| 1 | .TH "LTTNG-UST" "3" "February 16, 2012" "" "" |
| 2 | |
| 3 | .SH "NAME" |
| 4 | lttng-ust \(em Linux Trace Toolkit Next Generation User-Space Tracer 2.x |
| 5 | |
| 6 | .SH "SYNOPSIS" |
| 7 | |
| 8 | .PP |
| 9 | .nf |
| 10 | Link liblttng-ust.so with applications, following this manpage. |
| 11 | .fi |
| 12 | .SH "DESCRIPTION" |
| 13 | |
| 14 | .PP |
| 15 | LTTng-UST, the Linux Trace Toolkit Next Generation Userspace Tracer, is a |
| 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. |
| 19 | |
| 20 | .SH "USAGE WITH TRACEF" |
| 21 | .PP |
| 22 | The simplest way to add instrumentation to your code is by far the |
| 23 | tracef() API. To do it, in a nutshell: |
| 24 | |
| 25 | 1) #include <lttng/tracef.h> |
| 26 | |
| 27 | 2) /* in your code, use like a printf */ |
| 28 | tracef("my message, this integer %d", 1234); |
| 29 | |
| 30 | 3) Link your program against liblttng-ust.so. |
| 31 | |
| 32 | 4) Enable UST events when tracing with the following sequence of commands |
| 33 | from lttng-tools: |
| 34 | |
| 35 | lttng create |
| 36 | lttng enable-event -u -a |
| 37 | lttng start |
| 38 | [... run your program ...] |
| 39 | lttng stop |
| 40 | lttng view |
| 41 | |
| 42 | That's it! |
| 43 | |
| 44 | If you want to have more flexibility and control on the event names, |
| 45 | payload typing, etc, you can continue reading on and use the tracepoints |
| 46 | below. "tracef()" is there for quick and dirty ad hoc instrumentation, |
| 47 | whereas tracepoint.h is meant for thorough instrumentation of a code |
| 48 | base to be integrated with an upstream project. |
| 49 | .PP |
| 50 | |
| 51 | .SH "USAGE WITH TRACEPOINT" |
| 52 | .PP |
| 53 | The simple way to generate the lttng-ust tracepoint probes is to use the |
| 54 | lttng-gen-tp(1) tool. See the lttng-gen-tp(1) manpage for explanation. |
| 55 | .PP |
| 56 | |
| 57 | .PP |
| 58 | Here is the way to do it manually, without the lttng-gen-tp(1) helper |
| 59 | script, through an example: |
| 60 | .PP |
| 61 | |
| 62 | .SH "CREATION OF TRACEPOINT PROVIDER" |
| 63 | |
| 64 | .nf |
| 65 | |
| 66 | To create a tracepoint provider, within a build tree similar to |
| 67 | examples/easy-ust installed with lttng-ust documentation, see |
| 68 | sample_component_provider.h for the general layout. You will need to |
| 69 | define TRACEPOINT_CREATE_PROBES before including your tracepoint |
| 70 | provider probe in one source file of your application. See tp.c from |
| 71 | easy-ust for an example of a tracepoint probe source file. This manpage |
| 72 | will focus on the various types that can be recorded into a trace |
| 73 | event: |
| 74 | |
| 75 | TRACEPOINT_EVENT( |
| 76 | /* |
| 77 | * provider name, not a variable but a string starting with a |
| 78 | * letter and containing either letters, numbers or underscores. |
| 79 | * Needs to be the same as TRACEPOINT_PROVIDER. Needs to |
| 80 | * follow the namespacing guide-lines in lttng/tracepoint.h: |
| 81 | * |
| 82 | * Must be included before include tracepoint provider |
| 83 | * ex.: project_event |
| 84 | * ex.: project_component_event |
| 85 | * |
| 86 | * Optional company name goes here |
| 87 | * ex.: com_efficios_project_component_event |
| 88 | * |
| 89 | * In this example, "sample" is the project, and "component" is the |
| 90 | * component. |
| 91 | */ |
| 92 | sample_component, |
| 93 | |
| 94 | /* |
| 95 | * tracepoint name, characters permitted follow the same |
| 96 | * constraints as the provider name. The name of this example |
| 97 | * event is "sample_event". |
| 98 | */ |
| 99 | sample_event, |
| 100 | |
| 101 | /* |
| 102 | * TP_ARGS macro contains the arguments passed for the tracepoint |
| 103 | * it is in the following format |
| 104 | * TP_ARGS(type1, name1, type2, name2, ... type10, |
| 105 | name10) |
| 106 | * where there can be from zero to ten elements. |
| 107 | * typeN is the datatype, such as int, struct or double **. |
