1TRACE-CMD-SET(1) libtracefs Manual TRACE-CMD-SET(1)
2
6 trace-cmd-set - set a configuration parameter of the Ftrace Linux
7 internal tracer
8
10 trace-cmd set [OPTIONS] [command]
11
13 The trace-cmd(1) set command will set a configuration parameter of the
14 Ftrace Linux kernel tracer. The specified command will be run after the
15 ftrace state is set. The configured ftrace state can be restored to
16 default using the trace-cmd-reset(1) command.
17
19 -p tracer
20 Specify a tracer. Tracers usually do more than just trace an event.
21 Common tracers are: function, function_graph, preemptirqsoff,
22 irqsoff, preemptoff and wakeup. A tracer must be supported by the
23 running kernel. To see a list of available tracers, see
24 trace-cmd-list(1).
25 -e event
27 Specify an event to trace. Various static trace points have been
28 added to the Linux kernel. They are grouped by subsystem where you
29 can enable all events of a given subsystem or specify specific
30 events to be enabled. The event is of the format
31 "subsystem:event-name". You can also just specify the subsystem
32 without the :event-name or the event-name without the "subsystem:".
33 Using "-e sched_switch" will enable the "sched_switch" event where
34 as, "-e sched" will enable all events under the "sched" subsystem.
35 The 'event' can also contain glob expressions. That is, "*stat*" will
37 select all events (or subsystems) that have the characters "stat" in their
38 names.
39 The keyword 'all' can be used to enable all events.
41 -T
43 Enable a stacktrace on each event. For example:
44 <idle>-0 [003] 58549.289091: sched_switch: kworker/0:1:0 [120] R ==> trace-cmd:2603 [120]
46 <idle>-0 [003] 58549.289092: kernel_stack: <stack trace>
47 => schedule (ffffffff814b260e)
48 => cpu_idle (ffffffff8100a38c)
49 => start_secondary (ffffffff814ab828)
50 --func-stack
52 Enable a stack trace on all functions. Note this is only applicable
53 for the "function" plugin tracer, and will only take effect if the
54 -l option is used and succeeds in limiting functions. If the
55 function tracer is not filtered, and the stack trace is enabled,
56 you can live lock the machine.
57 -f filter
59 Specify a filter for the previous event. This must come after a -e.
60 This will filter what events get recorded based on the content of
61 the event. Filtering is passed to the kernel directly so what
62 filtering is allowed may depend on what version of the kernel you
63 have. Basically, it will let you use C notation to check if an
64 event should be processed or not.
65 .ft C
68 ==, >=, <=, >, <, &, |, && and ||
69 .ft
70 The above are usually safe to use to compare fields.
73 -R trigger
75 Specify a trigger for the previous event. This must come after a
76 -e. This will add a given trigger to the given event. To only
77 enable the trigger and not the event itself, then place the event
78 after the -v option.
79 See Documentation/trace/events.txt in the Linux kernel source for more
81 information on triggers.
82 -v
84 This will negate options specified after it on the command line. It
85 affects:
86 -e: Causes all specified events to not be traced. This is useful for
88 selecting a subsystem to be traced but to leave out various events.
89 For example: "-e sched -v -e "*stat*"" will enable all events in
90 the sched subsystem except those that have "stat" in their names.
91 -B: Deletes the specified ftrace instance. There must be no
92 configuration options related to this instance in the command line.
93 For example: "-v -B bar -B foo" will delete instance bar and create
94 a new instance foo.
95 Note: the -v option was taken from the way grep(1) inverts the following
96 matches.
97 -P pid
99 This will filter only the specified process IDs. Using -P will let
100 you trace only events that are caused by the process.
101 -c
103 Used -P to trace the process' children too (if kernel supports it).
104 --user
106 Execute the specified command as given user.
107 -C clock
109 Set the trace clock to "clock".
110 Use trace-cmd(1) list -C to see what clocks are available.
112 -l function-name
114 This will limit the function and function_graph tracers to only
115 trace the given function name. More than one -l may be specified on
116 the command line to trace more than one function. The limited use
117 of glob expressions are also allowed. These are match* to only
118 filter functions that start with match. *match to only filter
119 functions that end with match. *match\* to only filter on
120 functions that contain match.
121 -g function-name
123 This option is for the function_graph plugin. It will graph the
124 given function. That is, it will only trace the function and all
125 functions that it calls. You can have more than one -g on the
126 command line.
127 -n function-name
129 This has the opposite effect of -l. The function given with the -n
130 option will not be traced. This takes precedence, that is, if you
131 include the same function for both -n and -l, it will not be
132 traced.
133 -d
135 Some tracer plugins enable the function tracer by default. Like the
136 latency tracers. This option prevents the function tracer from
137 being enabled at start up.
138 -D
140 The option -d will try to use the function-trace option to disable
141 the function tracer (if available), otherwise it defaults to the
142 proc file: /proc/sys/kernel/ftrace_enabled, but will not touch it
143 if the function-trace option is available. The -D option will
144 disable both the ftrace_enabled proc file as well as the
145 function-trace option if it exists.
