open(tracing_file("current_tracer"), O_WRONLY);
2326 write(ffd, "nop", 3);
2328 fd = open(tracing_file("set_ftrace_pid"), O_WRONLY);
2329 s = sprintf(line, "%d\n", getpid());
2332 write(ffd, "function", 8);
2337 execvp(argv[1], argv+1);
2343 Or this simple script!
2348 tracefs=`sed -ne 's/^tracefs \(.*\) tracefs.*/\1/p' /proc/mounts`
2349 echo nop > $tracefs/tracing/current_tracer
2350 echo 0 > $tracefs/tracing/tracing_on
2351 echo $$ > $tracefs/tracing/set_ftrace_pid
2352 echo function > $tracefs/tracing/current_tracer
2353 echo 1 > $tracefs/tracing/tracing_on
2357 function graph tracer
2358 ---------------------------
2360 This tracer is similar to the function tracer except that it
2361 probes a function on its entry and its exit. This is done by
2362 using a dynamically allocated stack of return addresses in each
2363 task_struct. On function entry the tracer overwrites the return
2364 address of each function traced to set a custom probe. Thus the
2365 original return address is stored on the stack of return address
2368 Probing on both ends of a function leads to special features
2371 - measure of a function's time execution
2372 - having a reliable call stack to draw function calls graph
2374 This tracer is useful in several situations:
2376 - you want to find the reason of a strange kernel behavior and
2377 need to see what happens in detail on any areas (or specific
2380 - you are experiencing weird latencies but it's difficult to
2383 - you want to find quickly which path is taken by a specific
2386 - you just want to peek inside a working kernel and want to see
2391 # tracer: function_graph
2393 # CPU DURATION FUNCTION CALLS
2397 0) | do_sys_open() {
2399 0) | kmem_cache_alloc() {
2400 0) 1.382 us | __might_sleep();
2402 0) | strncpy_from_user() {
2403 0) | might_fault() {
2404 0) 1.389 us | __might_sleep();
2409 0) 0.668 us | _spin_lock();
2410 0) 0.570 us | expand_files();
2411 0) 0.586 us | _spin_unlock();
2414 There are several columns that can be dynamically
2415 enabled/disabled. You can use every combination of options you
2416 want, depending on your needs.
2418 - The cpu number on which the function executed is default
2419 enabled. It is sometimes better to only trace one cpu (see
2420 tracing_cpu_mask file) or you might sometimes see unordered
2421 function calls while cpu tracing switch.
2423 - hide: echo nofuncgraph-cpu > trace_options
2424 - show: echo funcgraph-cpu > trace_options
2426 - The duration (function's time of execution) is displayed on
2427 the closing bracket line of a function or on the same line
2428 than the current function in case of a leaf one. It is default
2431 - hide: echo nofuncgraph-duration > trace_options
2432 - show: echo funcgraph-duration > trace_options
2434 - The overhead field precedes the duration field in case of
2435 reached duration thresholds.
2437 - hide: echo nofuncgraph-overhead > trace_options
2438 - show: echo funcgraph-overhead > trace_options
2439 - depends on: funcgraph-duration
2443 3) # 1837.709 us | } /* __switch_to */
2444 3) | finish_task_switch() {
2445 3) 0.313 us | _raw_spin_unlock_irq();
2447 3) # 1889.063 us | } /* __schedule */
2448 3) ! 140.417 us | } /* __schedule */
2449 3) # 2034.948 us | } /* schedule */
2450 3) * 33998.59 us | } /* schedule_preempt_disabled */
2454 1) 0.260 us | msecs_to_jiffies();
2455 1) 0.313 us | __rcu_read_unlock();
2458 1) 0.313 us | rcu_bh_qs();
2459 1) 0.313 us | __local_bh_enable();
2461 1) 0.365 us | idle_cpu();
2462 1) | rcu_irq_exit() {
2463 1) 0.417 us | rcu_eqs_enter_common.isra.47();
2467 1) @ 119760.2 us | }
2473 2) 0.417 us | scheduler_ipi();
2483 + means that the function exceeded 10 usecs.
2484 ! means that the function exceeded 100 usecs.
2485 # means that the function exceeded 1000 usecs.
2486 * means that the function exceeded 10 msecs.
2487 @ means that the function exceeded 100 msecs.
2488 $ means that the function exceeded 1 sec.
2491 - The task/pid field displays the thread cmdline and pid which
2492 executed the function. It is default disabled.
