Documentation/trace/ftrace.rst

27 is really a framework of several assorted tracing utilities.
28 There's latency tracing to examine what occurs between interrupts
32 One of the most common uses of ftrace is the event tracing.
53 option will do) the directory /sys/kernel/tracing will be created. To mount
56  tracefs       /sys/kernel/tracing       tracefs defaults        0       0
60  mount -t tracefs nodev /sys/kernel/tracing
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.
74   /sys/kernel/debug/tracing
83   (cd /sys/kernel/tracing) and will only concentrate on the files within that
85   "/sys/kernel/tracing" path name.
113 	writing to the ring buffer, the tracing overhead may
118 	set this file to "0". User space can re-enable tracing by
122 	set this file to zero and stop tracing. Which can also
130         Note, this file is not a consumer. If tracing is off
132         the same output each time it is read. When tracing is on,
139 	file is meant to be streamed with live tracing.
239 	Note: When changing the sub-buffer size, tracing is stopped and any
244 	If a process is performing tracing, and the ring buffer	should be
248 	Having a process that is tracing also open this file, when the process
252 	It may also stop tracing if disable_on_free option is set.
264 	function profiler (mcount). This lets tracing be configured
279 	and requires a check of all functions registered to tracing, instead
351 	tracing when the function is hit until it exits the function.
352 	This makes it possible to ignore tracing functions that are called
427 	grep ' M ' /sys/kernel/tracing/touched_functions
442 	A directory that holds different tracing stats.
525 	Usual clocks for tracing::
578 		tracing in the suspend path, some side effects are possible
591 		seconds. Since the clock access is designed for use in tracing,
637 	      that are written into /sys/kernel/tracing/events/ftrace/print/trigger
681 	A list of events that can be enabled in tracing.
694 	Usual timestamp modes for tracing:
872 	This tracer can be configured when tracing likely/unlikely
880 	tracers from tracing simply echo "nop" into
890   available via the tracing/error_log file.  For the commands that
891   support it, reading the tracing/error_log file after an error will
893   information is available.  The tracing/error_log file is a circular
900     # echo xxx > /sys/kernel/tracing/events/sched/sched_wakeup/trigger
903     # cat /sys/kernel/tracing/error_log
913     # echo > /sys/kernel/tracing/error_log
1007 This shows that the current tracer is "irqsoff" tracing the time
1245 	When the file is closed, tracing will be enabled again.
1259 	impact of tracing. See "saved_cmdlines".
1275 	When the free_buffer is closed, tracing will
1307 	The latency tracers will enable function tracing
1324 	use function graph tracing instead of function tracing.
1331 	Enable branch tracing with the tracer. This enables branch
1582 enabling function tracing, we incur an added overhead. This
1860 function tracing, we do not know if interrupts were enabled
1968 One common case that people are interested in tracing is the
1970 Now for non Real-Time tasks, this can be arbitrary. But tracing
1973 Without function tracing::
2005 to avoid tracing the normal circumstances. Here we see that
2032 tracing non-RT tasks that are unpredictable will overwrite the
2076 is before the actual "switch", we stop the tracing when the recorded task
2203 This isn't that big of a trace, even with function tracing enabled,
2210 Latency tracing and events
2212 As function tracing can induce a much larger latency, but without
2393 the tracing could have overwritten the data that you wanted to
2395 tracing directly from a program. This allows you to stop the
2396 tracing at the point that you hit the part that you are
2397 interested in. To disable the tracing directly from a C program,
2413 Single thread tracing
2494 	       strcat(tracefs, "/tracing/");
2614   function calls while cpu tracing switch.
2890 following "dynamic ftrace" section such as tracing only specific
2897 virtually no overhead when function tracing is disabled. The way
2931 When tracing is enabled, the process of modifying the function
2956 tracing of specified functions. They are:
3148 We can see that there's no more lock or preempt tracing.
3248 function tracer. By default it is enabled (when function tracing is
3249 enabled in the kernel). If it is disabled, all function tracing is
3251 also for any other uses (perf, kprobes, stack tracing, profiling, etc). It
3293   Mod command supports module globbing. Disable tracing for all
3298   Disable tracing for all modules, but still trace kernel::
3311   These commands turn tracing on and off when the specified
3313   tracing system is turned on and off. If unspecified, there is
3314   no limit. For example, to disable tracing when a schedule bug
3319   To always disable tracing when __schedule_bug is hit::
3352   function tracing callbacks are very sensitive, when these commands
3394 the effect on the tracing is different. Every read from
3433 the trace file for reading, it will actually disable tracing and
3504 in time without stopping tracing. Ftrace swaps the current
3505 buffer with a spare buffer, and tracing continues in the new
3508 The following tracefs files in "tracing" are related to this
3518 	System"). Both reads snapshot and tracing are executable
3584 In the tracefs tracing directory, there is a directory called "instances".
3597 As you can see, the new directory looks similar to the tracing directory
3603 same name in the tracing directory except the buffer that is used
3695 the function tracing. The foo instance displays wakeups and task
3723 CONFIG_STACK_TRACER enables the ftrace stack tracing functionality.