| 108 | * name is the variable name (in "int myInt" the name would be |
| 109 | * myint) |
| 110 | * TP_ARGS() is valid to mean no arguments |
| 111 | * TP_ARGS(void) is valid too |
| 112 | */ |
| 113 | TP_ARGS(int, anint, int, netint, long *, values, |
| 114 | char *, text, size_t, textlen, |
| 115 | double, doublearg, float, floatarg), |
| 116 | |
| 117 | /* |
| 118 | * TP_FIELDS describes how to write the fields of the trace event. |
| 119 | * You can put expressions in the "argument expression" area, |
| 120 | * typically using the input arguments from TP_ARGS. |
| 121 | */ |
| 122 | TP_FIELDS( |
| 123 | /* |
| 124 | * ctf_integer: standard integer field. |
| 125 | * args: (type, field name, argument expression) |
| 126 | */ |
| 127 | ctf_integer(int, intfield, anint) |
| 128 | ctf_integer(long, longfield, anint) |
| 129 | |
| 130 | /* |
| 131 | * ctf_integer_hex: integer field printed as hexadecimal. |
| 132 | * args: (type, field name, argument expression) |
| 133 | */ |
| 134 | ctf_integer_hex(int, intfield2, anint) |
| 135 | |
| 136 | /* |
| 137 | * ctf_integer_network: integer field in network byte |
| 138 | * order. (_hex: printed as hexadecimal too) |
| 139 | * args: (type, field name, argument expression) |
| 140 | */ |
| 141 | ctf_integer_network(int, netintfield, netint) |
| 142 | ctf_integer_network_hex(int, netintfieldhex, netint) |
| 143 | |
| 144 | /* |
| 145 | * ctf_array: a statically-sized array. |
| 146 | * args: (type, field name, argument expression, value) |
| 147 | */ |
| 148 | ctf_array(long, arrfield1, values, 3) |
| 149 | |
| 150 | /* |
| 151 | * ctf_array_text: a statically-sized array, printed as |
| 152 | * a string. No need to be terminated by a null |
| 153 | * character. |
| 154 | * Behavior is undefined if "text" argument is NULL. |
| 155 | */ |
| 156 | ctf_array_text(char, arrfield2, text, 10) |
| 157 | |
| 158 | /* |
| 159 | * ctf_sequence: a dynamically-sized array. |
| 160 | * args: (type, field name, argument expression, |
| 161 | * type of length expression, length expression) |
| 162 | * The "type of length expression" needs to be an |
| 163 | * unsigned type. As a reminder, "unsigned char" should |
| 164 | * be preferred to "char", since the signedness of |
| 165 | * "char" is implementation-defined. |
| 166 | * Behavior is undefined if "text" argument is NULL. |
| 167 | */ |
| 168 | ctf_sequence(char, seqfield1, text, |
| 169 | size_t, textlen) |
| 170 | |
| 171 | /* |
| 172 | * ctf_sequence_text: a dynamically-sized array, printed |
| 173 | * as string. No need to be null-terminated. |
| 174 | * Behavior is undefined if "text" argument is NULL. |
| 175 | */ |
| 176 | ctf_sequence_text(char, seqfield2, text, |
| 177 | size_t, textlen) |
| 178 | |
| 179 | /* |
| 180 | * ctf_string: null-terminated string. |
| 181 | * args: (field name, argument expression) |
| 182 | * Behavior is undefined if "text" argument is NULL. |
| 183 | */ |
| 184 | ctf_string(stringfield, text) |
| 185 | |
| 186 | /* |
| 187 | * ctf_float: floating-point number. |
| 188 | * args: (type, field name, argument expression) |
| 189 | */ |
| 190 | ctf_float(float, floatfield, floatarg) |
| 191 | ctf_float(double, doublefield, doublearg) |
| 192 | ) |
| 193 | ) |
| 194 | |
| 195 | There can be an arbitrary number of tracepoint providers within an |
| 196 | application, but they must each have their own provider name. Duplicate |
| 197 | provider names are not allowed. |
| 198 | |
| 199 | .fi |
| 200 | |
| 201 | .SH "ASSIGNING LOGLEVEL TO EVENTS" |
| 202 | |
| 203 | .nf |
| 204 | |
| 205 | Optionally, a loglevel can be assigned to a TRACEPOINT_EVENT using the |
| 206 | following construct: |
| 207 | |
| 208 | TRACEPOINT_LOGLEVEL(< [com_company_]project[_component] >, |
| 209 | < event >, < loglevel_name >) |
| 210 | |
| 211 | The first field is the provider name, the second field is the name of |
| 212 | the tracepoint, and the third field is the loglevel name. A |
