2.13/lttng-docs-2.13.txt

   1 The LTTng Documentation
   2 =======================
   3 Philippe Proulx <pproulx@efficios.com>
   4 v2.13, 15 June 2021
   5
   6
   7 include::../common/copyright.txt[]
   8
   9
  10 include::../common/welcome.txt[]
  11
  12
  13 include::../common/audience.txt[]
  14
  15
  16 [[chapters]]
  17 === What's in this documentation?
  18
  19 The LTTng Documentation is divided into the following sections:
  20
  21 * ``**<<nuts-and-bolts,Nuts and bolts>>**'' explains the
  22   rudiments of software tracing and the rationale behind the
  23   LTTng project.
  24 +
  25 Skip this section if you’re familiar with software tracing and with the
  26 LTTng project.
  27
  28 * ``**<<installing-lttng,Installation>>**'' describes the steps to
  29   install the LTTng packages on common Linux distributions and from
  30   their sources.
  31 +
  32 Skip this section if you already properly installed LTTng on your target
  33 system.
  34
  35 * ``**<<getting-started,Quick start>>**'' is a concise guide to
  36   get started quickly with LTTng kernel and user space tracing.
  37 +
  38 We recommend this section if you're new to LTTng or to software tracing
  39 in general.
  40 +
  41 Skip this section if you're not new to LTTng.
  42
  43 * ``**<<core-concepts,Core concepts>>**'' explains the concepts at
  44   the heart of LTTng.
  45 +
  46 It's a good idea to become familiar with the core concepts
  47 before attempting to use the toolkit.
  48
  49 * ``**<<plumbing,Components of LTTng>>**'' describes the various
  50   components of the LTTng machinery, like the daemons, the libraries,
  51   and the command-line interface.
  52
  53 * ``**<<instrumenting,Instrumentation>>**'' shows different ways to
  54   instrument user applications and the Linux kernel for LTTng tracing.
  55 +
  56 Instrumenting source code is essential to provide a meaningful
  57 source of events.
  58 +
  59 Skip this section if you don't have a programming background.
  60
  61 * ``**<<controlling-tracing,Tracing control>>**'' is divided into topics
  62   which demonstrate how to use the vast array of features that
  63   LTTng{nbsp}{revision} offers.
  64
  65 * ``**<<reference,Reference>>**'' contains API reference tables.
  66
  67 * ``**<<glossary,Glossary>>**'' is a specialized dictionary of terms
  68   related to LTTng or to the field of software tracing.
  69
  70
  71 include::../common/convention.txt[]
  72
  73
  74 include::../common/acknowledgements.txt[]
  75
  76
  77 [[whats-new]]
  78 == What's new in LTTng{nbsp}{revision}?
  79
  80 LTTng{nbsp}{revision} bears the name _Nordicité_, the product of a
  81 collaboration between https://champlibre.co/[Champ Libre] and
  82 https://www.boreale.com/[Boréale]. This farmhouse IPA is brewed with
  83 https://en.wikipedia.org/wiki/Kveik[Kveik] yeast and Québec-grown
  84 barley, oats, and juniper branches. The result is a remarkable, fruity,
  85 hazy golden IPA that offers a balanced touch of resinous and woodsy
  86 bitterness.
  87
  88 New features and changes in LTTng{nbsp}{revision}:
  89
  90 General::
  91 +
  92 * The LTTng trigger API of <<liblttng-ctl-lttng,`liblttng-ctl`>> now
  93   offers the ``__event rule matches__'' condition (an <<event-rule,event
  94   rule>> matches an event) as well as the following new actions:
  95 +
  96 --
  97 * <<basic-tracing-session-control,Start or stop>> a recording session.
  98 * <<session-rotation,Archive the current trace chunk>> of a
  99   recording session (rotate).
 100 * <<taking-a-snapshot,Take a snapshot>> of a recording session.
 101 --
 102 +
 103 As a reminder, a <<trigger,trigger>> is a condition-actions pair. When
 104 the condition of a trigger is satisfied, LTTng attempts to execute its
 105 actions.
 106 +
 107 This feature is also available with the new man:lttng-add-trigger(1),
 108 man:lttng-remove-trigger(1), and man:lttng-list-triggers(1)
 109 <<lttng-cli,cmd:lttng>> commands.
 110 +
 111 Starting from LTTng{nbsp}{revision}, a trigger may have more than one
 112 action.
 113 +
 114 See “<<add-event-rule-matches-trigger,Add an ``event rule matches''
 115 trigger to a session daemon>>” to learn more.
 116
 117 * The LTTng <<lttng-ust,user space>> and <<lttng-modules,kernel>>
 118   tracers offer the new namespace context field `time_ns`, which is the
 119   inode number, in the proc file system, of the current clock namespace.
 120 +
 121 See man:lttng-add-context(1), man:lttng-ust(3), and
 122 man:time_namespaces(7).
 123
 124 * The link:/man[manual pages] of LTTng-tools now have a terminology and
 125   style which match the LTTng Documentation, many fixes, more internal
 126   and manual page links, clearer lists and procedures, superior
 127   consistency, and usage examples.
 128 +
 129 The new man:lttng-event-rule(7) manual page explains the new, common
 130 way to specify an event rule on the command line.
 131 +
 132 The new man:lttng-concepts(7) manual page explains the core concepts of
 133 LTTng. Its contents is essentially the ``<<core-concepts,Core
 134 concepts>>'' section of this documentation, but more adapted to the
 135 manual page style.
 136
 137 User space tracing::
 138 +
 139 [IMPORTANT]
 140 ====
 141 The major version part of the `liblttng-ust`
 142 https://en.wikipedia.org/wiki/Soname[soname] is bumped, which means you
 143 **must recompile** your instrumented applications/libraries and
 144 <<tracepoint-provider,tracepoint provider packages>> to use
 145 LTTng-UST{nbsp}{revision}.
 146
 147 This change became a necessity to clean up the library and for
 148 `liblttng-ust` to stop exporting private symbols.
 149
 150 Also, LTTng{nbsp}{revision} prepends the `lttng_ust_` and `LTTNG_UST_`
 151 prefix to all public macro/definition/function names to offer a
 152 consistent API namespace. The LTTng{nbsp}2.12 API is still available;
 153 see the ``Compatibility with previous APIs'' section of
 154 man:lttng-ust(3).
 155 ====
 156 +
 157 Other notable changes:
 158 +
 159 * The `liblttng-ust` C{nbsp}API offers the new man:lttng_ust_vtracef(3)
 160   and man:lttng_ust_vtracelog(3) macros which are to
 161   man:lttng_ust_tracef(3) and man:lttng_ust_tracelog(3) what
 162   man:vprintf(3) is to man:printf(3).
 163
 164 * LTTng-UST now only depends on https://liburcu.org/[`liburcu`] at build
 165   time, not at run time.
 166
 167 Kernel tracing::
 168 +
 169 * The preferred display base of event record integer fields which
 170   contain memory addresses is now hexadecimal instead of decimal.
 171
 172 * The `pid` field is removed from `lttng_statedump_file_descriptor`
 173   event records and the `file_table_address` field is added.
 174 +
 175 This new field is the address of the `files_struct` structure which
 176 contains the file descriptor.
 177 +
 178 See the
 179 ``https://github.com/lttng/lttng-modules/commit/e7a0ca7205fd4be7c829d171baa8823fe4784c90[statedump: introduce `file_table_address`]''
 180 patch to learn more.
 181
 182 * The `flags` field of `syscall_entry_clone` event records is now a
 183   structure containing two enumerations (exit signal and options).
 184 +
 185 This change makes the flag values more readable and meaningful.
 186 +
 187 See the
 188 ``https://github.com/lttng/lttng-modules/commit/d775625e2ba4825b73b5897e7701ad6e2bdba115[syscalls: Make `clone()`'s `flags` field a 2 enum struct]''
 189 patch to learn more.
 190
 191 * The memory footprint of the kernel tracer is improved: the latter only
 192   generates metadata for the specific system call recording event rules
 193   that you <<enabling-disabling-events,create>>.
 194
 195
 196 [[nuts-and-bolts]]
 197 == Nuts and bolts
 198
 199 What is LTTng? As its name suggests, the _Linux Trace Toolkit: next
 200 generation_ is a modern toolkit for tracing Linux systems and
 201 applications. So your first question might be:
 202 **what is tracing?**
 203
 204
 205 [[what-is-tracing]]
 206 === What is tracing?
 207
 208 As the history of software engineering progressed and led to what
 209 we now take for granted--complex, numerous and
 210 interdependent software applications running in parallel on
 211 sophisticated operating systems like Linux--the authors of such
 212 components, software developers, began feeling a natural
 213 urge to have tools that would ensure the robustness and good performance
 214 of their masterpieces.
 215
 216 One major achievement in this field is, inarguably, the
 217 https://www.gnu.org/software/gdb/[GNU debugger (GDB)],
 218 an essential tool for developers to find and fix bugs. But even the best
 219 debugger won't help make your software run faster, and nowadays, faster
 220 software means either more work done by the same hardware, or cheaper
 221 hardware for the same work.
 222
 223 A _profiler_ is often the tool of choice to identify performance
 224 bottlenecks. Profiling is suitable to identify _where_ performance is
 225 lost in a given piece of software. The profiler outputs a profile, a
 226 statistical summary of observed events, which you may use to discover
 227 which functions took the most time to execute. However, a profiler won't
 228 report _why_ some identified functions are the bottleneck. Bottlenecks
 229 might only occur when specific conditions are met, conditions that are
 230 sometimes impossible to capture by a statistical profiler, or impossible
 231 to reproduce with an application altered by the overhead of an
 232 event-based profiler. For a thorough investigation of software
 233 performance issues, a history of execution is essential, with the
 234 recorded values of variables and context fields you choose, and with as
 235 little influence as possible on the instrumented application. This is
 236 where tracing comes in handy.
 237
 238 _Tracing_ is a technique used to understand what goes on in a running
 239 software system. The piece of software used for tracing is called a
 240 _tracer_, which is conceptually similar to a tape recorder. When
 241 recording, specific instrumentation points placed in the software source
 242 code generate events that are saved on a giant tape: a _trace_ file. You
 243 can record user application and operating system events at the same
 244 time, opening the possibility of resolving a wide range of problems that
 245 would otherwise be extremely challenging.
 246
 247 Tracing is often compared to _logging_. However, tracers and loggers are
 248 two different tools, serving two different purposes. Tracers are
 249 designed to record much lower-level events that occur much more
 250 frequently than log messages, often in the range of thousands per
 251 second, with very little execution overhead. Logging is more appropriate
 252 for a very high-level analysis of less frequent events: user accesses,
 253 exceptional conditions (errors and warnings, for example), database
 254 transactions, instant messaging communications, and such. Simply put,
 255 logging is one of the many use cases that can be satisfied with tracing.
 256
 257 The list of recorded events inside a trace file can be read manually
 258 like a log file for the maximum level of detail, but it's generally
 259 much more interesting to perform application-specific analyses to
 260 produce reduced statistics and graphs that are useful to resolve a
 261 given problem. Trace viewers and analyzers are specialized tools
 262 designed to do this.
 263
 264 In the end, this is what LTTng is: a powerful, open source set of
 265 tools to trace the Linux kernel and user applications at the same time.
 266 LTTng is composed of several components actively maintained and
 267 developed by its link:/community/#where[community].
 268
 269
 270 [[lttng-alternatives]]
 271 === Alternatives to noch:{LTTng}
 272
 273 Excluding proprietary solutions, a few competing software tracers
 274 exist for Linux:
 275
 276 https://github.com/dtrace4linux/linux[dtrace4linux]::
 277     A port of Sun Microsystems' DTrace to Linux.
 278 +
 279 The cmd:dtrace tool interprets user scripts and is responsible for
 280 loading code into the Linux kernel for further execution and collecting
 281 the outputted data.
 282
 283 https://en.wikipedia.org/wiki/Berkeley_Packet_Filter[eBPF]::
 284     A subsystem in the Linux kernel in which a virtual machine can
 285     execute programs passed from the user space to the kernel.
 286 +
 287 You can attach such programs to tracepoints and kprobes thanks to a
 288 system call, and they can output data to the user space when executed
 289 thanks to different mechanisms (pipe, VM register values, and eBPF maps,
 290 to name a few).
 291
 292 https://www.kernel.org/doc/Documentation/trace/ftrace.txt[ftrace]::
 293     The de facto function tracer of the Linux kernel.
 294 +
 295 Its user interface is a set of special files in sysfs.
 296
 297 https://perf.wiki.kernel.org/[perf]::
 298     A performance analysis tool for Linux which supports hardware
 299     performance counters, tracepoints, as well as other counters and
 300     types of probes.
 301 +
 302 The controlling utility of perf is the cmd:perf command line/text UI
 303 tool.
 304
 305 https://linux.die.net/man/1/strace[strace]::
 306     A command-line utility which records system calls made by a
 307     user process, as well as signal deliveries and changes of process
 308     state.
 309 +
 310 strace makes use of https://en.wikipedia.org/wiki/Ptrace[ptrace] to
 311 fulfill its function.
 312
 313 https://www.sysdig.org/[sysdig]::
 314     Like SystemTap, uses scripts to analyze Linux kernel events.
 315 +
 316 You write scripts, or _chisels_ in the jargon of sysdig, in Lua and
 317 sysdig executes them while it traces the system or afterwards. The
 318 interface of sysdig is the cmd:sysdig command-line tool as well as the
 319 text UI-based cmd:csysdig tool.
 320
 321 https://sourceware.org/systemtap/[SystemTap]::
 322     A Linux kernel and user space tracer which uses custom user scripts
 323     to produce plain text traces.
 324 +
 325 SystemTap converts the scripts to the C language, and then compiles them
 326 as Linux kernel modules which are loaded to produce trace data. The
 327 primary user interface of SystemTap is the cmd:stap command-line tool.
 328
 329 The main distinctive features of LTTng is that it produces correlated
 330 kernel and user space traces, as well as doing so with the lowest
 331 overhead amongst other solutions. It produces trace files in the
 332 https://diamon.org/ctf[CTF] format, a file format optimized
 333 for the production and analyses of multi-gigabyte data.
 334
 335 LTTng is the result of more than 10{nbsp}years of active open source
 336 development by a community of passionate developers. LTTng is currently
 337 available on major desktop and server Linux distributions.
 338
 339 The main interface for tracing control is a single command-line tool
 340 named cmd:lttng. The latter can create several recording sessions, enable
 341 and disable recording event rules on the fly, filter events efficiently
 342 with custom user expressions, start and stop tracing, and much more.
 343 LTTng can write the traces on the file system or send them over the
 344 network, and keep them totally or partially. You can make LTTng execute
 345 user-defined actions when LTTng emits an event. You can view the traces
 346 once tracing becomes inactive or as LTTng records events.
 347
 348 <<installing-lttng,Install LTTng now>> and
 349 <<getting-started,start tracing>>!
 350
 351
 352 [[installing-lttng]]
 353 == Installation
 354
 355 **LTTng** is a set of software <<plumbing,components>> which interact to
 356 <<instrumenting,instrument>> the Linux kernel and user applications, and
 357 to <<controlling-tracing,control tracing>> (start and stop
 358 recording, create recording event rules, and the rest). Those
 359 components are bundled into the following packages:
 360
 361 LTTng-tools::
 362     Libraries and command-line interface to control tracing.
 363
 364 LTTng-modules::
 365     Linux kernel modules to instrument and trace the kernel.
 366
 367 LTTng-UST::
 368     Libraries and Java/Python packages to instrument and trace user
 369     applications.
 370
 371 Most distributions mark the LTTng-modules and LTTng-UST packages as
 372 optional when installing LTTng-tools (which is always required). In the
 373 following sections, we always provide the steps to install all three,
 374 but note that:
 375
 376 * You only need to install LTTng-modules if you intend to use
 377   the Linux kernel LTTng tracer.
 378
 379 * You only need to install LTTng-UST if you intend to use the user
 380   space LTTng tracer.
 381
 382 [NOTE]
 383 ====
 384 As of 10{nbsp}June{nbsp}2021, LTTng{nbsp}{revision} is not yet available
 385 in any major non-enterprise Linux distribution.
 386
 387 For https://www.redhat.com/[RHEL] and https://www.suse.com/[SLES]
 388 packages, see https://packages.efficios.com/[EfficiOS Enterprise
 389 Packages].
 390
 391 For other distributions, <<building-from-source,build LTTng from
 392 source>>.
 393 ====
 394
 395
 396 [[building-from-source]]
 397 === Build from source
 398
 399 To build and install LTTng{nbsp}{revision} from source:
 400
 401 . Using the package manager of your distribution, or from source,
 402   install the following dependencies of LTTng-tools and LTTng-UST:
 403 +
 404 --
 405 * https://sourceforge.net/projects/libuuid/[libuuid]
 406 * https://directory.fsf.org/wiki/Popt[popt]
 407 * https://liburcu.org/[Userspace RCU]
 408 * http://www.xmlsoft.org/[libxml2]
 409 * **Optional**: https://github.com/numactl/numactl[numactl]
 410 --
 411
 412 . Download, build, and install the latest LTTng-modules{nbsp}{revision}:
 413 +
 414 --
 415 [role="term"]
 416 ----
 417 $ cd $(mktemp -d) &&
 418   wget https://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
 419   tar -xf lttng-modules-latest-2.13.tar.bz2 &&
 420   cd lttng-modules-2.13.* &&
 421   make &&
 422   sudo make modules_install &&
 423   sudo depmod -a
 424 ----
 425 --
 426
 427 . Download, build, and install the latest LTTng-UST{nbsp}{revision}:
 428 +
 429 --
 430 [role="term"]
 431 ----
 432 $ cd $(mktemp -d) &&
 433   wget https://lttng.org/files/lttng-ust/lttng-ust-latest-2.13.tar.bz2 &&
 434   tar -xf lttng-ust-latest-2.13.tar.bz2 &&
 435   cd lttng-ust-2.13.* &&
 436   ./configure &&
 437   make &&
 438   sudo make install &&
 439   sudo ldconfig
 440 ----
 441 --
 442 +
 443 Add `--disable-numa` to `./configure` if you don't have
 444 https://github.com/numactl/numactl[numactl].
 445 +
 446 --
 447 [IMPORTANT]
 448 .Java and Python application tracing
 449 ====
 450 If you need to instrument and have LTTng trace <<java-application,Java
 451 applications>>, pass the `--enable-java-agent-jul`,
 452 `--enable-java-agent-log4j`, or `--enable-java-agent-all` options to the
 453 `configure` script, depending on which Java logging framework you use.
 454
 455 If you need to instrument and have LTTng trace
 456 <<python-application,Python applications>>, pass the
 457 `--enable-python-agent` option to the `configure` script. You can set
 458 the env:PYTHON environment variable to the path to the Python interpreter
 459 for which to install the LTTng-UST Python agent package.
 460 ====
 461 --
 462 +
 463 --
 464 [NOTE]
 465 ====
 466 By default, LTTng-UST libraries are installed to
 467 dir:{/usr/local/lib}, which is the de facto directory in which to
 468 keep self-compiled and third-party libraries.
 469
 470 When <<building-tracepoint-providers-and-user-application,linking an
 471 instrumented user application with `liblttng-ust`>>:
 472
 473 * Append `/usr/local/lib` to the env:LD_LIBRARY_PATH environment
 474   variable.
 475
 476 * Pass the `-L/usr/local/lib` and `-Wl,-rpath,/usr/local/lib` options to
 477   man:gcc(1), man:g++(1), or man:clang(1).
 478 ====
 479 --
 480
 481 . Download, build, and install the latest LTTng-tools{nbsp}{revision}:
 482 +
 483 --
 484 [role="term"]
 485 ----
 486 $ cd $(mktemp -d) &&
 487   wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
 488   tar -xf lttng-tools-latest-2.13.tar.bz2 &&
 489   cd lttng-tools-2.13.* &&
 490   ./configure &&
 491   make &&
 492   sudo make install &&
 493   sudo ldconfig
 494 ----
 495 --
 496
 497 TIP: The https://github.com/eepp/vlttng[vlttng tool] can do all the
 498 previous steps automatically for a given version of LTTng and confine
 499 the installed files to a specific directory. This can be useful to try
 500 LTTng without installing it on your system.
 501
 502
 503 [[getting-started]]
 504 == Quick start
 505
 506 This is a short guide to get started quickly with LTTng kernel and user
 507 space tracing.
 508
 509 Before you follow this guide, make sure to <<installing-lttng,install>>
 510 LTTng.
 511
 512 This tutorial walks you through the steps to:
 513
 514 . <<tracing-the-linux-kernel,Record Linux kernel events>>.
 515
 516 . <<tracing-your-own-user-application,Record the events of a user
 517   application>> written in C.
 518
 519 . <<viewing-and-analyzing-your-traces,View and analyze the
 520   recorded events>>.
 521
 522
 523 [[tracing-the-linux-kernel]]
 524 === Record Linux kernel events
 525
 526 NOTE: The following command lines start with the `#` prompt because you
 527 need root privileges to control the Linux kernel LTTng tracer. You can
 528 also control the kernel tracer as a regular user if your Unix user is a
 529 member of the <<tracing-group,tracing group>>.
 530
 531 . Create a <<tracing-session,recording session>> to write LTTng traces
 532   to dir:{/tmp/my-kernel-trace}:
 533 +
 534 --
 535 [role="term"]
 536 ----
 537 # lttng create my-kernel-session --output=/tmp/my-kernel-trace
 538 ----
 539 --
 540
 541 . List the available kernel tracepoints and system calls:
 542 +
 543 --
 544 [role="term"]
 545 ----
 546 # lttng list --kernel
 547 # lttng list --kernel --syscall
 548 ----
 549 --
 550
 551 . Create <<event,recording event rules>> which match events having
 552   the desired names, for example the `sched_switch` and
 553   `sched_process_fork` tracepoints, and the man:open(2) and man:close(2)
 554   system calls:
 555 +
 556 --
 557 [role="term"]
 558 ----
 559 # lttng enable-event --kernel sched_switch,sched_process_fork
 560 # lttng enable-event --kernel --syscall open,close
 561 ----
 562 --
 563 +
 564 Create a recording event rule which matches _all_ the Linux kernel
 565 tracepoint events with the opt:lttng-enable-event(1):--all option
 566 (recording with such a recording event rule generates a lot of data):
 567 +
 568 --
 569 [role="term"]
 570 ----
 571 # lttng enable-event --kernel --all
 572 ----
 573 --
 574
 575 . <<basic-tracing-session-control,Start recording>>:
 576 +
 577 --
 578 [role="term"]
 579 ----
 580 # lttng start
 581 ----
 582 --
 583
 584 . Do some operation on your system for a few seconds. For example,
 585   load a website, or list the files of a directory.
 586
 587 . <<creating-destroying-tracing-sessions,Destroy>> the current
 588   recording session:
 589 +
 590 --
 591 [role="term"]
 592 ----
 593 # lttng destroy
 594 ----
 595 --
 596 +
 597 The man:lttng-destroy(1) command doesn't destroy the trace data; it
 598 only destroys the state of the recording session.
 599 +
 600 The man:lttng-destroy(1) command also runs the man:lttng-stop(1) command
 601 implicitly (see ``<<basic-tracing-session-control,Start and stop a
 602 recording session>>''). You need to stop recording to make LTTng flush
 603 the remaining trace data and make the trace readable.
 604
 605 . For the sake of this example, make the recorded trace accessible to
 606   the non-root users:
 607 +
 608 --
 609 [role="term"]
 610 ----
 611 # chown -R $(whoami) /tmp/my-kernel-trace
 612 ----
 613 --
 614
 615 See ``<<viewing-and-analyzing-your-traces,View and analyze the
 616 recorded events>>'' to view the recorded events.
 617
 618
 619 [[tracing-your-own-user-application]]
 620 === Record user application events
 621
 622 This section walks you through a simple example to record the events of
 623 a _Hello world_ program written in{nbsp}C.
 624
 625 To create the traceable user application:
 626
 627 . Create the tracepoint provider header file, which defines the
 628   tracepoints and the events they can generate:
 629 +
 630 --
 631 [source,c]
 632 .path:{hello-tp.h}
 633 ----
 634 #undef LTTNG_UST_TRACEPOINT_PROVIDER
 635 #define LTTNG_UST_TRACEPOINT_PROVIDER hello_world
 636
 637 #undef LTTNG_UST_TRACEPOINT_INCLUDE
 638 #define LTTNG_UST_TRACEPOINT_INCLUDE "./hello-tp.h"
 639
 640 #if !defined(_HELLO_TP_H) || defined(LTTNG_UST_TRACEPOINT_HEADER_MULTI_READ)
 641 #define _HELLO_TP_H
 642
 643 #include <lttng/tracepoint.h>
 644
 645 LTTNG_UST_TRACEPOINT_EVENT(
 646     hello_world,
 647     my_first_tracepoint,
 648     LTTNG_UST_TP_ARGS(
 649         int, my_integer_arg,
 650         char *, my_string_arg
 651     ),
 652     LTTNG_UST_TP_FIELDS(
 653         lttng_ust_field_string(my_string_field, my_string_arg)
 654         lttng_ust_field_integer(int, my_integer_field, my_integer_arg)
 655     )
 656 )
 657
 658 #endif /* _HELLO_TP_H */
 659
 660 #include <lttng/tracepoint-event.h>
 661 ----
 662 --
 663
 664 . Create the tracepoint provider package source file:
 665 +
 666 --
 667 [source,c]
 668 .path:{hello-tp.c}
 669 ----
 670 #define LTTNG_UST_TRACEPOINT_CREATE_PROBES
 671 #define LTTNG_UST_TRACEPOINT_DEFINE
 672
 673 #include "hello-tp.h"
 674 ----
 675 --
 676
 677 . Build the tracepoint provider package:
 678 +
 679 --
 680 [role="term"]
 681 ----
 682 $ gcc -c -I. hello-tp.c
 683 ----
 684 --
 685
 686 . Create the _Hello World_ application source file:
 687 +
 688 --
 689 [source,c]
 690 .path:{hello.c}
 691 ----
 692 #include <stdio.h>
 693 #include "hello-tp.h"
 694
 695 int main(int argc, char *argv[])
 696 {
 697     unsigned int i;
 698
 699     puts("Hello, World!\nPress Enter to continue...");
 700
 701     /*
 702      * The following getchar() call only exists for the purpose of this
 703      * demonstration, to pause the application in order for you to have
 704      * time to list its tracepoints. You don't need it otherwise.
 705      */
 706     getchar();
 707
 708     /*
 709      * An lttng_ust_tracepoint() call.
 710      *
 711      * Arguments, as defined in `hello-tp.h`:
 712      *
 713      * 1. Tracepoint provider name (required)
 714      * 2. Tracepoint name (required)
 715      * 3. `my_integer_arg` (first user-defined argument)
 716      * 4. `my_string_arg` (second user-defined argument)
 717      *
 718      * Notice the tracepoint provider and tracepoint names are
 719      * C identifiers, NOT strings: they're in fact parts of variables
 720      * that the macros in `hello-tp.h` create.
 721      */
 722     lttng_ust_tracepoint(hello_world, my_first_tracepoint, 23,
 723                          "hi there!");
 724
 725     for (i = 0; i < argc; i++) {
 726         lttng_ust_tracepoint(hello_world, my_first_tracepoint,
 727                              i, argv[i]);
 728     }
 729
 730     puts("Quitting now!");
 731     lttng_ust_tracepoint(hello_world, my_first_tracepoint,
 732                          i * i, "i^2");
 733     return 0;
 734 }
 735 ----
 736 --
 737
 738 . Build the application:
 739 +
 740 --
 741 [role="term"]
 742 ----
 743 $ gcc -c hello.c
 744 ----
 745 --
 746
 747 . Link the application with the tracepoint provider package,
 748   `liblttng-ust` and `libdl`:
 749 +
 750 --
 751 [role="term"]
 752 ----
 753 $ gcc -o hello hello.o hello-tp.o -llttng-ust -ldl
 754 ----
 755 --
 756
 757 Here's the whole build process:
 758
 759 [role="img-100"]
 760 .Build steps of the user space tracing tutorial.
 761 image::ust-flow.png[]
 762
 763 To record the events of the user application:
 764
 765 . Run the application with a few arguments:
 766 +
 767 --
 768 [role="term"]
 769 ----
 770 $ ./hello world and beyond
 771 ----
 772 --
 773 +
 774 You see:
 775 +
 776 --
 777 ----
 778 Hello, World!
 779 Press Enter to continue...
 780 ----
 781 --
 782
 783 . Start an LTTng <<lttng-sessiond,session daemon>>:
 784 +
 785 --
 786 [role="term"]
 787 ----
 788 $ lttng-sessiond --daemonize
 789 ----
 790 --
 791 +
 792 NOTE: A session daemon might already be running, for example as a
 793 service that the service manager of your distribution started.
 794
 795 . List the available user space tracepoints:
 796 +
 797 --
 798 [role="term"]
 799 ----
 800 $ lttng list --userspace
 801 ----
 802 --
 803 +
 804 You see the `hello_world:my_first_tracepoint` tracepoint listed
 805 under the `./hello` process.
 806
 807 . Create a <<tracing-session,recording session>>:
 808 +
 809 --
 810 [role="term"]
 811 ----
 812 $ lttng create my-user-space-session
 813 ----
 814 --
 815
 816 . Create a <<event,recording event rule>> which matches user space
 817   tracepoint events named `hello_world:my_first_tracepoint`:
 818 +
 819 --
 820 [role="term"]
 821 ----
 822 $ lttng enable-event --userspace hello_world:my_first_tracepoint
 823 ----
 824 --
 825
 826 . <<basic-tracing-session-control,Start recording>>:
 827 +
 828 --
 829 [role="term"]
 830 ----
 831 $ lttng start
 832 ----
 833 --
 834
 835 . Go back to the running `hello` application and press **Enter**.
 836 +
 837 The program executes all `lttng_ust_tracepoint()` instrumentation
 838 points, emitting events as the event rule you created in step{nbsp}5
 839 matches them, and
 840 exits.
 841
 842 . <<creating-destroying-tracing-sessions,Destroy>> the current
 843   recording session:
 844 +
 845 --
 846 [role="term"]
 847 ----
 848 $ lttng destroy
 849 ----
 850 --
 851 +
 852 The man:lttng-destroy(1) command doesn't destroy the trace data; it
 853 only destroys the state of the recording session.
 854 +
 855 The man:lttng-destroy(1) command also runs the man:lttng-stop(1) command
 856 implicitly (see ``<<basic-tracing-session-control,Start and stop a
 857 recording session>>''). You need to stop recording to make LTTng flush
 858 the remaining trace data and make the trace readable.
 859
 860 By default, LTTng saves the traces to the
 861 +$LTTNG_HOME/lttng-traces/__NAME__-__DATE__-__TIME__+ directory, where
 862 +__NAME__+ is the recording session name. The env:LTTNG_HOME environment
 863 variable defaults to `$HOME` if not set.
 864
 865
 866 [[viewing-and-analyzing-your-traces]]
 867 === View and analyze the recorded events
 868
 869 Once you have completed the <<tracing-the-linux-kernel,Record Linux
 870 kernel events>> and <<tracing-your-own-user-application,Record user
 871 application events>> tutorials, you can inspect the recorded events.
 872
 873 There are many tools you can use to read LTTng traces:
 874
 875 https://babeltrace.org/[Babeltrace{nbsp}2]::
 876     A rich, flexible trace manipulation toolkit which includes
 877     a versatile command-line interface
 878     (man:babeltrace2(1)),
 879     a https://babeltrace.org/docs/v2.0/libbabeltrace2/[C{nbsp}library],
 880     and https://babeltrace.org/docs/v2.0/python/bt2/[Python{nbsp}3 bindings]
 881     so that you can easily process or convert an LTTng trace with
 882     your own script.
 883 +
 884 The Babeltrace{nbsp}2 project ships with a plugin
 885 (man:babeltrace2-plugin-ctf(7)) which supports the format of the traces
 886 which LTTng produces, https://diamon.org/ctf/[CTF].
 887
 888 http://tracecompass.org/[Trace Compass]::
 889     A graphical user interface for viewing and analyzing any type of
 890     logs or traces, including those of LTTng.
 891
 892 https://github.com/lttng/lttng-analyses[LTTng analyses]::
 893     An experimental project which includes many high-level analyses of
 894     LTTng kernel traces, like scheduling statistics, interrupt
 895     frequency distribution, top CPU usage, and more.
 896
 897 NOTE: This section assumes that LTTng wrote the traces it recorded
 898 during the previous tutorials to their default location, in the
 899 dir:{$LTTNG_HOME/lttng-traces} directory. The env:LTTNG_HOME
 900 environment variable defaults to `$HOME` if not set.
 901
 902
 903 [[viewing-and-analyzing-your-traces-bt]]
 904 ==== Use the cmd:babeltrace2 command-line tool
 905
 906 The simplest way to list all the recorded events of an LTTng trace is to
 907 pass its path to man:babeltrace2(1), without options:
 908
 909 [role="term"]
 910 ----
 911 $ babeltrace2 ~/lttng-traces/my-user-space-session*
 912 ----
 913
 914 The cmd:babeltrace2 command finds all traces recursively within the
 915 given path and prints all their events, sorting them chronologically.
 916
 917 Pipe the output of cmd:babeltrace2 into a tool like man:grep(1) for
 918 further filtering:
 919
 920 [role="term"]
 921 ----
 922 $ babeltrace2 /tmp/my-kernel-trace | grep _switch
 923 ----
 924
 925 Pipe the output of cmd:babeltrace2 into a tool like man:wc(1) to count
 926 the recorded events:
 927
 928 [role="term"]
 929 ----
 930 $ babeltrace2 /tmp/my-kernel-trace | grep _open | wc --lines
 931 ----
 932
 933
 934 [[viewing-and-analyzing-your-traces-bt-python]]
 935 ==== Use the Babeltrace{nbsp}2 Python bindings
 936
 937 The <<viewing-and-analyzing-your-traces-bt,text output of
 938 cmd:babeltrace2>> is useful to isolate event records by simple matching
 939 using man:grep(1) and similar utilities. However, more elaborate
 940 filters, such as keeping only event records with a field value falling
 941 within a specific range, are not trivial to write using a shell.
 942 Moreover, reductions and even the most basic computations involving
 943 multiple event records are virtually impossible to implement.
 944
 945 Fortunately, Babeltrace{nbsp}2 ships with
 946 https://babeltrace.org/docs/v2.0/python/bt2/[Python{nbsp}3 bindings]
 947 which make it easy to read the event records of an LTTng trace
 948 sequentially and compute the desired information.