2494 - hide: echo nofuncgraph-proc > trace_options
2495 - show: echo funcgraph-proc > trace_options
2499 # tracer: function_graph
2501 # CPU TASK/PID DURATION FUNCTION CALLS
2503 0) sh-4802 | | d_free() {
2504 0) sh-4802 | | call_rcu() {
2505 0) sh-4802 | | __call_rcu() {
2506 0) sh-4802 | 0.616 us | rcu_process_gp_end();
2507 0) sh-4802 | 0.586 us | check_for_new_grace_period();
2508 0) sh-4802 | 2.899 us | }
2509 0) sh-4802 | 4.040 us | }
2510 0) sh-4802 | 5.151 us | }
2511 0) sh-4802 | + 49.370 us | }
2514 - The absolute time field is an absolute timestamp given by the
2515 system clock since it started. A snapshot of this time is
2516 given on each entry/exit of functions
2518 - hide: echo nofuncgraph-abstime > trace_options
2519 - show: echo funcgraph-abstime > trace_options
2524 # TIME CPU DURATION FUNCTION CALLS
2526 360.774522 | 1) 0.541 us | }
2527 360.774522 | 1) 4.663 us | }
2528 360.774523 | 1) 0.541 us | __wake_up_bit();
2529 360.774524 | 1) 6.796 us | }
2530 360.774524 | 1) 7.952 us | }
2531 360.774525 | 1) 9.063 us | }
2532 360.774525 | 1) 0.615 us | journal_mark_dirty();
2533 360.774527 | 1) 0.578 us | __brelse();
2534 360.774528 | 1) | reiserfs_prepare_for_journal() {
2535 360.774528 | 1) | unlock_buffer() {
2536 360.774529 | 1) | wake_up_bit() {
2537 360.774529 | 1) | bit_waitqueue() {
2538 360.774530 | 1) 0.594 us | __phys_addr();
2541 The function name is always displayed after the closing bracket
2542 for a function if the start of that function is not in the
2545 Display of the function name after the closing bracket may be
2546 enabled for functions whose start is in the trace buffer,
2547 allowing easier searching with grep for function durations.
2548 It is default disabled.
2550 - hide: echo nofuncgraph-tail > trace_options
2551 - show: echo funcgraph-tail > trace_options
2553 Example with nofuncgraph-tail (default)::
2556 0) | kmem_cache_free() {
2557 0) 0.518 us | __phys_addr();
2561 Example with funcgraph-tail::
2564 0) | kmem_cache_free() {
2565 0) 0.518 us | __phys_addr();
2566 0) 1.757 us | } /* kmem_cache_free() */
2567 0) 2.861 us | } /* putname() */
2569 You can put some comments on specific functions by using
2570 trace_printk() For example, if you want to put a comment inside
2571 the __might_sleep() function, you just have to include
2572 <linux/ftrace.h> and call trace_printk() inside __might_sleep()::
2574 trace_printk("I'm a comment!\n")
2578 1) | __might_sleep() {
2579 1) | /* I'm a comment! */
2583 You might find other useful features for this tracer in the
2584 following "dynamic ftrace" section such as tracing only specific
2590 If CONFIG_DYNAMIC_FTRACE is set, the system will run with
2591 virtually no overhead when function tracing is disabled. The way
2592 this works is the mcount function call (placed at the start of
2593 every kernel function, produced by the -pg switch in gcc),
2594 starts of pointing to a simple return. (Enabling FTRACE will
2595 include the -pg switch in the compiling of the kernel.)
2597 At compile time every C file object is run through the
2598 recordmcount program (located in the scripts directory). This
2599 program will parse the ELF headers in the C object to find all
2600 the locations in the .text section that call mcount. Starting
2601 with gcc version 4.6, the -mfentry has been added for x86, which
2602 calls "__fentry__" instead of "mcount". Which is called before
2603 the creation of the stack frame.
2605 Note, not all sections are traced. They may be prevented by either
2606 a notrace, or blocked another way and all inline functions are not
2607 traced. Check the "available_filter_functions" file to see what functions
2610 A section called "__mcount_loc" is created that holds
2611 references to all the mcount/fentry call sites in the .text section.
2612 The recordmcount program re-links this section back into the
2613 original object. The final linking stage of the kernel will add all these
2614 references into a single table.
2616 On boot up, before SMP is initialized, the dynamic ftrace code
2617 scans this table and updates all the locations into nops. It
2618 also records the locations, which are added to the
2619 available_filter_functions list. Modules are processed as they
2620 are loaded and before they are executed. When a module is
2621 unloaded, it also removes its functions from the ftrace function
2622 list. This is automatic in the module unload code, and the
2623 module author does not need to worry about it.