| 213 | TRACEPOINT_EVENT should be declared prior to the the TRACEPOINT_LOGLEVEL |
| 214 | for a given tracepoint name. The TRACEPOINT_PROVIDER must be already |
| 215 | declared before declaring a TRACEPOINT_LOGLEVEL. |
| 216 | |
| 217 | The loglevels go from 0 to 14. Higher numbers imply the most verbosity |
| 218 | (higher event throughput expected. |
| 219 | |
| 220 | Loglevels 0 through 6, and loglevel 14, match syslog(3) loglevels |
| 221 | semantic. Loglevels 7 through 13 offer more fine-grained selection of |
| 222 | debug information. |
| 223 | |
| 224 | TRACE_EMERG 0 |
| 225 | system is unusable |
| 226 | |
| 227 | TRACE_ALERT 1 |
| 228 | action must be taken immediately |
| 229 | |
| 230 | TRACE_CRIT 2 |
| 231 | critical conditions |
| 232 | |
| 233 | TRACE_ERR 3 |
| 234 | error conditions |
| 235 | |
| 236 | TRACE_WARNING 4 |
| 237 | warning conditions |
| 238 | |
| 239 | TRACE_NOTICE 5 |
| 240 | normal, but significant, condition |
| 241 | |
| 242 | TRACE_INFO 6 |
| 243 | informational message |
| 244 | |
| 245 | TRACE_DEBUG_SYSTEM 7 |
| 246 | debug information with system-level scope (set of programs) |
| 247 | |
| 248 | TRACE_DEBUG_PROGRAM 8 |
| 249 | debug information with program-level scope (set of processes) |
| 250 | |
| 251 | TRACE_DEBUG_PROCESS 9 |
| 252 | debug information with process-level scope (set of modules) |
| 253 | |
| 254 | TRACE_DEBUG_MODULE 10 |
| 255 | debug information with module (executable/library) scope (set of |
| 256 | units) |
| 257 | |
| 258 | TRACE_DEBUG_UNIT 11 |
| 259 | debug information with compilation unit scope (set of functions) |
| 260 | |
| 261 | TRACE_DEBUG_FUNCTION 12 |
| 262 | debug information with function-level scope |
| 263 | |
| 264 | TRACE_DEBUG_LINE 13 |
| 265 | debug information with line-level scope (TRACEPOINT_EVENT default) |
| 266 | |
| 267 | TRACE_DEBUG 14 |
| 268 | debug-level message |
| 269 | |
| 270 | See lttng(1) for information on how to use LTTng-UST loglevels. |
| 271 | |
| 272 | .fi |
| 273 | |
| 274 | .SH "ADDING TRACEPOINTS TO YOUR CODE" |
| 275 | |
| 276 | .nf |
| 277 | |
| 278 | Include the provider header in each C files you plan to instrument, |
| 279 | following the building/linking directives in the next section. |
| 280 | |
| 281 | For instance, add within a function: |
| 282 | |
| 283 | tracepoint(ust_tests_hello, tptest, i, netint, values, |
| 284 | text, strlen(text), dbl, flt); |
| 285 | |
| 286 | As a call to the tracepoint. It will only be activated when requested by |
| 287 | lttng(1) through lttng-sessiond(8). |
| 288 | |
| 289 | Even though LTTng-UST supports tracepoint() call site duplicates having |
| 290 | the same provider and event name, it is recommended to use a |
| 291 | provider event name pair only once within the source code to help |
| 292 | map events back to their call sites when analyzing the trace. |
| 293 | .fi |
| 294 | |
| 295 | .SH "BUILDING/LINKING THE TRACEPOINT PROVIDER" |
| 296 | |
| 297 | .nf |
| 298 | There are 2 ways to compile the Tracepoint Provider with the |
| 299 | application: either statically or dynamically. Please follow |
| 300 | carefully: |
| 301 | |
| 302 | 1.1) Compile the Tracepoint provider with the application, either |
| 303 | directly or through a static library (.a): |
| 304 | - Into exactly one object of your application: define |
| 305 | "TRACEPOINT_DEFINE" and include the tracepoint provider. |
| 306 | - Use "\-I." for the compilation unit containing the tracepoint |
| 307 | provider include (e.g. tp.c). |
| 308 | - Link application with "\-ldl". |
| 309 | - If building the provider directly into the application, |
| 310 | link the application with "\-llttng-ust". |
| 311 | - If building a static library for the provider, link the static |
| 312 | library with "\-llttng-ust". |
| 313 | - Include the tracepoint provider header into all C files using |
| 314 | the provider. |
| 315 | - Examples: |
| 316 | - doc/examples/easy-ust/ sample.c sample_component_provider.h tp.c |