 949
 950 The following script accepts an LTTng Linux kernel trace path as its
 951 first argument and prints the short names of the top five running
 952 processes on CPU{nbsp}0 during the whole trace:
 953
 954 [source,python]
 955 .path:{top5proc.py}
 956 ----
 957 import bt2
 958 import sys
 959 import collections
 960
 961
 962 def top5proc():
 963     # Get the trace path from the first command-line argument
 964     it = bt2.TraceCollectionMessageIterator(sys.argv[1])
 965
 966     # This counter dictionary will hold execution times:
 967     #
 968     #     Task command name -> Total execution time (ns)
 969     exec_times = collections.Counter()
 970
 971     # This holds the last `sched_switch` timestamp
 972     last_ts = None
 973
 974     for msg in it:
 975         # We only care about event messages
 976         if type(msg) is not bt2._EventMessageConst:
 977             continue
 978
 979         # Event of the event message
 980         event = msg.event
 981
 982         # Keep only `sched_switch` events
 983         if event.cls.name != 'sched_switch':
 984             continue
 985
 986         # Keep only records of events which LTTng emitted from CPU 0
 987         if event.packet.context_field['cpu_id'] != 0:
 988             continue
 989
 990         # Event timestamp (ns)
 991         cur_ts = msg.default_clock_snapshot.ns_from_origin
 992
 993         if last_ts is None:
 994             # Start here
 995             last_ts = cur_ts
 996
 997         # (Short) name of the previous task command
 998         prev_comm = str(event.payload_field['prev_comm'])
 999
1000         # Initialize an entry in our dictionary if not done yet
1001         if prev_comm not in exec_times:
1002             exec_times[prev_comm] = 0
1003
1004         # Compute previous command execution time
1005         diff = cur_ts - last_ts
1006
1007         # Update execution time of this command
1008         exec_times[prev_comm] += diff
1009
1010         # Update last timestamp
1011         last_ts = cur_ts
1012
1013     # Print top 5
1014     for name, ns in exec_times.most_common(5):
1015         print('{:20}{} s'.format(name, ns / 1e9))
1016
1017
1018 if __name__ == '__main__':
1019     top5proc()
1020 ----
1021
1022 Run this script:
1023
1024 [role="term"]
1025 ----
1026 $ python3 top5proc.py /tmp/my-kernel-trace/kernel
1027 ----
1028
1029 Output example:
1030
1031 ----
1032 swapper/0           48.607245889 s
1033 chromium            7.192738188 s
1034 pavucontrol         0.709894415 s
1035 Compositor          0.660867933 s
1036 Xorg.bin            0.616753786 s
1037 ----
1038
1039 Note that `swapper/0` is the ``idle'' process of CPU{nbsp}0 on Linux;
1040 since we weren't using the CPU that much when recording, its first
1041 position in the list makes sense.
1042
1043
1044 [[core-concepts]]
1045 == [[understanding-lttng]]Core concepts
1046
1047 From a user's perspective, the LTTng system is built on a few concepts,
1048 or objects, on which the <<lttng-cli,cmd:lttng command-line tool>>
1049 operates by sending commands to the <<lttng-sessiond,session daemon>>
1050 (through <<liblttng-ctl-lttng,`liblttng-ctl`>>).
1051
1052 Understanding how those objects relate to each other is key to master
1053 the toolkit.
1054
1055 The core concepts of LTTng are:
1056
1057 * <<"event-rule","Instrumentation point, event rule, and event">>
1058 * <<trigger,Trigger>>
1059 * <<tracing-session,Recording session>>
1060 * <<domain,Tracing domain>>
1061 * <<channel,Channel and ring buffer>>
1062 * <<event,Recording event rule and event record>>
1063
1064 NOTE: The man:lttng-concepts(7) manual page also documents the core
1065 concepts of LTTng, with more links to other LTTng-tools manual pages.
1066
1067
1068 [[event-rule]]
1069 === Instrumentation point, event rule, and event
1070
1071 An _instrumentation point_ is a point, within a piece of software,
1072 which, when executed, creates an LTTng _event_.
1073
1074 LTTng offers various <<instrumentation-point-types,types of
1075 instrumentation>>.
1076
1077 An _event rule_ is a set of conditions to match a set of events.
1078
1079 When LTTng creates an event{nbsp}__E__, an event rule{nbsp}__ER__ is
1080 said to __match__{nbsp}__E__ when{nbsp}__E__ satisfies _all_ the
1081 conditions of{nbsp}__ER__. This concept is similar to a
1082 https://en.wikipedia.org/wiki/Regular_expression[regular expression]
1083 which matches a set of strings.
1084
1085 When an event rule matches an event, LTTng _emits_ the event, therefore
1086 attempting to execute one or more actions.
1087
1088 [IMPORTANT]
1089 ====
1090 [[event-creation-emission-opti]]The event creation and emission
1091 processes are documentation concepts to help understand the journey from
1092 an instrumentation point to the execution of actions.
1093
1094 The actual creation of an event can be costly because LTTng needs to
1095 evaluate the arguments of the instrumentation point.
1096
1097 In practice, LTTng implements various optimizations for the Linux kernel
1098 and user space <<domain,tracing domains>> to avoid actually creating an
1099 event when the tracer knows, thanks to properties which are independent
1100 from the event payload and current context, that it would never emit
1101 such an event. Those properties are:
1102
1103 * The <<instrumentation-point-types,instrumentation point type>>.
1104
1105 * The instrumentation point name.
1106
1107 * The instrumentation point log level.
1108
1109 * For a <<event,recording event rule>>:
1110 ** The status of the rule itself.
1111 ** The status of the <<channel,channel>>.
1112 ** The activity of the <<tracing-session,recording session>>.
1113 ** Whether or not the process for which LTTng would create the event is
1114    <<pid-tracking,allowed to record events>>.
1115
1116 In other words: if, for a given instrumentation point{nbsp}__IP__, the
1117 LTTng tracer knows that it would never emit an event,
1118 executing{nbsp}__IP__ represents a simple boolean variable check and,
1119 for a Linux kernel recording event rule, a few process attribute checks.
1120 ====
1121
1122 As of LTTng{nbsp}{revision}, there are two places where you can find an
1123 event rule:
1124
1125 <<event,Recording event rule>>::
1126     A specific type of event rule of which the action is to record the
1127     matched event as an event record.
1128 +
1129 See ``<<enabling-disabling-events,Create and enable a recording event
1130 rule>>'' to learn more.
1131
1132 ``Event rule matches'' <<trigger,trigger>> condition (since LTTng{nbsp}2.13)::
1133     When the event rule of the trigger condition matches an event, LTTng
1134     can execute user-defined actions such as sending an LTTng
1135     notification,
1136     <<basic-tracing-session-control,starting a recording session>>,
1137     and more.
1138 +
1139 See “<<add-event-rule-matches-trigger,Add an ``event rule matches''
1140 trigger to a session daemon>>” to learn more.
1141
1142 For LTTng to emit an event{nbsp}__E__,{nbsp}__E__ must satisfy _all_ the
1143 basic conditions of an event rule{nbsp}__ER__, that is:
1144
1145 * The instrumentation point from which LTTng
1146   creates{nbsp}__E__ has a specific
1147   <<instrumentation-point-types,type>>.
1148
1149 * A pattern matches the name of{nbsp}__E__ while another pattern
1150   doesn't.
1151
1152 * The log level of the instrumentation point from which LTTng
1153   creates{nbsp}__E__ is at least as severe as some value, or is exactly
1154   some value.
1155
1156 * The fields of the payload of{nbsp}__E__ and the current context fields
1157   satisfy a filter expression.
1158
1159 A <<event,recording event rule>> has additional, implicit conditions to
1160 satisfy.
1161
1162
1163 [[instrumentation-point-types]]
1164 ==== Instrumentation point types
1165
1166 As of LTTng{nbsp}{revision}, the available instrumentation point
1167 types are, depending on the <<domain,tracing domain>>:
1168
1169 Linux kernel::
1170     LTTng tracepoint:::
1171         A statically defined point in the source code of the kernel
1172         image or of a kernel module using the
1173         <<lttng-modules,LTTng-modules>> macros.
1174
1175     Linux kernel system call:::
1176         Entry, exit, or both of a Linux kernel system call.
1177
1178     Linux https://www.kernel.org/doc/html/latest/trace/kprobes.html[kprobe]:::
1179         A single probe dynamically placed in the compiled kernel code.
1180 +
1181 When you create such an instrumentation point, you set its memory
1182 address or symbol name.
1183
1184     Linux user space probe:::
1185         A single probe dynamically placed at the entry of a compiled
1186         user space application/library function through the kernel.
1187 +
1188 When you create such an instrumentation point, you set:
1189 +
1190 --
1191 With the ELF method::
1192     Its application/library path and its symbol name.
1193
1194 With the USDT method::
1195     Its application/library path, its provider name, and its probe name.
1196 +
1197 ``USDT'' stands for _SystemTap User-level Statically Defined Tracing_,
1198 a http://dtrace.org/blogs/about/[DTrace]-style marker.
1199 --
1200 +
1201 As of LTTng{nbsp}{revision}, LTTng only supports USDT probes which
1202 are _not_ reference-counted.
1203
1204     Linux https://www.kernel.org/doc/html/latest/trace/kprobes.html[kretprobe]:::
1205         Entry, exit, or both of a Linux kernel function.
1206 +
1207 When you create such an instrumentation point, you set the memory
1208 address or symbol name of its function.
1209
1210 User space::
1211     LTTng tracepoint:::
1212         A statically defined point in the source code of a C/$$C++$$
1213         application/library using the
1214         <<lttng-ust,LTTng-UST>> macros.
1215
1216 `java.util.logging`, Apache log4j, and Python::
1217     Java or Python logging statement:::
1218         A method call on a Java or Python logger attached to an
1219         LTTng-UST handler.
1220
1221 See ``<<list-instrumentation-points,List the available instrumentation
1222 points>>'' to learn how to list available Linux kernel, user space, and
1223 logging instrumentation points.
1224
1225
1226 [[trigger]]
1227 === Trigger
1228
1229 A _trigger_ associates a condition to one or more actions.
1230
1231 When the condition of a trigger is satisfied, LTTng attempts to execute
1232 its actions.
1233
1234 As of LTTng{nbsp}{revision}, the available trigger conditions and
1235 actions are:
1236
1237 Conditions::
1238 +
1239 * The consumed buffer size of a given <<tracing-session,recording
1240   session>> becomes greater than some value.
1241
1242 * The buffer usage of a given <<channel,channel>> becomes greater than
1243   some value.
1244
1245 * The buffer usage of a given channel becomes less than some value.
1246
1247 * There's an ongoing <<session-rotation,recording session rotation>>.
1248
1249 * A recording session rotation becomes completed.
1250
1251 * An <<add-event-rule-matches-trigger,event rule matches>> an event.
1252
1253 Actions::
1254 +
1255 * <<trigger-event-notif,Send a notification>> to a user application.
1256 * <<basic-tracing-session-control,Start>> a given recording session.
1257 * <<basic-tracing-session-control,Stop>> a given recording session.
1258 * <<session-rotation,Archive the current trace chunk>> of a given
1259   recording session (rotate).
1260 * <<taking-a-snapshot,Take a snapshot>> of a given recording session.
1261
1262 A trigger belongs to a <<lttng-sessiond,session daemon>>, not to a
1263 specific recording session. For a given session daemon, each Unix user has
1264 its own, private triggers. Note, however, that the `root` Unix user may,
1265 for the root session daemon:
1266
1267 * Add a trigger as another Unix user.
1268
1269 * List all the triggers, regardless of their owner.
1270
1271 * Remove a trigger which belongs to another Unix user.
1272
1273 For a given session daemon and Unix user, a trigger has a unique name.
1274
1275
1276 [[tracing-session]]
1277 === Recording session
1278
1279 A _recording session_ (named ``tracing session'' prior to
1280 LTTng{nbsp}2.13) is a stateful dialogue between you and a
1281 <<lttng-sessiond,session daemon>> for everything related to
1282 <<event,event recording>>.
1283
1284 Everything that you do when you control LTTng tracers to record events
1285 happens within a recording session. In particular, a recording session:
1286
1287 * Has its own name, unique for a given session daemon.
1288
1289 * Has its own set of trace files, if any.
1290
1291 * Has its own state of activity (started or stopped).
1292 +
1293 An active recording session is an implicit <<event,recording event rule>>
1294 condition.
1295
1296 * Has its own <<tracing-session-mode,mode>> (local, network streaming,
1297   snapshot, or live).
1298
1299 * Has its own <<channel,channels>> to which are attached their own
1300   recording event rules.
1301
1302 * Has its own <<pid-tracking,process attribute inclusion sets>>.
1303
1304 [role="img-100"]
1305 .A _recording session_ contains <<channel,channels>> that are members of <<domain,tracing domains>> and contain <<event,recording event rules>>.
1306 image::concepts.png[]
1307
1308 Those attributes and objects are completely isolated between different
1309 recording sessions.
1310
1311 A recording session is like an
1312 https://en.wikipedia.org/wiki/Automated_teller_machine[ATM] session: the
1313 operations you do on the banking system through the ATM don't alter the
1314 data of other users of the same system. In the case of the ATM, a
1315 session lasts as long as your bank card is inside. In the case of LTTng,
1316 a recording session lasts from the man:lttng-create(1) command to the
1317 man:lttng-destroy(1) command.
1318
1319 [role="img-100"]
1320 .Each Unix user has its own set of recording sessions.
1321 image::many-sessions.png[]
1322
1323 A recording session belongs to a <<lttng-sessiond,session daemon>>. For a
1324 given session daemon, each Unix user has its own, private recording
1325 sessions. Note, however, that the `root` Unix user may operate on or
1326 destroy another user's recording session.
1327
1328
1329 [[tracing-session-mode]]
1330 ==== Recording session mode
1331
1332 LTTng offers four recording session modes:
1333
1334 [[local-mode]]Local mode::
1335     Write the trace data to the local file system.
1336
1337 [[net-streaming-mode]]Network streaming mode::
1338     Send the trace data over the network to a listening
1339     <<lttng-relayd,relay daemon>>.
1340
1341 [[snapshot-mode]]Snapshot mode::
1342     Only write the trace data to the local file system or send it to a
1343     listening relay daemon when LTTng <<taking-a-snapshot,takes a
1344     snapshot>>.
1345 +
1346 LTTng forces all the <<channel,channels>>
1347 to be created to be configured to be snapshot-ready.
1348 +
1349 LTTng takes a snapshot of such a recording session when:
1350 +
1351 --
1352 * You run the man:lttng-snapshot(1) command.
1353
1354 * LTTng executes a `snapshot-session` <<trigger,trigger>> action.
1355 --
1356
1357 [[live-mode]]Live mode::
1358     Send the trace data over the network to a listening relay daemon
1359     for <<lttng-live,live reading>>.
1360 +
1361 An LTTng live reader (for example, man:babeltrace2(1)) can connect to
1362 the same relay daemon to receive trace data while the recording session is
1363 active.
1364
1365
1366 [[domain]]
1367 === Tracing domain
1368
1369 A _tracing domain_ identifies a type of LTTng tracer.
1370
1371 A tracing domain has its own properties and features.
1372
1373 There are currently five available tracing domains:
1374
1375 * Linux kernel
1376 * User space
1377 * `java.util.logging` (JUL)
1378 * log4j
1379 * Python
1380
1381 You must specify a tracing domain to target a type of LTTng tracer when
1382 using some <<lttng-cli,cmd:lttng>> commands to avoid ambiguity. For
1383 example, because the Linux kernel and user space tracing domains support
1384 named tracepoints as <<event-rule,instrumentation points>>, you need to
1385 specify a tracing domain when you <<enabling-disabling-events,create
1386 an event rule>> because both tracing domains could have tracepoints
1387 sharing the same name.
1388
1389 You can create <<channel,channels>> in the Linux kernel and user space
1390 tracing domains. The other tracing domains have a single, default
1391 channel.
1392
1393
1394 [[channel]]
1395 === Channel and ring buffer
1396
1397 A _channel_ is an object which is responsible for a set of
1398 _ring buffers_.
1399
1400 Each ring buffer is divided into multiple _sub-buffers_. When a
1401 <<event,recording event rule>>
1402 matches an event, LTTng can record it to one or more sub-buffers of one
1403 or more channels.
1404
1405 When you <<enabling-disabling-channels,create a channel>>, you set its
1406 final attributes, that is:
1407
1408 * Its <<channel-buffering-schemes,buffering scheme>>.
1409
1410 * What to do <<channel-overwrite-mode-vs-discard-mode,when there's no
1411   space left>> for a new event record because all sub-buffers are full.
1412
1413 * The <<channel-subbuf-size-vs-subbuf-count,size of each ring buffer and
1414   how many sub-buffers>> a ring buffer has.
1415
1416 * The <<tracefile-rotation,size of each trace file LTTng writes for this
1417   channel and the maximum count>> of trace files.
1418
1419 * The periods of its <<channel-read-timer,read>>,
1420   <<channel-switch-timer,switch>>, and <<channel-monitor-timer,monitor>>
1421   timers.
1422
1423 * For a Linux kernel channel: its output type.
1424 +
1425 See the opt:lttng-enable-channel(1):--output option of the
1426 man:lttng-enable-channel(1) command.
1427
1428 * For a user space channel: the value of its
1429   <<blocking-timeout-example,blocking timeout>>.
1430
1431 A channel is always associated to a <<domain,tracing domain>>. The
1432 `java.util.logging` (JUL), log4j, and Python tracing domains each have a
1433 default channel which you can't configure.
1434
1435 A channel owns <<event,recording event rules>>.
1436
1437
1438 [[channel-buffering-schemes]]
1439 ==== Buffering scheme
1440
1441 A channel has at least one ring buffer _per CPU_. LTTng always records
1442 an event to the ring buffer dedicated to the CPU which emits it.
1443
1444 The buffering scheme of a user space channel determines what has its own
1445 set of per-CPU ring buffers:
1446
1447 Per-user buffering::
1448   Allocate one set of ring buffers--one per CPU--shared by all the
1449   instrumented processes of:
1450     If your Unix user is `root`:::
1451       Each Unix user.
1452 +
1453 --
1454 [role="img-100"]
1455 .Per-user buffering scheme (recording session belongs to the `root` Unix user).
1456 image::per-user-buffering-root.png[]
1457 --
1458
1459     Otherwise:::
1460       Your Unix user.
1461 +
1462 --
1463 [role="img-100"]
1464 .Per-user buffering scheme (recording session belongs to the `Bob` Unix user).
1465 image::per-user-buffering.png[]
1466 --
1467
1468 Per-process buffering::
1469   Allocate one set of ring buffers--one per CPU--for each
1470   instrumented process of:
1471     If your Unix user is `root`:::
1472       All Unix users.
1473 +
1474 --
1475 [role="img-100"]
1476 .Per-process buffering scheme (recording session belongs to the `root` Unix user).
1477 image::per-process-buffering-root.png[]
1478 --
1479
1480     Otherwise:::
1481       Your Unix user.
1482 +
1483 --
1484 [role="img-100"]
1485 .Per-process buffering scheme (recording session belongs to the `Bob` Unix user).
1486 image::per-process-buffering.png[]
1487 --
1488
1489 The per-process buffering scheme tends to consume more memory than the
1490 per-user option because systems generally have more instrumented
1491 processes than Unix users running instrumented processes. However, the
1492 per-process buffering scheme ensures that one process having a high
1493 event throughput won't fill all the shared sub-buffers of the same Unix
1494 user, only its own.
1495
1496 The buffering scheme of a Linux kernel channel is always to allocate a
1497 single set of ring buffers for the whole system. This scheme is similar
1498 to the per-user option, but with a single, global user ``running'' the
1499 kernel.
1500
1501
1502 [[channel-overwrite-mode-vs-discard-mode]]
1503 ==== Event record loss mode
1504
1505 When LTTng emits an event, LTTng can record it to a specific, available
1506 sub-buffer within the ring buffers of specific channels. When there's no
1507 space left in a sub-buffer, the tracer marks it as consumable and
1508 another, available sub-buffer starts receiving the following event
1509 records. An LTTng <<lttng-consumerd,consumer daemon>> eventually
1510 consumes the marked sub-buffer, which returns to the available state.
1511
1512 [NOTE]
1513 [role="docsvg-channel-subbuf-anim"]
1514 ====
1515 {note-no-anim}
1516 ====
1517
1518 In an ideal world, sub-buffers are consumed faster than they're filled.
1519 In the real world, however, all sub-buffers can be full at some point,
1520 leaving no space to record the following events.
1521
1522 In an ideal world, sub-buffers are consumed faster than they're filled,
1523 as it's the case in the previous animation. In the real world,
1524 however, all sub-buffers can be full at some point, leaving no space to
1525 record the following events.
1526
1527 By default, <<lttng-modules,LTTng-modules>> and <<lttng-ust,LTTng-UST>>
1528 are _non-blocking_ tracers: when there's no available sub-buffer to
1529 record an event, it's acceptable to lose event records when the
1530 alternative would be to cause substantial delays in the execution of the
1531 instrumented application. LTTng privileges performance over integrity;
1532 it aims at perturbing the instrumented application as little as possible
1533 in order to make the detection of subtle race conditions and rare
1534 interrupt cascades possible.
1535
1536 Since LTTng{nbsp}2.10, the LTTng user space tracer, LTTng-UST, supports
1537 a _blocking mode_. See the <<blocking-timeout-example,blocking timeout
1538 example>> to learn how to use the blocking mode.
1539
1540 When it comes to losing event records because there's no available
1541 sub-buffer, or because the blocking timeout of
1542 the channel is reached, the _event record loss mode_ of the channel
1543 determines what to do. The available event record loss modes are:
1544
1545 [[discard-mode]]Discard mode::
1546     Drop the newest event records until a sub-buffer becomes available.
1547 +
1548 This is the only available mode when you specify a blocking timeout.
1549 +
1550 With this mode, LTTng increments a count of lost event records when an
1551 event record is lost and saves this count to the trace. A trace reader
1552 can use the saved discarded event record count of the trace to decide
1553 whether or not to perform some analysis even if trace data is known to
1554 be missing.
1555
1556 [[overwrite-mode]]Overwrite mode::
1557     Clear the sub-buffer containing the oldest event records and start
1558     writing the newest event records there.
1559 +
1560 This mode is sometimes called _flight recorder mode_ because it's
1561 similar to a https://en.wikipedia.org/wiki/Flight_recorder[flight
1562 recorder]: always keep a fixed amount of the latest data. It's also
1563 similar to the roll mode of an oscilloscope.
1564 +
1565 Since LTTng{nbsp}2.8, with this mode, LTTng writes to a given sub-buffer
1566 its sequence number within its data stream. With a <<local-mode,local>>,
1567 <<net-streaming-mode,network streaming>>, or <<live-mode,live>> recording
1568 session, a trace reader can use such sequence numbers to report lost
1569 packets. A trace reader can use the saved discarded sub-buffer (packet)
1570 count of the trace to decide whether or not to perform some analysis
1571 even if trace data is known to be missing.
1572 +
1573 With this mode, LTTng doesn't write to the trace the exact number of
1574 lost event records in the lost sub-buffers.
1575
1576 Which mechanism you should choose depends on your context: prioritize
1577 the newest or the oldest event records in the ring buffer?
1578
1579 Beware that, in overwrite mode, the tracer abandons a _whole sub-buffer_
1580 as soon as a there's no space left for a new event record, whereas in
1581 discard mode, the tracer only discards the event record that doesn't
1582 fit.
1583
1584 There are a few ways to decrease your probability of losing event
1585 records. The ``<<channel-subbuf-size-vs-subbuf-count,Sub-buffer size and
1586 count>>'' section shows how to fine-tune the sub-buffer size and count
1587 of a channel to virtually stop losing event records, though at the cost
1588 of greater memory usage.
1589
1590
1591 [[channel-subbuf-size-vs-subbuf-count]]
1592 ==== Sub-buffer size and count
1593
1594 A channel has one or more ring buffer for each CPU of the target system.
1595
1596 See the ``<<channel-buffering-schemes,Buffering scheme>>'' section to
1597 learn how many ring buffers of a given channel are dedicated to each CPU
1598 depending on its buffering scheme.
1599
1600 Set the size of each sub-buffer the ring buffers of a channel contain
1601 and how many there are
1602 when you <<enabling-disabling-channels,create it>>.
1603
1604 Note that LTTng switching the current sub-buffer of a ring buffer
1605 (marking a full one as consumable and switching to an available one for
1606 LTTng to record the next events) introduces noticeable CPU overhead.
1607 Knowing this, the following list presents a few practical situations
1608 along with how to configure the sub-buffer size and count for them:
1609
1610 High event throughput::
1611     In general, prefer large sub-buffers to lower the risk of losing
1612     event records.
1613 +
1614 Having larger sub-buffers also ensures a lower sub-buffer switching
1615 frequency.
1616 +
1617 The sub-buffer count is only meaningful if you create the channel in
1618 <<overwrite-mode,overwrite mode>>: in this case, if LTTng overwrites a
1619 sub-buffer, then the other sub-buffers are left unaltered.
1620
1621 Low event throughput::
1622     In general, prefer smaller sub-buffers since the risk of losing
1623     event records is low.
1624 +
1625 Because LTTng emits events less frequently, the sub-buffer switching
1626 frequency should remain low and therefore the overhead of the tracer
1627 shouldn't be a problem.
1628
1629 Low memory system::
1630     If your target system has a low memory limit, prefer fewer first,
1631     then smaller sub-buffers.
1632 +
1633 Even if the system is limited in memory, you want to keep the
1634 sub-buffers as large as possible to avoid a high sub-buffer switching
1635 frequency.
1636
1637 Note that LTTng uses https://diamon.org/ctf/[CTF] as its trace format,
1638 which means event record data is very compact. For example, the average
1639 LTTng kernel event record weights about 32{nbsp}bytes. Therefore, a
1640 sub-buffer size of 1{nbsp}MiB is considered large.
1641
1642 The previous scenarios highlight the major trade-off between a few large
1643 sub-buffers and more, smaller sub-buffers: sub-buffer switching
1644 frequency vs. how many event records are lost in overwrite mode.
1645 Assuming a constant event throughput and using the overwrite mode, the
1646 two following configurations have the same ring buffer total size:
1647
1648 [NOTE]
1649 [role="docsvg-channel-subbuf-size-vs-count-anim"]
1650 ====
1651 {note-no-anim}
1652 ====
1653
1654 Two sub-buffers of 4{nbsp}MiB each::
1655     Expect a very low sub-buffer switching frequency, but if LTTng
1656     ever needs to overwrite a sub-buffer, half of the event records so
1657     far (4{nbsp}MiB) are definitely lost.
1658
1659 Eight sub-buffers of 1{nbsp}MiB each::
1660     Expect four times the tracer overhead of the configuration above,
1661     but if LTTng needs to overwrite a sub-buffer, only the eighth of
1662     event records so far (1{nbsp}MiB) are definitely lost.
1663
1664 In <<discard-mode,discard mode>>, the sub-buffer count parameter is
1665 pointless: use two sub-buffers and set their size according to your
1666 requirements.
1667
1668
1669 [[tracefile-rotation]]
1670 ==== Maximum trace file size and count (trace file rotation)
1671
1672 By default, trace files can grow as large as needed.
1673
1674 Set the maximum size of each trace file that LTTng writes of a given
1675 channel when you <<enabling-disabling-channels,create it>>.
1676
1677 When the size of a trace file reaches the fixed maximum size of the
1678 channel, LTTng creates another file to contain the next event records.
1679 LTTng appends a file count to each trace file name in this case.
1680
1681 If you set the trace file size attribute when you create a channel, the
1682 maximum number of trace files that LTTng creates is _unlimited_ by
1683 default. To limit them, set a maximum number of trace files. When the
1684 number of trace files reaches the fixed maximum count of the channel,
1685 LTTng overwrites the oldest trace file. This mechanism is called _trace
1686 file rotation_.
1687
1688 [IMPORTANT]
1689 ====
1690 Even if you don't limit the trace file count, always assume that LTTng
1691 manages all the trace files of the recording session.
1692
1693 In other words, there's no safe way to know if LTTng still holds a given
1694 trace file open with the trace file rotation feature.
1695
1696 The only way to obtain an unmanaged, self-contained LTTng trace before
1697 you <<creating-destroying-tracing-sessions,destroy the recording session>>
1698 is with the <<session-rotation,recording session rotation>> feature, which
1699 is available since LTTng{nbsp}2.11.
1700 ====
1701
1702
1703 [[channel-timers]]
1704 ==== Timers
1705
1706 Each channel can have up to three optional timers:
1707
1708 [[channel-switch-timer]]Switch timer::
1709     When this timer expires, a sub-buffer switch happens: for each ring
1710     buffer of the channel, LTTng marks the current sub-buffer as
1711     consumable and _switches_ to an available one to record the next
1712     events.
1713 +
1714 [NOTE]
1715 [role="docsvg-channel-switch-timer"]
1716 ====
1717 {note-no-anim}
1718 ====
1719 +
1720 A switch timer is useful to ensure that LTTng consumes and commits trace
1721 data to trace files or to a distant <<lttng-relayd,relay daemon>>
1722 periodically in case of a low event throughput.
1723 +
1724 Such a timer is also convenient when you use large
1725 <<channel-subbuf-size-vs-subbuf-count,sub-buffers>> to cope with a
1726 sporadic high event throughput, even if the throughput is otherwise low.
1727 +
1728 Set the period of the switch timer of a channel when you
1729 <<enabling-disabling-channels,create it>> with
1730 the opt:lttng-enable-channel(1):--switch-timer option.
1731
1732 [[channel-read-timer]]Read timer::
1733     When this timer expires, LTTng checks for full, consumable
1734     sub-buffers.
1735 +
1736 By default, the LTTng tracers use an asynchronous message mechanism to
1737 signal a full sub-buffer so that a <<lttng-consumerd,consumer daemon>>
1738 can consume it.
1739 +
1740 When such messages must be avoided, for example in real-time
1741 applications, use this timer instead.
1742 +
1743 Set the period of the read timer of a channel when you
1744 <<enabling-disabling-channels,create it>> with the
1745 opt:lttng-enable-channel(1):--read-timer option.
1746
1747 [[channel-monitor-timer]]Monitor timer::
1748     When this timer expires, the consumer daemon samples some channel
1749     statistics to evaluate the following <<trigger,trigger>>
1750     conditions:
1751 +
1752 --
1753 . The consumed buffer size of a given <<tracing-session,recording
1754   session>> becomes greater than some value.
1755 . The buffer usage of a given channel becomes greater than some value.
1756 . The buffer usage of a given channel becomes less than some value.
1757 --
1758 +
1759 If you disable the monitor timer of a channel{nbsp}__C__:
1760 +
1761 --
1762 * The consumed buffer size value of the recording session of{nbsp}__C__
1763   could be wrong for trigger condition type{nbsp}1: the consumed buffer
1764   size of{nbsp}__C__ won't be part of the grand total.
1765
1766 * The buffer usage trigger conditions (types{nbsp}2 and{nbsp}3)
1767   for{nbsp}__C__ will never be satisfied.
1768 --
1769 +
1770 Set the period of the monitor timer of a channel when you
1771 <<enabling-disabling-channels,create it>> with the
1772 opt:lttng-enable-channel(1):--monitor-timer option.
1773
1774
1775 [[event]]
1776 === Recording event rule and event record
1777
1778 A _recording event rule_ is a specific type of <<event-rule,event rule>>
1779 of which the action is to serialize and record the matched event as an
1780 _event record_.
1781
1782 Set the explicit conditions of a recording event rule when you
1783 <<enabling-disabling-events,create it>>. A recording event rule also has
1784 the following implicit conditions:
1785
1786 * The recording event rule itself is enabled.
1787 +
1788 A recording event rule is enabled on creation.
1789
1790 * The <<channel,channel>> to which the recording event rule is attached
1791   is enabled.
1792 +
1793 A channel is enabled on creation.
1794
1795 * The <<tracing-session,recording session>> of the recording event rule is
1796   <<basic-tracing-session-control,active>> (started).
1797 +
1798 A recording session is inactive (stopped) on creation.
1799
1800 * The process for which LTTng creates an event to match is
1801   <<pid-tracking,allowed to record events>>.
1802 +
1803 All processes are allowed to record events on recording session
1804 creation.
1805
1806 You always attach a recording event rule to a channel, which belongs to
1807 a recording session, when you create it.
1808
1809 When a recording event rule{nbsp}__ER__ matches an event{nbsp}__E__,
1810 LTTng attempts to serialize and record{nbsp}__E__ to one of the
1811 available sub-buffers of the channel to which{nbsp}__E__ is attached.
1812
1813 When multiple matching recording event rules are attached to the same
1814 channel, LTTng attempts to serialize and record the matched event
1815 _once_. In the following example, the second recording event rule is
1816 redundant when both are enabled:
1817
1818 [role="term"]
1819 ----
1820 $ lttng enable-event --userspace hello:world
1821 $ lttng enable-event --userspace hello:world --loglevel=INFO
1822 ----
1823
1824 [role="img-100"]
1825 .Logical path from an instrumentation point to an event record.
1826 image::event-rule.png[]
1827
1828 As of LTTng{nbsp}{revision}, you cannot remove a recording event
1829 rule: it exists as long as its recording session exists.
1830
1831
1832 [[plumbing]]
1833 == Components of noch:{LTTng}
1834
1835 The second _T_ in _LTTng_ stands for _toolkit_: it would be wrong
1836 to call LTTng a simple _tool_ since it's composed of multiple
1837 interacting components.
1838
1839 This section describes those components, explains their respective
1840 roles, and shows how they connect together to form the LTTng ecosystem.
1841
1842 The following diagram shows how the most important components of LTTng
1843 interact with user applications, the Linux kernel, and you:
1844
1845 [role="img-100"]
1846 .Control and trace data paths between LTTng components.
1847 image::plumbing.png[]
1848
1849 The LTTng project integrates:
1850
1851 LTTng-tools::
1852     Libraries and command-line interface to control recording sessions:
1853 +
1854 * <<lttng-sessiond,Session daemon>> (man:lttng-sessiond(8)).
1855 * <<lttng-consumerd,Consumer daemon>> (cmd:lttng-consumerd).
1856 * <<lttng-relayd,Relay daemon>> (man:lttng-relayd(8)).