2625 When tracing is enabled, the process of modifying the function
2626 tracepoints is dependent on architecture. The old method is to use
2627 kstop_machine to prevent races with the CPUs executing code being
2628 modified (which can cause the CPU to do undesirable things, especially
2629 if the modified code crosses cache (or page) boundaries), and the nops are
2630 patched back to calls. But this time, they do not call mcount
2631 (which is just a function stub). They now call into the ftrace
2634 The new method of modifying the function tracepoints is to place
2635 a breakpoint at the location to be modified, sync all CPUs, modify
2636 the rest of the instruction not covered by the breakpoint. Sync
2637 all CPUs again, and then remove the breakpoint with the finished
2638 version to the ftrace call site.
2640 Some archs do not even need to monkey around with the synchronization,
2641 and can just slap the new code on top of the old without any
2642 problems with other CPUs executing it at the same time.
2644 One special side-effect to the recording of the functions being
2645 traced is that we can now selectively choose which functions we
2646 wish to trace and which ones we want the mcount calls to remain
2649 Two files are used, one for enabling and one for disabling the
2650 tracing of specified functions. They are:
2658 A list of available functions that you can add to these files is
2661 available_filter_functions
2665 # cat available_filter_functions
2674 If I am only interested in sys_nanosleep and hrtimer_interrupt::
2676 # echo sys_nanosleep hrtimer_interrupt > set_ftrace_filter
2677 # echo function > current_tracer
2678 # echo 1 > tracing_on
2680 # echo 0 > tracing_on
2684 # entries-in-buffer/entries-written: 5/5 #P:4
2687 # / _----=> need-resched
2688 # | / _---=> hardirq/softirq
2689 # || / _--=> preempt-depth
2691 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2693 usleep-2665 [001] .... 4186.475355: sys_nanosleep <-system_call_fastpath
2694 <idle>-0 [001] d.h1 4186.475409: hrtimer_interrupt <-smp_apic_timer_interrupt
2695 usleep-2665 [001] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt
2696 <idle>-0 [003] d.h1 4186.475426: hrtimer_interrupt <-smp_apic_timer_interrupt
2697 <idle>-0 [002] d.h1 4186.475427: hrtimer_interrupt <-smp_apic_timer_interrupt
2699 To see which functions are being traced, you can cat the file:
2702 # cat set_ftrace_filter
2707 Perhaps this is not enough. The filters also allow glob(7) matching.
2710 will match functions that begin with <match>
2712 will match functions that end with <match>
2714 will match functions that have <match> in it
2715 ``<match1>*<match2>``
2716 will match functions that begin with <match1> and end with <match2>
2719 It is better to use quotes to enclose the wild cards,
2720 otherwise the shell may expand the parameters into names
2721 of files in the local directory.
2725 # echo 'hrtimer_*' > set_ftrace_filter
2731 # entries-in-buffer/entries-written: 897/897 #P:4
2734 # / _----=> need-resched
2735 # | / _---=> hardirq/softirq
2736 # || / _--=> preempt-depth
2738 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2740 <idle>-0 [003] dN.1 4228.547803: hrtimer_cancel <-tick_nohz_idle_exit
2741 <idle>-0 [003] dN.1 4228.547804: hrtimer_try_to_cancel <-hrtimer_cancel
2742 <idle>-0 [003] dN.2 4228.547805: hrtimer_force_reprogram <-__remove_hrtimer
2743 <idle>-0 [003] dN.1 4228.547805: hrtimer_forward <-tick_nohz_idle_exit
2744 <idle>-0 [003] dN.1 4228.547805: hrtimer_start_range_ns <-hrtimer_start_expires.constprop.11
2745 <idle>-0 [003] d..1 4228.547858: hrtimer_get_next_event <-get_next_timer_interrupt