| 317 | Makefile |
| 318 | - doc/examples/hello-static-lib/ hello.c tp.c ust_test_hello.h Makefile |
| 319 | |
| 320 | 2) Compile the Tracepoint Provider separately from the application, |
| 321 | using dynamic linking: |
| 322 | - Into exactly one object of your application: define |
| 323 | "TRACEPOINT_DEFINE" _and_ also define |
| 324 | "TRACEPOINT_PROBE_DYNAMIC_LINKAGE", then include the tracepoint |
| 325 | provider header. |
| 326 | - Include the tracepoint provider header into all instrumented C |
| 327 | files that use the provider. |
| 328 | - Compile the tracepoint provider with "\-I.". |
| 329 | - Link the tracepoint provider with "\-llttng-ust". |
| 330 | - Link application with "\-ldl". |
| 331 | - Set a LD_PRELOAD environment to preload the tracepoint provider |
| 332 | shared object before starting the application when tracing is |
| 333 | needed. Another way is to dlopen the tracepoint probe when needed |
| 334 | by the application. |
| 335 | - Example: |
| 336 | - doc/examples/demo demo.c tp*.c ust_tests_demo*.h demo-trace Makefile |
| 337 | |
| 338 | - Note about dlclose() usage: it is not safe to use dlclose on a |
| 339 | provider shared object that is being actively used for tracing due |
| 340 | to a lack of reference counting from lttng-ust to the used shared |
| 341 | object. |
| 342 | - Enable instrumentation and control tracing with the "lttng" command |
| 343 | from lttng-tools. See lttng-tools doc/quickstart.txt. |
| 344 | - Note for C++ support: although an application instrumented with |
| 345 | tracepoints can be compiled with g++, tracepoint probes should be |
| 346 | compiled with gcc (only tested with gcc so far). |
| 347 | |
| 348 | .fi |
| 349 | |
| 350 | .SH "USING LTTNG UST WITH DAEMONS" |
| 351 | |
| 352 | .nf |
| 353 | Some extra care is needed when using liblttng-ust with daemon |
| 354 | applications that call fork(), clone(), or BSD rfork() without a |
| 355 | following exec() family system call. The library "liblttng-ust-fork.so" |
| 356 | needs to be preloaded for the application (launch with e.g. |
| 357 | LD_PRELOAD=liblttng-ust-fork.so appname). |
| 358 | |
| 359 | .fi |
| 360 | |
| 361 | .SH "CONTEXT" |
| 362 | |
| 363 | .PP |
| 364 | Context information can be prepended by the tracer before each, or some, |
| 365 | events. The following context information is supported by LTTng-UST: |
| 366 | .PP |
| 367 | |
| 368 | .PP |
| 369 | .IP "vtid" |
| 370 | Virtual thread ID: thread ID as seen from the point of view of the |
| 371 | process namespace. |
| 372 | .PP |
| 373 | |
| 374 | .PP |
| 375 | .IP "vpid" |
| 376 | Virtual process ID: process ID as seen from the point of view of the |
| 377 | process namespace. |
| 378 | .PP |
| 379 | |
| 380 | .PP |
| 381 | .IP "ip" |
| 382 | Instruction pointer: Enables recording of the exact location where a tracepoint |
| 383 | was emitted. Can be used to reverse-lookup the source location that caused the |
| 384 | event to be emitted. |
| 385 | .PP |
| 386 | |
| 387 | .PP |
| 388 | .IP "procname" |
| 389 | Thread name, as set by exec() or prctl(). It is recommended that |
| 390 | programs set their thread name with prctl() before hitting the first |
| 391 | tracepoint for that thread. |
| 392 | .PP |
| 393 | |
| 394 | .PP |
| 395 | .IP "pthread_id" |
| 396 | Pthread identifier. Can be used on architectures where pthread_t maps |
| 397 | nicely to an unsigned long type. |
| 398 | .PP |
| 399 | |
| 400 | .SH "BASE ADDRESS STATEDUMP" |
| 401 | |
| 402 | .PP |
| 403 | If an application that uses liblttng-ust.so becomes part of a session, |
| 404 | information about its currently loaded shared objects will be traced to the |
| 405 | session at session-enable time. To record this information, the following event |
| 406 | needs to be enabled: |
| 407 | .PP |
| 408 | .IP "ust_baddr_statedump:soinfo" |
| 409 | This event is used to trace a currently loaded shared object. The base address |