1857 * <<liblttng-ctl-lttng,Tracing control library>> (`liblttng-ctl`).
1858 * <<lttng-cli,Tracing control command-line tool>> (man:lttng(1)).
1859 * <<persistent-memory-file-systems,`lttng-crash` command-line tool>>
1860   (man:lttng-crash(1)).
1861
1862 LTTng-UST::
1863     Libraries and Java/Python packages to instrument and trace user
1864     applications:
1865 +
1866 * <<lttng-ust,User space tracing library>> (`liblttng-ust`) and its
1867   headers to instrument and trace any native user application.
1868 * <<prebuilt-ust-helpers,Preloadable user space tracing helpers>>:
1869 ** `liblttng-ust-libc-wrapper`
1870 ** `liblttng-ust-pthread-wrapper`
1871 ** `liblttng-ust-cyg-profile`
1872 ** `liblttng-ust-cyg-profile-fast`
1873 ** `liblttng-ust-dl`
1874 * <<lttng-ust-agents,LTTng-UST Java agent>> to instrument and trace
1875   Java applications using `java.util.logging` or
1876   Apache log4j{nbsp}1.2 logging.
1877 * <<lttng-ust-agents,LTTng-UST Python agent>> to instrument
1878   Python applications using the standard `logging` package.
1879
1880 LTTng-modules::
1881     <<lttng-modules,Linux kernel modules>> to instrument and trace the
1882     kernel:
1883 +
1884 * LTTng kernel tracer module.
1885 * Recording ring buffer kernel modules.
1886 * Probe kernel modules.
1887 * LTTng logger kernel module.
1888
1889
1890 [[lttng-cli]]
1891 === Tracing control command-line interface
1892
1893 The _man:lttng(1) command-line tool_ is the standard user interface to
1894 control LTTng <<tracing-session,recording sessions>>.
1895
1896 The cmd:lttng tool is part of LTTng-tools.
1897
1898 The cmd:lttng tool is linked with
1899 <<liblttng-ctl-lttng,`liblttng-ctl`>> to communicate with
1900 one or more <<lttng-sessiond,session daemons>> behind the scenes.
1901
1902 The cmd:lttng tool has a Git-like interface:
1903
1904 [role="term"]
1905 ----
1906 $ lttng [GENERAL OPTIONS] <COMMAND> [COMMAND OPTIONS]
1907 ----
1908
1909 The ``<<controlling-tracing,Tracing control>>'' section explores the
1910 available features of LTTng through its cmd:lttng tool.
1911
1912
1913 [[liblttng-ctl-lttng]]
1914 === Tracing control library
1915
1916 [role="img-100"]
1917 .The tracing control library.
1918 image::plumbing-liblttng-ctl.png[]
1919
1920 The _LTTng control library_, `liblttng-ctl`, is used to communicate with
1921 a <<lttng-sessiond,session daemon>> using a C{nbsp}API that hides the
1922 underlying details of the protocol.
1923
1924 `liblttng-ctl` is part of LTTng-tools.
1925
1926 The <<lttng-cli,cmd:lttng command-line tool>> is linked with
1927 `liblttng-ctl`.
1928
1929 Use `liblttng-ctl` in C or $$C++$$ source code by including its
1930 ``master'' header:
1931
1932 [source,c]
1933 ----
1934 #include <lttng/lttng.h>
1935 ----
1936
1937 As of LTTng{nbsp}{revision}, the best available developer documentation
1938 for `liblttng-ctl` is its installed header files. Functions and
1939 structures are documented with header comments.
1940
1941
1942 [[lttng-ust]]
1943 === User space tracing library
1944
1945 [role="img-100"]
1946 .The user space tracing library.
1947 image::plumbing-liblttng-ust.png[]
1948
1949 The _user space tracing library_, `liblttng-ust` (see man:lttng-ust(3)),
1950 is the LTTng user space tracer.
1951
1952 `liblttng-ust` receives commands from a <<lttng-sessiond,session
1953 daemon>>, for example to allow specific instrumentation points to emit
1954 LTTng <<event-rule,events>>, and writes event records to <<channel,ring
1955 buffers>> shared with a <<lttng-consumerd,consumer daemon>>.
1956
1957 `liblttng-ust` is part of LTTng-UST.
1958
1959 `liblttng-ust` can also send asynchronous messages to the session daemon
1960 when it emits an event. This supports the ``event rule matches''
1961 <<trigger,trigger>> condition feature (see
1962 “<<add-event-rule-matches-trigger,Add an ``event rule matches'' trigger
1963 to a session daemon>>”).
1964
1965 Public C{nbsp}header files are installed beside `liblttng-ust` to
1966 instrument any <<c-application,C or $$C++$$ application>>.
1967
1968 <<lttng-ust-agents,LTTng-UST agents>>, which are regular Java and Python
1969 packages, use their own <<tracepoint-provider,tracepoint provider
1970 package>> which is linked with `liblttng-ust`.
1971
1972 An application or library doesn't have to initialize `liblttng-ust`
1973 manually: its constructor does the necessary tasks to register the
1974 application to a session daemon. The initialization phase also
1975 configures instrumentation points depending on the <<event-rule,event
1976 rules>> that you already created.
1977
1978
1979 [[lttng-ust-agents]]
1980 === User space tracing agents
1981
1982 [role="img-100"]
1983 .The user space tracing agents.
1984 image::plumbing-lttng-ust-agents.png[]
1985
1986 The _LTTng-UST Java and Python agents_ are regular Java and Python
1987 packages which add LTTng tracing capabilities to the
1988 native logging frameworks.
1989
1990 The LTTng-UST agents are part of LTTng-UST.
1991
1992 In the case of Java, the
1993 https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[`java.util.logging`
1994 core logging facilities] and
1995 https://logging.apache.org/log4j/1.2/[Apache log4j{nbsp}1.2] are supported.
1996 Note that Apache Log4j{nbsp}2 isn't supported.
1997
1998 In the case of Python, the standard
1999 https://docs.python.org/3/library/logging.html[`logging`] package
2000 is supported. Both Python{nbsp}2 and Python{nbsp}3 modules can import the
2001 LTTng-UST Python agent package.
2002
2003 The applications using the LTTng-UST agents are in the
2004 `java.util.logging` (JUL), log4j, and Python <<domain,tracing domains>>.
2005
2006 Both agents use the same mechanism to convert log statements to LTTng
2007 events. When an agent initializes, it creates a log handler that
2008 attaches to the root logger. The agent also registers to a
2009 <<lttng-sessiond,session daemon>>. When the user application executes a
2010 log statement, the root logger passes it to the log handler of the
2011 agent. The custom log handler of the agent calls a native function in a
2012 tracepoint provider package shared library linked with
2013 <<lttng-ust,`liblttng-ust`>>, passing the formatted log message and
2014 other fields, like its logger name and its log level. This native
2015 function contains a user space instrumentation point, therefore tracing
2016 the log statement.
2017
2018 The log level condition of a <<event,recording event rule>> is
2019 considered when tracing a Java or a Python application, and it's
2020 compatible with the standard `java.util.logging`, log4j, and Python log
2021 levels.
2022
2023
2024 [[lttng-modules]]
2025 === LTTng kernel modules
2026
2027 [role="img-100"]
2028 .The LTTng kernel modules.
2029 image::plumbing-lttng-modules.png[]
2030
2031 The _LTTng kernel modules_ are a set of Linux kernel modules
2032 which implement the kernel tracer of the LTTng project.
2033
2034 The LTTng kernel modules are part of LTTng-modules.
2035
2036 The LTTng kernel modules include:
2037
2038 * A set of _probe_ modules.
2039 +
2040 Each module attaches to a specific subsystem
2041 of the Linux kernel using its tracepoint instrument points.
2042 +
2043 There are also modules to attach to the entry and return points of the
2044 Linux system call functions.
2045
2046 * _Ring buffer_ modules.
2047 +
2048 A ring buffer implementation is provided as kernel modules. The LTTng
2049 kernel tracer writes to ring buffers; a
2050 <<lttng-consumerd,consumer daemon>> reads from ring buffers.
2051
2052 * The _LTTng kernel tracer_ module.
2053 * The <<proc-lttng-logger-abi,_LTTng logger_>> module.
2054 +
2055 The LTTng logger module implements the special path:{/proc/lttng-logger}
2056 (and path:{/dev/lttng-logger}, since LTTng{nbsp}2.11) files so that any
2057 executable can generate LTTng events by opening those files and
2058 writing to them.
2059
2060 The LTTng kernel tracer can also send asynchronous messages to the
2061 <<lttng-sessiond,session daemon>> when it emits an event.
2062 This supports the ``event rule matches''
2063 <<trigger,trigger>> condition feature (see
2064 “<<add-event-rule-matches-trigger,Add an ``event rule matches'' trigger
2065 to a session daemon>>”).
2066
2067 Generally, you don't have to load the LTTng kernel modules manually
2068 (using man:modprobe(8), for example): a root session daemon loads the
2069 necessary modules when starting. If you have extra probe modules, you
2070 can specify to load them to the session daemon on the command line
2071 (see the opt:lttng-sessiond(8):--extra-kmod-probes option).
2072
2073 The LTTng kernel modules are installed in
2074 +/usr/lib/modules/__release__/extra+ by default, where +__release__+ is
2075 the kernel release (output of `uname --kernel-release`).
2076
2077
2078 [[lttng-sessiond]]
2079 === Session daemon
2080
2081 [role="img-100"]
2082 .The session daemon.
2083 image::plumbing-sessiond.png[]
2084
2085 The _session daemon_, man:lttng-sessiond(8), is a
2086 https://en.wikipedia.org/wiki/Daemon_(computing)[daemon] which:
2087
2088 * Manages <<tracing-session,recording sessions>>.
2089
2090 * Controls the various components (like tracers and
2091   <<lttng-consumerd,consumer daemons>>) of LTTng.
2092
2093 * Sends <<notif-trigger-api,asynchronous notifications>> to user
2094   applications.
2095
2096 The session daemon is part of LTTng-tools.
2097
2098 The session daemon sends control requests to and receives control
2099 responses from:
2100
2101 * The <<lttng-ust,user space tracing library>>.
2102 +
2103 Any instance of the user space tracing library first registers to
2104 a session daemon. Then, the session daemon can send requests to
2105 this instance, such as:
2106 +
2107 --
2108 ** Get the list of tracepoints.
2109 ** Share a <<event,recording event rule>> so that the user space tracing
2110    library can decide whether or not a given tracepoint can emit events.
2111    Amongst the possible conditions of a recording event rule is a filter
2112    expression which `liblttng-ust` evaluates before it emits an event.
2113 ** Share <<channel,channel>> attributes and ring buffer locations.
2114 --
2115 +
2116 The session daemon and the user space tracing library use a Unix
2117 domain socket to communicate.
2118
2119 * The <<lttng-ust-agents,user space tracing agents>>.
2120 +
2121 Any instance of a user space tracing agent first registers to
2122 a session daemon. Then, the session daemon can send requests to
2123 this instance, such as:
2124 +
2125 --
2126 ** Get the list of loggers.
2127 ** Enable or disable a specific logger.
2128 --
2129 +
2130 The session daemon and the user space tracing agent use a TCP connection
2131 to communicate.
2132
2133 * The <<lttng-modules,LTTng kernel tracer>>.
2134 * The <<lttng-consumerd,consumer daemon>>.
2135 +
2136 The session daemon sends requests to the consumer daemon to instruct
2137 it where to send the trace data streams, amongst other information.
2138
2139 * The <<lttng-relayd,relay daemon>>.
2140
2141 The session daemon receives commands from the
2142 <<liblttng-ctl-lttng,tracing control library>>.
2143
2144 The session daemon can receive asynchronous messages from the
2145 <<lttng-ust,user space>> and <<lttng-modules,kernel>> tracers
2146 when they emit events. This supports the ``event rule matches''
2147 <<trigger,trigger>> condition feature (see
2148 “<<add-event-rule-matches-trigger,Add an ``event rule matches'' trigger
2149 to a session daemon>>”).
2150
2151 The root session daemon loads the appropriate
2152 <<lttng-modules,LTTng kernel modules>> on startup. It also spawns
2153 one or more <<lttng-consumerd,consumer daemons>> as soon as you create
2154 a <<event,recording event rule>>.
2155
2156 The session daemon doesn't send and receive trace data: this is the
2157 role of the <<lttng-consumerd,consumer daemon>> and
2158 <<lttng-relayd,relay daemon>>. It does, however, generate the
2159 https://diamon.org/ctf/[CTF] metadata stream.
2160
2161 Each Unix user can have its own session daemon instance. The
2162 recording sessions which different session daemons manage are completely
2163 independent.
2164
2165 The root user's session daemon is the only one which is
2166 allowed to control the LTTng kernel tracer, and its spawned consumer
2167 daemon is the only one which is allowed to consume trace data from the
2168 LTTng kernel tracer. Note, however, that any Unix user which is a member
2169 of the <<tracing-group,tracing group>> is allowed
2170 to create <<channel,channels>> in the
2171 Linux kernel <<domain,tracing domain>>, and therefore to use the Linux
2172 kernel LTTng tracer.
2173
2174 The <<lttng-cli,cmd:lttng command-line tool>> automatically starts a
2175 session daemon when using its `create` command if none is currently
2176 running. You can also start the session daemon manually.
2177
2178
2179 [[lttng-consumerd]]
2180 === Consumer daemon
2181
2182 [role="img-100"]
2183 .The consumer daemon.
2184 image::plumbing-consumerd.png[]
2185
2186 The _consumer daemon_, cmd:lttng-consumerd, is a
2187 https://en.wikipedia.org/wiki/Daemon_(computing)[daemon] which shares
2188 ring buffers with user applications or with the LTTng kernel modules to
2189 collect trace data and send it to some location (file system or to a
2190 <<lttng-relayd,relay daemon>> over the network).
2191
2192 The consumer daemon is part of LTTng-tools.
2193
2194 You don't start a consumer daemon manually: a consumer daemon is always
2195 spawned by a <<lttng-sessiond,session daemon>> as soon as you create a
2196 <<event,recording event rule>>, that is, before you start recording. When
2197 you kill its owner session daemon, the consumer daemon also exits
2198 because it's the child process of the session daemon. Command-line
2199 options of man:lttng-sessiond(8) target the consumer daemon process.
2200
2201 There are up to two running consumer daemons per Unix user, whereas only
2202 one session daemon can run per user. This is because each process can be
2203 either 32-bit or 64-bit: if the target system runs a mixture of 32-bit
2204 and 64-bit processes, it's more efficient to have separate
2205 corresponding 32-bit and 64-bit consumer daemons. The root user is an
2206 exception: it can have up to _three_ running consumer daemons: 32-bit
2207 and 64-bit instances for its user applications, and one more
2208 reserved for collecting kernel trace data.
2209
2210
2211 [[lttng-relayd]]
2212 === Relay daemon
2213
2214 [role="img-100"]
2215 .The relay daemon.
2216 image::plumbing-relayd.png[]
2217
2218 The _relay daemon_, man:lttng-relayd(8), is a
2219 https://en.wikipedia.org/wiki/Daemon_(computing)[daemon] acting as a bridge
2220 between remote session and consumer daemons, local trace files, and a
2221 remote live trace reader.
2222
2223 The relay daemon is part of LTTng-tools.
2224
2225 The main purpose of the relay daemon is to implement a receiver of
2226 <<sending-trace-data-over-the-network,trace data over the network>>.
2227 This is useful when the target system doesn't have much file system
2228 space to write trace files locally.
2229
2230 The relay daemon is also a server to which a
2231 <<lttng-live,live trace reader>> can
2232 connect. The live trace reader sends requests to the relay daemon to
2233 receive trace data as the target system records events. The
2234 communication protocol is named _LTTng live_; it's used over TCP
2235 connections.
2236
2237 Note that you can start the relay daemon on the target system directly.
2238 This is the setup of choice when the use case is to view/analyze events
2239 as the target system records them without the need of a remote system.
2240
2241
2242 [[instrumenting]]
2243 == [[using-lttng]]Instrumentation
2244
2245 There are many examples of tracing and monitoring in our everyday life:
2246
2247 * You have access to real-time and historical weather reports and
2248   forecasts thanks to weather stations installed around the country.
2249 * You know your heart is safe thanks to an electrocardiogram.
2250 * You make sure not to drive your car too fast and to have enough fuel
2251   to reach your destination thanks to gauges visible on your dashboard.
2252
2253 All the previous examples have something in common: they rely on
2254 **instruments**. Without the electrodes attached to the surface of your
2255 body skin, cardiac monitoring is futile.
2256
2257 LTTng, as a tracer, is no different from those real life examples. If
2258 you're about to trace a software system or, in other words, record its
2259 history of execution, you better have **instrumentation points** in the
2260 subject you're tracing, that is, the actual software system.
2261
2262 <<instrumentation-point-types,Various ways>> were developed to
2263 instrument a piece of software for LTTng tracing. The most
2264 straightforward one is to manually place static instrumentation points,
2265 called _tracepoints_, in the source code of the application. The Linux
2266 kernel <<domain,tracing domain>> also makes it possible to dynamically
2267 add instrumentation points.
2268
2269 If you're only interested in tracing the Linux kernel, your
2270 instrumentation needs are probably already covered by the built-in
2271 <<lttng-modules,Linux kernel instrumentation points>> of LTTng. You may
2272 also wish to have LTTng trace a user application which is already
2273 instrumented for LTTng tracing. In such cases, skip this whole section
2274 and read the topics of the ``<<controlling-tracing,Tracing control>>''
2275 section.
2276
2277 Many methods are available to instrument a piece of software for LTTng
2278 tracing:
2279
2280 * <<c-application,Instrument a C/$$C++$$ user application>>.
2281 * <<prebuilt-ust-helpers,Load a prebuilt user space tracing helper>>.
2282 * <<java-application,Instrument a Java application>>.
2283 * <<python-application,Instrument a Python application>>.
2284 * <<proc-lttng-logger-abi,Use the LTTng logger>>.
2285 * <<instrumenting-linux-kernel,Instrument a Linux kernel image or module>>.
2286
2287
2288 [[c-application]]
2289 === [[cxx-application]]Instrument a C/$$C++$$ user application
2290
2291 The high level procedure to instrument a C or $$C++$$ user application
2292 with the <<lttng-ust,LTTng user space tracing library>>, `liblttng-ust`,
2293 is:
2294
2295 . <<tracepoint-provider,Create the source files of a tracepoint provider
2296   package>>.
2297
2298 . <<probing-the-application-source-code,Add tracepoints to
2299   the source code of the application>>.
2300
2301 . <<building-tracepoint-providers-and-user-application,Build and link
2302   a tracepoint provider package and the user application>>.
2303
2304 If you need quick, man:printf(3)-like instrumentation, skip those steps
2305 and use <<tracef,`lttng_ust_tracef()`>> or
2306 <<tracelog,`lttng_ust_tracelog()`>> instead.
2307
2308 IMPORTANT: You need to <<installing-lttng,install>> LTTng-UST to
2309 instrument a user application with `liblttng-ust`.
2310
2311
2312 [[tracepoint-provider]]
2313 ==== Create the source files of a tracepoint provider package
2314
2315 A _tracepoint provider_ is a set of compiled functions which provide
2316 **tracepoints** to an application, the type of instrumentation point
2317 which LTTng-UST provides.
2318
2319 Those functions can make LTTng emit events with user-defined fields and
2320 serialize those events as event records to one or more LTTng-UST
2321 <<channel,channel>> sub-buffers. The `lttng_ust_tracepoint()` macro,
2322 which you <<probing-the-application-source-code,insert in the source
2323 code of a user application>>, calls those functions.
2324
2325 A _tracepoint provider package_ is an object file (`.o`) or a shared
2326 library (`.so`) which contains one or more tracepoint providers. Its
2327 source files are:
2328
2329 * One or more <<tpp-header,tracepoint provider header>> (`.h`).
2330 * A <<tpp-source,tracepoint provider package source>> (`.c`).
2331
2332 A tracepoint provider package is dynamically linked with `liblttng-ust`,
2333 the LTTng user space tracer, at run time.
2334
2335 [role="img-100"]
2336 .User application linked with `liblttng-ust` and containing a tracepoint provider.
2337 image::ust-app.png[]
2338
2339 NOTE: If you need quick, man:printf(3)-like instrumentation, skip
2340 creating and using a tracepoint provider and use
2341 <<tracef,`lttng_ust_tracef()`>> or <<tracelog,`lttng_ust_tracelog()`>>
2342 instead.
2343
2344
2345 [[tpp-header]]
2346 ===== Create a tracepoint provider header file template
2347
2348 A _tracepoint provider header file_ contains the tracepoint definitions
2349 of a tracepoint provider.
2350
2351 To create a tracepoint provider header file:
2352
2353 . Start from this template:
2354 +
2355 --
2356 [source,c]
2357 .Tracepoint provider header file template (`.h` file extension).
2358 ----
2359 #undef LTTNG_UST_TRACEPOINT_PROVIDER
2360 #define LTTNG_UST_TRACEPOINT_PROVIDER provider_name
2361
2362 #undef LTTNG_UST_TRACEPOINT_INCLUDE
2363 #define LTTNG_UST_TRACEPOINT_INCLUDE "./tp.h"
2364
2365 #if !defined(_TP_H) || defined(LTTNG_UST_TRACEPOINT_HEADER_MULTI_READ)
2366 #define _TP_H
2367
2368 #include <lttng/tracepoint.h>
2369
2370 /*
2371  * Use LTTNG_UST_TRACEPOINT_EVENT(), LTTNG_UST_TRACEPOINT_EVENT_CLASS(),
2372  * LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(), and
2373  * LTTNG_UST_TRACEPOINT_LOGLEVEL() here.
2374  */
2375
2376 #endif /* _TP_H */
2377
2378 #include <lttng/tracepoint-event.h>
2379 ----
2380 --
2381 +
2382 Replace:
2383 +
2384 * +__provider_name__+ with the name of your tracepoint provider.
2385 * `"tp.h"` with the name of your tracepoint provider header file.
2386
2387 . Below the `#include <lttng/tracepoint.h>` line, put your
2388   <<defining-tracepoints,tracepoint definitions>>.
2389
2390 Your tracepoint provider name must be unique amongst all the possible
2391 tracepoint provider names used on the same target system. We suggest to
2392 include the name of your project or company in the name, for example,
2393 `org_lttng_my_project_tpp`.
2394
2395
2396 [[defining-tracepoints]]
2397 ===== Create a tracepoint definition
2398
2399 A _tracepoint definition_ defines, for a given tracepoint:
2400
2401 * Its **input arguments**.
2402 +
2403 They're the macro parameters that the `lttng_ust_tracepoint()` macro
2404 accepts for this particular tracepoint in the source code of the user
2405 application.
2406
2407 * Its **output event fields**.
2408 +
2409 They're the sources of event fields that form the payload of any event
2410 that the execution of the `lttng_ust_tracepoint()` macro emits for this
2411 particular tracepoint.
2412
2413 Create a tracepoint definition with the
2414 `LTTNG_UST_TRACEPOINT_EVENT()` macro below the `#include <lttng/tracepoint.h>`
2415 line in the
2416 <<tpp-header,tracepoint provider header file template>>.
2417
2418 The syntax of the `LTTNG_UST_TRACEPOINT_EVENT()` macro is:
2419
2420 [source,c]
2421 .`LTTNG_UST_TRACEPOINT_EVENT()` macro syntax.
2422 ----
2423 LTTNG_UST_TRACEPOINT_EVENT(
2424     /* Tracepoint provider name */
2425     provider_name,
2426
2427     /* Tracepoint name */
2428     tracepoint_name,
2429
2430     /* Input arguments */
2431     LTTNG_UST_TP_ARGS(
2432         arguments
2433     ),
2434
2435     /* Output event fields */
2436     LTTNG_UST_TP_FIELDS(
2437         fields
2438     )
2439 )
2440 ----
2441
2442 Replace:
2443
2444 * +__provider_name__+ with your tracepoint provider name.
2445 * +__tracepoint_name__+ with your tracepoint name.
2446 * +__arguments__+ with the <<tpp-def-input-args,input arguments>>.
2447 * +__fields__+ with the <<tpp-def-output-fields,output event field>>
2448   definitions.
2449
2450 The full name of this tracepoint is `provider_name:tracepoint_name`.
2451
2452 [IMPORTANT]
2453 .Event name length limitation
2454 ====
2455 The concatenation of the tracepoint provider name and the tracepoint
2456 name must not exceed **254{nbsp}characters**. If it does, the
2457 instrumented application compiles and runs, but LTTng throws multiple
2458 warnings and you could experience serious issues.
2459 ====
2460
2461 [[tpp-def-input-args]]The syntax of the `LTTNG_UST_TP_ARGS()` macro is:
2462
2463 [source,c]
2464 .`LTTNG_UST_TP_ARGS()` macro syntax.
2465 ----
2466 LTTNG_UST_TP_ARGS(
2467     type, arg_name
2468 )
2469 ----
2470
2471 Replace:
2472
2473 * +__type__+ with the C{nbsp}type of the argument.
2474 * +__arg_name__+ with the argument name.
2475
2476 You can repeat +__type__+ and +__arg_name__+ up to 10{nbsp}times to have
2477 more than one argument.
2478
2479 .`LTTNG_UST_TP_ARGS()` usage with three arguments.
2480 ====
2481 [source,c]
2482 ----
2483 LTTNG_UST_TP_ARGS(
2484     int, count,
2485     float, ratio,
2486     const char*, query
2487 )
2488 ----
2489 ====
2490
2491 The `LTTNG_UST_TP_ARGS()` and `LTTNG_UST_TP_ARGS(void)` forms are valid
2492 to create a tracepoint definition with no input arguments.
2493
2494 [[tpp-def-output-fields]]The `LTTNG_UST_TP_FIELDS()` macro contains a
2495 list of `lttng_ust_field_*()` macros. Each `lttng_ust_field_*()` macro
2496 defines one event field. See man:lttng-ust(3) for a complete description
2497 of the available `lttng_ust_field_*()` macros. A `lttng_ust_field_*()`
2498 macro specifies the type, size, and byte order of one event field.
2499
2500 Each `lttng_ust_field_*()` macro takes an _argument expression_
2501 parameter. This is a C{nbsp}expression that the tracer evaluates at the
2502 `lttng_ust_tracepoint()` macro site in the source code of the
2503 application. This expression provides the source of data of a field. The
2504 argument expression can include input argument names listed in the
2505 `LTTNG_UST_TP_ARGS()` macro.
2506
2507 Each `lttng_ust_field_*()` macro also takes a _field name_ parameter.
2508 Field names must be unique within a given tracepoint definition.
2509
2510 Here's a complete tracepoint definition example:
2511
2512 .Tracepoint definition.
2513 ====
2514 The following tracepoint definition defines a tracepoint which takes
2515 three input arguments and has four output event fields.
2516
2517 [source,c]
2518 ----
2519 #include "my-custom-structure.h"
2520
2521 LTTNG_UST_TRACEPOINT_EVENT(
2522     my_provider,
2523     my_tracepoint,
2524     LTTNG_UST_TP_ARGS(
2525         const struct my_custom_structure *, my_custom_structure,
2526         float, ratio,
2527         const char *, query
2528     ),
2529     LTTNG_UST_TP_FIELDS(
2530         lttng_ust_field_string(query_field, query)
2531         lttng_ust_field_float(double, ratio_field, ratio)
2532         lttng_ust_field_integer(int, recv_size,
2533                                 my_custom_structure->recv_size)
2534         lttng_ust_field_integer(int, send_size,
2535                                 my_custom_structure->send_size)
2536     )
2537 )
2538 ----
2539
2540 Refer to this tracepoint definition with the `lttng_ust_tracepoint()`
2541 macro in the source code of your application like this:
2542
2543 [source,c]
2544 ----
2545 lttng_ust_tracepoint(my_provider, my_tracepoint,
2546                      my_structure, some_ratio, the_query);
2547 ----
2548 ====
2549
2550 NOTE: The LTTng-UST tracer only evaluates the arguments of a tracepoint
2551 at run time when such a tracepoint _could_ emit an event. See
2552 <<event-creation-emission-opti,this note>> to learn more.
2553
2554
2555 [[using-tracepoint-classes]]
2556 ===== Use a tracepoint class
2557
2558 A _tracepoint class_ is a class of tracepoints which share the same
2559 output event field definitions. A _tracepoint instance_ is one
2560 instance of such a defined tracepoint class, with its own tracepoint
2561 name.
2562
2563 The <<defining-tracepoints,`LTTNG_UST_TRACEPOINT_EVENT()` macro>> is
2564 actually a shorthand which defines both a tracepoint class and a
2565 tracepoint instance at the same time.
2566
2567 When you build a tracepoint provider package, the C or $$C++$$ compiler
2568 creates one serialization function for each **tracepoint class**. A
2569 serialization function is responsible for serializing the event fields
2570 of a tracepoint to a sub-buffer when recording.
2571
2572 For various performance reasons, when your situation requires multiple
2573 tracepoint definitions with different names, but with the same event
2574 fields, we recommend that you manually create a tracepoint class and
2575 instantiate as many tracepoint instances as needed. One positive effect
2576 of such a design, amongst other advantages, is that all tracepoint
2577 instances of the same tracepoint class reuse the same serialization
2578 function, thus reducing
2579 https://en.wikipedia.org/wiki/Cache_pollution[cache pollution].
2580
2581 .Use a tracepoint class and tracepoint instances.
2582 ====
2583 Consider the following three tracepoint definitions:
2584
2585 [source,c]
2586 ----
2587 LTTNG_UST_TRACEPOINT_EVENT(
2588     my_app,
2589     get_account,
2590     LTTNG_UST_TP_ARGS(
2591         int, userid,
2592         size_t, len
2593     ),
2594     LTTNG_UST_TP_FIELDS(
2595         lttng_ust_field_integer(int, userid, userid)
2596         lttng_ust_field_integer(size_t, len, len)
2597     )
2598 )
2599
2600 LTTNG_UST_TRACEPOINT_EVENT(
2601     my_app,
2602     get_settings,
2603     LTTNG_UST_TP_ARGS(
2604         int, userid,
2605         size_t, len
2606     ),
2607     LTTNG_UST_TP_FIELDS(
2608         lttng_ust_field_integer(int, userid, userid)
2609         lttng_ust_field_integer(size_t, len, len)
2610     )
2611 )
2612
2613 LTTNG_UST_TRACEPOINT_EVENT(
2614     my_app,
2615     get_transaction,
2616     LTTNG_UST_TP_ARGS(
2617         int, userid,
2618         size_t, len
2619     ),
2620     LTTNG_UST_TP_FIELDS(
2621         lttng_ust_field_integer(int, userid, userid)
2622         lttng_ust_field_integer(size_t, len, len)
2623     )
2624 )
2625 ----
2626
2627 In this case, we create three tracepoint classes, with one implicit
2628 tracepoint instance for each of them: `get_account`, `get_settings`, and
2629 `get_transaction`. However, they all share the same event field names
2630 and types. Hence three identical, yet independent serialization
2631 functions are created when you build the tracepoint provider package.
2632
2633 A better design choice is to define a single tracepoint class and three
2634 tracepoint instances:
2635
2636 [source,c]
2637 ----
2638 /* The tracepoint class */
2639 LTTNG_UST_TRACEPOINT_EVENT_CLASS(
2640     /* Tracepoint class provider name */
2641     my_app,
2642
2643     /* Tracepoint class name */
2644     my_class,
2645
2646     /* Input arguments */
2647     LTTNG_UST_TP_ARGS(
2648         int, userid,
2649         size_t, len
2650     ),
2651
2652     /* Output event fields */
2653     LTTNG_UST_TP_FIELDS(
2654         lttng_ust_field_integer(int, userid, userid)
2655         lttng_ust_field_integer(size_t, len, len)
2656     )
2657 )
2658
2659 /* The tracepoint instances */
2660 LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
2661     /* Tracepoint class provider name */
2662     my_app,
2663
2664     /* Tracepoint class name */
2665     my_class,
2666
2667     /* Instance provider name */
2668     my_app,
2669
2670     /* Tracepoint name */
2671     get_account,
2672
2673     /* Input arguments */
2674     LTTNG_UST_TP_ARGS(
2675         int, userid,
2676         size_t, len
2677     )
2678 )
2679 LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
2680     my_app,
2681     my_class,
2682     get_settings,
2683     LTTNG_UST_TP_ARGS(
2684         int, userid,
2685         size_t, len
2686     )
2687 )
2688 LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
2689     my_app,
2690     my_class,
2691     get_transaction,
2692     LTTNG_UST_TP_ARGS(
2693         int, userid,
2694         size_t, len
2695     )
2696 )
2697 ----
2698 ====
2699
2700 The tracepoint class and instance provider names must be the same if the
2701 `LTTNG_UST_TRACEPOINT_EVENT_CLASS()` and
2702 `LTTNG_UST_TRACEPOINT_EVENT_INSTANCE()` expansions are part of the same
2703 translation unit. See man:lttng-ust(3) to learn more.
2704
2705
2706 [[assigning-log-levels]]
2707 ===== Assign a log level to a tracepoint definition
2708
2709 Assign a _log level_ to a <<defining-tracepoints,tracepoint definition>>
2710 with the `LTTNG_UST_TRACEPOINT_LOGLEVEL()` macro.
2711
2712 Assigning different levels of severity to tracepoint definitions can be
2713 useful: when you <<enabling-disabling-events,create a recording event
2714 rule>>, you can target tracepoints having a log level at least as severe
2715 as a specific value.
2716
2717 The concept of LTTng-UST log levels is similar to the levels found
2718 in typical logging frameworks:
2719
2720 * In a logging framework, the log level is given by the function
2721   or method name you use at the log statement site: `debug()`,
2722   `info()`, `warn()`, `error()`, and so on.
2723
2724 * In LTTng-UST, you statically assign the log level to a tracepoint
2725   definition; any `lttng_ust_tracepoint()` macro invocation which refers
2726   to this definition has this log level.
2727
2728 You must use `LTTNG_UST_TRACEPOINT_LOGLEVEL()` _after_ the
2729 <<defining-tracepoints,`LTTNG_UST_TRACEPOINT_EVENT()`>> or
2730 <<using-tracepoint-classes,`LTTNG_UST_TRACEPOINT_INSTANCE()`>> macro for
2731 a given tracepoint.
2732
2733 The syntax of the `LTTNG_UST_TRACEPOINT_LOGLEVEL()` macro is:
2734
2735 [source,c]
2736 .`LTTNG_UST_TRACEPOINT_LOGLEVEL()` macro syntax.