2746 <idle>-0 [003] d..1 4228.547859: hrtimer_start <-__tick_nohz_idle_enter
2747 <idle>-0 [003] d..2 4228.547860: hrtimer_force_reprogram <-__rem
2749 Notice that we lost the sys_nanosleep.
2752 # cat set_ftrace_filter
2757 hrtimer_try_to_cancel
2761 hrtimer_force_reprogram
2762 hrtimer_get_next_event
2766 hrtimer_get_remaining
2768 hrtimer_init_sleeper
2771 This is because the '>' and '>>' act just like they do in bash.
2772 To rewrite the filters, use '>'
2773 To append to the filters, use '>>'
2775 To clear out a filter so that all functions will be recorded
2778 # echo > set_ftrace_filter
2779 # cat set_ftrace_filter
2782 Again, now we want to append.
2786 # echo sys_nanosleep > set_ftrace_filter
2787 # cat set_ftrace_filter
2789 # echo 'hrtimer_*' >> set_ftrace_filter
2790 # cat set_ftrace_filter
2795 hrtimer_try_to_cancel
2799 hrtimer_force_reprogram
2800 hrtimer_get_next_event
2805 hrtimer_get_remaining
2807 hrtimer_init_sleeper
2810 The set_ftrace_notrace prevents those functions from being
2814 # echo '*preempt*' '*lock*' > set_ftrace_notrace
2820 # entries-in-buffer/entries-written: 39608/39608 #P:4
2823 # / _----=> need-resched
2824 # | / _---=> hardirq/softirq
2825 # || / _--=> preempt-depth
2827 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
2829 bash-1994 [000] .... 4342.324896: file_ra_state_init <-do_dentry_open
2830 bash-1994 [000] .... 4342.324897: open_check_o_direct <-do_last
2831 bash-1994 [000] .... 4342.324897: ima_file_check <-do_last
2832 bash-1994 [000] .... 4342.324898: process_measurement <-ima_file_check
2833 bash-1994 [000] .... 4342.324898: ima_get_action <-process_measurement
2834 bash-1994 [000] .... 4342.324898: ima_match_policy <-ima_get_action
2835 bash-1994 [000] .... 4342.324899: do_truncate <-do_last
2836 bash-1994 [000] .... 4342.324899: should_remove_suid <-do_truncate
2837 bash-1994 [000] .... 4342.324899: notify_change <-do_truncate
2838 bash-1994 [000] .... 4342.324900: current_fs_time <-notify_change
2839 bash-1994 [000] .... 4342.324900: current_kernel_time <-current_fs_time
2840 bash-1994 [000] .... 4342.324900: timespec_trunc <-current_fs_time
2842 We can see that there's no more lock or preempt tracing.
2844 Selecting function filters via index
2845 ------------------------------------
2847 Because processing of strings is expensive (the address of the function
2848 needs to be looked up before comparing to the string being passed in),
2849 an index can be used as well to enable functions. This is useful in the
2850 case of setting thousands of specific functions at a time. By passing
2851 in a list of numbers, no string processing will occur. Instead, the function
2852 at the specific location in the internal array (which corresponds to the
2853 functions in the "available_filter_functions" file), is selected.
2857 # echo 1 > set_ftrace_filter
2859 Will select the first function listed in "available_filter_functions"
2863 # head -1 available_filter_functions
2864 trace_initcall_finish_cb
2866 # cat set_ftrace_filter
2867 trace_initcall_finish_cb
2869 # head -50 available_filter_functions | tail -1
2872 # echo 1 50 > set_ftrace_filter
2873 # cat set_ftrace_filter
2874 trace_initcall_finish_cb
2877 Dynamic ftrace with the function graph tracer
2878 ---------------------------------------------
2880 Although what has been explained above concerns both the
2881 function tracer and the function-graph-tracer, there are some
2882 special features only available in the function-graph tracer.
2884 If you want to trace only one function and all of its children,
2885 you just have to echo its name into set_graph_function::
2887 echo __do_fault > set_graph_function
2889 will produce the following "expanded" trace of the __do_fault()
2893 0) | filemap_fault() {
2894 0) | find_lock_page() {
2895 0) 0.804 us | find_get_page();
2896 0) | __might_sleep() {
2900 0) 0.653 us | _spin_lock();
2901 0) 0.578 us | page_add_file_rmap();
2902 0) 0.525 us | native_set_pte_at();
2903 0) 0.585 us | _spin_unlock();
2904 0) | unlock_page() {
2905 0) 0.541 us | page_waitqueue();
2906 0) 0.639 us | __wake_up_bit();
2910 0) | filemap_fault() {
2911 0) | find_lock_page() {
2912 0) 0.698 us | find_get_page();
2913 0) | __might_sleep() {
2917 0) 0.631 us | _spin_lock();
2918 0) 0.571 us | page_add_file_rmap();
2919 0) 0.526 us | native_set_pte_at();
2920 0) 0.586 us | _spin_unlock();
2921 0) | unlock_page() {
2922 0) 0.533 us | page_waitqueue();
2923 0) 0.638 us | __wake_up_bit();
2927 You can also expand several functions at once::
2929 echo sys_open > set_graph_function
2930 echo sys_close >> set_graph_function
2932 Now if you want to go back to trace all functions you can clear
2933 this special filter via::
2935 echo > set_graph_function
2941 Note, the proc sysctl ftrace_enable is a big on/off switch for the
2942 function tracer. By default it is enabled (when function tracing is
2943 enabled in the kernel). If it is disabled, all function tracing is
2944 disabled. This includes not only the function tracers for ftrace, but
2945 also for any other uses (perf, kprobes, stack tracing, profiling, etc).
