1 .TH "LTTNG-UST" "3" "February 16, 2012" "" ""
4 lttng-ust \(em Linux Trace Toolkit Next Generation User-Space Tracer
10 Link liblttng-ust.so with applications, following this manpage.
15 LTTng-UST, the Linux Trace Toolkit Next Generation Userspace Tracer, is
16 port of the low-overhead tracing capabilities of the LTTng kernel tracer
17 to user-space. The library "liblttng-ust" enables tracing of
18 applications and libraries.
22 The simple way to generate the lttng-ust tracepoint probes is to use the
23 lttng-gen-tp(1) tool. See the lttng-gen-tp(1) manpage for explanation.
27 Here is the way to do it manually, without the lttng-gen-tp(1) helper
28 script, through an example:
31 .SH "CREATION OF TRACEPOINT PROVIDER"
35 To create a tracepoint provider, within a build tree similar to
36 examples/easy-ust installed with lttng-ust documentation, a
37 sample_component_provider.h for the general layout. This manpage will
38 focus on the various types that can be recorded into a trace event:
42 * provider name, not a variable but a string starting with a
43 * letter and containing either letters, numbers or underscores.
44 * Needs to be the same as TRACEPOINT_PROVIDER. Needs to
45 * follow the namespacing guide-lines in lttng/tracepoint.h:
47 * Must be included before include tracepoint provider
49 * ex.: project_component_event
51 * Optional company name goes here
52 * ex.: com_efficios_project_component_event
54 * In this example, "sample" is the project, and "component" is the
60 * tracepoint name, same format as sample provider. Does not
61 * need to be declared before. in this case the name is
67 * TP_ARGS macro contains the arguments passed for the tracepoint
68 * it is in the following format
69 * TP_ARGS(type1, name1, type2, name2, ... type10,
71 * where there can be from zero to ten elements.
72 * typeN is the datatype, such as int, struct or double **.
73 * name is the variable name (in "int myInt" the name would be
75 * TP_ARGS() is valid to mean no arguments
76 * TP_ARGS(void) is valid too
78 TP_ARGS(int, anint, int, netint, long *, values,
79 char *, text, size_t, textlen,
80 double, doublearg, float, floatarg),
83 * TP_FIELDS describes how to write the fields of the trace event.
84 * You can put expressions in the "argument expression" area,
85 * typically using the input arguments from TP_ARGS.
89 * ctf_integer: standard integer field.
90 * args: (type, field name, argument expression)
92 ctf_integer(int, intfield, anint)
93 ctf_integer(long, longfield, anint)
96 * ctf_integer_hex: integer field printed as hexadecimal.
97 * args: (type, field name, argument expression)
99 ctf_integer_hex(int, intfield2, anint)
102 * ctf_integer_network: integer field in network byte
103 * order. (_hex: printed as hexadecimal too)
104 * args: (type, field name, argument expression)
106 ctf_integer_network(int, netintfield, netint)
107 ctf_integer_network_hex(int, netintfieldhex, netint)
110 * ctf_array: a statically-sized array.
111 * args: (type, field name, argument expression, value)
113 ctf_array(long, arrfield1, values, 3)
116 * ctf_array_text: a statically-sized array, printed as
117 * a string. No need to be terminated by a null
120 ctf_array_text(char, arrfield2, text, 10)
123 * ctf_sequence: a dynamically-sized array.
124 * args: (type, field name, argument expression,
125 * type of length expression, length expression)
127 ctf_sequence(char, seqfield1, text,
131 * ctf_sequence_text: a dynamically-sized array, printed
132 * as string. No need to be null-terminated.
134 ctf_sequence_text(char, seqfield2, text,
138 * ctf_string: null-terminated string.
139 * args: (field name, argument expression)
141 ctf_string(stringfield, text)
144 * ctf_float: floating-point number.
145 * args: (type, field name, argument expression)
147 ctf_float(float, floatfield, floatarg)
148 ctf_float(double, doublefield, doublearg)
152 There can be an arbitrary number of tracepoint providers within an
153 application, but they must each have their own provider name. Duplicate
154 provider names are not allowed.
158 .SH "ASSIGNING LOGLEVEL TO EVENTS"
162 Optionally, a loglevel can be assigned to a TRACEPOINT_EVENT using the
165 TRACEPOINT_LOGLEVEL(< [com_company_]project[_component] >,
166 < event >, < loglevel_name >)
168 The first field is the provider name, the second field is the name of
169 the tracepoint, and the third field is the loglevel name. A
170 TRACEPOINT_EVENT should be declared prior to the the TRACEPOINT_LOGLEVEL
171 for a given tracepoint name. The TRACEPOINT_PROVIDER must be already
172 declared before declaring a TRACEPOINT_LOGLEVEL.
174 The loglevels go from 0 to 14. Higher numbers imply the most verbosity
175 (higher event throughput expected.
177 Loglevels 0 through 6, and loglevel 14, match syslog(3) loglevels
178 semantic. Loglevels 7 through 13 offer more fine-grained selection of
185 action must be taken immediately
197 normal, but significant, condition
200 informational message
203 debug information with system-level scope (set of programs)
205 TRACE_DEBUG_PROGRAM 8
206 debug information with program-level scope (set of processes)
208 TRACE_DEBUG_PROCESS 9
209 debug information with process-level scope (set of modules)
211 TRACE_DEBUG_MODULE 10
212 debug information with module (executable/library) scope (set of
216 debug information with compilation unit scope (set of functions)
218 TRACE_DEBUG_FUNCTION 12
219 debug information with function-level scope
222 debug information with line-level scope (TRACEPOINT_EVENT default)
225 debug-level message (trace_printf default)
227 See lttng(1) for information on how to use LTTng-UST loglevels.
231 .SH "ADDING TRACEPOINTS TO YOUR CODE"
235 Include the provider header in each C files you plan to instrument,
236 following the building/linking directives in the next section.