2737 ----
2738 LTTNG_UST_TRACEPOINT_LOGLEVEL(provider_name, tracepoint_name, log_level)
2739 ----
2740
2741 Replace:
2742
2743 * +__provider_name__+ with the tracepoint provider name.
2744 * +__tracepoint_name__+ with the tracepoint name.
2745 * +__log_level__+ with the log level to assign to the tracepoint
2746   definition named +__tracepoint_name__+ in the +__provider_name__+
2747   tracepoint provider.
2748 +
2749 See man:lttng-ust(3) for a list of available log level names.
2750
2751 .Assign the `LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_UNIT` log level to a tracepoint definition.
2752 ====
2753 [source,c]
2754 ----
2755 /* Tracepoint definition */
2756 LTTNG_UST_TRACEPOINT_EVENT(
2757     my_app,
2758     get_transaction,
2759     LTTNG_UST_TP_ARGS(
2760         int, userid,
2761         size_t, len
2762     ),
2763     LTTNG_UST_TP_FIELDS(
2764         lttng_ust_field_integer(int, userid, userid)
2765         lttng_ust_field_integer(size_t, len, len)
2766     )
2767 )
2768
2769 /* Log level assignment */
2770 LTTNG_UST_TRACEPOINT_LOGLEVEL(my_app, get_transaction,
2771                               LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_UNIT)
2772 ----
2773 ====
2774
2775
2776 [[tpp-source]]
2777 ===== Create a tracepoint provider package source file
2778
2779 A _tracepoint provider package source file_ is a C source file which
2780 includes a <<tpp-header,tracepoint provider header file>> to expand its
2781 macros into event serialization and other functions.
2782
2783 Use the following tracepoint provider package source file template:
2784
2785 [source,c]
2786 .Tracepoint provider package source file template.
2787 ----
2788 #define LTTNG_UST_TRACEPOINT_CREATE_PROBES
2789
2790 #include "tp.h"
2791 ----
2792
2793 Replace `tp.h` with the name of your <<tpp-header,tracepoint provider
2794 header file>> name. You may also include more than one tracepoint
2795 provider header file here to create a tracepoint provider package
2796 holding more than one tracepoint providers.
2797
2798
2799 [[probing-the-application-source-code]]
2800 ==== Add tracepoints to the source code of an application
2801
2802 Once you <<tpp-header,create a tracepoint provider header file>>, use
2803 the `lttng_ust_tracepoint()` macro in the source code of your
2804 application to insert the tracepoints that this header
2805 <<defining-tracepoints,defines>>.
2806
2807 The `lttng_ust_tracepoint()` macro takes at least two parameters: the
2808 tracepoint provider name and the tracepoint name. The corresponding
2809 tracepoint definition defines the other parameters.
2810
2811 .`lttng_ust_tracepoint()` usage.
2812 ====
2813 The following <<defining-tracepoints,tracepoint definition>> defines a
2814 tracepoint which takes two input arguments and has two output event
2815 fields.
2816
2817 [source,c]
2818 .Tracepoint provider header file.
2819 ----
2820 #include "my-custom-structure.h"
2821
2822 LTTNG_UST_TRACEPOINT_EVENT(
2823     my_provider,
2824     my_tracepoint,
2825     LTTNG_UST_TP_ARGS(
2826         int, argc,
2827         const char *, cmd_name
2828     ),
2829     LTTNG_UST_TP_FIELDS(
2830         lttng_ust_field_string(cmd_name, cmd_name)
2831         lttng_ust_field_integer(int, number_of_args, argc)
2832     )
2833 )
2834 ----
2835
2836 Refer to this tracepoint definition with the `lttng_ust_tracepoint()`
2837 macro in the source code of your application like this:
2838
2839 [source,c]
2840 .Application source file.
2841 ----
2842 #include "tp.h"
2843
2844 int main(int argc, char* argv[])
2845 {
2846     lttng_ust_tracepoint(my_provider, my_tracepoint, argc, argv[0]);
2847     return 0;
2848 }
2849 ----
2850
2851 Note how the source code of the application includes
2852 the tracepoint provider header file containing the tracepoint
2853 definitions to use, path:{tp.h}.
2854 ====
2855
2856 .`lttng_ust_tracepoint()` usage with a complex tracepoint definition.
2857 ====
2858 Consider this complex tracepoint definition, where multiple event
2859 fields refer to the same input arguments in their argument expression
2860 parameter:
2861
2862 [source,c]
2863 .Tracepoint provider header file.
2864 ----
2865 /* For `struct stat` */
2866 #include <sys/types.h>
2867 #include <sys/stat.h>
2868 #include <unistd.h>
2869
2870 LTTNG_UST_TRACEPOINT_EVENT(
2871     my_provider,
2872     my_tracepoint,
2873     LTTNG_UST_TP_ARGS(
2874         int, my_int_arg,
2875         char *, my_str_arg,
2876         struct stat *, st
2877     ),
2878     LTTNG_UST_TP_FIELDS(
2879         lttng_ust_field_integer(int, my_constant_field, 23 + 17)
2880         lttng_ust_field_integer(int, my_int_arg_field, my_int_arg)
2881         lttng_ust_field_integer(int, my_int_arg_field2,
2882                                 my_int_arg * my_int_arg)
2883         lttng_ust_field_integer(int, sum4_field,
2884                                 my_str_arg[0] + my_str_arg[1] +
2885                                 my_str_arg[2] + my_str_arg[3])
2886         lttng_ust_field_string(my_str_arg_field, my_str_arg)
2887         lttng_ust_field_integer_hex(off_t, size_field, st->st_size)
2888         lttng_ust_field_float(double, size_dbl_field, (double) st->st_size)
2889         lttng_ust_field_sequence_text(char, half_my_str_arg_field,
2890                                       my_str_arg, size_t,
2891                                       strlen(my_str_arg) / 2)
2892     )
2893 )
2894 ----
2895
2896 Refer to this tracepoint definition with the `lttng_ust_tracepoint()`
2897 macro in the source code of your application like this:
2898
2899 [source,c]
2900 .Application source file.
2901 ----
2902 #define LTTNG_UST_TRACEPOINT_DEFINE
2903 #include "tp.h"
2904
2905 int main(void)
2906 {
2907     struct stat s;
2908
2909     stat("/etc/fstab", &s);
2910     lttng_ust_tracepoint(my_provider, my_tracepoint, 23,
2911                          "Hello, World!", &s);
2912
2913     return 0;
2914 }
2915 ----
2916
2917 If you look at the event record that LTTng writes when recording this
2918 program, assuming the file size of path:{/etc/fstab} is 301{nbsp}bytes,
2919 it should look like this:
2920
2921 .Event record fields
2922 |====
2923 |Field name              |Field value
2924 |`my_constant_field`     |40
2925 |`my_int_arg_field`      |23
2926 |`my_int_arg_field2`     |529
2927 |`sum4_field`            |389
2928 |`my_str_arg_field`      |`Hello, World!`
2929 |`size_field`            |0x12d
2930 |`size_dbl_field`        |301.0
2931 |`half_my_str_arg_field` |`Hello,`
2932 |====
2933 ====
2934
2935 Sometimes, the arguments you pass to `lttng_ust_tracepoint()` are
2936 expensive to evaluate--they use the call stack, for example. To avoid
2937 this computation when LTTng wouldn't emit any event anyway, use the
2938 `lttng_ust_tracepoint_enabled()` and `lttng_ust_do_tracepoint()` macros.
2939
2940 The syntax of the `lttng_ust_tracepoint_enabled()` and
2941 `lttng_ust_do_tracepoint()` macros is:
2942
2943 [source,c]
2944 .`lttng_ust_tracepoint_enabled()` and `lttng_ust_do_tracepoint()` macros syntax.
2945 ----
2946 lttng_ust_tracepoint_enabled(provider_name, tracepoint_name)
2947
2948 lttng_ust_do_tracepoint(provider_name, tracepoint_name, ...)
2949 ----
2950
2951 Replace:
2952
2953 * +__provider_name__+ with the tracepoint provider name.
2954 * +__tracepoint_name__+ with the tracepoint name.
2955
2956 `lttng_ust_tracepoint_enabled()` returns a non-zero value if executing
2957 the tracepoint named `tracepoint_name` from the provider named
2958 `provider_name` _could_ make LTTng emit an event, depending on the
2959 payload of said event.
2960
2961 `lttng_ust_do_tracepoint()` is like `lttng_ust_tracepoint()`, except
2962 that it doesn't check what `lttng_ust_tracepoint_enabled()` checks.
2963 Using `lttng_ust_tracepoint()` with `lttng_ust_tracepoint_enabled()` is
2964 dangerous because `lttng_ust_tracepoint()` also contains the
2965 `lttng_ust_tracepoint_enabled()` check; therefore, a race condition is
2966 possible in this situation:
2967
2968 [source,c]
2969 .Possible race condition when using `lttng_ust_tracepoint_enabled()` with `lttng_ust_tracepoint()`.
2970 ----
2971 if (lttng_ust_tracepoint_enabled(my_provider, my_tracepoint)) {
2972     stuff = prepare_stuff();
2973 }
2974
2975 lttng_ust_tracepoint(my_provider, my_tracepoint, stuff);
2976 ----
2977
2978 If `lttng_ust_tracepoint_enabled()` is false, but would be true after
2979 the conditional block, then `stuff` isn't prepared: the emitted event
2980 will either contain wrong data, or the whole application could crash
2981 (with a segmentation fault, for example).
2982
2983 NOTE: Neither `lttng_ust_tracepoint_enabled()` nor
2984 `lttng_ust_do_tracepoint()` have an `STAP_PROBEV()` call. If you need
2985 it, you must emit this call yourself.
2986
2987
2988 [[building-tracepoint-providers-and-user-application]]
2989 ==== Build and link a tracepoint provider package and an application
2990
2991 Once you have one or more <<tpp-header,tracepoint provider header
2992 files>> and a <<tpp-source,tracepoint provider package source file>>,
2993 create the tracepoint provider package by compiling its source
2994 file. From here, multiple build and run scenarios are possible. The
2995 following table shows common application and library configurations
2996 along with the required command lines to achieve them.
2997
2998 In the following diagrams, we use the following file names:
2999
3000 `app`::
3001   Executable application.
3002
3003 `app.o`::
3004   Application object file.
3005
3006 `tpp.o`::
3007   Tracepoint provider package object file.
3008
3009 `tpp.a`::
3010   Tracepoint provider package archive file.
3011
3012 `libtpp.so`::
3013   Tracepoint provider package shared object file.
3014
3015 `emon.o`::
3016   User library object file.
3017
3018 `libemon.so`::
3019   User library shared object file.
3020
3021 We use the following symbols in the diagrams of table below:
3022
3023 [role="img-100"]
3024 .Symbols used in the build scenario diagrams.
3025 image::ust-sit-symbols.png[]
3026
3027 We assume that path:{.} is part of the env:LD_LIBRARY_PATH environment
3028 variable in the following instructions.
3029
3030 [role="growable ust-scenarios",cols="asciidoc,asciidoc"]
3031 .Common tracepoint provider package scenarios.
3032 |====
3033 |Scenario |Instructions
3034
3035 |
3036 The instrumented application is statically linked with
3037 the tracepoint provider package object.
3038
3039 image::ust-sit+app-linked-with-tp-o+app-instrumented.png[]
3040
3041 |
3042 include::../common/ust-sit-step-tp-o.txt[]
3043
3044 To build the instrumented application:
3045
3046 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3047 +
3048 --
3049 [source,c]
3050 ----
3051 #define LTTNG_UST_TRACEPOINT_DEFINE
3052 ----
3053 --
3054
3055 . Compile the application source file:
3056 +
3057 --
3058 [role="term"]
3059 ----
3060 $ gcc -c app.c
3061 ----
3062 --
3063
3064 . Build the application:
3065 +
3066 --
3067 [role="term"]
3068 ----
3069 $ gcc -o app app.o tpp.o -llttng-ust -ldl
3070 ----
3071 --
3072
3073 To run the instrumented application:
3074
3075 * Start the application:
3076 +
3077 --
3078 [role="term"]
3079 ----
3080 $ ./app
3081 ----
3082 --
3083
3084 |
3085 The instrumented application is statically linked with the
3086 tracepoint provider package archive file.
3087
3088 image::ust-sit+app-linked-with-tp-a+app-instrumented.png[]
3089
3090 |
3091 To create the tracepoint provider package archive file:
3092
3093 . Compile the <<tpp-source,tracepoint provider package source file>>:
3094 +
3095 --
3096 [role="term"]
3097 ----
3098 $ gcc -I. -c tpp.c
3099 ----
3100 --
3101
3102 . Create the tracepoint provider package archive file:
3103 +
3104 --
3105 [role="term"]
3106 ----
3107 $ ar rcs tpp.a tpp.o
3108 ----
3109 --
3110
3111 To build the instrumented application:
3112
3113 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3114 +
3115 --
3116 [source,c]
3117 ----
3118 #define LTTNG_UST_TRACEPOINT_DEFINE
3119 ----
3120 --
3121
3122 . Compile the application source file:
3123 +
3124 --
3125 [role="term"]
3126 ----
3127 $ gcc -c app.c
3128 ----
3129 --
3130
3131 . Build the application:
3132 +
3133 --
3134 [role="term"]
3135 ----
3136 $ gcc -o app app.o tpp.a -llttng-ust -ldl
3137 ----
3138 --
3139
3140 To run the instrumented application:
3141
3142 * Start the application:
3143 +
3144 --
3145 [role="term"]
3146 ----
3147 $ ./app
3148 ----
3149 --
3150
3151 |
3152 The instrumented application is linked with the tracepoint provider
3153 package shared object.
3154
3155 image::ust-sit+app-linked-with-tp-so+app-instrumented.png[]
3156
3157 |
3158 include::../common/ust-sit-step-tp-so.txt[]
3159
3160 To build the instrumented application:
3161
3162 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3163 +
3164 --
3165 [source,c]
3166 ----
3167 #define LTTNG_UST_TRACEPOINT_DEFINE
3168 ----
3169 --
3170
3171 . Compile the application source file:
3172 +
3173 --
3174 [role="term"]
3175 ----
3176 $ gcc -c app.c
3177 ----
3178 --
3179
3180 . Build the application:
3181 +
3182 --
3183 [role="term"]
3184 ----
3185 $ gcc -o app app.o -ldl -L. -ltpp
3186 ----
3187 --
3188
3189 To run the instrumented application:
3190
3191 * Start the application:
3192 +
3193 --
3194 [role="term"]
3195 ----
3196 $ ./app
3197 ----
3198 --
3199
3200 |
3201 The tracepoint provider package shared object is preloaded before the
3202 instrumented application starts.
3203
3204 image::ust-sit+tp-so-preloaded+app-instrumented.png[]
3205
3206 |
3207 include::../common/ust-sit-step-tp-so.txt[]
3208
3209 To build the instrumented application:
3210
3211 . In path:{app.c}, before including path:{tpp.h}, add the
3212   following lines:
3213 +
3214 --
3215 [source,c]
3216 ----
3217 #define LTTNG_UST_TRACEPOINT_DEFINE
3218 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3219 ----
3220 --
3221
3222 . Compile the application source file:
3223 +
3224 --
3225 [role="term"]
3226 ----
3227 $ gcc -c app.c
3228 ----
3229 --
3230
3231 . Build the application:
3232 +
3233 --
3234 [role="term"]
3235 ----
3236 $ gcc -o app app.o -ldl
3237 ----
3238 --
3239
3240 To run the instrumented application with tracing support:
3241
3242 * Preload the tracepoint provider package shared object and
3243   start the application:
3244 +
3245 --
3246 [role="term"]
3247 ----
3248 $ LD_PRELOAD=./libtpp.so ./app
3249 ----
3250 --
3251
3252 To run the instrumented application without tracing support:
3253
3254 * Start the application:
3255 +
3256 --
3257 [role="term"]
3258 ----
3259 $ ./app
3260 ----
3261 --
3262
3263 |
3264 The instrumented application dynamically loads the tracepoint provider
3265 package shared object.
3266
3267 image::ust-sit+app-dlopens-tp-so+app-instrumented.png[]
3268
3269 |
3270 include::../common/ust-sit-step-tp-so.txt[]
3271
3272 To build the instrumented application:
3273
3274 . In path:{app.c}, before including path:{tpp.h}, add the
3275   following lines:
3276 +
3277 --
3278 [source,c]
3279 ----
3280 #define LTTNG_UST_TRACEPOINT_DEFINE
3281 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3282 ----
3283 --
3284
3285 . Compile the application source file:
3286 +
3287 --
3288 [role="term"]
3289 ----
3290 $ gcc -c app.c
3291 ----
3292 --
3293
3294 . Build the application:
3295 +
3296 --
3297 [role="term"]
3298 ----
3299 $ gcc -o app app.o -ldl
3300 ----
3301 --
3302
3303 To run the instrumented application:
3304
3305 * Start the application:
3306 +
3307 --
3308 [role="term"]
3309 ----
3310 $ ./app
3311 ----
3312 --
3313
3314 |
3315 The application is linked with the instrumented user library.
3316
3317 The instrumented user library is statically linked with the tracepoint
3318 provider package object file.
3319
3320 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-o+lib-instrumented.png[]
3321
3322 |
3323 include::../common/ust-sit-step-tp-o-fpic.txt[]
3324
3325 To build the instrumented user library:
3326
3327 . In path:{emon.c}, before including path:{tpp.h}, add the
3328   following line:
3329 +
3330 --
3331 [source,c]
3332 ----
3333 #define LTTNG_UST_TRACEPOINT_DEFINE
3334 ----
3335 --
3336
3337 . Compile the user library source file:
3338 +
3339 --
3340 [role="term"]
3341 ----
3342 $ gcc -I. -fpic -c emon.c
3343 ----
3344 --
3345
3346 . Build the user library shared object:
3347 +
3348 --
3349 [role="term"]
3350 ----
3351 $ gcc -shared -o libemon.so emon.o tpp.o -llttng-ust -ldl
3352 ----
3353 --
3354
3355 To build the application:
3356
3357 . Compile the application source file:
3358 +
3359 --
3360 [role="term"]
3361 ----
3362 $ gcc -c app.c
3363 ----
3364 --
3365
3366 . Build the application:
3367 +
3368 --
3369 [role="term"]
3370 ----
3371 $ gcc -o app app.o -L. -lemon
3372 ----
3373 --
3374
3375 To run the application:
3376
3377 * Start the application:
3378 +
3379 --
3380 [role="term"]
3381 ----
3382 $ ./app
3383 ----
3384 --
3385
3386 |
3387 The application is linked with the instrumented user library.
3388
3389 The instrumented user library is linked with the tracepoint provider
3390 package shared object.
3391
3392 image::ust-sit+app-linked-with-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3393
3394 |
3395 include::../common/ust-sit-step-tp-so.txt[]
3396
3397 To build the instrumented user library:
3398
3399 . In path:{emon.c}, before including path:{tpp.h}, add the
3400   following line:
3401 +
3402 --
3403 [source,c]
3404 ----
3405 #define LTTNG_UST_TRACEPOINT_DEFINE
3406 ----
3407 --
3408
3409 . Compile the user library source file:
3410 +
3411 --
3412 [role="term"]
3413 ----
3414 $ gcc -I. -fpic -c emon.c
3415 ----
3416 --
3417
3418 . Build the user library shared object:
3419 +
3420 --
3421 [role="term"]
3422 ----
3423 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3424 ----
3425 --
3426
3427 To build the application:
3428
3429 . Compile the application source file:
3430 +
3431 --
3432 [role="term"]
3433 ----
3434 $ gcc -c app.c
3435 ----
3436 --
3437
3438 . Build the application:
3439 +
3440 --
3441 [role="term"]
3442 ----
3443 $ gcc -o app app.o -L. -lemon
3444 ----
3445 --
3446
3447 To run the application:
3448
3449 * Start the application:
3450 +
3451 --
3452 [role="term"]
3453 ----
3454 $ ./app
3455 ----
3456 --
3457
3458 |
3459 The tracepoint provider package shared object is preloaded before the
3460 application starts.
3461
3462 The application is linked with the instrumented user library.
3463
3464 image::ust-sit+tp-so-preloaded+app-linked-with-lib+lib-instrumented.png[]
3465
3466 |
3467 include::../common/ust-sit-step-tp-so.txt[]
3468
3469 To build the instrumented user library:
3470
3471 . In path:{emon.c}, before including path:{tpp.h}, add the
3472   following lines:
3473 +
3474 --
3475 [source,c]
3476 ----
3477 #define LTTNG_UST_TRACEPOINT_DEFINE
3478 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3479 ----
3480 --
3481
3482 . Compile the user library source file:
3483 +
3484 --
3485 [role="term"]
3486 ----
3487 $ gcc -I. -fpic -c emon.c
3488 ----
3489 --
3490
3491 . Build the user library shared object:
3492 +
3493 --
3494 [role="term"]
3495 ----
3496 $ gcc -shared -o libemon.so emon.o -ldl
3497 ----
3498 --
3499
3500 To build the application:
3501
3502 . Compile the application source file:
3503 +
3504 --
3505 [role="term"]
3506 ----
3507 $ gcc -c app.c
3508 ----
3509 --
3510
3511 . Build the application:
3512 +
3513 --
3514 [role="term"]
3515 ----
3516 $ gcc -o app app.o -L. -lemon
3517 ----
3518 --
3519
3520 To run the application with tracing support:
3521
3522 * Preload the tracepoint provider package shared object and
3523   start the application:
3524 +
3525 --
3526 [role="term"]
3527 ----
3528 $ LD_PRELOAD=./libtpp.so ./app
3529 ----
3530 --
3531
3532 To run the application without tracing support:
3533
3534 * Start the application:
3535 +
3536 --
3537 [role="term"]
3538 ----
3539 $ ./app
3540 ----
3541 --
3542
3543 |
3544 The application is linked with the instrumented user library.
3545
3546 The instrumented user library dynamically loads the tracepoint provider
3547 package shared object.
3548
3549 image::ust-sit+app-linked-with-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3550
3551 |
3552 include::../common/ust-sit-step-tp-so.txt[]
3553
3554 To build the instrumented user library:
3555
3556 . In path:{emon.c}, before including path:{tpp.h}, add the
3557   following lines:
3558 +
3559 --
3560 [source,c]
3561 ----
3562 #define LTTNG_UST_TRACEPOINT_DEFINE
3563 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3564 ----
3565 --
3566
3567 . Compile the user library source file:
3568 +
3569 --
3570 [role="term"]
3571 ----
3572 $ gcc -I. -fpic -c emon.c
3573 ----
3574 --
3575
3576 . Build the user library shared object:
3577 +
3578 --
3579 [role="term"]
3580 ----
3581 $ gcc -shared -o libemon.so emon.o -ldl
3582 ----
3583 --
3584
3585 To build the application:
3586
3587 . Compile the application source file:
3588 +
3589 --
3590 [role="term"]
3591 ----
3592 $ gcc -c app.c
3593 ----
3594 --
3595
3596 . Build the application:
3597 +
3598 --
3599 [role="term"]
3600 ----
3601 $ gcc -o app app.o -L. -lemon
3602 ----
3603 --
3604
3605 To run the application:
3606
3607 * Start the application:
3608 +
3609 --
3610 [role="term"]
3611 ----
3612 $ ./app
3613 ----
3614 --
3615
3616 |
3617 The application dynamically loads the instrumented user library.
3618
3619 The instrumented user library is linked with the tracepoint provider
3620 package shared object.
3621
3622 image::ust-sit+app-dlopens-lib+lib-linked-with-tp-so+lib-instrumented.png[]
3623
3624 |
3625 include::../common/ust-sit-step-tp-so.txt[]
3626
3627 To build the instrumented user library:
3628
3629 . In path:{emon.c}, before including path:{tpp.h}, add the
3630   following line:
3631 +
3632 --
3633 [source,c]
3634 ----
3635 #define LTTNG_UST_TRACEPOINT_DEFINE
3636 ----
3637 --
3638
3639 . Compile the user library source file:
3640 +
3641 --
3642 [role="term"]
3643 ----
3644 $ gcc -I. -fpic -c emon.c
3645 ----
3646 --
3647
3648 . Build the user library shared object:
3649 +
3650 --
3651 [role="term"]
3652 ----
3653 $ gcc -shared -o libemon.so emon.o -ldl -L. -ltpp
3654 ----
3655 --
3656
3657 To build the application:
3658
3659 . Compile the application source file:
3660 +
3661 --
3662 [role="term"]
3663 ----
3664 $ gcc -c app.c
3665 ----
3666 --
3667
3668 . Build the application:
3669 +
3670 --
3671 [role="term"]
3672 ----
3673 $ gcc -o app app.o -ldl -L. -lemon
3674 ----
3675 --
3676
3677 To run the application:
3678
3679 * Start the application:
3680 +
3681 --
3682 [role="term"]
3683 ----
3684 $ ./app
3685 ----
3686 --
3687
3688 |
3689 The application dynamically loads the instrumented user library.
3690
3691 The instrumented user library dynamically loads the tracepoint provider
3692 package shared object.
3693
3694 image::ust-sit+app-dlopens-lib+lib-dlopens-tp-so+lib-instrumented.png[]
3695
3696 |
3697 include::../common/ust-sit-step-tp-so.txt[]
3698
3699 To build the instrumented user library:
3700
3701 . In path:{emon.c}, before including path:{tpp.h}, add the
3702   following lines:
3703 +
3704 --
3705 [source,c]
3706 ----
3707 #define LTTNG_UST_TRACEPOINT_DEFINE
3708 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3709 ----
3710 --
3711
3712 . Compile the user library source file:
3713 +
3714 --
3715 [role="term"]
3716 ----
3717 $ gcc -I. -fpic -c emon.c
3718 ----
3719 --
3720
3721 . Build the user library shared object:
3722 +
3723 --
3724 [role="term"]
3725 ----
3726 $ gcc -shared -o libemon.so emon.o -ldl
3727 ----
3728 --
3729
3730 To build the application:
3731
3732 . Compile the application source file:
3733 +
3734 --
3735 [role="term"]
3736 ----
3737 $ gcc -c app.c
3738 ----
3739 --
3740
3741 . Build the application:
3742 +
3743 --
3744 [role="term"]
3745 ----
3746 $ gcc -o app app.o -ldl -L. -lemon
3747 ----
3748 --
3749
3750 To run the application:
3751
3752 * Start the application:
3753 +
3754 --
3755 [role="term"]
3756 ----
3757 $ ./app
3758 ----
3759 --
3760
3761 |
3762 The tracepoint provider package shared object is preloaded before the
3763 application starts.
3764
3765 The application dynamically loads the instrumented user library.
3766
3767 image::ust-sit+tp-so-preloaded+app-dlopens-lib+lib-instrumented.png[]
3768
3769 |
3770 include::../common/ust-sit-step-tp-so.txt[]
3771
3772 To build the instrumented user library:
3773
3774 . In path:{emon.c}, before including path:{tpp.h}, add the
3775   following lines:
3776 +
3777 --
3778 [source,c]
3779 ----
3780 #define LTTNG_UST_TRACEPOINT_DEFINE
3781 #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
3782 ----
3783 --
3784
3785 . Compile the user library source file:
3786 +
3787 --
3788 [role="term"]
3789 ----
3790 $ gcc -I. -fpic -c emon.c
3791 ----
3792 --
3793
3794 . Build the user library shared object:
3795 +
3796 --
3797 [role="term"]
3798 ----
3799 $ gcc -shared -o libemon.so emon.o -ldl
3800 ----
3801 --
3802
3803 To build the application:
3804
3805 . Compile the application source file:
3806 +
3807 --
3808 [role="term"]
3809 ----
3810 $ gcc -c app.c
3811 ----
3812 --
3813
3814 . Build the application:
3815 +
3816 --
3817 [role="term"]
3818 ----
3819 $ gcc -o app app.o -L. -lemon
3820 ----
3821 --
3822
3823 To run the application with tracing support:
3824
3825 * Preload the tracepoint provider package shared object and
3826   start the application:
3827 +
3828 --
3829 [role="term"]
3830 ----
3831 $ LD_PRELOAD=./libtpp.so ./app
3832 ----
3833 --
3834
3835 To run the application without tracing support:
3836
3837 * Start the application:
3838 +
3839 --
3840 [role="term"]
3841 ----
3842 $ ./app
3843 ----
3844 --
3845
3846 |
3847 The application is statically linked with the tracepoint provider
3848 package object file.
3849
3850 The application is linked with the instrumented user library.
3851
3852 image::ust-sit+app-linked-with-tp-o+app-linked-with-lib+lib-instrumented.png[]
3853
3854 |
3855 include::../common/ust-sit-step-tp-o.txt[]
3856
3857 To build the instrumented user library:
3858
3859 . In path:{emon.c}, before including path:{tpp.h}, add the
3860   following line:
3861 +
3862 --
3863 [source,c]
3864 ----
3865 #define LTTNG_UST_TRACEPOINT_DEFINE
3866 ----
3867 --
3868
3869 . Compile the user library source file:
3870 +
3871 --
3872 [role="term"]
3873 ----
3874 $ gcc -I. -fpic -c emon.c
3875 ----
3876 --
3877
3878 . Build the user library shared object:
3879 +
3880 --
3881 [role="term"]
3882 ----
3883 $ gcc -shared -o libemon.so emon.o
3884 ----
3885 --
3886
3887 To build the application:
3888
3889 . Compile the application source file:
3890 +
3891 --
3892 [role="term"]
3893 ----
3894 $ gcc -c app.c
3895 ----
3896 --
3897
3898 . Build the application:
3899 +
3900 --
3901 [role="term"]
3902 ----
3903 $ gcc -o app app.o tpp.o -llttng-ust -ldl -L. -lemon
3904 ----
3905 --
3906
3907 To run the instrumented application:
3908
3909 * Start the application:
3910 +
3911 --
3912 [role="term"]
3913 ----
3914 $ ./app
3915 ----
3916 --
3917
3918 |
3919 The application is statically linked with the tracepoint provider
3920 package object file.
3921
3922 The application dynamically loads the instrumented user library.
3923
3924 image::ust-sit+app-linked-with-tp-o+app-dlopens-lib+lib-instrumented.png[]
3925
3926 |
3927 include::../common/ust-sit-step-tp-o.txt[]
3928
3929 To build the application:
3930
3931 . In path:{app.c}, before including path:{tpp.h}, add the following line:
3932 +
3933 --
3934 [source,c]
3935 ----
3936 #define LTTNG_UST_TRACEPOINT_DEFINE
3937 ----
3938 --
3939
3940 . Compile the application source file:
3941 +
3942 --
3943 [role="term"]
3944 ----
3945 $ gcc -c app.c
3946 ----
3947 --
3948
3949 . Build the application:
3950 +
3951 --
3952 [role="term"]
3953 ----
3954 $ gcc -Wl,--export-dynamic -o app app.o tpp.o \
3955       -llttng-ust -ldl
3956 ----
3957 --
3958 +
3959 The `--export-dynamic` option passed to the linker is necessary for the
3960 dynamically loaded library to ``see'' the tracepoint symbols defined in
3961 the application.
3962
3963 To build the instrumented user library:
3964
3965 . Compile the user library source file:
3966 +
3967 --
3968 [role="term"]
3969 ----
3970 $ gcc -I. -fpic -c emon.c
3971 ----
3972 --
3973
3974 . Build the user library shared object:
3975 +
3976 --
3977 [role="term"]
3978 ----
3979 $ gcc -shared -o libemon.so emon.o
3980 ----
3981 --
3982
3983 To run the application:
3984
3985 * Start the application:
3986 +
3987 --
3988 [role="term"]
3989 ----
3990 $ ./app
3991 ----
3992 --
3993 |====
3994
3995
3996 [[using-lttng-ust-with-daemons]]
3997 ===== Use noch:{LTTng-UST} with daemons
3998
3999 If your instrumented application calls man:fork(2), man:clone(2),
4000 or BSD's man:rfork(2), without a following man:exec(3)-family
4001 system call, you must preload the path:{liblttng-ust-fork.so} shared
4002 object when you start the application.
4003
4004 [role="term"]
4005 ----
4006 $ LD_PRELOAD=liblttng-ust-fork.so ./my-app
4007 ----
4008
4009 If your tracepoint provider package is
4010 a shared library which you also preload, you must put both
4011 shared objects in env:LD_PRELOAD:
4012
4013 [role="term"]
4014 ----
4015 $ LD_PRELOAD=liblttng-ust-fork.so:/path/to/tp.so ./my-app
4016 ----
4017
4018
4019 [role="since-2.9"]
4020 [[liblttng-ust-fd]]
4021 ===== Use noch:{LTTng-UST} with applications which close file descriptors that don't belong to them
4022
4023 If your instrumented application closes one or more file descriptors
4024 which it did not open itself, you must preload the
4025 path:{liblttng-ust-fd.so} shared object when you start the application:
4026
4027 [role="term"]
4028 ----
4029 $ LD_PRELOAD=liblttng-ust-fd.so ./my-app
4030 ----
4031
4032 Typical use cases include closing all the file descriptors after
4033 man:fork(2) or man:rfork(2) and buggy applications doing
4034 ``double closes''.
4035
4036
4037 [[lttng-ust-pkg-config]]
4038 ===== Use noch:{pkg-config}
4039
4040 On some distributions, LTTng-UST ships with a
4041 https://www.freedesktop.org/wiki/Software/pkg-config/[pkg-config]
4042 metadata file. If this is your case, then use cmd:pkg-config to
4043 build an application on the command line:
4044
4045 [role="term"]
4046 ----
4047 $ gcc -o my-app my-app.o tp.o $(pkg-config --cflags --libs lttng-ust)
4048 ----
4049
4050
4051 [[instrumenting-32-bit-app-on-64-bit-system]]
4052 ===== [[advanced-instrumenting-techniques]]Build a 32-bit instrumented application for a 64-bit target system
4053
4054 In order to trace a 32-bit application running on a 64-bit system,
4055 LTTng must use a dedicated 32-bit
4056 <<lttng-consumerd,consumer daemon>>.
4057
4058 The following steps show how to build and install a 32-bit consumer
4059 daemon, which is _not_ part of the default 64-bit LTTng build, how to
4060 build and install the 32-bit LTTng-UST libraries, and how to build and
4061 link an instrumented 32-bit application in that context.