2947 Please disable this with care.
2949 This can be disable (and enabled) with::
2951 sysctl kernel.ftrace_enabled=0
2952 sysctl kernel.ftrace_enabled=1
2956 echo 0 > /proc/sys/kernel/ftrace_enabled
2957 echo 1 > /proc/sys/kernel/ftrace_enabled
2963 A few commands are supported by the set_ftrace_filter interface.
2964 Trace commands have the following format::
2966 <function>:<command>:<parameter>
2968 The following commands are supported:
2971 This command enables function filtering per module. The
2972 parameter defines the module. For example, if only the write*
2973 functions in the ext3 module are desired, run:
2975 echo 'write*:mod:ext3' > set_ftrace_filter
2977 This command interacts with the filter in the same way as
2978 filtering based on function names. Thus, adding more functions
2979 in a different module is accomplished by appending (>>) to the
2980 filter file. Remove specific module functions by prepending
2983 echo '!writeback*:mod:ext3' >> set_ftrace_filter
2985 Mod command supports module globbing. Disable tracing for all
2986 functions except a specific module::
2988 echo '!*:mod:!ext3' >> set_ftrace_filter
2990 Disable tracing for all modules, but still trace kernel::
2992 echo '!*:mod:*' >> set_ftrace_filter
2994 Enable filter only for kernel::
2996 echo '*write*:mod:!*' >> set_ftrace_filter
2998 Enable filter for module globbing::
3000 echo '*write*:mod:*snd*' >> set_ftrace_filter
3003 These commands turn tracing on and off when the specified
3004 functions are hit. The parameter determines how many times the
3005 tracing system is turned on and off. If unspecified, there is
3006 no limit. For example, to disable tracing when a schedule bug
3007 is hit the first 5 times, run::
3009 echo '__schedule_bug:traceoff:5' > set_ftrace_filter
3011 To always disable tracing when __schedule_bug is hit::
3013 echo '__schedule_bug:traceoff' > set_ftrace_filter
3015 These commands are cumulative whether or not they are appended
3016 to set_ftrace_filter. To remove a command, prepend it by '!'
3017 and drop the parameter::
3019 echo '!__schedule_bug:traceoff:0' > set_ftrace_filter
3021 The above removes the traceoff command for __schedule_bug
3022 that have a counter. To remove commands without counters::
3024 echo '!__schedule_bug:traceoff' > set_ftrace_filter
3027 Will cause a snapshot to be triggered when the function is hit.
3030 echo 'native_flush_tlb_others:snapshot' > set_ftrace_filter
3032 To only snapshot once:
3035 echo 'native_flush_tlb_others:snapshot:1' > set_ftrace_filter
3037 To remove the above commands::
3039 echo '!native_flush_tlb_others:snapshot' > set_ftrace_filter
3040 echo '!native_flush_tlb_others:snapshot:0' > set_ftrace_filter
3042 - enable_event/disable_event:
3043 These commands can enable or disable a trace event. Note, because
3044 function tracing callbacks are very sensitive, when these commands
3045 are registered, the trace point is activated, but disabled in
3046 a "soft" mode. That is, the tracepoint will be called, but
3047 just will not be traced. The event tracepoint stays in this mode
3048 as long as there's a command that triggers it.
3051 echo 'try_to_wake_up:enable_event:sched:sched_switch:2' > \
3056 <function>:enable_event:<system>:<event>[:count]
3057 <function>:disable_event:<system>:<event>[:count]
3059 To remove the events commands::
3061 echo '!try_to_wake_up:enable_event:sched:sched_switch:0' > \
3063 echo '!schedule:disable_event:sched:sched_switch' > \
3067 When the function is hit, it will dump the contents of the ftrace
3068 ring buffer to the console. This is useful if you need to debug
3069 something, and want to dump the trace when a certain function
3070 is hit. Perhaps it's a function that is called before a triple
3071 fault happens and does not allow you to get a regular dump.
3074 When the function is hit, it will dump the contents of the ftrace
3075 ring buffer for the current CPU to the console. Unlike the "dump"
3076 command, it only prints out the contents of the ring buffer for the
3077 CPU that executed the function that triggered the dump.
3080 When the function is hit, a stack trace is recorded.
3085 The trace_pipe outputs the same content as the trace file, but
3086 the effect on the tracing is different. Every read from
3087 trace_pipe is consumed. This means that subsequent reads will be
3088 different. The trace is live.