146 Note, this disable function tracing for all users, which includes users
148 outside of ftrace tracers (stack_tracer, perf, etc).
149 -O option
151 Ftrace has various options that can be enabled or disabled. This
152 allows you to set them. Appending the text no to an option disables
153 it. For example: "-O nograph-time" will disable the "graph-time"
154 Ftrace option.
155 -b size
157 This sets the ring buffer size to size kilobytes. Because the
158 Ftrace ring buffer is per CPU, this size is the size of each per
159 CPU ring buffer inside the kernel. Using "-b 10000" on a machine
160 with 4 CPUs will make Ftrace have a total buffer size of 40 Megs.
161 -B buffer-name
163 If the kernel supports multiple buffers, this will add a buffer
164 with the given name. If the buffer name already exists, that buffer
165 is just reset.
166 After a buffer name is stated, all events added after that will be
168 associated with that buffer. If no buffer is specified, or an event
169 is specified before a buffer name, it will be associated with the
170 main (toplevel) buffer.
171 trace-cmd set -e sched -B block -e block -B time -e timer sleep 1
173 The above is will enable all sched events in the main buffer. It will
175 then create a 'block' buffer instance and enable all block events within
176 that buffer. A 'time' buffer instance is created and all timer events
177 will be enabled for that event.
178 -m size
180 The max size in kilobytes that a per cpu buffer should be. Note,
181 due to rounding to page size, the number may not be totally
182 correct. Also, this is performed by switching between two buffers
183 that are half the given size thus the output may not be of the
184 given size even if much more was written.
185 Use this to prevent running out of diskspace for long runs.
187 -M cpumask
189 Set the cpumask for to trace. It only affects the last buffer
190 instance given. If supplied before any buffer instance, then it
191 affects the main buffer. The value supplied must be a hex number.
192 trace-cmd set -p function -M c -B events13 -e all -M 5
194 If the -M is left out, then the mask stays the same. To enable all
196 CPUs, pass in a value of '-1'.
197 -i
199 By default, if an event is listed that trace-cmd does not find, it
200 will exit with an error. This option will just ignore events that
201 are listed on the command line but are not found on the system.
202 -q | --quiet
204 Suppresses normal output, except for errors.
205 --max-graph-depth depth
207 Set the maximum depth the function_graph tracer will trace into a
208 function. A value of one will only show where userspace enters the
209 kernel but not any functions called in the kernel. The default is
210 zero, which means no limit.
211 --cmdlines-size size
213 Set the number of entries the kernel tracing file "saved_cmdlines"
214 can contain. This file is a circular buffer which stores the
215 mapping between cmdlines and PIDs. If full, it leads to unresolved
216 cmdlines ("<...>") within the trace. The kernel default value is
217 128.
218 --module module
220 Filter a module’s name in function tracing. It is equivalent to
221 adding :mod:module after all other functions being filtered. If no
222 other function filter is listed, then all modules functions will be
223 filtered in the filter.
224 '--module snd' is equivalent to '-l :mod:snd'
226 '--module snd -l "*jack*"' is equivalent to '-l "*jack*:mod:snd"'
228 '--module snd -n "*"' is equivalent to '-n :mod:snd'
230 --stderr
232 Have output go to stderr instead of stdout, but the output of the
233 command executed will not be changed. This is useful if you want to
234 monitor the output of the command being executed, but not see the
235 output from trace-cmd.
236 --fork
238 If a command is listed, then trace-cmd will wait for that command
239 to finish, unless the --fork option is specified. Then it will fork
240 the command and return immediately.
241 --verbose[=level]
243 Set the log level. Supported log levels are "none", "critical",
244 "error", "warning", "info", "debug", "all" or their identifiers
245 "0", "1", "2", "3", "4", "5", "6". Setting the log level to
246 specific value enables all logs from that and all previous levels.
247 The level will default to "info" if one is not specified.
248 Example: enable all critical, error and warning logs
250 trace-cmd set --verbose=warning
252
254 Enable all events for tracing:
255 .ft C
258 # trace-cmd set -e all
259 .ft
260 Set the function tracer:
263 .ft C
266 # trace-cmd set -p function
267 .ft
268
271 trace-cmd(1), trace-cmd-report(1), trace-cmd-start(1),
272 trace-cmd-stop(1), trace-cmd-extract(1), trace-cmd-reset(1),
273 trace-cmd-split(1), trace-cmd-list(1), trace-cmd-listen(1),
274 trace-cmd-profile(1)
275
277 Written by Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com[1]>
278
280 https://git.kernel.org/pub/scm/utils/trace-cmd/trace-cmd.git/
281
283 Copyright (C) 2010 Red Hat, Inc. Free use of this software is granted
284 under the terms of the GNU Public License (GPL).
285
287 1. tz.stoyanov@gmail.com
288 mailto:tz.stoyanov@gmail.com
289libtracefs 07/23/2022 TRACE-CMD-SET(1)