4062
4063 To build a 32-bit instrumented application for a 64-bit target system,
4064 assuming you have a fresh target system with no installed Userspace RCU
4065 or LTTng packages:
4066
4067 . Download, build, and install a 32-bit version of Userspace RCU:
4068 +
4069 --
4070 [role="term"]
4071 ----
4072 $ cd $(mktemp -d) &&
4073   wget https://lttng.org/files/urcu/userspace-rcu-latest-0.13.tar.bz2 &&
4074   tar -xf userspace-rcu-latest-0.13.tar.bz2 &&
4075   cd userspace-rcu-0.13.* &&
4076   ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 &&
4077   make &&
4078   sudo make install &&
4079   sudo ldconfig
4080 ----
4081 --
4082
4083 . Using the package manager of your distribution, or from source,
4084   install the 32-bit versions of the following dependencies of
4085   LTTng-tools and LTTng-UST:
4086 +
4087 --
4088 * https://sourceforge.net/projects/libuuid/[libuuid]
4089 * https://directory.fsf.org/wiki/Popt[popt]
4090 * https://www.xmlsoft.org/[libxml2]
4091 * **Optional**: https://github.com/numactl/numactl[numactl]
4092 --
4093
4094 . Download, build, and install a 32-bit version of the latest
4095   LTTng-UST{nbsp}{revision}:
4096 +
4097 --
4098 [role="term"]
4099 ----
4100 $ cd $(mktemp -d) &&
4101   wget https://lttng.org/files/lttng-ust/lttng-ust-latest-2.13.tar.bz2 &&
4102   tar -xf lttng-ust-latest-2.13.tar.bz2 &&
4103   cd lttng-ust-2.13.* &&
4104   ./configure --libdir=/usr/local/lib32 \
4105               CFLAGS=-m32 CXXFLAGS=-m32 \
4106               LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' &&
4107   make &&
4108   sudo make install &&
4109   sudo ldconfig
4110 ----
4111 --
4112 +
4113 Add `--disable-numa` to `./configure` if you don't have
4114 https://github.com/numactl/numactl[numactl].
4115 +
4116 [NOTE]
4117 ====
4118 Depending on your distribution, 32-bit libraries could be installed at a
4119 different location than `/usr/lib32`. For example, Debian is known to
4120 install some 32-bit libraries in `/usr/lib/i386-linux-gnu`.
4121
4122 In this case, make sure to set `LDFLAGS` to all the
4123 relevant 32-bit library paths, for example:
4124
4125 [role="term"]
4126 ----
4127 $ LDFLAGS='-L/usr/lib/i386-linux-gnu -L/usr/lib32'
4128 ----
4129 ====
4130
4131 . Download the latest LTTng-tools{nbsp}{revision}, build, and install
4132   the 32-bit consumer daemon:
4133 +
4134 --
4135 [role="term"]
4136 ----
4137 $ cd $(mktemp -d) &&
4138   wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
4139   tar -xf lttng-tools-latest-2.13.tar.bz2 &&
4140   cd lttng-tools-2.13.* &&
4141   ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 CXXFLAGS=-m32 \
4142               LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' \
4143               --disable-bin-lttng --disable-bin-lttng-crash \
4144               --disable-bin-lttng-relayd --disable-bin-lttng-sessiond &&
4145   make &&
4146   cd src/bin/lttng-consumerd &&
4147   sudo make install &&
4148   sudo ldconfig
4149 ----
4150 --
4151
4152 . From your distribution or from source, <<installing-lttng,install>>
4153   the 64-bit versions of LTTng-UST and Userspace RCU.
4154
4155 . Download, build, and install the 64-bit version of the
4156   latest LTTng-tools{nbsp}{revision}:
4157 +
4158 --
4159 [role="term"]
4160 ----
4161 $ cd $(mktemp -d) &&
4162   wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
4163   tar -xf lttng-tools-latest-2.13.tar.bz2 &&
4164   cd lttng-tools-2.13.* &&
4165   ./configure --with-consumerd32-libdir=/usr/local/lib32 \
4166               --with-consumerd32-bin=/usr/local/lib32/lttng/libexec/lttng-consumerd &&
4167   make &&
4168   sudo make install &&
4169   sudo ldconfig
4170 ----
4171 --
4172
4173 . Pass the following options to man:gcc(1), man:g++(1), or man:clang(1)
4174   when linking your 32-bit application:
4175 +
4176 ----
4177 -m32 -L/usr/lib32 -L/usr/local/lib32 \
4178 -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32
4179 ----
4180 +
4181 For example, let's rebuild the quick start example in
4182 ``<<tracing-your-own-user-application,Record user application events>>''
4183 as an instrumented 32-bit application:
4184 +
4185 --
4186 [role="term"]
4187 ----
4188 $ gcc -m32 -c -I. hello-tp.c
4189 $ gcc -m32 -c hello.c
4190 $ gcc -m32 -o hello hello.o hello-tp.o \
4191       -L/usr/lib32 -L/usr/local/lib32 \
4192       -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32 \
4193       -llttng-ust -ldl
4194 ----
4195 --
4196
4197 No special action is required to execute the 32-bit application and
4198 for LTTng to trace it: use the command-line man:lttng(1) tool as usual.
4199
4200
4201 [role="since-2.5"]
4202 [[tracef]]
4203 ==== Use `lttng_ust_tracef()`
4204
4205 man:lttng_ust_tracef(3) is a small LTTng-UST API designed for quick,
4206 man:printf(3)-like instrumentation without the burden of
4207 <<tracepoint-provider,creating>> and
4208 <<building-tracepoint-providers-and-user-application,building>>
4209 a tracepoint provider package.
4210
4211 To use `lttng_ust_tracef()` in your application:
4212
4213 . In the C or $$C++$$ source files where you need to use
4214   `lttng_ust_tracef()`, include `<lttng/tracef.h>`:
4215 +
4216 --
4217 [source,c]
4218 ----
4219 #include <lttng/tracef.h>
4220 ----
4221 --
4222
4223 . In the source code of the application, use `lttng_ust_tracef()` like
4224   you would use man:printf(3):
4225 +
4226 --
4227 [source,c]
4228 ----
4229     /* ... */
4230
4231     lttng_ust_tracef("my message: %d (%s)", my_integer, my_string);
4232
4233     /* ... */
4234 ----
4235 --
4236
4237 . Link your application with `liblttng-ust`:
4238 +
4239 --
4240 [role="term"]
4241 ----
4242 $ gcc -o app app.c -llttng-ust
4243 ----
4244 --
4245
4246 To record the events that `lttng_ust_tracef()` calls emit:
4247
4248 * <<enabling-disabling-events,Create a recording event rule>> which
4249   matches user space events named `lttng_ust_tracef:*`:
4250 +
4251 --
4252 [role="term"]
4253 ----
4254 $ lttng enable-event --userspace 'lttng_ust_tracef:*'
4255 ----
4256 --
4257
4258 [IMPORTANT]
4259 .Limitations of `lttng_ust_tracef()`
4260 ====
4261 The `lttng_ust_tracef()` utility function was developed to make user
4262 space tracing super simple, albeit with notable disadvantages compared
4263 to <<defining-tracepoints,user-defined tracepoints>>:
4264
4265 * All the created events have the same tracepoint provider and
4266   tracepoint names, respectively `lttng_ust_tracef` and `event`.
4267 * There's no static type checking.
4268 * The only event record field you actually get, named `msg`, is a string
4269   potentially containing the values you passed to `lttng_ust_tracef()`
4270   using your own format string. This also means that you can't filter
4271   events with a custom expression at run time because there are no
4272   isolated fields.
4273 * Since `lttng_ust_tracef()` uses the man:vasprintf(3) function of the
4274   C{nbsp}standard library behind the scenes to format the strings at run
4275   time, its expected performance is lower than with user-defined
4276   tracepoints, which don't require a conversion to a string.
4277
4278 Taking this into consideration, `lttng_ust_tracef()` is useful for some
4279 quick prototyping and debugging, but you shouldn't consider it for any
4280 permanent and serious applicative instrumentation.
4281 ====
4282
4283
4284 [role="since-2.7"]
4285 [[tracelog]]
4286 ==== Use `lttng_ust_tracelog()`
4287
4288 The man:tracelog(3) API is very similar to
4289 <<tracef,`lttng_ust_tracef()`>>, with the difference that it accepts an
4290 additional log level parameter.
4291
4292 The goal of `lttng_ust_tracelog()` is to ease the migration from logging
4293 to tracing.
4294
4295 To use `lttng_ust_tracelog()` in your application:
4296
4297 . In the C or $$C++$$ source files where you need to use `tracelog()`,
4298   include `<lttng/tracelog.h>`:
4299 +
4300 --
4301 [source,c]
4302 ----
4303 #include <lttng/tracelog.h>
4304 ----
4305 --
4306
4307 . In the source code of the application, use `lttng_ust_tracelog()` like
4308   you would use man:printf(3), except for the first parameter which is
4309   the log level:
4310 +
4311 --
4312 [source,c]
4313 ----
4314     /* ... */
4315
4316     tracelog(LTTNG_UST_TRACEPOINT_LOGLEVEL_WARNING,
4317             "my message: %d (%s)", my_integer, my_string);
4318
4319     /* ... */
4320 ----
4321 --
4322 +
4323 See man:lttng-ust(3) for a list of available log level names.
4324
4325 . Link your application with `liblttng-ust`:
4326 +
4327 --
4328 [role="term"]
4329 ----
4330 $ gcc -o app app.c -llttng-ust
4331 ----
4332 --
4333
4334 To record the events that `lttng_ust_tracelog()` calls emit with a log
4335 level _at least as severe as_ a specific log level:
4336
4337 * <<enabling-disabling-events,Create a recording event rule>> which
4338   matches user space tracepoint events named `lttng_ust_tracelog:*` and
4339   with some minimum level of severity:
4340 +
4341 --
4342 [role="term"]
4343 ----
4344 $ lttng enable-event --userspace 'lttng_ust_tracelog:*' \
4345                      --loglevel=WARNING
4346 ----
4347 --
4348
4349 To record the events that `lttng_ust_tracelog()` calls emit with a
4350 _specific log level_:
4351
4352 * Create a recording event rule which matches tracepoint events named
4353   `lttng_ust_tracelog:*` and with a specific log level:
4354 +
4355 --
4356 [role="term"]
4357 ----
4358 $ lttng enable-event --userspace 'lttng_ust_tracelog:*' \
4359                      --loglevel-only=INFO
4360 ----
4361 --
4362
4363
4364 [[prebuilt-ust-helpers]]
4365 === Load a prebuilt user space tracing helper
4366
4367 The LTTng-UST package provides a few helpers in the form of preloadable
4368 shared objects which automatically instrument system functions and
4369 calls.
4370
4371 The helper shared objects are normally found in dir:{/usr/lib}. If you
4372 built LTTng-UST <<building-from-source,from source>>, they're probably
4373 located in dir:{/usr/local/lib}.
4374
4375 The installed user space tracing helpers in LTTng-UST{nbsp}{revision}
4376 are:
4377
4378 path:{liblttng-ust-libc-wrapper.so}::
4379 path:{liblttng-ust-pthread-wrapper.so}::
4380   <<liblttng-ust-libc-pthread-wrapper,C{nbsp}standard library
4381   memory and POSIX threads function tracing>>.
4382
4383 path:{liblttng-ust-cyg-profile.so}::
4384 path:{liblttng-ust-cyg-profile-fast.so}::
4385   <<liblttng-ust-cyg-profile,Function entry and exit tracing>>.
4386
4387 path:{liblttng-ust-dl.so}::
4388   <<liblttng-ust-dl,Dynamic linker tracing>>.
4389
4390 To use a user space tracing helper with any user application:
4391
4392 * Preload the helper shared object when you start the application:
4393 +
4394 --
4395 [role="term"]
4396 ----
4397 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so my-app
4398 ----
4399 --
4400 +
4401 You can preload more than one helper:
4402 +
4403 --
4404 [role="term"]
4405 ----
4406 $ LD_PRELOAD=liblttng-ust-libc-wrapper.so:liblttng-ust-dl.so my-app
4407 ----
4408 --
4409
4410
4411 [role="since-2.3"]
4412 [[liblttng-ust-libc-pthread-wrapper]]
4413 ==== Instrument C standard library memory and POSIX threads functions
4414
4415 The path:{liblttng-ust-libc-wrapper.so} and
4416 path:{liblttng-ust-pthread-wrapper.so} helpers
4417 add instrumentation to some C standard library and POSIX
4418 threads functions.
4419
4420 [role="growable"]
4421 .Functions instrumented by preloading path:{liblttng-ust-libc-wrapper.so}.
4422 |====
4423 |TP provider name |TP name |Instrumented function
4424
4425 .6+|`lttng_ust_libc` |`malloc`         |man:malloc(3)
4426                      |`calloc`         |man:calloc(3)
4427                      |`realloc`        |man:realloc(3)
4428                      |`free`           |man:free(3)
4429                      |`memalign`       |man:memalign(3)
4430                      |`posix_memalign` |man:posix_memalign(3)
4431 |====
4432
4433 [role="growable"]
4434 .Functions instrumented by preloading path:{liblttng-ust-pthread-wrapper.so}.
4435 |====
4436 |TP provider name |TP name |Instrumented function
4437
4438 .4+|`lttng_ust_pthread` |`pthread_mutex_lock_req` |man:pthread_mutex_lock(3p) (request time)
4439                         |`pthread_mutex_lock_acq` |man:pthread_mutex_lock(3p) (acquire time)
4440                         |`pthread_mutex_trylock`  |man:pthread_mutex_trylock(3p)
4441                         |`pthread_mutex_unlock`   |man:pthread_mutex_unlock(3p)
4442 |====
4443
4444 When you preload the shared object, it replaces the functions listed
4445 in the previous tables by wrappers which contain tracepoints and call
4446 the replaced functions.
4447
4448
4449 [[liblttng-ust-cyg-profile]]
4450 ==== Instrument function entry and exit
4451
4452 The path:{liblttng-ust-cyg-profile*.so} helpers can add instrumentation
4453 to the entry and exit points of functions.
4454
4455 man:gcc(1) and man:clang(1) have an option named
4456 https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html[`-finstrument-functions`]
4457 which generates instrumentation calls for entry and exit to functions.
4458 The LTTng-UST function tracing helpers,
4459 path:{liblttng-ust-cyg-profile.so} and
4460 path:{liblttng-ust-cyg-profile-fast.so}, take advantage of this feature
4461 to add tracepoints to the two generated functions (which contain
4462 `cyg_profile` in their names, hence the name of the helper).
4463
4464 To use the LTTng-UST function tracing helper, the source files to
4465 instrument must be built using the `-finstrument-functions` compiler
4466 flag.
4467
4468 There are two versions of the LTTng-UST function tracing helper:
4469
4470 * **path:{liblttng-ust-cyg-profile-fast.so}** is a lightweight variant
4471   that you should only use when it can be _guaranteed_ that the
4472   complete event stream is recorded without any lost event record.
4473   Any kind of duplicate information is left out.
4474 +
4475 Assuming no event record is lost, having only the function addresses on
4476 entry is enough to create a call graph, since an event record always
4477 contains the ID of the CPU that generated it.
4478 +
4479 Use a tool like man:addr2line(1) to convert function addresses back to
4480 source file names and line numbers.
4481
4482 * **path:{liblttng-ust-cyg-profile.so}** is a more robust variant
4483 which also works in use cases where event records might get discarded or
4484 not recorded from application startup.
4485 In these cases, the trace analyzer needs more information to be
4486 able to reconstruct the program flow.
4487
4488 See man:lttng-ust-cyg-profile(3) to learn more about the instrumentation
4489 points of this helper.
4490
4491 All the tracepoints that this helper provides have the log level
4492 `LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_FUNCTION` (see man:lttng-ust(3)).
4493
4494 TIP: It's sometimes a good idea to limit the number of source files that
4495 you compile with the `-finstrument-functions` option to prevent LTTng
4496 from writing an excessive amount of trace data at run time. When using
4497 man:gcc(1), use the
4498 `-finstrument-functions-exclude-function-list` option to avoid
4499 instrument entries and exits of specific function names.
4500
4501
4502 [role="since-2.4"]
4503 [[liblttng-ust-dl]]
4504 ==== Instrument the dynamic linker
4505
4506 The path:{liblttng-ust-dl.so} helper adds instrumentation to the
4507 man:dlopen(3) and man:dlclose(3) function calls.
4508
4509 See man:lttng-ust-dl(3) to learn more about the instrumentation points
4510 of this helper.
4511
4512
4513 [role="since-2.4"]
4514 [[java-application]]
4515 === Instrument a Java application
4516
4517 You can instrument any Java application which uses one of the following
4518 logging frameworks:
4519
4520 * The https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[**`java.util.logging`**]
4521   (JUL) core logging facilities.
4522
4523 * https://logging.apache.org/log4j/1.2/[**Apache log4j{nbsp}1.2**], since
4524   LTTng{nbsp}2.6. Note that Apache Log4j{nbsp}2 isn't supported.
4525
4526 [role="img-100"]
4527 .LTTng-UST Java agent imported by a Java application.
4528 image::java-app.png[]
4529
4530 Note that the methods described below are new in LTTng{nbsp}2.8.
4531 Previous LTTng versions use another technique.
4532
4533 NOTE: We use https://openjdk.java.net/[OpenJDK]{nbsp}8 for development
4534 and https://ci.lttng.org/[continuous integration], thus this version is
4535 directly supported. However, the LTTng-UST Java agent is also tested
4536 with OpenJDK{nbsp}7.
4537
4538
4539 [role="since-2.8"]
4540 [[jul]]
4541 ==== Use the LTTng-UST Java agent for `java.util.logging`
4542
4543 To use the LTTng-UST Java agent in a Java application which uses
4544 `java.util.logging` (JUL):
4545
4546 . In the source code of the Java application, import the LTTng-UST log
4547   handler package for `java.util.logging`:
4548 +
4549 --
4550 [source,java]
4551 ----
4552 import org.lttng.ust.agent.jul.LttngLogHandler;
4553 ----
4554 --
4555
4556 . Create an LTTng-UST `java.util.logging` log handler:
4557 +
4558 --
4559 [source,java]
4560 ----
4561 Handler lttngUstLogHandler = new LttngLogHandler();
4562 ----
4563 --
4564
4565 . Add this handler to the `java.util.logging` loggers which should emit
4566   LTTng events:
4567 +
4568 --
4569 [source,java]
4570 ----
4571 Logger myLogger = Logger.getLogger("some-logger");
4572
4573 myLogger.addHandler(lttngUstLogHandler);
4574 ----
4575 --
4576
4577 . Use `java.util.logging` log statements and configuration as usual.
4578   The loggers with an attached LTTng-UST log handler can emit
4579   LTTng events.
4580
4581 . Before exiting the application, remove the LTTng-UST log handler from
4582   the loggers attached to it and call its `close()` method:
4583 +
4584 --
4585 [source,java]
4586 ----
4587 myLogger.removeHandler(lttngUstLogHandler);
4588 lttngUstLogHandler.close();
4589 ----
4590 --
4591 +
4592 This isn't strictly necessary, but it's recommended for a clean
4593 disposal of the resources of the handler.
4594
4595 . Include the common and JUL-specific JAR files of the LTTng-UST Java agent,
4596   path:{lttng-ust-agent-common.jar} and path:{lttng-ust-agent-jul.jar},
4597   in the
4598   https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4599   path] when you build the Java application.
4600 +
4601 The JAR files are typically located in dir:{/usr/share/java}.
4602 +
4603 IMPORTANT: The LTTng-UST Java agent must be
4604 <<installing-lttng,installed>> for the logging framework your
4605 application uses.
4606
4607 .Use the LTTng-UST Java agent for `java.util.logging`.
4608 ====
4609 [source,java]
4610 .path:{Test.java}
4611 ----
4612 import java.io.IOException;
4613 import java.util.logging.Handler;
4614 import java.util.logging.Logger;
4615 import org.lttng.ust.agent.jul.LttngLogHandler;
4616
4617 public class Test
4618 {
4619     private static final int answer = 42;
4620
4621     public static void main(String[] argv) throws Exception
4622     {
4623         // Create a logger
4624         Logger logger = Logger.getLogger("jello");
4625
4626         // Create an LTTng-UST log handler
4627         Handler lttngUstLogHandler = new LttngLogHandler();
4628
4629         // Add the LTTng-UST log handler to our logger
4630         logger.addHandler(lttngUstLogHandler);
4631
4632         // Log at will!
4633         logger.info("some info");
4634         logger.warning("some warning");
4635         Thread.sleep(500);
4636         logger.finer("finer information; the answer is " + answer);
4637         Thread.sleep(123);
4638         logger.severe("error!");
4639
4640         // Not mandatory, but cleaner
4641         logger.removeHandler(lttngUstLogHandler);
4642         lttngUstLogHandler.close();
4643     }
4644 }
4645 ----
4646
4647 Build this example:
4648
4649 [role="term"]
4650 ----
4651 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
4652 ----
4653
4654 <<creating-destroying-tracing-sessions,Create a recording session>>,
4655 <<enabling-disabling-events,create a recording event rule>> matching JUL
4656 events named `jello`, and <<basic-tracing-session-control,start
4657 recording>>:
4658
4659 [role="term"]
4660 ----
4661 $ lttng create
4662 $ lttng enable-event --jul jello
4663 $ lttng start
4664 ----
4665
4666 Run the compiled class:
4667
4668 [role="term"]
4669 ----
4670 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
4671 ----
4672
4673 <<basic-tracing-session-control,Stop recording>> and inspect the
4674 recorded events:
4675
4676 [role="term"]
4677 ----
4678 $ lttng stop
4679 $ lttng view
4680 ----
4681 ====
4682
4683 In the resulting trace, an <<event,event record>> which a Java
4684 application using `java.util.logging` generated is named
4685 `lttng_jul:event` and has the following fields:
4686
4687 `msg`::
4688   Log record message.
4689
4690 `logger_name`::
4691   Logger name.
4692
4693 `class_name`::
4694   Name of the class in which the log statement was executed.
4695
4696 `method_name`::
4697   Name of the method in which the log statement was executed.
4698
4699 `long_millis`::
4700   Logging time (timestamp in milliseconds).
4701
4702 `int_loglevel`::
4703   Log level integer value.
4704
4705 `int_threadid`::
4706   ID of the thread in which the log statement was executed.
4707
4708 Use the opt:lttng-enable-event(1):--loglevel or
4709 opt:lttng-enable-event(1):--loglevel-only option of the
4710 man:lttng-enable-event(1) command to target a range of
4711 `java.util.logging` log levels or a specific `java.util.logging` log
4712 level.
4713
4714
4715 [role="since-2.8"]
4716 [[log4j]]
4717 ==== Use the LTTng-UST Java agent for Apache log4j
4718
4719 To use the LTTng-UST Java agent in a Java application which uses
4720 Apache log4j{nbsp}1.2:
4721
4722 . In the source code of the Java application, import the LTTng-UST log
4723   appender package for Apache log4j:
4724 +
4725 --
4726 [source,java]
4727 ----
4728 import org.lttng.ust.agent.log4j.LttngLogAppender;
4729 ----
4730 --
4731
4732 . Create an LTTng-UST log4j log appender:
4733 +
4734 --
4735 [source,java]
4736 ----
4737 Appender lttngUstLogAppender = new LttngLogAppender();
4738 ----
4739 --
4740
4741 . Add this appender to the log4j loggers which should emit LTTng events:
4742 +
4743 --
4744 [source,java]
4745 ----
4746 Logger myLogger = Logger.getLogger("some-logger");
4747
4748 myLogger.addAppender(lttngUstLogAppender);
4749 ----
4750 --
4751
4752 . Use Apache log4j log statements and configuration as usual. The
4753   loggers with an attached LTTng-UST log appender can emit LTTng events.
4754
4755 . Before exiting the application, remove the LTTng-UST log appender from
4756   the loggers attached to it and call its `close()` method:
4757 +
4758 --
4759 [source,java]
4760 ----
4761 myLogger.removeAppender(lttngUstLogAppender);
4762 lttngUstLogAppender.close();
4763 ----
4764 --
4765 +
4766 This isn't strictly necessary, but it's recommended for a clean
4767 disposal of the resources of the appender.
4768
4769 . Include the common and log4j-specific JAR
4770   files of the LTTng-UST Java agent, path:{lttng-ust-agent-common.jar} and
4771   path:{lttng-ust-agent-log4j.jar}, in the
4772   https://docs.oracle.com/javase/tutorial/essential/environment/paths.html[class
4773   path] when you build the Java application.
4774 +
4775 The JAR files are typically located in dir:{/usr/share/java}.
4776 +
4777 IMPORTANT: The LTTng-UST Java agent must be
4778 <<installing-lttng,installed>> for the logging framework your
4779 application uses.
4780
4781 .Use the LTTng-UST Java agent for Apache log4j.
4782 ====
4783 [source,java]
4784 .path:{Test.java}
4785 ----
4786 import org.apache.log4j.Appender;
4787 import org.apache.log4j.Logger;
4788 import org.lttng.ust.agent.log4j.LttngLogAppender;
4789
4790 public class Test
4791 {
4792     private static final int answer = 42;
4793
4794     public static void main(String[] argv) throws Exception
4795     {
4796         // Create a logger
4797         Logger logger = Logger.getLogger("jello");
4798
4799         // Create an LTTng-UST log appender
4800         Appender lttngUstLogAppender = new LttngLogAppender();
4801
4802         // Add the LTTng-UST log appender to our logger
4803         logger.addAppender(lttngUstLogAppender);
4804
4805         // Log at will!
4806         logger.info("some info");
4807         logger.warn("some warning");
4808         Thread.sleep(500);
4809         logger.debug("debug information; the answer is " + answer);
4810         Thread.sleep(123);
4811         logger.fatal("error!");
4812
4813         // Not mandatory, but cleaner
4814         logger.removeAppender(lttngUstLogAppender);
4815         lttngUstLogAppender.close();
4816     }
4817 }
4818
4819 ----
4820
4821 Build this example (`$LOG4JPATH` is the path to the Apache log4j JAR
4822 file):
4823
4824 [role="term"]
4825 ----
4826 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH Test.java
4827 ----
4828
4829 <<creating-destroying-tracing-sessions,Create a recording session>>,
4830 <<enabling-disabling-events,create a recording event rule>> matching
4831 log4j events named `jello`, and <<basic-tracing-session-control,start
4832 recording>>:
4833
4834 [role="term"]
4835 ----
4836 $ lttng create
4837 $ lttng enable-event --log4j jello
4838 $ lttng start
4839 ----
4840
4841 Run the compiled class:
4842
4843 [role="term"]
4844 ----
4845 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH:. Test
4846 ----
4847
4848 <<basic-tracing-session-control,Stop recording>> and inspect the
4849 recorded events:
4850
4851 [role="term"]
4852 ----
4853 $ lttng stop
4854 $ lttng view
4855 ----
4856 ====
4857
4858 In the resulting trace, an <<event,event record>> which a Java
4859 application using log4j generated is named `lttng_log4j:event` and
4860 has the following fields:
4861
4862 `msg`::
4863   Log record message.
4864
4865 `logger_name`::
4866   Logger name.
4867
4868 `class_name`::
4869   Name of the class in which the log statement was executed.
4870
4871 `method_name`::
4872   Name of the method in which the log statement was executed.
4873
4874 `filename`::
4875   Name of the file in which the executed log statement is located.
4876
4877 `line_number`::
4878   Line number at which the log statement was executed.
4879
4880 `timestamp`::
4881   Logging timestamp.
4882
4883 `int_loglevel`::
4884   Log level integer value.
4885
4886 `thread_name`::
4887   Name of the Java thread in which the log statement was executed.
4888
4889 Use the opt:lttng-enable-event(1):--loglevel or
4890 opt:lttng-enable-event(1):--loglevel-only option of the
4891 man:lttng-enable-event(1) command to target a range of Apache log4j
4892 log levels or a specific log4j log level.
4893
4894
4895 [role="since-2.8"]
4896 [[java-application-context]]
4897 ==== Provide application-specific context fields in a Java application
4898
4899 A Java application-specific context field is a piece of state which
4900 the Java application provides. You can <<adding-context,add>> such
4901 a context field to be recorded, using the
4902 man:lttng-add-context(1) command, to each <<event,event record>>
4903 which the log statements of this application produce.
4904
4905 For example, a given object might have a current request ID variable.
4906 You can create a context information retriever for this object and
4907 assign a name to this current request ID. You can then, using the
4908 man:lttng-add-context(1) command, add this context field by name so that
4909 LTTng writes it to the event records of a given `java.util.logging` or
4910 log4j <<channel,channel>>.
4911
4912 To provide application-specific context fields in a Java application:
4913
4914 . In the source code of the Java application, import the LTTng-UST
4915   Java agent context classes and interfaces:
4916 +
4917 --
4918 [source,java]
4919 ----
4920 import org.lttng.ust.agent.context.ContextInfoManager;
4921 import org.lttng.ust.agent.context.IContextInfoRetriever;
4922 ----
4923 --
4924
4925 . Create a context information retriever class, that is, a class which
4926   implements the `IContextInfoRetriever` interface:
4927 +
4928 --
4929 [source,java]
4930 ----
4931 class MyContextInfoRetriever implements IContextInfoRetriever
4932 {
4933     @Override
4934     public Object retrieveContextInfo(String key)
4935     {
4936         if (key.equals("intCtx")) {
4937             return (short) 17;
4938         } else if (key.equals("strContext")) {
4939             return "context value!";
4940         } else {
4941             return null;
4942         }
4943     }
4944 }
4945 ----
4946 --
4947 +
4948 This `retrieveContextInfo()` method is the only member of the
4949 `IContextInfoRetriever` interface. Its role is to return the current
4950 value of a state by name to create a context field. The names of the
4951 context fields and which state variables they return depends on your
4952 specific scenario.
4953 +
4954 All primitive types and objects are supported as context fields.
4955 When `retrieveContextInfo()` returns an object, the context field
4956 serializer calls its `toString()` method to add a string field to
4957 event records. The method can also return `null`, which means that
4958 no context field is available for the required name.
4959
4960 . Register an instance of your context information retriever class to
4961   the context information manager singleton:
4962 +
4963 --
4964 [source,java]
4965 ----
4966 IContextInfoRetriever cir = new MyContextInfoRetriever();
4967 ContextInfoManager cim = ContextInfoManager.getInstance();
4968 cim.registerContextInfoRetriever("retrieverName", cir);
4969 ----
4970 --
4971
4972 . Before exiting the application, remove your context information
4973   retriever from the context information manager singleton:
4974 +
4975 --
4976 [source,java]
4977 ----
4978 ContextInfoManager cim = ContextInfoManager.getInstance();
4979 cim.unregisterContextInfoRetriever("retrieverName");
4980 ----
4981 --
4982 +
4983 This isn't strictly necessary, but it's recommended for a clean
4984 disposal of some resources of the manager.
4985
4986 . Build your Java application with LTTng-UST Java agent support as
4987   usual, following the procedure for either the
4988   <<jul,`java.util.logging`>> or <<log4j,Apache log4j>> framework.
4989
4990 .Provide application-specific context fields in a Java application.
4991 ====
4992 [source,java]
4993 .path:{Test.java}
4994 ----
4995 import java.util.logging.Handler;
4996 import java.util.logging.Logger;
4997 import org.lttng.ust.agent.jul.LttngLogHandler;
4998 import org.lttng.ust.agent.context.ContextInfoManager;
4999 import org.lttng.ust.agent.context.IContextInfoRetriever;
5000
5001 public class Test
5002 {
5003     // Our context information retriever class
5004     private static class MyContextInfoRetriever
5005     implements IContextInfoRetriever
5006     {
5007         @Override
5008         public Object retrieveContextInfo(String key) {
5009             if (key.equals("intCtx")) {
5010                 return (short) 17;
5011             } else if (key.equals("strContext")) {
5012                 return "context value!";
5013             } else {
5014                 return null;
5015             }
5016         }
5017     }
5018
5019     private static final int answer = 42;
5020
5021     public static void main(String args[]) throws Exception
5022     {
5023         // Get the context information manager instance
5024         ContextInfoManager cim = ContextInfoManager.getInstance();
5025
5026         // Create and register our context information retriever
5027         IContextInfoRetriever cir = new MyContextInfoRetriever();
5028         cim.registerContextInfoRetriever("myRetriever", cir);
5029
5030         // Create a logger
5031         Logger logger = Logger.getLogger("jello");
5032
5033         // Create an LTTng-UST log handler
5034         Handler lttngUstLogHandler = new LttngLogHandler();
5035
5036         // Add the LTTng-UST log handler to our logger
5037         logger.addHandler(lttngUstLogHandler);
5038
5039         // Log at will!
5040         logger.info("some info");
5041         logger.warning("some warning");
5042         Thread.sleep(500);
5043         logger.finer("finer information; the answer is " + answer);
5044         Thread.sleep(123);
5045         logger.severe("error!");
5046
5047         // Not mandatory, but cleaner
5048         logger.removeHandler(lttngUstLogHandler);
5049         lttngUstLogHandler.close();
5050         cim.unregisterContextInfoRetriever("myRetriever");
5051     }
5052 }
5053 ----
5054
5055 Build this example:
5056
5057 [role="term"]
5058 ----
5059 $ javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java
5060 ----
5061
5062 <<creating-destroying-tracing-sessions,Create a recording session>> and
5063 <<enabling-disabling-events,create a recording event rule>> matching
5064 `java.util.logging` events named `jello`:
5065
5066 [role="term"]
5067 ----
5068 $ lttng create
5069 $ lttng enable-event --jul jello
5070 ----
5071
5072 <<adding-context,Add the application-specific context fields>> to be
5073 recorded to the event records of the `java.util.logging` channel:
5074
5075 [role="term"]
5076 ----
5077 $ lttng add-context --jul --type='$app.myRetriever:intCtx'
5078 $ lttng add-context --jul --type='$app.myRetriever:strContext'
5079 ----
5080
5081 <<basic-tracing-session-control,Start recording>>:
5082
5083 [role="term"]
5084 ----
5085 $ lttng start
5086 ----
5087
5088 Run the compiled class:
5089
5090 [role="term"]
5091 ----
5092 $ java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test
5093 ----
5094
5095 <<basic-tracing-session-control,Stop recording>> and inspect the
5096 recorded events:
5097
5098 [role="term"]
5099 ----
5100 $ lttng stop
5101 $ lttng view
5102 ----
5103 ====
5104
5105
5106 [role="since-2.7"]
5107 [[python-application]]
5108 === Instrument a Python application
5109
5110 You can instrument a Python{nbsp}2 or Python{nbsp}3 application which
5111 uses the standard
5112 https://docs.python.org/3/library/logging.html[`logging`] package.