3091 # echo function > current_tracer
3092 # cat trace_pipe > /tmp/trace.out &
3094 # echo 1 > tracing_on
3096 # echo 0 > tracing_on
3100 # entries-in-buffer/entries-written: 0/0 #P:4
3103 # / _----=> need-resched
3104 # | / _---=> hardirq/softirq
3105 # || / _--=> preempt-depth
3107 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3111 # cat /tmp/trace.out
3112 bash-1994 [000] .... 5281.568961: mutex_unlock <-rb_simple_write
3113 bash-1994 [000] .... 5281.568963: __mutex_unlock_slowpath <-mutex_unlock
3114 bash-1994 [000] .... 5281.568963: __fsnotify_parent <-fsnotify_modify
3115 bash-1994 [000] .... 5281.568964: fsnotify <-fsnotify_modify
3116 bash-1994 [000] .... 5281.568964: __srcu_read_lock <-fsnotify
3117 bash-1994 [000] .... 5281.568964: add_preempt_count <-__srcu_read_lock
3118 bash-1994 [000] ...1 5281.568965: sub_preempt_count <-__srcu_read_lock
3119 bash-1994 [000] .... 5281.568965: __srcu_read_unlock <-fsnotify
3120 bash-1994 [000] .... 5281.568967: sys_dup2 <-system_call_fastpath
3123 Note, reading the trace_pipe file will block until more input is
3129 Having too much or not enough data can be troublesome in
3130 diagnosing an issue in the kernel. The file buffer_size_kb is
3131 used to modify the size of the internal trace buffers. The
3132 number listed is the number of entries that can be recorded per
3133 CPU. To know the full size, multiply the number of possible CPUs
3134 with the number of entries.
3137 # cat buffer_size_kb
3138 1408 (units kilobytes)
3140 Or simply read buffer_total_size_kb
3143 # cat buffer_total_size_kb
3146 To modify the buffer, simple echo in a number (in 1024 byte segments).
3149 # echo 10000 > buffer_size_kb
3150 # cat buffer_size_kb
3151 10000 (units kilobytes)
3153 It will try to allocate as much as possible. If you allocate too
3154 much, it can cause Out-Of-Memory to trigger.
3157 # echo 1000000000000 > buffer_size_kb
3158 -bash: echo: write error: Cannot allocate memory
3159 # cat buffer_size_kb
3162 The per_cpu buffers can be changed individually as well:
3165 # echo 10000 > per_cpu/cpu0/buffer_size_kb
3166 # echo 100 > per_cpu/cpu1/buffer_size_kb
3168 When the per_cpu buffers are not the same, the buffer_size_kb
3169 at the top level will just show an X
3172 # cat buffer_size_kb
3175 This is where the buffer_total_size_kb is useful:
3178 # cat buffer_total_size_kb
3181 Writing to the top level buffer_size_kb will reset all the buffers
3182 to be the same again.
3186 CONFIG_TRACER_SNAPSHOT makes a generic snapshot feature
3187 available to all non latency tracers. (Latency tracers which
3188 record max latency, such as "irqsoff" or "wakeup", can't use
3189 this feature, since those are already using the snapshot
3190 mechanism internally.)
3192 Snapshot preserves a current trace buffer at a particular point
3193 in time without stopping tracing. Ftrace swaps the current
3194 buffer with a spare buffer, and tracing continues in the new
3195 current (=previous spare) buffer.
3197 The following tracefs files in "tracing" are related to this
3202 This is used to take a snapshot and to read the output
3203 of the snapshot. Echo 1 into this file to allocate a
3204 spare buffer and to take a snapshot (swap), then read
3205 the snapshot from this file in the same format as
3206 "trace" (described above in the section "The File
3207 System"). Both reads snapshot and tracing are executable
3208 in parallel. When the spare buffer is allocated, echoing
3209 0 frees it, and echoing else (positive) values clear the
3211 More details are shown in the table below.
3213 +--------------+------------+------------+------------+
3214 |status\\input | 0 | 1 | else |
3215 +==============+============+============+============+
3216 |not allocated |(do nothing)| alloc+swap |(do nothing)|
3217 +--------------+------------+------------+------------+
3218 |allocated | free | swap | clear |
3219 +--------------+------------+------------+------------+
3221 Here is an example of using the snapshot feature.
3224 # echo 1 > events/sched/enable
3229 # entries-in-buffer/entries-written: 71/71 #P:8
3232 # / _----=> need-resched
3233 # | / _---=> hardirq/softirq
3234 # || / _--=> preempt-depth
3236 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3238 <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
3239 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
3241 <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
3246 # entries-in-buffer/entries-written: 77/77 #P:8
3249 # / _----=> need-resched
3250 # | / _---=> hardirq/softirq
3251 # || / _--=> preempt-depth
3253 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3255 <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
3256 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
3260 If you try to use this snapshot feature when current tracer is
3261 one of the latency tracers, you will get the following results.
3264 # echo wakeup > current_tracer
3266 bash: echo: write error: Device or resource busy
3268 cat: snapshot: Device or resource busy
3273 In the tracefs tracing directory is a directory called "instances".
3274 This directory can have new directories created inside of it using
3275 mkdir, and removing directories with rmdir. The directory created
3276 with mkdir in this directory will already contain files and other
3277 directories after it is created.