238 For instance, add within a function:
240 tracepoint(ust_tests_hello, tptest, i, netint, values,
241 text, strlen(text), dbl, flt);
243 As a call to the tracepoint. It will only be activated when requested by
244 lttng(1) through lttng-sessiond(8).
246 Even though LTTng-UST supports tracepoint() call site duplicates having
247 the same provider and event name, it is recommended to use a
248 provider event name pair only once within the source code to help
249 mapping events back to their call sites when analyzing the trace.
252 .SH "BUILDING/LINKING THE TRACEPOINT PROVIDER"
255 There are 2 ways to compile the Tracepoint Provider with the
256 application: either statically or dynamically. Please follow
259 1.1) Compile the Tracepoint provider with the application, either
260 directly or through a static library (.a):
261 - Into exactly one object of your application: define
262 "TRACEPOINT_DEFINE" and include the tracepoint provider.
263 - Use "\-I." for the compilation unit containing the tracepoint
264 provider include (e.g. tp.c).
265 - Link application with "\-ldl".
266 - If building the provider directly into the application,
267 link the application with "\-llttng-ust".
268 - If building a static library for the provider, link the static
269 library with "\-llttng-ust".
270 - Include the tracepoint provider header into all C files using
273 - tests/hello/ hello.c tp.c ust_tests_hello.h Makefile.example
275 2) Compile the Tracepoint Provider separately from the application,
276 using dynamic linking:
277 - Into exactly one object of your application: define
278 "TRACEPOINT_DEFINE" _and_ also define
279 "TRACEPOINT_PROBE_DYNAMIC_LINKAGE", then include the tracepoint
281 - Include the tracepoint provider header into all instrumented C
282 files that use the provider.
283 - Compile the tracepoint provider with "\-I.".
284 - Link the tracepoint provider with "\-llttng-ust".
285 - Link application with "\-ldl".
286 - Set a LD_PRELOAD environment to preload the tracepoint provider
287 shared object before starting the application when tracing is
288 needed. Another way is to dlopen the tracepoint probe when needed
291 - tests/demo/ demo.c tp*.c ust_tests_demo*.h demo-trace
293 - Note about dlclose() usage: it is not safe to use dlclose on a
294 provider shared object that is being actively used for tracing due
295 to a lack of reference counting from lttng-ust to the used shared
297 - Enable instrumentation and control tracing with the "lttng" command
298 from lttng-tools. See lttng-tools doc/quickstart.txt.
299 - Note for C++ support: although an application instrumented with
300 tracepoints can be compiled with g++, tracepoint probes should be
301 compiled with gcc (only tested with gcc so far).
305 .SH "USING LTTNG UST WITH DAEMONS"
308 Some extra care is needed when using liblttng-ust with daemon
309 applications that call fork(), clone(), or BSD rfork() without a
310 following exec() family system call. The library "liblttng-ust-fork.so"
311 needs to be preloaded for the application (launch with e.g.
312 LD_PRELOAD=liblttng-ust-fork.so appname).
319 Context information can be prepended by the tracer before each, or some,
320 events. The following context information is supported by LTTng-UST:
325 Virtual thread ID: thread ID as seen from the point of view of the
331 Virtual process ID: process ID as seen from the point of view of the
337 Thread name, as set by exec() or prctl(). It is recommended that
338 programs set their thread name with prctl() before hitting the first
339 tracepoint for that thread.
344 Pthread identifier. Can be used on architectures where pthread_t maps
345 nicely to an unsigned long type.
348 .SH "ENVIRONMENT VARIABLES"
351 .IP "LTTNG_UST_DEBUG"
352 Activate liblttng-ust debug output.
354 .IP "LTTNG_UST_REGISTER_TIMEOUT"
355 The environment variable "LTTNG_UST_REGISTER_TIMEOUT" can be used to
356 specify how long the applications should wait for sessiond
357 "registration done" command before proceeding to execute the main
358 program. The default is 3000ms (3 seconds). The timeout value is
359 specified in milliseconds. The value 0 means "don't wait". The value
360 \-1 means "wait forever". Setting this environment variable to 0 is
361 recommended for applications with time constraints on the process
368 lttng-gen-tp(1), lttng(1), babeltrace(1), lttng-sessiond(8)
374 Older lttng-ust libraries reject more recent, and incompatible, probe
375 providers. Newer lttng-ust librairies accept older probe providers, even
376 though some newer features might not be available with those providers.
382 LTTng-UST 2.0 and 2.1 lttng-ust libraries do not check for probe
383 provider version compatibility. This can lead to out-of-bound accesses
384 when using a more recent probe provider with an older lttng-ust library.
385 These error only trigger when tracing is active. This issue has been
386 fixed in LTTng-UST 2.2.
388 If you encounter any issues or usability problem, please report it on
389 our mailing list <lttng-dev@lists.lttng.org> to help improve this
393 liblttng-ust is distributed under the GNU Lesser General Public License
394 version 2.1. The headers are distributed under the MIT license.
396 See http://lttng.org for more information on the LTTng project.
398 Mailing list for support and development: <lttng-dev@lists.lttng.org>.
400 You can find us on IRC server irc.oftc.net (OFTC) in #lttng.
404 Thanks to Ericsson for funding this work, providing real-life use-cases,
407 Special thanks to Michel Dagenais and the DORSAL laboratory at
408 Polytechnique de Montreal for the LTTng journey.
413 liblttng-ust was originally written by Mathieu Desnoyers, with additional
414 contributions from various other people. It is currently maintained by
415 Mathieu Desnoyers <mathieu.desnoyers@efficios.com>.