5113
5114 Each log statement creates an LTTng event once the application module
5115 imports the <<lttng-ust-agents,LTTng-UST Python agent>> package.
5116
5117 [role="img-100"]
5118 .A Python application importing the LTTng-UST Python agent.
5119 image::python-app.png[]
5120
5121 To use the LTTng-UST Python agent:
5122
5123 . In the source code of the Python application, import the LTTng-UST
5124   Python agent:
5125 +
5126 --
5127 [source,python]
5128 ----
5129 import lttngust
5130 ----
5131 --
5132 +
5133 The LTTng-UST Python agent automatically adds its logging handler to the
5134 root logger at import time.
5135 +
5136 A log statement that the application executes before this import doesn't
5137 create an LTTng event.
5138 +
5139 IMPORTANT: The LTTng-UST Python agent must be
5140 <<installing-lttng,installed>>.
5141
5142 . Use log statements and logging configuration as usual.
5143   Since the LTTng-UST Python agent adds a handler to the _root_
5144   logger, any log statement from any logger can emit an LTTng event.
5145
5146 .Use the LTTng-UST Python agent.
5147 ====
5148 [source,python]
5149 .path:{test.py}
5150 ----
5151 import lttngust
5152 import logging
5153 import time
5154
5155
5156 def example():
5157     logging.basicConfig()
5158     logger = logging.getLogger('my-logger')
5159
5160     while True:
5161         logger.debug('debug message')
5162         logger.info('info message')
5163         logger.warn('warn message')
5164         logger.error('error message')
5165         logger.critical('critical message')
5166         time.sleep(1)
5167
5168
5169 if __name__ == '__main__':
5170     example()
5171 ----
5172
5173 NOTE: `logging.basicConfig()`, which adds to the root logger a basic
5174 logging handler which prints to the standard error stream, isn't
5175 strictly required for LTTng-UST tracing to work, but in versions of
5176 Python preceding{nbsp}3.2, you could see a warning message which
5177 indicates that no handler exists for the logger `my-logger`.
5178
5179 <<creating-destroying-tracing-sessions,Create a recording session>>,
5180 <<enabling-disabling-events,create a recording event rule>> matching
5181 Python logging events named `my-logger`, and
5182 <<basic-tracing-session-control,start recording>>:
5183
5184 [role="term"]
5185 ----
5186 $ lttng create
5187 $ lttng enable-event --python my-logger
5188 $ lttng start
5189 ----
5190
5191 Run the Python script:
5192
5193 [role="term"]
5194 ----
5195 $ python test.py
5196 ----
5197
5198 <<basic-tracing-session-control,Stop recording>> and inspect the
5199 recorded events:
5200
5201 [role="term"]
5202 ----
5203 $ lttng stop
5204 $ lttng view
5205 ----
5206 ====
5207
5208 In the resulting trace, an <<event,event record>> which a Python
5209 application generated is named `lttng_python:event` and has the
5210 following fields:
5211
5212 `asctime`::
5213   Logging time (string).
5214
5215 `msg`::
5216   Log record message.
5217
5218 `logger_name`::
5219   Logger name.
5220
5221 `funcName`::
5222   Name of the function in which the log statement was executed.
5223
5224 `lineno`::
5225   Line number at which the log statement was executed.
5226
5227 `int_loglevel`::
5228   Log level integer value.
5229
5230 `thread`::
5231   ID of the Python thread in which the log statement was executed.
5232
5233 `threadName`::
5234   Name of the Python thread in which the log statement was executed.
5235
5236 Use the opt:lttng-enable-event(1):--loglevel or
5237 opt:lttng-enable-event(1):--loglevel-only option of the
5238 man:lttng-enable-event(1) command to target a range of Python log levels
5239 or a specific Python log level.
5240
5241 When an application imports the LTTng-UST Python agent, the agent tries
5242 to register to a <<lttng-sessiond,session daemon>>. Note that you must
5243 <<start-sessiond,start the session daemon>> _before_ you run the Python
5244 application. If a session daemon is found, the agent tries to register
5245 to it during five seconds, after which the application continues
5246 without LTTng tracing support. Override this timeout value with
5247 the env:LTTNG_UST_PYTHON_REGISTER_TIMEOUT environment variable
5248 (milliseconds).
5249
5250 If the session daemon stops while a Python application with an imported
5251 LTTng-UST Python agent runs, the agent retries to connect and to
5252 register to a session daemon every three seconds. Override this
5253 delay with the env:LTTNG_UST_PYTHON_REGISTER_RETRY_DELAY environment
5254 variable.
5255
5256
5257 [role="since-2.5"]
5258 [[proc-lttng-logger-abi]]
5259 === Use the LTTng logger
5260
5261 The `lttng-tracer` Linux kernel module, part of
5262 <<lttng-modules,LTTng-modules>>, creates the special LTTng logger files
5263 path:{/proc/lttng-logger} and path:{/dev/lttng-logger} (since
5264 LTTng{nbsp}2.11) when it's loaded. Any application can write text data
5265 to any of those files to create one or more LTTng events.
5266
5267 [role="img-100"]
5268 .An application writes to the LTTng logger file to create one or more LTTng events.
5269 image::lttng-logger.png[]
5270
5271 The LTTng logger is the quickest method--not the most efficient,
5272 however--to add instrumentation to an application. It's designed
5273 mostly to instrument shell scripts:
5274
5275 [role="term"]
5276 ----
5277 $ echo "Some message, some $variable" > /dev/lttng-logger
5278 ----
5279
5280 Any event that the LTTng logger creates is named `lttng_logger` and
5281 belongs to the Linux kernel <<domain,tracing domain>>. However, unlike
5282 other instrumentation points in the kernel tracing domain, **any Unix
5283 user** can <<enabling-disabling-events,create a recording event rule>>
5284 which matches events named `lttng_logger`, not only the root user or
5285 users in the <<tracing-group,tracing group>>.
5286
5287 To use the LTTng logger:
5288
5289 * From any application, write text data to the path:{/dev/lttng-logger}
5290   file.
5291
5292 The `msg` field of `lttng_logger` event records contains the
5293 recorded message.
5294
5295 NOTE: The maximum message length of an LTTng logger event is
5296 1024{nbsp}bytes. Writing more than this makes the LTTng logger emit more
5297 than one event to contain the remaining data.
5298
5299 You shouldn't use the LTTng logger to trace a user application which you
5300 can instrument in a more efficient way, namely:
5301
5302 * <<c-application,C and $$C++$$ applications>>.
5303 * <<java-application,Java applications>>.
5304 * <<python-application,Python applications>>.
5305
5306 .Use the LTTng logger.
5307 ====
5308 [source,bash]
5309 .path:{test.bash}
5310 ----
5311 echo 'Hello, World!' > /dev/lttng-logger
5312 sleep 2
5313 df --human-readable --print-type / > /dev/lttng-logger
5314 ----
5315
5316 <<creating-destroying-tracing-sessions,Create a recording session>>,
5317 <<enabling-disabling-events,create a recording event rule>> matching
5318 Linux kernel tracepoint events named `lttng_logger`, and
5319 <<basic-tracing-session-control,start recording>>:
5320
5321 [role="term"]
5322 ----
5323 $ lttng create
5324 $ lttng enable-event --kernel lttng_logger
5325 $ lttng start
5326 ----
5327
5328 Run the Bash script:
5329
5330 [role="term"]
5331 ----
5332 $ bash test.bash
5333 ----
5334
5335 <<basic-tracing-session-control,Stop recording>> and inspect the recorded
5336 events:
5337
5338 [role="term"]
5339 ----
5340 $ lttng stop
5341 $ lttng view
5342 ----
5343 ====
5344
5345
5346 [[instrumenting-linux-kernel]]
5347 === Instrument a Linux kernel image or module
5348
5349 NOTE: This section shows how to _add_ instrumentation points to the
5350 Linux kernel. The subsystems of the kernel are already thoroughly
5351 instrumented at strategic points for LTTng when you
5352 <<installing-lttng,install>> the <<lttng-modules,LTTng-modules>>
5353 package.
5354
5355
5356 [[linux-add-lttng-layer]]
5357 ==== [[instrumenting-linux-kernel-itself]][[mainline-trace-event]][[lttng-adaptation-layer]]Add an LTTng layer to an existing ftrace tracepoint
5358
5359 This section shows how to add an LTTng layer to existing ftrace
5360 instrumentation using the `TRACE_EVENT()` API.
5361
5362 This section doesn't document the `TRACE_EVENT()` macro. Read the
5363 following articles to learn more about this API:
5364
5365 * https://lwn.net/Articles/379903/[Using the TRACE_EVENT() macro (Part{nbsp}1)]
5366 * https://lwn.net/Articles/381064/[Using the TRACE_EVENT() macro (Part{nbsp}2)]
5367 * https://lwn.net/Articles/383362/[Using the TRACE_EVENT() macro (Part{nbsp}3)]
5368
5369 The following procedure assumes that your ftrace tracepoints are
5370 correctly defined in their own header and that they're created in
5371 one source file using the `CREATE_TRACE_POINTS` definition.
5372
5373 To add an LTTng layer over an existing ftrace tracepoint:
5374
5375 . Make sure the following kernel configuration options are
5376   enabled:
5377 +
5378 --
5379 * `CONFIG_MODULES`
5380 * `CONFIG_KALLSYMS`
5381 * `CONFIG_HIGH_RES_TIMERS`
5382 * `CONFIG_TRACEPOINTS`
5383 --
5384
5385 . Build the Linux source tree with your custom ftrace tracepoints.
5386 . Boot the resulting Linux image on your target system.
5387 +
5388 Confirm that the tracepoints exist by looking for their names in the
5389 dir:{/sys/kernel/debug/tracing/events/subsys} directory, where `subsys`
5390 is your subsystem name.
5391
5392 . Get a copy of the latest LTTng-modules{nbsp}{revision}:
5393 +
5394 --
5395 [role="term"]
5396 ----
5397 $ cd $(mktemp -d) &&
5398   wget https://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
5399   tar -xf lttng-modules-latest-2.13.tar.bz2 &&
5400   cd lttng-modules-2.13.*
5401 ----
5402 --
5403
5404 . In dir:{instrumentation/events/lttng-module}, relative to the root
5405   of the LTTng-modules source tree, create a header file named
5406   +__subsys__.h+ for your custom subsystem +__subsys__+ and write your
5407   LTTng-modules tracepoint definitions using the LTTng-modules
5408   macros in it.
5409 +
5410 Start with this template:
5411 +
5412 --
5413 [source,c]
5414 .path:{instrumentation/events/lttng-module/my_subsys.h}
5415 ----
5416 #undef TRACE_SYSTEM
5417 #define TRACE_SYSTEM my_subsys
5418
5419 #if !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ)
5420 #define _LTTNG_MY_SUBSYS_H
5421
5422 #include "../../../probes/lttng-tracepoint-event.h"
5423 #include <linux/tracepoint.h>
5424
5425 LTTNG_TRACEPOINT_EVENT(
5426     /*
5427      * Format is identical to the TRACE_EVENT() version for the three
5428      * following macro parameters:
5429      */
5430     my_subsys_my_event,
5431     TP_PROTO(int my_int, const char *my_string),
5432     TP_ARGS(my_int, my_string),
5433
5434     /* LTTng-modules specific macros */
5435     TP_FIELDS(
5436         ctf_integer(int, my_int_field, my_int)
5437         ctf_string(my_bar_field, my_bar)
5438     )
5439 )
5440
5441 #endif /* !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ) */
5442
5443 #include "../../../probes/define_trace.h"
5444 ----
5445 --
5446 +
5447 The entries in the `TP_FIELDS()` section are the list of fields for the
5448 LTTng tracepoint. This is similar to the `TP_STRUCT__entry()` part of
5449 the `TRACE_EVENT()` ftrace macro.
5450 +
5451 See ``<<lttng-modules-tp-fields,Tracepoint fields macros>>'' for a
5452 complete description of the available `ctf_*()` macros.
5453
5454 . Create the kernel module C{nbsp}source file of the LTTng-modules
5455   probe, +probes/lttng-probe-__subsys__.c+, where +__subsys__+ is your
5456   subsystem name:
5457 +
5458 --
5459 [source,c]
5460 .path:{probes/lttng-probe-my-subsys.c}
5461 ----
5462 #include <linux/module.h>
5463 #include "../lttng-tracer.h"
5464
5465 /*
5466  * Build-time verification of mismatch between mainline
5467  * TRACE_EVENT() arguments and the LTTng-modules adaptation
5468  * layer LTTNG_TRACEPOINT_EVENT() arguments.
5469  */
5470 #include <trace/events/my_subsys.h>
5471
5472 /* Create LTTng tracepoint probes */
5473 #define LTTNG_PACKAGE_BUILD
5474 #define CREATE_TRACE_POINTS
5475 #define TRACE_INCLUDE_PATH ../instrumentation/events/lttng-module
5476
5477 #include "../instrumentation/events/lttng-module/my_subsys.h"
5478
5479 MODULE_LICENSE("GPL and additional rights");
5480 MODULE_AUTHOR("Your name <your-email>");
5481 MODULE_DESCRIPTION("LTTng my_subsys probes");
5482 MODULE_VERSION(__stringify(LTTNG_MODULES_MAJOR_VERSION) "."
5483     __stringify(LTTNG_MODULES_MINOR_VERSION) "."
5484     __stringify(LTTNG_MODULES_PATCHLEVEL_VERSION)
5485     LTTNG_MODULES_EXTRAVERSION);
5486 ----
5487 --
5488
5489 . Edit path:{probes/KBuild} and add your new kernel module object
5490   next to the existing ones:
5491 +
5492 --
5493 [source,make]
5494 .path:{probes/KBuild}
5495 ----
5496 # ...
5497
5498 obj-m += lttng-probe-module.o
5499 obj-m += lttng-probe-power.o
5500
5501 obj-m += lttng-probe-my-subsys.o
5502
5503 # ...
5504 ----
5505 --
5506
5507 . Build and install the LTTng kernel modules:
5508 +
5509 --
5510 [role="term"]
5511 ----
5512 $ make KERNELDIR=/path/to/linux
5513 # make modules_install && depmod -a
5514 ----
5515 --
5516 +
5517 Replace `/path/to/linux` with the path to the Linux source tree where
5518 you defined and used tracepoints with the `TRACE_EVENT()` ftrace macro.
5519
5520 Note that you can also use the
5521 <<lttng-tracepoint-event-code,`LTTNG_TRACEPOINT_EVENT_CODE()` macro>>
5522 instead of `LTTNG_TRACEPOINT_EVENT()` to use custom local variables and
5523 C{nbsp}code that need to be executed before LTTng records the event
5524 fields.
5525
5526 The best way to learn how to use the previous LTTng-modules macros is to
5527 inspect the existing LTTng-modules tracepoint definitions in the
5528 dir:{instrumentation/events/lttng-module} header files. Compare them
5529 with the Linux kernel mainline versions in the
5530 dir:{include/trace/events} directory of the Linux source tree.
5531
5532
5533 [role="since-2.7"]
5534 [[lttng-tracepoint-event-code]]
5535 ===== Use custom C code to access the data for tracepoint fields
5536
5537 Although we recommended to always use the
5538 <<lttng-adaptation-layer,`LTTNG_TRACEPOINT_EVENT()`>> macro to describe
5539 the arguments and fields of an LTTng-modules tracepoint when possible,
5540 sometimes you need a more complex process to access the data that the
5541 tracer records as event record fields. In other words, you need local
5542 variables and multiple C{nbsp}statements instead of simple
5543 argument-based expressions that you pass to the
5544 <<lttng-modules-tp-fields,`ctf_*()` macros of `TP_FIELDS()`>>.
5545
5546 Use the `LTTNG_TRACEPOINT_EVENT_CODE()` macro instead of
5547 `LTTNG_TRACEPOINT_EVENT()` to declare custom local variables and define
5548 a block of C{nbsp}code to be executed before LTTng records the fields.
5549 The structure of this macro is:
5550
5551 [source,c]
5552 .`LTTNG_TRACEPOINT_EVENT_CODE()` macro syntax.
5553 ----
5554 LTTNG_TRACEPOINT_EVENT_CODE(
5555     /*
5556      * Format identical to the LTTNG_TRACEPOINT_EVENT()
5557      * version for the following three macro parameters:
5558      */
5559     my_subsys_my_event,
5560     TP_PROTO(int my_int, const char *my_string),
5561     TP_ARGS(my_int, my_string),
5562
5563     /* Declarations of custom local variables */
5564     TP_locvar(
5565         int a = 0;
5566         unsigned long b = 0;
5567         const char *name = "(undefined)";
5568         struct my_struct *my_struct;
5569     ),
5570
5571     /*
5572      * Custom code which uses both tracepoint arguments
5573      * (in TP_ARGS()) and local variables (in TP_locvar()).
5574      *
5575      * Local variables are actually members of a structure pointed
5576      * to by the special variable tp_locvar.
5577      */
5578     TP_code(
5579         if (my_int) {
5580             tp_locvar->a = my_int + 17;
5581             tp_locvar->my_struct = get_my_struct_at(tp_locvar->a);
5582             tp_locvar->b = my_struct_compute_b(tp_locvar->my_struct);
5583             tp_locvar->name = my_struct_get_name(tp_locvar->my_struct);
5584             put_my_struct(tp_locvar->my_struct);
5585
5586             if (tp_locvar->b) {
5587                 tp_locvar->a = 1;
5588             }
5589         }
5590     ),
5591
5592     /*
5593      * Format identical to the LTTNG_TRACEPOINT_EVENT()
5594      * version for this, except that tp_locvar members can be
5595      * used in the argument expression parameters of
5596      * the ctf_*() macros.
5597      */
5598     TP_FIELDS(
5599         ctf_integer(unsigned long, my_struct_b, tp_locvar->b)
5600         ctf_integer(int, my_struct_a, tp_locvar->a)
5601         ctf_string(my_string_field, my_string)
5602         ctf_string(my_struct_name, tp_locvar->name)
5603     )
5604 )
5605 ----
5606
5607 IMPORTANT: The C code defined in `TP_code()` must not have any side
5608 effects when executed. In particular, the code must not allocate
5609 memory or get resources without deallocating this memory or putting
5610 those resources afterwards.
5611
5612
5613 [[instrumenting-linux-kernel-tracing]]
5614 ==== Load and unload a custom probe kernel module
5615
5616 You must load a <<lttng-adaptation-layer,created LTTng-modules probe
5617 kernel module>> in the kernel before it can emit LTTng events.
5618
5619 To load the default probe kernel modules and a custom probe kernel
5620 module:
5621
5622 * Use the opt:lttng-sessiond(8):--extra-kmod-probes option to give extra
5623   probe modules to load when starting a root <<lttng-sessiond,session
5624   daemon>>:
5625 +
5626 --
5627 .Load the `my_subsys`, `usb`, and the default probe modules.
5628 ====
5629 [role="term"]
5630 ----
5631 # lttng-sessiond --extra-kmod-probes=my_subsys,usb
5632 ----
5633 ====
5634 --
5635 +
5636 You only need to pass the subsystem name, not the whole kernel module
5637 name.
5638
5639 To load _only_ a given custom probe kernel module:
5640
5641 * Use the opt:lttng-sessiond(8):--kmod-probes option to give the probe
5642   modules to load when starting a root session daemon:
5643 +
5644 --
5645 .Load only the `my_subsys` and `usb` probe modules.
5646 ====
5647 [role="term"]
5648 ----
5649 # lttng-sessiond --kmod-probes=my_subsys,usb
5650 ----
5651 ====
5652 --
5653
5654 To confirm that a probe module is loaded:
5655
5656 * Use man:lsmod(8):
5657 +
5658 --
5659 [role="term"]
5660 ----
5661 $ lsmod | grep lttng_probe_usb
5662 ----
5663 --
5664
5665 To unload the loaded probe modules:
5666
5667 * Kill the session daemon with `SIGTERM`:
5668 +
5669 --
5670 [role="term"]
5671 ----
5672 # pkill lttng-sessiond
5673 ----
5674 --
5675 +
5676 You can also use the `--remove` option of man:modprobe(8) if the session
5677 daemon terminates abnormally.
5678
5679
5680 [[controlling-tracing]]
5681 == Tracing control
5682
5683 Once an application or a Linux kernel is <<instrumenting,instrumented>>
5684 for LTTng tracing, you can _trace_ it.
5685
5686 In the LTTng context, _tracing_ means making sure that LTTng attempts to
5687 execute some action(s) when a CPU executes an instrumentation point.
5688
5689 This section is divided in topics on how to use the various
5690 <<plumbing,components of LTTng>>, in particular the
5691 <<lttng-cli,cmd:lttng command-line tool>>, to _control_ the LTTng
5692 daemons and tracers.
5693
5694 NOTE: In the following subsections, we refer to an man:lttng(1) command
5695 using its man page name. For example, instead of ``Run the `create`
5696 command to'', we write ``Run the man:lttng-create(1) command to''.
5697
5698
5699 [[start-sessiond]]
5700 === Start a session daemon
5701
5702 In some situations, you need to run a <<lttng-sessiond,session daemon>>
5703 (man:lttng-sessiond(8)) _before_ you can use the man:lttng(1)
5704 command-line tool.
5705
5706 You will see the following error when you run a command while no session
5707 daemon is running:
5708
5709 ----
5710 Error: No session daemon is available
5711 ----
5712
5713 The only command that automatically runs a session daemon is
5714 man:lttng-create(1), which you use to
5715 <<creating-destroying-tracing-sessions,create a recording session>>. While
5716 this could be your most used first operation, sometimes it's not. Some
5717 examples are:
5718
5719 * <<list-instrumentation-points,List the available instrumentation points>>.
5720 * <<saving-loading-tracing-session,Load a recording session configuration>>.
5721 * <<add-event-rule-matches-trigger,Add a trigger>>.
5722
5723 All the examples above don't require a recording session to operate on.
5724
5725 [[tracing-group]] Each Unix user can have its own running session daemon
5726 to use the user space LTTng tracer. The session daemon that the `root`
5727 user starts is the only one allowed to control the LTTng kernel tracer.
5728 Members of the Unix _tracing group_ may connect to and control the root
5729 session daemon, even for user space tracing. See the ``Session daemon
5730 connection'' section of man:lttng(1) to learn more about the Unix
5731 tracing group.
5732
5733 To start a user session daemon:
5734
5735 * Run man:lttng-sessiond(8):
5736 +
5737 --
5738 [role="term"]
5739 ----
5740 $ lttng-sessiond --daemonize
5741 ----
5742 --
5743
5744 To start the root session daemon:
5745
5746 * Run man:lttng-sessiond(8) as the `root` user:
5747 +
5748 --
5749 [role="term"]
5750 ----
5751 # lttng-sessiond --daemonize
5752 ----
5753 --
5754
5755 In both cases, remove the opt:lttng-sessiond(8):--daemonize option to
5756 start the session daemon in foreground.
5757
5758 To stop a session daemon, kill its process (see man:kill(1)) with the
5759 standard `TERM` signal.
5760
5761 Note that some Linux distributions could manage the LTTng session daemon
5762 as a service. In this case, we suggest that you use the service manager
5763 to start, restart, and stop session daemons.
5764
5765
5766 [[creating-destroying-tracing-sessions]]
5767 === Create and destroy a recording session
5768
5769 Many LTTng control operations happen in the scope of a
5770 <<tracing-session,recording session>>, which is the dialogue between the
5771 <<lttng-sessiond,session daemon>> and you for everything related to
5772 <<event,event recording>>.
5773
5774 To create a recording session with a generated name:
5775
5776 * Use the man:lttng-create(1) command:
5777 +
5778 --
5779 [role="term"]
5780 ----
5781 $ lttng create
5782 ----
5783 --
5784
5785 The name of the created recording session is `auto` followed by the
5786 creation date.
5787
5788 To create a recording session with a specific name:
5789
5790 * Use the optional argument of the man:lttng-create(1) command:
5791 +
5792 --
5793 [role="term"]
5794 ----
5795 $ lttng create SESSION
5796 ----
5797 --
5798 +
5799 Replace +__SESSION__+ with your specific recording session name.
5800
5801 In <<local-mode,local mode>>, LTTng writes the traces of a recording
5802 session to the +$LTTNG_HOME/lttng-traces/__NAME__-__DATE__-__TIME__+
5803 directory by default, where +__NAME__+ is the name of the recording
5804 session. Note that the env:LTTNG_HOME environment variable defaults to
5805 `$HOME` if not set.
5806
5807 To output LTTng traces to a non-default location:
5808
5809 * Use the opt:lttng-create(1):--output option of the man:lttng-create(1)
5810   command:
5811 +
5812 --
5813 [role="term"]
5814 ----
5815 $ lttng create my-session --output=/tmp/some-directory
5816 ----
5817 --
5818
5819 You may create as many recording sessions as you wish.
5820
5821 To list all the existing recording sessions for your Unix user, or for
5822 all users if your Unix user is `root`:
5823
5824 * Use the man:lttng-list(1) command:
5825 +
5826 --
5827 [role="term"]
5828 ----
5829 $ lttng list
5830 ----
5831 --
5832
5833 [[cur-tracing-session]]When you create a recording session, the
5834 man:lttng-create(1) command sets it as the _current recording session_.
5835 The following man:lttng(1) commands operate on the current recording
5836 session when you don't specify one:
5837
5838 [role="list-3-cols"]
5839 * man:lttng-add-context(1)
5840 * man:lttng-clear(1)
5841 * man:lttng-destroy(1)
5842 * man:lttng-disable-channel(1)
5843 * man:lttng-disable-event(1)
5844 * man:lttng-disable-rotation(1)
5845 * man:lttng-enable-channel(1)
5846 * man:lttng-enable-event(1)
5847 * man:lttng-enable-rotation(1)
5848 * man:lttng-load(1)
5849 * man:lttng-regenerate(1)
5850 * man:lttng-rotate(1)
5851 * man:lttng-save(1)
5852 * man:lttng-snapshot(1)
5853 * man:lttng-start(1)
5854 * man:lttng-status(1)
5855 * man:lttng-stop(1)
5856 * man:lttng-track(1)
5857 * man:lttng-untrack(1)
5858 * man:lttng-view(1)
5859
5860 To change the current recording session:
5861
5862 * Use the man:lttng-set-session(1) command:
5863 +
5864 --
5865 [role="term"]
5866 ----
5867 $ lttng set-session SESSION
5868 ----
5869 --
5870 +
5871 Replace +__SESSION__+ with the name of the new current recording session.
5872
5873 When you're done recording in a given recording session, destroy it.
5874 This operation frees the resources taken by the recording session to
5875 destroy; it doesn't destroy the trace data that LTTng wrote for this
5876 recording session (see ``<<clear,Clear a recording session>>'' for one
5877 way to do this).
5878
5879 To destroy the current recording session:
5880
5881 * Use the man:lttng-destroy(1) command:
5882 +
5883 --
5884 [role="term"]
5885 ----
5886 $ lttng destroy
5887 ----
5888 --
5889
5890 The man:lttng-destroy(1) command also runs the man:lttng-stop(1) command
5891 implicitly (see ``<<basic-tracing-session-control,Start and stop a
5892 recording session>>''). You need to stop recording to make LTTng flush the
5893 remaining trace data and make the trace readable.
5894
5895
5896 [[list-instrumentation-points]]
5897 === List the available instrumentation points
5898
5899 The <<lttng-sessiond,session daemon>> can query the running instrumented
5900 user applications and the Linux kernel to get a list of available
5901 instrumentation points:
5902
5903 * LTTng tracepoints and system calls for the Linux kernel
5904   <<domain,tracing domain>>.
5905
5906 * LTTng tracepoints for the user space tracing domain.
5907
5908 To list the available instrumentation points:
5909
5910 . <<start-sessiond,Make sure>> there's a running
5911   <<lttng-sessiond,session daemon>> to which your Unix user can
5912   connect.
5913
5914 . Use the man:lttng-list(1) command with the option of the requested
5915   tracing domain amongst:
5916 +
5917 --
5918 opt:lttng-list(1):--kernel::
5919     Linux kernel tracepoints.
5920 +
5921 Your Unix user must be `root`, or it must be a member of the Unix
5922 <<tracing-group,tracing group>>.
5923
5924 opt:lttng-list(1):--kernel with opt:lttng-list(1):--syscall::
5925     Linux kernel system calls.
5926 +
5927 Your Unix user must be `root`, or it must be a member of the Unix
5928 <<tracing-group,tracing group>>.
5929
5930 opt:lttng-list(1):--userspace::
5931     User space tracepoints.
5932
5933 opt:lttng-list(1):--jul::
5934    `java.util.logging` loggers.
5935
5936 opt:lttng-list(1):--log4j::
5937     Apache log4j loggers.
5938
5939 opt:lttng-list(1):--python::
5940     Python loggers.
5941 --
5942
5943 .List the available user space tracepoints.
5944 ====
5945 [role="term"]
5946 ----
5947 $ lttng list --userspace
5948 ----
5949 ====
5950
5951 .List the available Linux kernel system calls.
5952 ====
5953 [role="term"]
5954 ----
5955 $ lttng list --kernel --syscall
5956 ----
5957 ====
5958
5959
5960 [[enabling-disabling-events]]
5961 === Create and enable a recording event rule
5962
5963 Once you <<creating-destroying-tracing-sessions,create a recording
5964 session>>, you can create <<event,recording event rules>> with the
5965 man:lttng-enable-event(1) command.
5966
5967 The man:lttng-enable-event(1) command always attaches an event rule to a
5968 <<channel,channel>> on creation. The command can create a _default
5969 channel_, named `channel0`, for you. The man:lttng-enable-event(1)
5970 command reuses the default channel each time you run it for the same
5971 tracing domain and session.
5972
5973 A recording event rule is always enabled at creation time.
5974
5975 The following examples show how to combine the command-line arguments of
5976 the man:lttng-enable-event(1) command to create simple to more complex
5977 recording event rules within the <<cur-tracing-session,current recording
5978 session>>.
5979
5980 .Create a recording event rule matching specific Linux kernel tracepoint events (default channel).
5981 ====
5982 [role="term"]
5983 ----
5984 # lttng enable-event --kernel sched_switch
5985 ----
5986 ====
5987
5988 .Create a recording event rule matching Linux kernel system call events with four specific names (default channel).
5989 ====
5990 [role="term"]
5991 ----
5992 # lttng enable-event --kernel --syscall open,write,read,close
5993 ----
5994 ====
5995
5996 .Create recording event rules matching tracepoint events which satisfy a filter expressions (default channel).
5997 ====
5998 [role="term"]
5999 ----
6000 # lttng enable-event --kernel sched_switch --filter='prev_comm == "bash"'
6001 ----
6002
6003 [role="term"]
6004 ----
6005 # lttng enable-event --kernel --all \
6006                      --filter='$ctx.tid == 1988 || $ctx.tid == 1534'
6007 ----
6008
6009 [role="term"]
6010 ----
6011 $ lttng enable-event --jul my_logger \
6012                      --filter='$app.retriever:cur_msg_id > 3'
6013 ----
6014
6015 IMPORTANT: Make sure to always single-quote the filter string when you
6016 run man:lttng(1) from a shell.
6017
6018 See also ``<<pid-tracking,Allow specific processes to record events>>''
6019 which offers another, more efficient filtering mechanism for process ID,
6020 user ID, and group ID attributes.
6021 ====
6022
6023 .Create a recording event rule matching any user space event from the `my_app` tracepoint provider and with a log level range (default channel).
6024 ====
6025 [role="term"]
6026 ----
6027 $ lttng enable-event --userspace my_app:'*' --loglevel=INFO
6028 ----
6029
6030 IMPORTANT: Make sure to always single-quote the wildcard character when
6031 you run man:lttng(1) from a shell.
6032 ====
6033
6034 .Create a recording event rule matching user space events named specifically, but with name exclusions (default channel).
6035 ====
6036 [role="term"]
6037 ----
6038 $ lttng enable-event --userspace my_app:'*' \
6039                      --exclude=my_app:set_user,my_app:handle_sig
6040 ----
6041 ====
6042
6043 .Create a recording event rule matching any Apache log4j event with a specific log level (default channel).
6044 ====
6045 [role="term"]
6046 ----
6047 $ lttng enable-event --log4j --all --loglevel-only=WARN
6048 ----
6049 ====
6050
6051 .Create a recording event rule, attached to a specific channel, and matching user space tracepoint events named `my_app:my_tracepoint`.
6052 ====
6053 [role="term"]
6054 ----
6055 $ lttng enable-event --userspace my_app:my_tracepoint \
6056                      --channel=my-channel
6057 ----
6058 ====
6059
6060 .Create a recording event rule matching user space probe events for the `malloc` function entry in path:{/usr/lib/libc.so.6}:
6061 ====
6062 [role="term"]
6063 ----
6064 # lttng enable-event --kernel \
6065                      --userspace-probe=/usr/lib/libc.so.6:malloc \
6066                      libc_malloc
6067 ----
6068 ====
6069
6070 .Create a recording event rule matching user space probe events for the `server`/`accept_request` https://www.sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps[USDT probe] in path:{/usr/bin/serv}:
6071 ====
6072 [role="term"]
6073 ----
6074 # lttng enable-event --kernel \
6075                      --userspace-probe=sdt:serv:server:accept_request \
6076                      server_accept_request
6077 ----
6078 ====
6079
6080 The recording event rules of a given channel form a whitelist: as soon
6081 as an event rule matches an event, LTTng emits it _once_ and therefore
6082 <<channel-overwrite-mode-vs-discard-mode,can>> record it. For example,
6083 the following rules both match user space tracepoint events named
6084 `my_app:my_tracepoint` with an `INFO` log level:
6085
6086 [role="term"]
6087 ----
6088 $ lttng enable-event --userspace my_app:my_tracepoint
6089 $ lttng enable-event --userspace my_app:my_tracepoint \
6090                      --loglevel=INFO
6091 ----
6092
6093 The second recording event rule is redundant: the first one includes the
6094 second one.
6095
6096
6097 [[disable-event-rule]]
6098 === Disable a recording event rule
6099
6100 To disable a <<event,recording event rule>> that you
6101 <<enabling-disabling-events,created>> previously, use the
6102 man:lttng-disable-event(1) command.
6103
6104 man:lttng-disable-event(1) can only find recording event rules to
6105 disable by their <<instrumentation-point-types,instrumentation point
6106 type>> and event name conditions. Therefore, you cannot disable
6107 recording event rules having a specific instrumentation point log level
6108 condition, for example.