3280 # mkdir instances/foo
3282 buffer_size_kb buffer_total_size_kb events free_buffer per_cpu
3283 set_event snapshot trace trace_clock trace_marker trace_options
3284 trace_pipe tracing_on
3286 As you can see, the new directory looks similar to the tracing directory
3287 itself. In fact, it is very similar, except that the buffer and
3288 events are agnostic from the main director, or from any other
3289 instances that are created.
3291 The files in the new directory work just like the files with the
3292 same name in the tracing directory except the buffer that is used
3293 is a separate and new buffer. The files affect that buffer but do not
3294 affect the main buffer with the exception of trace_options. Currently,
3295 the trace_options affect all instances and the top level buffer
3296 the same, but this may change in future releases. That is, options
3297 may become specific to the instance they reside in.
3299 Notice that none of the function tracer files are there, nor is
3300 current_tracer and available_tracers. This is because the buffers
3301 can currently only have events enabled for them.
3304 # mkdir instances/foo
3305 # mkdir instances/bar
3306 # mkdir instances/zoot
3307 # echo 100000 > buffer_size_kb
3308 # echo 1000 > instances/foo/buffer_size_kb
3309 # echo 5000 > instances/bar/per_cpu/cpu1/buffer_size_kb
3310 # echo function > current_trace
3311 # echo 1 > instances/foo/events/sched/sched_wakeup/enable
3312 # echo 1 > instances/foo/events/sched/sched_wakeup_new/enable
3313 # echo 1 > instances/foo/events/sched/sched_switch/enable
3314 # echo 1 > instances/bar/events/irq/enable
3315 # echo 1 > instances/zoot/events/syscalls/enable
3317 CPU:2 [LOST 11745 EVENTS]
3318 bash-2044 [002] .... 10594.481032: _raw_spin_lock_irqsave <-get_page_from_freelist
3319 bash-2044 [002] d... 10594.481032: add_preempt_count <-_raw_spin_lock_irqsave
3320 bash-2044 [002] d..1 10594.481032: __rmqueue <-get_page_from_freelist
3321 bash-2044 [002] d..1 10594.481033: _raw_spin_unlock <-get_page_from_freelist
3322 bash-2044 [002] d..1 10594.481033: sub_preempt_count <-_raw_spin_unlock
3323 bash-2044 [002] d... 10594.481033: get_pageblock_flags_group <-get_pageblock_migratetype
3324 bash-2044 [002] d... 10594.481034: __mod_zone_page_state <-get_page_from_freelist
3325 bash-2044 [002] d... 10594.481034: zone_statistics <-get_page_from_freelist
3326 bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics
3327 bash-2044 [002] d... 10594.481034: __inc_zone_state <-zone_statistics
3328 bash-2044 [002] .... 10594.481035: arch_dup_task_struct <-copy_process
3331 # cat instances/foo/trace_pipe
3332 bash-1998 [000] d..4 136.676759: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000
3333 bash-1998 [000] dN.4 136.676760: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000
3334 <idle>-0 [003] d.h3 136.676906: sched_wakeup: comm=rcu_preempt pid=9 prio=120 success=1 target_cpu=003
3335 <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
3336 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
3337 bash-1998 [000] d..4 136.677014: sched_wakeup: comm=kworker/0:1 pid=59 prio=120 success=1 target_cpu=000
3338 bash-1998 [000] dN.4 136.677016: sched_wakeup: comm=bash pid=1998 prio=120 success=1 target_cpu=000
3339 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
3340 kworker/0:1-59 [000] d..4 136.677022: sched_wakeup: comm=sshd pid=1995 prio=120 success=1 target_cpu=001
3341 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
3344 # cat instances/bar/trace_pipe
3345 migration/1-14 [001] d.h3 138.732674: softirq_raise: vec=3 [action=NET_RX]
3346 <idle>-0 [001] dNh3 138.732725: softirq_raise: vec=3 [action=NET_RX]
3347 bash-1998 [000] d.h1 138.733101: softirq_raise: vec=1 [action=TIMER]
3348 bash-1998 [000] d.h1 138.733102: softirq_raise: vec=9 [action=RCU]
3349 bash-1998 [000] ..s2 138.733105: softirq_entry: vec=1 [action=TIMER]
3350 bash-1998 [000] ..s2 138.733106: softirq_exit: vec=1 [action=TIMER]
3351 bash-1998 [000] ..s2 138.733106: softirq_entry: vec=9 [action=RCU]