| 410 | (where the dynamic linker has placed the shared object) is recorded in the |
| 411 | "baddr" field. The path to the shared object gets recorded in the |
| 412 | "sopath" field (as string). The file size of the loaded object (in |
| 413 | bytes) is recorded to the "size" field and its time of last modification |
| 414 | (in seconds since Epoch) is recorded in the "mtime" field. |
| 415 | .PP |
| 416 | If the event above is enabled, a series of "ust_baddr_statedump:soinfo" |
| 417 | events is recorded at session-enable time. It represents the state of |
| 418 | currently loaded shared objects for the traced process. If this |
| 419 | information gets combined with the lttng-ust-dl(3) instrumentation, all |
| 420 | aspects of dynamic loading that are relevant for symbol and |
| 421 | line number lookup are traced by LTTng. |
| 422 | .PP |
| 423 | .SH "ENVIRONMENT VARIABLES" |
| 424 | |
| 425 | .PP |
| 426 | .IP "LTTNG_UST_DEBUG" |
| 427 | Activate liblttng-ust debug output. |
| 428 | .PP |
| 429 | .IP "LTTNG_UST_REGISTER_TIMEOUT" |
| 430 | The environment variable "LTTNG_UST_REGISTER_TIMEOUT" can be used to |
| 431 | specify how long the applications should wait for sessiond |
| 432 | "registration done" command before proceeding to execute the main |
| 433 | program. The default is 3000ms (3 seconds). The timeout value is |
| 434 | specified in milliseconds. The value 0 means "don't wait". The value |
| 435 | \-1 means "wait forever". Setting this environment variable to 0 is |
| 436 | recommended for applications with time constraints on the process |
| 437 | startup time. |
| 438 | .PP |
| 439 | .IP "LTTNG_UST_WITHOUT_BADDR_STATEDUMP" |
| 440 | Prevent liblttng-ust to perform a base-address statedump on session-enable. |
| 441 | .PP |
| 442 | |
| 443 | .SH "SEE ALSO" |
| 444 | |
| 445 | .PP |
| 446 | lttng-gen-tp(1), lttng(1), babeltrace(1), lttng-ust-cyg-profile(3), |
| 447 | lttng-ust-dl(3), lttng-sessiond(8) |
| 448 | .PP |
| 449 | |
| 450 | .SH "COMPATIBILITY" |
| 451 | |
| 452 | .PP |
| 453 | Older lttng-ust libraries reject more recent, and incompatible, probe |
| 454 | providers. Newer lttng-ust libraries accept older probe providers, even |
| 455 | though some newer features might not be available with those providers. |
| 456 | .PP |
| 457 | |
| 458 | .SH "BUGS" |
| 459 | |
| 460 | .PP |
| 461 | LTTng-UST 2.0 and 2.1 lttng-ust libraries do not check for probe |
| 462 | provider version compatibility. This can lead to out-of-bound accesses |
| 463 | when using a more recent probe provider with an older lttng-ust library. |
| 464 | These error only trigger when tracing is active. This issue has been |
| 465 | fixed in LTTng-UST 2.2. |
| 466 | |
| 467 | If you encounter any issues or usability problem, please report it on |
| 468 | our mailing list <lttng-dev@lists.lttng.org> to help improve this |
| 469 | project. |
| 470 | .SH "CREDITS" |
| 471 | |
| 472 | liblttng-ust is distributed under the GNU Lesser General Public License |
| 473 | version 2.1. The headers are distributed under the MIT license. |
| 474 | .PP |
| 475 | See http://lttng.org for more information on the LTTng project. |
| 476 | .PP |
| 477 | Mailing list for support and development: <lttng-dev@lists.lttng.org>. |
| 478 | .PP |
| 479 | You can find us on IRC server irc.oftc.net (OFTC) in #lttng. |
| 480 | .PP |
| 481 | .SH "THANKS" |
| 482 | |
| 483 | Thanks to Ericsson for funding this work, providing real-life use-cases, |
| 484 | and testing. |
| 485 | |
| 486 | Special thanks to Michel Dagenais and the DORSAL laboratory at |
| 487 | Polytechnique de Montreal for the LTTng journey. |
| 488 | .PP |
| 489 | .SH "AUTHORS" |
| 490 | |
| 491 | .PP |
| 492 | liblttng-ust was originally written by Mathieu Desnoyers, with additional |
| 493 | contributions from various other people. It is currently maintained by |
| 494 | Mathieu Desnoyers <mathieu.desnoyers@efficios.com>. |
| 495 | .PP |