6109
6110 LTTng doesn't emit (and, therefore, won't record) an event which only
6111 _disabled_ recording event rules match.
6112
6113 .Disable event rules matching Python logging events from the `my-logger` logger (default <<channel,channel>>, <<cur-tracing-session,current recording session>>).
6114 ====
6115 [role="term"]
6116 ----
6117 $ lttng disable-event --python my-logger
6118 ----
6119 ====
6120
6121 .Disable event rules matching all `java.util.logging` events (default channel, recording session `my-session`).
6122 ====
6123 [role="term"]
6124 ----
6125 $ lttng disable-event --jul --session=my-session '*'
6126 ----
6127 ====
6128
6129 .Disable _all_ the Linux kernel recording event rules (channel `my-chan`, current recording session).
6130 ====
6131 The opt:lttng-disable-event(1):--all-events option isn't, like the
6132 opt:lttng-enable-event(1):--all option of the man:lttng-enable-event(1)
6133 command, an alias for the event name globbing pattern `*`: it disables
6134 _all_ the recording event rules of a given channel.
6135
6136 [role="term"]
6137 ----
6138 # lttng disable-event --kernel --channel=my-chan --all-events
6139 ----
6140 ====
6141
6142 NOTE: You can't _remove_ a recording event rule once you create it.
6143
6144
6145 [[status]]
6146 === Get the status of a recording session
6147
6148 To get the status of the <<cur-tracing-session,current recording
6149 session>>, that is, its parameters, its channels, recording event rules,
6150 and their attributes:
6151
6152 * Use the man:lttng-status(1) command:
6153 +
6154 --
6155 [role="term"]
6156 ----
6157 $ lttng status
6158 ----
6159 --
6160
6161 To get the status of any recording session:
6162
6163 * Use the man:lttng-list(1) command with the name of the recording
6164   session:
6165 +
6166 --
6167 [role="term"]
6168 ----
6169 $ lttng list SESSION
6170 ----
6171 --
6172 +
6173 Replace +__SESSION__+ with the recording session name.
6174
6175
6176 [[basic-tracing-session-control]]
6177 === Start and stop a recording session
6178
6179 Once you <<creating-destroying-tracing-sessions,create a recording
6180 session>> and <<enabling-disabling-events,create one or more recording
6181 event rules>>, you can start and stop the tracers for this recording
6182 session.
6183
6184 To start the <<cur-tracing-session,current recording session>>:
6185
6186 * Use the man:lttng-start(1) command:
6187 +
6188 --
6189 [role="term"]
6190 ----
6191 $ lttng start
6192 ----
6193 --
6194
6195 LTTng is flexible: you can launch user applications before or after you
6196 start the tracers. An LTTng tracer only <<event,records an event>> if a
6197 recording event rule matches it, which means the tracer is active.
6198
6199 The `start-session` <<trigger,trigger>> action can also start a recording
6200 session.
6201
6202 To stop the current recording session:
6203
6204 * Use the man:lttng-stop(1) command:
6205 +
6206 --
6207 [role="term"]
6208 ----
6209 $ lttng stop
6210 ----
6211 --
6212 +
6213 If there were <<channel-overwrite-mode-vs-discard-mode,lost event
6214 records>> or lost sub-buffers since the last time you ran
6215 man:lttng-start(1), the man:lttng-stop(1) command prints corresponding
6216 warnings.
6217
6218 IMPORTANT: You need to stop recording to make LTTng flush the remaining
6219 trace data and make the trace readable. Note that the
6220 man:lttng-destroy(1) command (see
6221 ``<<creating-destroying-tracing-sessions,Create and destroy a recording
6222 session>>'') also runs the man:lttng-stop(1) command implicitly.
6223
6224 The `stop-session` <<trigger,trigger>> action can also stop a recording
6225 session.
6226
6227 [role="since-2.12"]
6228 [[clear]]
6229 === Clear a recording session
6230
6231 You might need to remove all the current tracing data of one or more
6232 <<tracing-session,recording sessions>> between multiple attempts to
6233 reproduce a problem without interrupting the LTTng recording activity.
6234
6235 To clear the tracing data of the
6236 <<cur-tracing-session,current recording session>>:
6237
6238 * Use the man:lttng-clear(1) command:
6239 +
6240 --
6241 [role="term"]
6242 ----
6243 $ lttng clear
6244 ----
6245 --
6246
6247 To clear the tracing data of all the recording sessions:
6248
6249 * Use the `lttng clear` command with its opt:lttng-clear(1):--all
6250   option:
6251 +
6252 --
6253 [role="term"]
6254 ----
6255 $ lttng clear --all
6256 ----
6257 --
6258
6259
6260 [[enabling-disabling-channels]]
6261 === Create a channel
6262
6263 Once you <<creating-destroying-tracing-sessions,create a recording
6264 session>>, you can create a <<channel,channel>> with the
6265 man:lttng-enable-channel(1) command.
6266
6267 Note that LTTng can automatically create a default channel when you
6268 <<enabling-disabling-events,create a recording event rule>>.
6269 Therefore, you only need to create a channel when you need non-default
6270 attributes.
6271
6272 Specify each non-default channel attribute with a command-line
6273 option when you run the man:lttng-enable-channel(1) command.
6274
6275 You can only create a custom channel in the Linux kernel and user space
6276 <<domain,tracing domains>>: the Java/Python logging tracing domains have
6277 their own default channel which LTTng automatically creates when you
6278 <<enabling-disabling-events,create a recording event rule>>.
6279
6280 [IMPORTANT]
6281 ====
6282 As of LTTng{nbsp}{revision}, you may _not_ perform the
6283 following operations with the man:lttng-enable-channel(1) command:
6284
6285 * Change an attribute of an existing channel.
6286
6287 * Enable a disabled channel once its recording session has been
6288   <<basic-tracing-session-control,active>> at least once.
6289
6290 * Create a channel once its recording session has been active at
6291   least once.
6292
6293 * Create a user space channel with a given
6294   <<channel-buffering-schemes,buffering scheme>> and create a second
6295   user space channel with a different buffering scheme in the same
6296   recording session.
6297 ====
6298
6299 The following examples show how to combine the command-line options of
6300 the man:lttng-enable-channel(1) command to create simple to more complex
6301 channels within the <<cur-tracing-session,current recording session>>.
6302
6303 .Create a Linux kernel channel with default attributes.
6304 ====
6305 [role="term"]
6306 ----
6307 # lttng enable-channel --kernel my-channel
6308 ----
6309 ====
6310
6311 .Create a user space channel with four sub-buffers or 1{nbsp}MiB each, per CPU, per instrumented process.
6312 ====
6313 [role="term"]
6314 ----
6315 $ lttng enable-channel --userspace --num-subbuf=4 --subbuf-size=1M \
6316                        --buffers-pid my-channel
6317 ----
6318 ====
6319
6320 .[[blocking-timeout-example]]Create a default user space channel with an infinite blocking timeout.
6321 ====
6322 <<creating-destroying-tracing-sessions,Create a recording session>>,
6323 create the channel, <<enabling-disabling-events,create a recording event
6324 rule>>, and <<basic-tracing-session-control,start recording>>:
6325
6326 [role="term"]
6327 ----
6328 $ lttng create
6329 $ lttng enable-channel --userspace --blocking-timeout=inf blocking-chan
6330 $ lttng enable-event --userspace --channel=blocking-chan --all
6331 $ lttng start
6332 ----
6333
6334 Run an application instrumented with LTTng-UST tracepoints and allow it
6335 to block:
6336
6337 [role="term"]
6338 ----
6339 $ LTTNG_UST_ALLOW_BLOCKING=1 my-app
6340 ----
6341 ====
6342
6343 .Create a Linux kernel channel which rotates eight trace files of 4{nbsp}MiB each for each stream.
6344 ====
6345 [role="term"]
6346 ----
6347 # lttng enable-channel --kernel --tracefile-count=8 \
6348                        --tracefile-size=4194304 my-channel
6349 ----
6350 ====
6351
6352 .Create a user space channel in <<overwrite-mode,overwrite>> (or ``flight recorder'') mode.
6353 ====
6354 [role="term"]
6355 ----
6356 $ lttng enable-channel --userspace --overwrite my-channel
6357 ----
6358 ====
6359
6360 .<<enabling-disabling-events,Create>> the same <<event,recording event rule>> attached to two different channels.
6361 ====
6362 [role="term"]
6363 ----
6364 $ lttng enable-event --userspace --channel=my-channel app:tp
6365 $ lttng enable-event --userspace --channel=other-channel app:tp
6366 ----
6367
6368 When a CPU executes the `app:tp` <<c-application,user space
6369 tracepoint>>, the two recording event rules above match the created
6370 event, making LTTng emit the event. Because the recording event rules
6371 are not attached to the same channel, LTTng records the event twice.
6372 ====
6373
6374
6375 [[disable-channel]]
6376 === Disable a channel
6377
6378 To disable a specific channel that you
6379 <<enabling-disabling-channels,created>> previously, use the
6380 man:lttng-disable-channel(1) command.
6381
6382 .Disable a specific Linux kernel channel (<<cur-tracing-session,current recording session>>).
6383 ====
6384 [role="term"]
6385 ----
6386 # lttng disable-channel --kernel my-channel
6387 ----
6388 ====
6389
6390 An enabled channel is an implicit <<event,recording event rule>>
6391 condition.
6392
6393 NOTE: As of LTTng{nbsp}{revision}, you may _not_ enable a disabled
6394 channel once its recording session has been
6395 <<basic-tracing-session-control,started>> at least once.
6396
6397
6398 [[adding-context]]
6399 === Add context fields to be recorded to the event records of a channel
6400
6401 <<event,Event record>> fields in trace files provide important
6402 information about previously emitted events, but sometimes some external
6403 context may help you solve a problem faster.
6404
6405 Examples of context fields are:
6406
6407 * The **process ID**, **thread ID**, **process name**, and
6408   **process priority** of the thread from which LTTng emits the event.
6409
6410 * The **hostname** of the system on which LTTng emits the event.
6411
6412 * The Linux kernel and user call stacks (since LTTng{nbsp}2.11).
6413
6414 * The current values of many possible **performance counters** using
6415   perf, for example:
6416
6417 ** CPU cycles, stalled cycles, idle cycles, and the other cycle types.
6418 ** Cache misses.
6419 ** Branch instructions, misses, and loads.
6420 ** CPU faults.
6421
6422 * Any state defined at the application level (supported for the
6423   `java.util.logging` and Apache log4j <<domain,tracing domains>>).
6424
6425 To get the full list of available context fields:
6426
6427 * Use the opt:lttng-add-context(1):--list option of the
6428   man:lttng-add-context(1) command:
6429 +
6430 [role="term"]
6431 ----
6432 $ lttng add-context --list
6433 ----
6434
6435 .Add context fields to be recorded to the event records of all the <<channel,channels>> of the <<cur-tracing-session,current recording session>>.
6436 ====
6437 The following command line adds the virtual process identifier and the
6438 per-thread CPU cycles count fields to all the user space channels of the
6439 current recording session.
6440
6441 [role="term"]
6442 ----
6443 $ lttng add-context --userspace --type=vpid --type=perf:thread:cpu-cycles
6444 ----
6445 ====
6446
6447 .Add performance counter context fields by raw ID
6448 ====
6449 See man:lttng-add-context(1) for the exact format of the context field
6450 type, which is partly compatible with the format used in
6451 man:perf-record(1).
6452
6453 [role="term"]
6454 ----
6455 # lttng add-context --userspace --type=perf:thread:raw:r0110:test
6456 # lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted
6457 ----
6458 ====
6459
6460 .Add context fields to be recorded to the event records of a specific channel.
6461 ====
6462 The following command line adds the thread identifier and user call
6463 stack context fields to the Linux kernel channel named `my-channel` of
6464 the <<cur-tracing-session,current recording session>>.
6465
6466 [role="term"]
6467 ----
6468 # lttng add-context --kernel --channel=my-channel \
6469                     --type=tid --type=callstack-user
6470 ----
6471 ====
6472
6473 .Add an <<java-application-context,application-specific context field>> to be recorded to the event records of a specific channel.
6474 ====
6475 The following command line makes sure LTTng writes the `cur_msg_id`
6476 context field of the `retriever` context retriever to all the Java
6477 logging <<event,event records>> of the channel named `my-channel`:
6478
6479 [role="term"]
6480 ----
6481 # lttng add-context --kernel --channel=my-channel \
6482                     --type='$app:retriever:cur_msg_id'
6483 ----
6484
6485 IMPORTANT: Make sure to always single-quote the `$` character when you
6486 run man:lttng-add-context(1) from a shell.
6487 ====
6488
6489 NOTE: You can't undo what the man:lttng-add-context(1) command does.
6490
6491
6492 [role="since-2.7"]
6493 [[pid-tracking]]
6494 === Allow specific processes to record events
6495
6496 It's often useful to only allow processes with specific attributes to
6497 record events. For example, you may wish to record all the system calls
6498 which a given process makes (à la man:strace(1)).
6499
6500 The man:lttng-track(1) and man:lttng-untrack(1) commands serve this
6501 purpose. Both commands operate on _inclusion sets_ of process
6502 attributes. The available process attribute types are:
6503
6504 Linux kernel <<domain,tracing domain>>::
6505 +
6506 * Process ID (PID).
6507
6508 * Virtual process ID (VPID).
6509 +
6510 This is the PID as seen by the application.
6511
6512 * Unix user ID (UID).
6513
6514 * Virtual Unix user ID (VUID).
6515 +
6516 This is the UID as seen by the application.
6517
6518 * Unix group ID (GID).
6519
6520 * Virtual Unix group ID (VGID).
6521 +
6522 This is the GID as seen by the application.
6523
6524 User space tracing domain::
6525 +
6526 * VPID
6527 * VUID
6528 * VGID
6529
6530 A <<tracing-session,recording session>> has nine process
6531 attribute inclusion sets: six for the Linux kernel <<domain,tracing domain>>
6532 and three for the user space tracing domain.
6533
6534 For a given recording session, a process{nbsp}__P__ is allowed to record
6535 LTTng events for a given <<domain,tracing domain>>{nbsp}__D__ if _all_
6536 the attributes of{nbsp}__P__ are part of the inclusion sets
6537 of{nbsp}__D__.
6538
6539 Whether a process is allowed or not to record LTTng events is an
6540 implicit condition of all <<event,recording event rules>>. Therefore, if
6541 LTTng creates an event{nbsp}__E__ for a given process, but this process
6542 may not record events, then no recording event rule matches{nbsp}__E__,
6543 which means LTTng won't emit and record{nbsp}__E__.
6544
6545 When you <<creating-destroying-tracing-sessions,create a recording
6546 session>>, all its process attribute inclusion sets contain all the
6547 possible values. In other words, all processes are allowed to record
6548 events.
6549
6550 Add values to an inclusion set with the man:lttng-track(1) command and
6551 remove values with the man:lttng-untrack(1) command.
6552
6553 [NOTE]
6554 ====
6555 The process attribute values are _numeric_.
6556
6557 Should a process with a given ID (part of an inclusion set), for
6558 example, exit, and then a new process be given this same ID, then the
6559 latter would also be allowed to record events.
6560
6561 With the man:lttng-track(1) command, you can add Unix user and group
6562 _names_ to the user and group inclusion sets: the
6563 <<lttng-sessiond,session daemon>> finds the corresponding UID, VUID,
6564 GID, or VGID once on _addition_ to the inclusion set. This means that if
6565 you rename the user or group after you run the man:lttng-track(1)
6566 command, its user/group ID remains part of the inclusion sets.
6567 ====
6568
6569 .Allow processes to record events based on their virtual process ID (VPID).
6570 ====
6571 For the sake of the following example, assume the target system has
6572 16{nbsp}possible VPIDs.
6573
6574 When you
6575 <<creating-destroying-tracing-sessions,create a recording session>>,
6576 the user space VPID inclusion set contains _all_ the possible VPIDs:
6577
6578 [role="img-100"]
6579 .The VPID inclusion set is full.
6580 image::track-all.png[]
6581
6582 When the inclusion set is full and you run the man:lttng-track(1)
6583 command to specify some VPIDs, LTTng:
6584
6585 . Clears the inclusion set.
6586 . Adds the specific VPIDs to the inclusion set.
6587
6588 After:
6589
6590 [role="term"]
6591 ----
6592 $ lttng track --userspace --vpid=3,4,7,10,13
6593 ----
6594
6595 the VPID inclusion set is:
6596
6597 [role="img-100"]
6598 .The VPID inclusion set contains the VPIDs 3, 4, 7, 10, and 13.
6599 image::track-3-4-7-10-13.png[]
6600
6601 Add more VPIDs to the inclusion set afterwards:
6602
6603 [role="term"]
6604 ----
6605 $ lttng track --userspace --vpid=1,15,16
6606 ----
6607
6608 The result is:
6609
6610 [role="img-100"]
6611 .VPIDs 1, 15, and 16 are added to the inclusion set.
6612 image::track-1-3-4-7-10-13-15-16.png[]
6613
6614 The man:lttng-untrack(1) command removes entries from process attribute
6615 inclusion sets. Given the previous example, the following command:
6616
6617 [role="term"]
6618 ----
6619 $ lttng untrack --userspace --vpid=3,7,10,13
6620 ----
6621
6622 leads to this VPID inclusion set:
6623
6624 [role="img-100"]
6625 .VPIDs 3, 7, 10, and 13 are removed from the inclusion set.
6626 image::track-1-4-15-16.png[]
6627
6628 You can make the VPID inclusion set full again with the
6629 opt:lttng-track(1):--all option:
6630
6631 [role="term"]
6632 ----
6633 $ lttng track --userspace --vpid --all
6634 ----
6635
6636 The result is, again:
6637
6638 [role="img-100"]
6639 .The VPID inclusion set is full.
6640 image::track-all.png[]
6641 ====
6642
6643 .Allow specific processes to record events based on their user ID (UID).
6644 ====
6645 A typical use case with process attribute inclusion sets is to start
6646 with an empty inclusion set, then <<basic-tracing-session-control,start
6647 the tracers>>, and finally add values manually while the tracers are
6648 active.
6649
6650 Use the opt:lttng-untrack(1):--all option of the
6651 man:lttng-untrack(1) command to clear the inclusion set after you
6652 <<creating-destroying-tracing-sessions,create a recording session>>, for
6653 example (with UIDs):
6654
6655 [role="term"]
6656 ----
6657 # lttng untrack --kernel --uid --all
6658 ----
6659
6660 gives:
6661
6662 [role="img-100"]
6663 .The UID inclusion set is empty.
6664 image::untrack-all.png[]
6665
6666 If the LTTng tracer runs with this inclusion set configuration, it
6667 records no events within the <<cur-tracing-session,current recording
6668 session>> because no processes is allowed to do so. Use the
6669 man:lttng-track(1) command as usual to add specific values to the UID
6670 inclusion set when you need to, for example:
6671
6672 [role="term"]
6673 ----
6674 # lttng track --kernel --uid=http,11
6675 ----
6676
6677 Result:
6678
6679 [role="img-100"]
6680 .UIDs 6 (`http`) and 11 are part of the UID inclusion set.
6681 image::track-6-11.png[]
6682 ====
6683
6684
6685 [role="since-2.5"]
6686 [[saving-loading-tracing-session]]
6687 === Save and load recording session configurations
6688
6689 Configuring a <<tracing-session,recording session>> can be long. Some of
6690 the tasks involved are:
6691
6692 * <<enabling-disabling-channels,Create channels>> with
6693   specific attributes.
6694
6695 * <<adding-context,Add context fields>> to be recorded to the
6696   <<event,event records>> of specific channels.
6697
6698 * <<enabling-disabling-events,Create recording event rules>> with
6699   specific log level, filter, and other conditions.
6700
6701 If you use LTTng to solve real world problems, chances are you have to
6702 record events using the same recording session setup over and over,
6703 modifying a few variables each time in your instrumented program or
6704 environment.
6705
6706 To avoid constant recording session reconfiguration, the man:lttng(1)
6707 command-line tool can save and load recording session configurations
6708 to/from XML files.
6709
6710 To save a given recording session configuration:
6711
6712 * Use the man:lttng-save(1) command:
6713 +
6714 --
6715 [role="term"]
6716 ----
6717 $ lttng save SESSION
6718 ----
6719 --
6720 +
6721 Replace +__SESSION__+ with the name of the recording session to save.
6722
6723 LTTng saves recording session configurations to
6724 dir:{$LTTNG_HOME/.lttng/sessions} by default. Note that the
6725 env:LTTNG_HOME environment variable defaults to `$HOME` if not set. See
6726 man:lttng-save(1) to learn more about the recording session configuration
6727 output path.
6728
6729 LTTng saves all configuration parameters, for example:
6730
6731 * The recording session name.
6732 * The trace data output path.
6733 * The <<channel,channels>>, with their state and all their attributes.
6734 * The context fields you added to channels.
6735 * The <<event,recording event rules>> with their state and conditions.
6736
6737 To load a recording session:
6738
6739 * Use the man:lttng-load(1) command:
6740 +
6741 --
6742 [role="term"]
6743 ----
6744 $ lttng load SESSION
6745 ----
6746 --
6747 +
6748 Replace +__SESSION__+ with the name of the recording session to load.
6749
6750 When LTTng loads a configuration, it restores your saved recording session
6751 as if you just configured it manually.
6752
6753 You can also save and load many sessions at a time; see
6754 man:lttng-save(1) and man:lttng-load(1) to learn more.
6755
6756
6757 [[sending-trace-data-over-the-network]]
6758 === Send trace data over the network
6759
6760 LTTng can send the recorded trace data of a <<tracing-session,recording
6761 session>> to a remote system over the network instead of writing it to
6762 the local file system.
6763
6764 To send the trace data over the network:
6765
6766 . On the _remote_ system (which can also be the target system),
6767   start an LTTng <<lttng-relayd,relay daemon>> (man:lttng-relayd(8)):
6768 +
6769 --
6770 [role="term"]
6771 ----
6772 $ lttng-relayd
6773 ----
6774 --
6775
6776 . On the _target_ system, create a recording session
6777   <<net-streaming-mode,configured>> to send trace data over the network:
6778 +
6779 --
6780 [role="term"]
6781 ----
6782 $ lttng create my-session --set-url=net://remote-system
6783 ----
6784 --
6785 +
6786 Replace +__remote-system__+ with the host name or IP address of the
6787 remote system. See man:lttng-create(1) for the exact URL format.
6788
6789 . On the target system, use the man:lttng(1) command-line tool as usual.
6790 +
6791 When recording is <<basic-tracing-session-control,active>>, the
6792 <<lttng-consumerd,consumer daemon>> of the target sends the contents of
6793 <<channel,sub-buffers>> to the remote relay daemon instead of flushing
6794 them to the local file system. The relay daemon writes the received
6795 packets to its local file system.
6796
6797 See the ``Output directory'' section of man:lttng-relayd(8) to learn
6798 where a relay daemon writes its received trace data.
6799
6800
6801 [role="since-2.4"]
6802 [[lttng-live]]
6803 === View events as LTTng records them (noch:{LTTng} live)
6804
6805 _LTTng live_ is a network protocol implemented by the
6806 <<lttng-relayd,relay daemon>> (man:lttng-relayd(8)) to allow compatible
6807 trace readers to display or analyze <<event,event records>> as LTTng
6808 records events on the target system while recording is
6809 <<basic-tracing-session-control,active>>.
6810
6811 The relay daemon creates a _tee_: it forwards the trace data to both the
6812 local file system and to connected live readers:
6813
6814 [role="img-90"]
6815 .The relay daemon creates a _tee_, forwarding the trace data to both trace files and a connected live reader.
6816 image::live.png[]
6817
6818 To use LTTng live:
6819
6820 . On the _target system_, create a <<tracing-session,recording session>>
6821   in _live mode_:
6822 +
6823 --
6824 [role="term"]
6825 ----
6826 $ lttng create my-session --live
6827 ----
6828 --
6829 +
6830 This operation spawns a local relay daemon.
6831
6832 . Start the live reader and configure it to connect to the relay daemon.
6833 +
6834 For example, with man:babeltrace2(1):
6835 +
6836 --
6837 [role="term"]
6838 ----
6839 $ babeltrace2 net://localhost/host/HOSTNAME/my-session
6840 ----
6841 --
6842 +
6843 Replace +__HOSTNAME__+ with the host name of the target system.
6844
6845 . Configure the recording session as usual with the man:lttng(1)
6846   command-line tool, and <<basic-tracing-session-control,start recording>>.
6847
6848 List the available live recording sessions with man:babeltrace2(1):
6849
6850 [role="term"]
6851 ----
6852 $ babeltrace2 net://localhost
6853 ----
6854
6855 You can start the relay daemon on another system. In this case, you need
6856 to specify the URL of the relay daemon when you
6857 <<creating-destroying-tracing-sessions,create the recording session>> with
6858 the opt:lttng-create(1):--set-url option of the man:lttng-create(1)
6859 command. You also need to replace +__localhost__+ in the procedure above
6860 with the host name of the system on which the relay daemon runs.
6861
6862
6863 [role="since-2.3"]
6864 [[taking-a-snapshot]]
6865 === Take a snapshot of the current sub-buffers of a recording session
6866
6867 The normal behavior of LTTng is to append full sub-buffers to growing
6868 trace data files. This is ideal to keep a full history of the events
6869 which the target system emitted, but it can represent too much data in
6870 some situations.
6871
6872 For example, you may wish to have LTTng record your application
6873 continuously until some critical situation happens, in which case you
6874 only need the latest few recorded events to perform the desired
6875 analysis, not multi-gigabyte trace files.
6876
6877 With the man:lttng-snapshot(1) command, you can take a _snapshot_ of the
6878 current <<channel,sub-buffers>> of a given <<tracing-session,recording
6879 session>>. LTTng can write the snapshot to the local file system or send
6880 it over the network.
6881
6882 [role="img-100"]
6883 .A snapshot is a copy of the current sub-buffers, which LTTng does _not_ clear after the operation.
6884 image::snapshot.png[]
6885
6886 The snapshot feature of LTTng is similar to how a
6887 https://en.wikipedia.org/wiki/Flight_recorder[flight recorder] or the
6888 ``roll'' mode of an oscilloscope work.
6889
6890 TIP: If you wish to create unmanaged, self-contained, non-overlapping
6891 trace chunk archives instead of a simple copy of the current
6892 sub-buffers, see the <<session-rotation,recording session rotation>>
6893 feature (available since LTTng{nbsp}2.11).
6894
6895 To take a snapshot of the <<cur-tracing-session,current recording
6896 session>>:
6897
6898 . Create a recording session in <<snapshot-mode,snapshot mode>>:
6899 +
6900 --
6901 [role="term"]
6902 ----
6903 $ lttng create my-session --snapshot
6904 ----
6905 --
6906 +
6907 The <<channel-overwrite-mode-vs-discard-mode,event record loss mode>> of
6908 <<channel,channels>> created in this mode is automatically set to
6909 <<overwrite-mode,_overwrite_>>.
6910
6911 . Configure the recording session as usual with the man:lttng(1)
6912   command-line tool, and <<basic-tracing-session-control,start
6913   recording>>.
6914
6915 . **Optional**: When you need to take a snapshot,
6916   <<basic-tracing-session-control,stop recording>>.
6917 +
6918 You can take a snapshot when the tracers are active, but if you stop
6919 them first, you're guaranteed that the trace data in the sub-buffers
6920 doesn't change before you actually take the snapshot.
6921
6922 . Take a snapshot:
6923 +
6924 --
6925 [role="term"]
6926 ----
6927 $ lttng snapshot record --name=my-first-snapshot
6928 ----
6929 --
6930 +
6931 LTTng writes the current sub-buffers of all the channels of the
6932 <<cur-tracing-session,current recording session>> to
6933 trace files on the local file system. Those trace files have
6934 `my-first-snapshot` in their name.
6935
6936 There's no difference between the format of a normal trace file and the
6937 format of a snapshot: LTTng trace readers also support LTTng snapshots.
6938
6939 By default, LTTng writes snapshot files to the path shown by
6940
6941 [role="term"]
6942 ----
6943 $ lttng snapshot list-output
6944 ----
6945
6946 You can change this path or decide to send snapshots over the network
6947 using either:
6948
6949 . An output path or URL that you specify when you
6950   <<creating-destroying-tracing-sessions,create the recording session>>.
6951
6952 . A snapshot output path or URL that you add using the
6953   `add-output` action of the man:lttng-snapshot(1) command.
6954
6955 . An output path or URL that you provide directly to the
6956   `record` action of the man:lttng-snapshot(1) command.
6957
6958 Method{nbsp}3 overrides method{nbsp}2, which overrides method 1. When
6959 you specify a URL, a <<lttng-relayd,relay daemon>> must listen on a
6960 remote system (see ``<<sending-trace-data-over-the-network,Send trace
6961 data over the network>>'').
6962
6963 The `snapshot-session` <<trigger,trigger>> action can also take
6964 a recording session snapshot.
6965
6966
6967 [role="since-2.11"]
6968 [[session-rotation]]
6969 === Archive the current trace chunk (rotate a recording session)
6970
6971 The <<taking-a-snapshot,snapshot user guide>> shows how to dump the
6972 current sub-buffers of a recording session to the file system or send them
6973 over the network. When you take a snapshot, LTTng doesn't clear the ring
6974 buffers of the recording session: if you take another snapshot immediately
6975 after, both snapshots could contain overlapping trace data.
6976
6977 Inspired by https://en.wikipedia.org/wiki/Log_rotation[log rotation],
6978 _recording session rotation_ is a feature which appends the content of the
6979 ring buffers to what's already on the file system or sent over the
6980 network since the creation of the recording session or since the last
6981 rotation, and then clears those ring buffers to avoid trace data
6982 overlaps.
6983
6984 What LTTng is about to write when performing a recording session rotation
6985 is called the _current trace chunk_. When LTTng writes or sends over the
6986 network this current trace chunk, it becomes a _trace chunk archive_.
6987 Therefore, a recording session rotation operation _archives_ the current
6988 trace chunk.
6989
6990 [role="img-100"]
6991 .A recording session rotation operation _archives_ the current trace chunk.
6992 image::rotation.png[]
6993
6994 A trace chunk archive is a self-contained LTTng trace which LTTng
6995 doesn't manage anymore: you can read it, modify it, move it, or remove
6996 it.
6997
6998 As of LTTng{nbsp}{revision}, there are three methods to perform a
6999 recording session rotation:
7000
7001 * <<immediate-rotation,Immediately>>.
7002
7003 * With a <<rotation-schedule,rotation schedule>>.
7004
7005 * Through the execution of a `rotate-session` <<trigger,trigger>>
7006   action.
7007
7008 [[immediate-rotation]]To perform an immediate rotation of the
7009 <<cur-tracing-session,current recording session>>:
7010
7011 . <<creating-destroying-tracing-sessions,Create a recording session>> in
7012   <<local-mode,local mode>> or <<net-streaming-mode,network streaming
7013   mode>> (only those two recording session modes support recording session
7014   rotation):
7015 +
7016 --
7017 [role="term"]
7018 ----
7019 # lttng create my-session
7020 ----
7021 --
7022
7023 . <<enabling-disabling-events,Create one or more recording event rules>>
7024   and <<basic-tracing-session-control,start recording>>:
7025 +
7026 --
7027 [role="term"]
7028 ----
7029 # lttng enable-event --kernel sched_'*'
7030 # lttng start
7031 ----
7032 --
7033
7034 . When needed, immediately rotate the current recording session:
7035 +
7036 --
7037 [role="term"]
7038 ----
7039 # lttng rotate
7040 ----
7041 --
7042 +
7043 The man:lttng-rotate(1) command prints the path to the created trace
7044 chunk archive. See its manual page to learn about the format of trace
7045 chunk archive directory names.
7046 +
7047 Perform other immediate rotations while the recording session is active.
7048 It's guaranteed that all the trace chunk archives don't contain
7049 overlapping trace data. You can also perform an immediate rotation once
7050 you have <<basic-tracing-session-control,stopped>> the recording session.
7051
7052 . When you're done recording,
7053   <<creating-destroying-tracing-sessions,destroy the current recording
7054   session>>:
7055 +
7056 --
7057 [role="term"]
7058 ----
7059 # lttng destroy
7060 ----
7061 --
7062 +
7063 The recording session destruction operation creates one last trace chunk
7064 archive from the current trace chunk.
7065
7066 [[rotation-schedule]]A recording session rotation schedule is a planned
7067 rotation which LTTng performs automatically based on one of the
7068 following conditions:
7069
7070 * A timer with a configured period expires.
7071
7072 * The total size of the _flushed_ part of the current trace chunk
7073   becomes greater than or equal to a configured value.
7074
7075 To schedule a rotation of the <<cur-tracing-session,current recording
7076 session>>, set a _rotation schedule_:
7077
7078 . <<creating-destroying-tracing-sessions,Create a recording session>> in
7079   <<local-mode,local mode>> or <<net-streaming-mode,network streaming
7080   mode>> (only those two creation modes support recording session
7081   rotation):
7082 +
7083 --
7084 [role="term"]
7085 ----
7086 # lttng create my-session
7087 ----
7088 --
7089
7090 . <<enabling-disabling-events,Create one or more recording event rules>>:
7091 +
7092 --
7093 [role="term"]
7094 ----
7095 # lttng enable-event --kernel sched_'*'
7096 ----
7097 --
7098
7099 . Set a recording session rotation schedule:
7100 +
7101 --
7102 [role="term"]
7103 ----
7104 # lttng enable-rotation --timer=10s
7105 ----
7106 --
7107 +
7108 In this example, we set a rotation schedule so that LTTng performs a
7109 recording session rotation every ten seconds.
7110 +
7111 See man:lttng-enable-rotation(1) to learn more about other ways to set a
7112 rotation schedule.
7113
7114 . <<basic-tracing-session-control,Start recording>>:
7115 +
7116 --
7117 [role="term"]
7118 ----
7119 # lttng start
7120 ----
7121 --
7122 +
7123 LTTng performs recording session rotations automatically while the
7124 recording session is active thanks to the rotation schedule.
7125
7126 . When you're done recording,
7127   <<creating-destroying-tracing-sessions,destroy the current recording
7128   session>>:
7129 +
7130 --
7131 [role="term"]
7132 ----
7133 # lttng destroy
7134 ----
7135 --
7136 +
7137 The recording session destruction operation creates one last trace chunk
7138 archive from the current trace chunk.