3352 bash-1998 [000] ..s2 138.733109: softirq_exit: vec=9 [action=RCU]
3353 sshd-1995 [001] d.h1 138.733278: irq_handler_entry: irq=21 name=uhci_hcd:usb4
3354 sshd-1995 [001] d.h1 138.733280: irq_handler_exit: irq=21 ret=unhandled
3355 sshd-1995 [001] d.h1 138.733281: irq_handler_entry: irq=21 name=eth0
3356 sshd-1995 [001] d.h1 138.733283: irq_handler_exit: irq=21 ret=handled
3359 # cat instances/zoot/trace
3362 # entries-in-buffer/entries-written: 18996/18996 #P:4
3365 # / _----=> need-resched
3366 # | / _---=> hardirq/softirq
3367 # || / _--=> preempt-depth
3369 # TASK-PID CPU# |||| TIMESTAMP FUNCTION
3371 bash-1998 [000] d... 140.733501: sys_write -> 0x2
3372 bash-1998 [000] d... 140.733504: sys_dup2(oldfd: a, newfd: 1)
3373 bash-1998 [000] d... 140.733506: sys_dup2 -> 0x1
3374 bash-1998 [000] d... 140.733508: sys_fcntl(fd: a, cmd: 1, arg: 0)
3375 bash-1998 [000] d... 140.733509: sys_fcntl -> 0x1
3376 bash-1998 [000] d... 140.733510: sys_close(fd: a)
3377 bash-1998 [000] d... 140.733510: sys_close -> 0x0
3378 bash-1998 [000] d... 140.733514: sys_rt_sigprocmask(how: 0, nset: 0, oset: 6e2768, sigsetsize: 8)
3379 bash-1998 [000] d... 140.733515: sys_rt_sigprocmask -> 0x0
3380 bash-1998 [000] d... 140.733516: sys_rt_sigaction(sig: 2, act: 7fff718846f0, oact: 7fff71884650, sigsetsize: 8)
3381 bash-1998 [000] d... 140.733516: sys_rt_sigaction -> 0x0
3383 You can see that the trace of the top most trace buffer shows only
3384 the function tracing. The foo instance displays wakeups and task
3387 To remove the instances, simply delete their directories:
3390 # rmdir instances/foo
3391 # rmdir instances/bar
3392 # rmdir instances/zoot
3394 Note, if a process has a trace file open in one of the instance
3395 directories, the rmdir will fail with EBUSY.
3400 Since the kernel has a fixed sized stack, it is important not to
3401 waste it in functions. A kernel developer must be conscience of
3402 what they allocate on the stack. If they add too much, the system
3403 can be in danger of a stack overflow, and corruption will occur,
3404 usually leading to a system panic.
3406 There are some tools that check this, usually with interrupts
3407 periodically checking usage. But if you can perform a check
3408 at every function call that will become very useful. As ftrace provides
3409 a function tracer, it makes it convenient to check the stack size
3410 at every function call. This is enabled via the stack tracer.
3412 CONFIG_STACK_TRACER enables the ftrace stack tracing functionality.
3413 To enable it, write a '1' into /proc/sys/kernel/stack_tracer_enabled.
3416 # echo 1 > /proc/sys/kernel/stack_tracer_enabled
3418 You can also enable it from the kernel command line to trace
3419 the stack size of the kernel during boot up, by adding "stacktrace"
3420 to the kernel command line parameter.
3422 After running it for a few minutes, the output looks like:
3425 # cat stack_max_size
3429 Depth Size Location (18 entries)
3431 0) 2928 224 update_sd_lb_stats+0xbc/0x4ac
3432 1) 2704 160 find_busiest_group+0x31/0x1f1
3433 2) 2544 256 load_balance+0xd9/0x662
3434 3) 2288 80 idle_balance+0xbb/0x130
3435 4) 2208 128 __schedule+0x26e/0x5b9
3436 5) 2080 16 schedule+0x64/0x66
3437 6) 2064 128 schedule_timeout+0x34/0xe0
3438 7) 1936 112 wait_for_common+0x97/0xf1
3439 8) 1824 16 wait_for_completion+0x1d/0x1f
3440 9) 1808 128 flush_work+0xfe/0x119
3441 10) 1680 16 tty_flush_to_ldisc+0x1e/0x20
3442 11) 1664 48 input_available_p+0x1d/0x5c
3443 12) 1616 48 n_tty_poll+0x6d/0x134
3444 13) 1568 64 tty_poll+0x64/0x7f
3445 14) 1504 880 do_select+0x31e/0x511
3446 15) 624 400 core_sys_select+0x177/0x216
3447 16) 224 96 sys_select+0x91/0xb9
3448 17) 128 128 system_call_fastpath+0x16/0x1b
3450 Note, if -mfentry is being used by gcc, functions get traced before
3451 they set up the stack frame. This means that leaf level functions
3452 are not tested by the stack tracer when -mfentry is used.
3454 Currently, -mfentry is used by gcc 4.6.0 and above on x86 only.
3458 More details can be found in the source code, in the `kernel/trace/*.c` files.