7139
7140 Unset a recording session rotation schedule with the
7141 man:lttng-disable-rotation(1) command.
7142
7143
7144 [role="since-2.13"]
7145 [[add-event-rule-matches-trigger]]
7146 === Add an ``event rule matches'' trigger to a session daemon
7147
7148 With the man:lttng-add-trigger(1) command, you can add a
7149 <<trigger,trigger>> to a <<lttng-sessiond,session daemon>>.
7150
7151 A trigger associates an LTTng tracing condition to one or more actions:
7152 when the condition is satisfied, LTTng attempts to execute the actions.
7153
7154 A trigger doesn't need any <<tracing-session,recording session>> to exist:
7155 it belongs to a session daemon.
7156
7157 As of LTTng{nbsp}{revision}, many condition types are available through
7158 the <<liblttng-ctl-lttng,`liblttng-ctl`>> C{nbsp}API, but the
7159 man:lttng-add-trigger(1) command only accepts the ``event rule matches''
7160 condition.
7161
7162 An ``event rule matches'' condition is satisfied when its event rule
7163 matches an event.
7164
7165 Unlike a <<event,recording event rule>>, the event rule of an
7166 ``event rule matches'' trigger condition has no implicit conditions,
7167 that is:
7168
7169 * It has no enabled/disabled state.
7170 * It has no attached <<channel,channel>>.
7171 * It doesn't belong to a <<tracing-session,recording session>>.
7172
7173 Both the man:lttng-add-trigger(1) and man:lttng-enable-event(1) commands
7174 accept command-line arguments to specify an <<event-rule,event rule>>.
7175 That being said, the former is a more recent command and therefore
7176 follows the common event rule specification format (see
7177 man:lttng-event-rule(7)).
7178
7179 .Start a <<tracing-session,recording session>> when an event rule matches.
7180 ====
7181 This example shows how to add the following trigger to the root
7182 <<lttng-sessiond,session daemon>>:
7183
7184 Condition::
7185     An event rule matches a Linux kernel system call event of which the
7186     name starts with `exec` and `*/ls` matches the `filename` payload
7187     field.
7188 +
7189 With such an event rule, LTTng emits an event when the cmd:ls program
7190 starts.
7191
7192 Action::
7193     <<basic-tracing-session-control,Start the recording session>>
7194     named `pitou`.
7195
7196 To add such a trigger to the root session daemon:
7197
7198 . **If there's no currently running LTTng root session daemon**, start
7199   one:
7200 +
7201 [role="term"]
7202 ----
7203 # lttng-sessiond --daemonize
7204 ----
7205
7206 . <<creating-destroying-tracing-sessions,Create a recording session>>
7207   named `pitou` and
7208   <<enabling-disabling-events,create a recording event rule>> matching
7209   all the system call events:
7210 +
7211 [role="term"]
7212 ----
7213 # lttng create pitou
7214 # lttng enable-event --kernel --syscall --all
7215 ----
7216
7217 . Add the trigger to the root session daemon:
7218 +
7219 [role="term"]
7220 ----
7221 # lttng add-trigger --condition=event-rule-matches \
7222                     --type=syscall --name='exec*' \
7223                     --filter='filename == "*/ls"' \
7224                     --action=start-session pitou
7225 ----
7226 +
7227 Confirm that the trigger exists with the man:lttng-list-triggers(1)
7228 command:
7229 +
7230 [role="term"]
7231 ----
7232 # lttng list-triggers
7233 ----
7234
7235 . Make sure the `pitou` recording session is still inactive (stopped):
7236 +
7237 [role="term"]
7238 ----
7239 # lttng list pitou
7240 ----
7241 +
7242 The first line should be something like:
7243 +
7244 ----
7245 Recording session pitou: [inactive]
7246 ----
7247
7248 Run the cmd:ls program to fire the LTTng trigger above:
7249
7250 [role="term"]
7251 ----
7252 $ ls ~
7253 ----
7254
7255 At this point, the `pitou` recording session should be active
7256 (started). Confirm this with the man:lttng-list(1) command again:
7257
7258 [role="term"]
7259 ----
7260 # lttng list pitou
7261 ----
7262
7263 The first line should now look like:
7264
7265 ----
7266 Recording session pitou: [active]
7267 ----
7268
7269 This line confirms that the LTTng trigger you added fired, therefore
7270 starting the `pitou` recording session.
7271 ====
7272
7273 .[[trigger-event-notif]]Send a notification to a user application when an event rule matches.
7274 ====
7275 This example shows how to add the following trigger to the root
7276 <<lttng-sessiond,session daemon>>:
7277
7278 Condition::
7279     An event rule matches a Linux kernel tracepoint event named
7280     `sched_switch` and of which the value of the `next_comm` payload
7281     field is `bash`.
7282 +
7283 With such an event rule, LTTng emits an event when Linux gives access to
7284 the processor to a process named `bash`.
7285
7286 Action::
7287     Send an LTTng notification to a user application.
7288
7289 Moreover, we'll specify a _capture descriptor_ with the
7290 `event-rule-matches` trigger condition so that the user application can
7291 get the value of a specific `sched_switch` event payload field.
7292
7293 First, write and build the user application:
7294
7295 . Create the C{nbsp}source file of the application:
7296 +
7297 --
7298 [source,c]
7299 .path:{notif-app.c}
7300 ----
7301 #include <stdlib.h>
7302 #include <stdio.h>
7303 #include <stdbool.h>
7304 #include <assert.h>
7305 #include <string.h>
7306 #include <lttng/lttng.h>
7307
7308 /*
7309  * Subscribes to notifications, through the notification channel
7310  * `notification_channel`, which match the condition of the trigger
7311  * named `trigger_name`.
7312  *
7313  * Returns `true` on success.
7314  */
7315 static bool subscribe(struct lttng_notification_channel *notification_channel,
7316                       const char *trigger_name)
7317 {
7318     const struct lttng_condition *condition = NULL;
7319     struct lttng_triggers *triggers = NULL;
7320     unsigned int trigger_count;
7321     unsigned int i;
7322     enum lttng_error_code error_code;
7323     enum lttng_trigger_status trigger_status;
7324     bool ret = false;
7325
7326     /* Get all LTTng triggers */
7327     error_code = lttng_list_triggers(&triggers);
7328     assert(error_code == LTTNG_OK);
7329
7330     /* Get the number of triggers */
7331     trigger_status = lttng_triggers_get_count(triggers, &trigger_count);
7332     assert(trigger_status == LTTNG_TRIGGER_STATUS_OK);
7333
7334     /* Find the trigger named `trigger_name` */
7335     for (i = 0; i < trigger_count; i++) {
7336         const struct lttng_trigger *trigger;
7337         const char *this_trigger_name;
7338
7339         trigger = lttng_triggers_get_at_index(triggers, i);
7340         trigger_status = lttng_trigger_get_name(trigger, &this_trigger_name);
7341         assert(trigger_status == LTTNG_TRIGGER_STATUS_OK);
7342
7343         if (strcmp(this_trigger_name, trigger_name) == 0) {
7344             /* Trigger found: subscribe with its condition */
7345             enum lttng_notification_channel_status notification_channel_status;
7346
7347             notification_channel_status = lttng_notification_channel_subscribe(
7348                 notification_channel,
7349                 lttng_trigger_get_const_condition(trigger));
7350             assert(notification_channel_status ==
7351                    LTTNG_NOTIFICATION_CHANNEL_STATUS_OK);
7352             ret = true;
7353             break;
7354         }
7355     }
7356
7357     lttng_triggers_destroy(triggers);
7358     return ret;
7359 }
7360
7361 /*
7362  * Handles the evaluation `evaluation` of a single notification.
7363  */
7364 static void handle_evaluation(const struct lttng_evaluation *evaluation)
7365 {
7366     enum lttng_evaluation_status evaluation_status;
7367     const struct lttng_event_field_value *array_field_value;
7368     const struct lttng_event_field_value *string_field_value;
7369     enum lttng_event_field_value_status event_field_value_status;
7370     const char *string_field_string_value;
7371
7372     /* Get the value of the first captured (string) field */
7373     evaluation_status = lttng_evaluation_event_rule_matches_get_captured_values(
7374         evaluation, &array_field_value);
7375     assert(evaluation_status == LTTNG_EVALUATION_STATUS_OK);
7376     event_field_value_status =
7377         lttng_event_field_value_array_get_element_at_index(
7378             array_field_value, 0, &string_field_value);
7379     assert(event_field_value_status == LTTNG_EVENT_FIELD_VALUE_STATUS_OK);
7380     assert(lttng_event_field_value_get_type(string_field_value) ==
7381            LTTNG_EVENT_FIELD_VALUE_TYPE_STRING);
7382     event_field_value_status = lttng_event_field_value_string_get_value(
7383             string_field_value, &string_field_string_value);
7384     assert(event_field_value_status == LTTNG_EVENT_FIELD_VALUE_STATUS_OK);
7385
7386     /* Print the string value of the field */
7387     puts(string_field_string_value);
7388 }
7389
7390 int main(int argc, char *argv[])
7391 {
7392     int exit_status = EXIT_SUCCESS;
7393     struct lttng_notification_channel *notification_channel;
7394     enum lttng_notification_channel_status notification_channel_status;
7395     const struct lttng_condition *condition;
7396     const char *trigger_name;
7397     bool subscribe_res;
7398
7399     assert(argc >= 2);
7400     trigger_name = argv[1];
7401
7402     /*
7403      * Create a notification channel.
7404      *
7405      * A notification channel connects the user application to the LTTng
7406      * session daemon.
7407      *
7408      * You can use this notification channel to listen to various types
7409      * of notifications.
7410      */
7411     notification_channel = lttng_notification_channel_create(
7412         lttng_session_daemon_notification_endpoint);
7413     assert(notification_channel);
7414
7415     /*
7416      * Subscribe to notifications which match the condition of the
7417      * trigger named `trigger_name`.
7418      */
7419     if (!subscribe(notification_channel, trigger_name)) {
7420         fprintf(stderr,
7421                 "Error: Failed to subscribe to notifications (trigger `%s`).\n",
7422                 trigger_name);
7423         exit_status = EXIT_FAILURE;
7424         goto end;
7425     }
7426
7427     /*
7428      * Notification loop.
7429      *
7430      * Put this in a dedicated thread to avoid blocking the main thread.
7431      */
7432     while (true) {
7433         struct lttng_notification *notification;
7434         enum lttng_notification_channel_status status;
7435         const struct lttng_evaluation *notification_evaluation;
7436
7437         /* Receive the next notification */
7438         status = lttng_notification_channel_get_next_notification(
7439             notification_channel, &notification);
7440
7441         switch (status) {
7442         case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
7443             break;
7444         case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
7445             /*
7446              * The session daemon can drop notifications if a receiving
7447              * application doesn't consume the notifications fast
7448              * enough.
7449              */
7450             continue;
7451         case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
7452             /*
7453              * The session daemon closed the notification channel.
7454              *
7455              * This is typically caused by a session daemon shutting
7456              * down.
7457              */
7458             goto end;
7459         default:
7460             /* Unhandled conditions or errors */
7461             exit_status = EXIT_FAILURE;
7462             goto end;
7463         }
7464
7465         /*
7466          * Handle the condition evaluation.
7467          *
7468          * A notification provides, amongst other things:
7469          *
7470          * * The condition that caused LTTng to send this notification.
7471          *
7472          * * The condition evaluation, which provides more specific
7473          *   information on the evaluation of the condition.
7474          */
7475         handle_evaluation(lttng_notification_get_evaluation(notification));
7476
7477         /* Destroy the notification object */
7478         lttng_notification_destroy(notification);
7479     }
7480
7481 end:
7482     lttng_notification_channel_destroy(notification_channel);
7483     return exit_status;
7484 }
7485 ----
7486 --
7487 +
7488 This application prints the first captured string field value of the
7489 condition evaluation of each LTTng notification it receives.
7490
7491 . Build the `notif-app` application,
7492   using https://www.freedesktop.org/wiki/Software/pkg-config/[pkg-config]
7493   to provide the right compiler and linker flags:
7494 +
7495 --
7496 [role="term"]
7497 ----
7498 $ gcc -o notif-app notif-app.c $(pkg-config --cflags --libs lttng-ctl)
7499 ----
7500 --
7501
7502 Now, to add the trigger to the root session daemon:
7503
7504 [start=3]
7505 . **If there's no currently running LTTng root session daemon**, start
7506   one:
7507 +
7508 [role="term"]
7509 ----
7510 # lttng-sessiond --daemonize
7511 ----
7512
7513 . Add the trigger, naming it `sched-switch-notif`, to the root
7514   session daemon:
7515 +
7516 [role="term"]
7517 ----
7518 # lttng add-trigger --name=sched-switch-notif \
7519                     --condition=event-rule-matches \
7520                     --type=kernel --name=sched_switch \
7521                     --filter='next_comm == "bash"' --capture=prev_comm \
7522                     --action=notify
7523 ----
7524 +
7525 Confirm that the `sched-switch-notif` trigger exists with the
7526 man:lttng-list-triggers(1) command:
7527 +
7528 [role="term"]
7529 ----
7530 # lttng list-triggers
7531 ----
7532
7533 Run the cmd:notif-app application, passing the name of the trigger
7534 of which to watch the notifications:
7535
7536 [role="term"]
7537 ----
7538 # ./notif-app sched-switch-notif
7539 ----
7540
7541 Now, in an interactive Bash, type a few keys to fire the
7542 `sched-switch-notif` trigger. Watch the `notif-app` application print
7543 the previous process names.
7544 ====
7545
7546 [role="since-2.6"]
7547 [[mi]]
7548 === Use the machine interface
7549
7550 With any command of the man:lttng(1) command-line tool, set the
7551 opt:lttng(1):--mi option to `xml` (before the command name) to get an
7552 XML machine interface output, for example:
7553
7554 [role="term"]
7555 ----
7556 $ lttng --mi=xml list my-session
7557 ----
7558
7559 A schema definition (XSD) is
7560 https://github.com/lttng/lttng-tools/blob/stable-{revision}/src/common/mi-lttng-4.0.xsd[available]
7561 to ease the integration with external tools as much as possible.
7562
7563
7564 [role="since-2.8"]
7565 [[metadata-regenerate]]
7566 === Regenerate the metadata of an LTTng trace
7567
7568 An LTTng trace, which is a https://diamon.org/ctf[CTF] trace, has both
7569 data stream files and a metadata stream file. This metadata file
7570 contains, amongst other things, information about the offset of the
7571 clock sources which LTTng uses to assign timestamps to <<event,event
7572 records>> when recording.
7573
7574 If, once a <<tracing-session,recording session>> is
7575 <<basic-tracing-session-control,started>>, a major
7576 https://en.wikipedia.org/wiki/Network_Time_Protocol[NTP] correction
7577 happens, the clock offset of the trace also needs to be updated. Use
7578 the `metadata` item of the man:lttng-regenerate(1) command to do so.
7579
7580 The main use case of this command is to allow a system to boot with
7581 an incorrect wall time and have LTTng trace it before its wall time
7582 is corrected. Once the system is known to be in a state where its
7583 wall time is correct, you can run `lttng regenerate metadata`.
7584
7585 To regenerate the metadata stream files of the
7586 <<cur-tracing-session,current recording session>>:
7587
7588 * Use the `metadata` item of the man:lttng-regenerate(1) command:
7589 +
7590 --
7591 [role="term"]
7592 ----
7593 $ lttng regenerate metadata
7594 ----
7595 --
7596
7597
7598 [role="since-2.9"]
7599 [[regenerate-statedump]]
7600 === Regenerate the state dump event records of a recording session
7601
7602 The LTTng kernel and user space tracers generate state dump
7603 <<event,event records>> when the application starts or when you
7604 <<basic-tracing-session-control,start a recording session>>.
7605
7606 An analysis can use the state dump event records to set an initial state
7607 before it builds the rest of the state from the subsequent event
7608 records. http://tracecompass.org/[Trace Compass] and
7609 https://github.com/lttng/lttng-analyses[LTTng analyses] are notable
7610 examples of applications which use the state dump of an LTTng trace.
7611
7612 When you <<taking-a-snapshot,take a snapshot>>, it's possible that the
7613 state dump event records aren't included in the snapshot trace files
7614 because they were recorded to a <<channel,sub-buffer>> that has been
7615 consumed or <<overwrite-mode,overwritten>> already.
7616
7617 Use the `statedump` item of the man:lttng-regenerate(1) command to emit
7618 and record the state dump events again.
7619
7620 To regenerate the state dump of the <<cur-tracing-session,current
7621 recording session>>, provided you created it in <<snapshot-mode,snapshot
7622 mode>>, before you take a snapshot:
7623
7624 . Use the `statedump` item of the man:lttng-regenerate(1) command:
7625 +
7626 --
7627 [role="term"]
7628 ----
7629 $ lttng regenerate statedump
7630 ----
7631 --
7632
7633 . <<basic-tracing-session-control,Stop the recording session>>:
7634 +
7635 --
7636 [role="term"]
7637 ----
7638 $ lttng stop
7639 ----
7640 --
7641
7642 . <<taking-a-snapshot,Take a snapshot>>:
7643 +
7644 --
7645 [role="term"]
7646 ----
7647 $ lttng snapshot record --name=my-snapshot
7648 ----
7649 --
7650
7651 Depending on the event throughput, you should run steps{nbsp}1
7652 and{nbsp}2 as closely as possible.
7653
7654 [NOTE]
7655 ====
7656 To record the state dump events, you need to
7657 <<enabling-disabling-events,create recording event rules>> which enable
7658 them:
7659
7660 * The names of LTTng-UST state dump tracepoints start with
7661   `lttng_ust_statedump:`.
7662
7663 * The names of LTTng-modules state dump tracepoints start with
7664   `lttng_statedump_`.
7665 ====
7666
7667
7668 [role="since-2.7"]
7669 [[persistent-memory-file-systems]]
7670 === Record trace data on persistent memory file systems
7671
7672 https://en.wikipedia.org/wiki/Non-volatile_random-access_memory[Non-volatile
7673 random-access memory] (NVRAM) is random-access memory that retains its
7674 information when power is turned off (non-volatile). Systems with such
7675 memory can store data structures in RAM and retrieve them after a
7676 reboot, without flushing to typical _storage_.
7677
7678 Linux supports NVRAM file systems thanks to either
7679 http://pramfs.sourceforge.net/[PRAMFS] or
7680 https://www.kernel.org/doc/Documentation/filesystems/dax.txt[DAX]{nbsp}+{nbsp}http://lkml.iu.edu/hypermail/linux/kernel/1504.1/03463.html[pmem]
7681 (requires Linux{nbsp}4.1+).
7682
7683 This section doesn't describe how to operate such file systems; we
7684 assume that you have a working persistent memory file system.
7685
7686 When you <<creating-destroying-tracing-sessions,create a recording
7687 session>>, you can specify the path of the shared memory holding the
7688 sub-buffers. If you specify a location on an NVRAM file system, then you
7689 can retrieve the latest recorded trace data when the system reboots
7690 after a crash.
7691
7692 To record trace data on a persistent memory file system and retrieve the
7693 trace data after a system crash:
7694
7695 . Create a recording session with a <<channel,sub-buffer>> shared memory
7696   path located on an NVRAM file system:
7697 +
7698 --
7699 [role="term"]
7700 ----
7701 $ lttng create my-session --shm-path=/path/to/shm/on/nvram
7702 ----
7703 --
7704
7705 . Configure the recording session as usual with the man:lttng(1)
7706   command-line tool, and <<basic-tracing-session-control,start
7707   recording>>.
7708
7709 . After a system crash, use the man:lttng-crash(1) command-line tool to
7710   read the trace data recorded on the NVRAM file system:
7711 +
7712 --
7713 [role="term"]
7714 ----
7715 $ lttng-crash /path/to/shm/on/nvram
7716 ----
7717 --
7718
7719 The binary layout of the ring buffer files isn't exactly the same as the
7720 trace files layout. This is why you need to use man:lttng-crash(1)
7721 instead of some standard LTTng trace reader.
7722
7723 To convert the ring buffer files to LTTng trace files:
7724
7725 * Use the opt:lttng-crash(1):--extract option of man:lttng-crash(1):
7726 +
7727 --
7728 [role="term"]
7729 ----
7730 $ lttng-crash --extract=/path/to/trace /path/to/shm/on/nvram
7731 ----
7732 --
7733
7734
7735 [role="since-2.10"]
7736 [[notif-trigger-api]]
7737 === Get notified when the buffer usage of a channel is too high or too low
7738
7739 With the notification and <<trigger,trigger>> C{nbsp}API of
7740 <<liblttng-ctl-lttng,`liblttng-ctl`>>, LTTng can notify your user
7741 application when the buffer usage of one or more <<channel,channels>>
7742 becomes too low or too high.
7743
7744 Use this API and enable or disable <<event,recording event rules>> while
7745 a recording session <<basic-tracing-session-control,is active>> to avoid
7746 <<channel-overwrite-mode-vs-discard-mode,discarded event records>>, for
7747 example.
7748
7749 .Send a notification to a user application when the buffer usage of an LTTng channel is too high.
7750 ====
7751 In this example, we create and build an application which gets notified
7752 when the buffer usage of a specific LTTng channel is higher than
7753 75{nbsp}%.
7754
7755 We only print that it's the case in this example, but we could as well
7756 use the `liblttng-ctl` C{nbsp}API to <<enabling-disabling-events,disable
7757 recording event rules>> when this happens, for example.
7758
7759 . Create the C{nbsp}source file of the application:
7760 +
7761 --
7762 [source,c]
7763 .path:{notif-app.c}
7764 ----
7765 #include <stdlib.h>
7766 #include <stdio.h>
7767 #include <assert.h>
7768 #include <lttng/lttng.h>
7769
7770 int main(int argc, char *argv[])
7771 {
7772     int exit_status = EXIT_SUCCESS;
7773     struct lttng_notification_channel *notification_channel;
7774     struct lttng_condition *condition;
7775     struct lttng_action *action;
7776     struct lttng_trigger *trigger;
7777     const char *recording_session_name;
7778     const char *channel_name;
7779
7780     assert(argc >= 3);
7781     recording_session_name = argv[1];
7782     channel_name = argv[2];
7783
7784     /*
7785      * Create a notification channel.
7786      *
7787      * A notification channel connects the user application to the LTTng
7788      * session daemon.
7789      *
7790      * You can use this notification channel to listen to various types
7791      * of notifications.
7792      */
7793     notification_channel = lttng_notification_channel_create(
7794         lttng_session_daemon_notification_endpoint);
7795
7796     /*
7797      * Create a "buffer usage becomes greater than" condition.
7798      *
7799      * In this case, the condition is satisfied when the buffer usage
7800      * becomes greater than or equal to 75 %.
7801      *
7802      * We create the condition for a specific recording session name,
7803      * channel name, and for the user space tracing domain.
7804      *
7805      * The following condition types also exist:
7806      *
7807      * * The buffer usage of a channel becomes less than a given value.
7808      *
7809      * * The consumed data size of a recording session becomes greater
7810      *   than a given value.
7811      *
7812      * * A recording session rotation becomes ongoing.
7813      *
7814      * * A recording session rotation becomes completed.
7815      *
7816      * * A given event rule matches an event.
7817      */
7818     condition = lttng_condition_buffer_usage_high_create();
7819     lttng_condition_buffer_usage_set_threshold_ratio(condition, .75);
7820     lttng_condition_buffer_usage_set_session_name(condition,
7821                                                   recording_session_name);
7822     lttng_condition_buffer_usage_set_channel_name(condition,
7823                                                   channel_name);
7824     lttng_condition_buffer_usage_set_domain_type(condition,
7825                                                  LTTNG_DOMAIN_UST);
7826
7827     /*
7828      * Create an action (receive a notification) to execute when the
7829      * condition created above is satisfied.
7830      */
7831     action = lttng_action_notify_create();
7832
7833     /*
7834      * Create a trigger.
7835      *
7836      * A trigger associates a condition to an action: LTTng executes
7837      * the action when the condition is satisfied.
7838      */
7839     trigger = lttng_trigger_create(condition, action);
7840
7841     /* Register the trigger to the LTTng session daemon. */
7842     lttng_register_trigger(trigger);
7843
7844     /*
7845      * Now that we have registered a trigger, LTTng will send a
7846      * notification every time its condition is met through a
7847      * notification channel.
7848      *
7849      * To receive this notification, we must subscribe to notifications
7850      * which match the same condition.
7851      */
7852     lttng_notification_channel_subscribe(notification_channel,
7853                                          condition);
7854
7855     /*
7856      * Notification loop.
7857      *
7858      * Put this in a dedicated thread to avoid blocking the main thread.
7859      */
7860     for (;;) {
7861         struct lttng_notification *notification;
7862         enum lttng_notification_channel_status status;
7863         const struct lttng_evaluation *notification_evaluation;
7864         const struct lttng_condition *notification_condition;
7865         double buffer_usage;
7866
7867         /* Receive the next notification. */
7868         status = lttng_notification_channel_get_next_notification(
7869             notification_channel, &notification);
7870
7871         switch (status) {
7872         case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
7873             break;
7874         case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
7875             /*
7876              * The session daemon can drop notifications if a monitoring
7877              * application isn't consuming the notifications fast
7878              * enough.
7879              */
7880             continue;
7881         case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
7882             /*
7883              * The session daemon closed the notification channel.
7884              *
7885              * This is typically caused by a session daemon shutting
7886              * down.
7887              */
7888             goto end;
7889         default:
7890             /* Unhandled conditions or errors. */
7891             exit_status = EXIT_FAILURE;
7892             goto end;
7893         }
7894
7895         /*
7896          * A notification provides, amongst other things:
7897          *
7898          * * The condition that caused LTTng to send this notification.
7899          *
7900          * * The condition evaluation, which provides more specific
7901          *   information on the evaluation of the condition.
7902          *
7903          * The condition evaluation provides the buffer usage
7904          * value at the moment the condition was satisfied.
7905          */
7906         notification_condition = lttng_notification_get_condition(
7907             notification);
7908         notification_evaluation = lttng_notification_get_evaluation(
7909             notification);
7910
7911         /* We're subscribed to only one condition. */
7912         assert(lttng_condition_get_type(notification_condition) ==
7913             LTTNG_CONDITION_TYPE_BUFFER_USAGE_HIGH);
7914
7915         /*
7916          * Get the exact sampled buffer usage from the condition
7917          * evaluation.
7918          */
7919         lttng_evaluation_buffer_usage_get_usage_ratio(
7920             notification_evaluation, &buffer_usage);
7921
7922         /*
7923          * At this point, instead of printing a message, we could do
7924          * something to reduce the buffer usage of the channel, like
7925          * disable specific events, for example.
7926          */
7927         printf("Buffer usage is %f %% in recording session \"%s\", "
7928                "user space channel \"%s\".\n", buffer_usage * 100,
7929                recording_session_name, channel_name);
7930
7931         /* Destroy the notification object. */
7932         lttng_notification_destroy(notification);
7933     }
7934
7935 end:
7936     lttng_action_destroy(action);
7937     lttng_condition_destroy(condition);
7938     lttng_trigger_destroy(trigger);
7939     lttng_notification_channel_destroy(notification_channel);
7940     return exit_status;
7941 }
7942 ----
7943 --
7944
7945 . Build the `notif-app` application, linking it with `liblttng-ctl`:
7946 +
7947 --
7948 [role="term"]
7949 ----
7950 $ gcc -o notif-app notif-app.c $(pkg-config --cflags --libs lttng-ctl)
7951 ----
7952 --
7953
7954 . <<creating-destroying-tracing-sessions,Create a recording session>>,
7955   <<enabling-disabling-events,create a recording event rule>> matching
7956   all the user space tracepoint events, and
7957   <<basic-tracing-session-control,start recording>>:
7958 +
7959 --
7960 [role="term"]
7961 ----
7962 $ lttng create my-session
7963 $ lttng enable-event --userspace --all
7964 $ lttng start
7965 ----
7966 --
7967 +
7968 If you create the channel manually with the man:lttng-enable-channel(1)
7969 command, you can set its <<channel-monitor-timer,monitor timer>> to
7970 control how frequently LTTng samples the current values of the channel
7971 properties to evaluate user conditions.
7972
7973 . Run the `notif-app` application.
7974 +
7975 This program accepts the <<tracing-session,recording session>> and
7976 user space channel names as its two first arguments. The channel
7977 which LTTng automatically creates with the man:lttng-enable-event(1)
7978 command above is named `channel0`:
7979 +
7980 --
7981 [role="term"]
7982 ----
7983 $ ./notif-app my-session channel0
7984 ----
7985 --
7986
7987 . In another terminal, run an application with a very high event
7988   throughput so that the 75{nbsp}% buffer usage condition is reached.
7989 +
7990 In the first terminal, the application should print lines like this:
7991 +
7992 ----
7993 Buffer usage is 81.45197 % in recording session "my-session", user space
7994 channel "channel0".
7995 ----
7996 +
7997 If you don't see anything, try to make the threshold of the condition in
7998 path:{notif-app.c} lower (0.1{nbsp}%, for example), and then rebuild the
7999 `notif-app` application (step{nbsp}2) and run it again (step{nbsp}4).
8000 ====
8001
8002
8003 [[reference]]
8004 == Reference
8005
8006 [[lttng-modules-ref]]
8007 === noch:{LTTng-modules}
8008
8009
8010 [role="since-2.9"]
8011 [[lttng-tracepoint-enum]]
8012 ==== `LTTNG_TRACEPOINT_ENUM()` usage
8013
8014 Use the `LTTNG_TRACEPOINT_ENUM()` macro to define an enumeration:
8015
8016 [source,c]
8017 ----
8018 LTTNG_TRACEPOINT_ENUM(name, TP_ENUM_VALUES(entries))
8019 ----
8020
8021 Replace:
8022
8023 * `name` with the name of the enumeration (C identifier, unique
8024   amongst all the defined enumerations).
8025 * `entries` with a list of enumeration entries.
8026
8027 The available enumeration entry macros are:
8028
8029 +ctf_enum_value(__name__, __value__)+::
8030   Entry named +__name__+ mapped to the integral value +__value__+.
8031
8032 +ctf_enum_range(__name__, __begin__, __end__)+::
8033   Entry named +__name__+ mapped to the range of integral values between
8034   +__begin__+ (included) and +__end__+ (included).
8035
8036 +ctf_enum_auto(__name__)+::
8037   Entry named +__name__+ mapped to the integral value following the
8038   last mapping value.
8039 +
8040 The last value of a `ctf_enum_value()` entry is its +__value__+
8041 parameter.
8042 +
8043 The last value of a `ctf_enum_range()` entry is its +__end__+ parameter.
8044 +
8045 If `ctf_enum_auto()` is the first entry in the list, its integral
8046 value is 0.
8047
8048 Use the `ctf_enum()` <<lttng-modules-tp-fields,field definition macro>>
8049 to use a defined enumeration as a tracepoint field.
8050
8051 .Define an enumeration with `LTTNG_TRACEPOINT_ENUM()`.
8052 ====
8053 [source,c]
8054 ----
8055 LTTNG_TRACEPOINT_ENUM(
8056     my_enum,
8057     TP_ENUM_VALUES(
8058         ctf_enum_auto("AUTO: EXPECT 0")
8059         ctf_enum_value("VALUE: 23", 23)
8060         ctf_enum_value("VALUE: 27", 27)
8061         ctf_enum_auto("AUTO: EXPECT 28")
8062         ctf_enum_range("RANGE: 101 TO 303", 101, 303)
8063         ctf_enum_auto("AUTO: EXPECT 304")
8064     )
8065 )
8066 ----
8067 ====
8068
8069
8070 [role="since-2.7"]
8071 [[lttng-modules-tp-fields]]
8072 ==== Tracepoint fields macros (for `TP_FIELDS()`)
8073
8074 [[tp-fast-assign]][[tp-struct-entry]]The available macros to define
8075 tracepoint fields, which must be listed within `TP_FIELDS()` in
8076 `LTTNG_TRACEPOINT_EVENT()`, are:
8077
8078 [role="func-desc growable",cols="asciidoc,asciidoc"]
8079 .Available macros to define LTTng-modules tracepoint fields
8080 |====
8081 |Macro |Description and parameters
8082
8083 |
8084 +ctf_integer(__t__, __n__, __e__)+
8085
8086 +ctf_integer_nowrite(__t__, __n__, __e__)+
8087
8088 +ctf_user_integer(__t__, __n__, __e__)+
8089
8090 +ctf_user_integer_nowrite(__t__, __n__, __e__)+
8091 |
8092 Standard integer, displayed in base{nbsp}10.
8093
8094 +__t__+::
8095   Integer C type (`int`, `long`, `size_t`, ...).
8096
8097 +__n__+::
8098   Field name.
8099
8100 +__e__+::
8101   Argument expression.
8102
8103 |
8104 +ctf_integer_hex(__t__, __n__, __e__)+
8105
8106 +ctf_user_integer_hex(__t__, __n__, __e__)+
8107 |
8108 Standard integer, displayed in base{nbsp}16.
8109
8110 +__t__+::
8111   Integer C type.
8112
8113 +__n__+::
8114   Field name.
8115
8116 +__e__+::
8117   Argument expression.
8118
8119 |+ctf_integer_oct(__t__, __n__, __e__)+
8120 |
8121 Standard integer, displayed in base{nbsp}8.
8122
8123 +__t__+::
8124   Integer C type.
8125
8126 +__n__+::
8127   Field name.
8128
8129 +__e__+::
8130   Argument expression.
8131
8132 |
8133 +ctf_integer_network(__t__, __n__, __e__)+
8134
8135 +ctf_user_integer_network(__t__, __n__, __e__)+
8136 |
8137 Integer in network byte order (big-endian), displayed in base{nbsp}10.
8138
8139 +__t__+::
8140   Integer C type.
8141
8142 +__n__+::
8143   Field name.
8144
8145 +__e__+::
8146   Argument expression.
8147
8148 |
8149 +ctf_integer_network_hex(__t__, __n__, __e__)+
8150
8151 +ctf_user_integer_network_hex(__t__, __n__, __e__)+
8152 |
8153 Integer in network byte order, displayed in base{nbsp}16.
8154
8155 +__t__+::
8156   Integer C type.
8157
8158 +__n__+::
8159   Field name.
8160
8161 +__e__+::
8162   Argument expression.
8163
8164 |
8165 +ctf_enum(__N__, __t__, __n__, __e__)+
8166
8167 +ctf_enum_nowrite(__N__, __t__, __n__, __e__)+
8168
8169 +ctf_user_enum(__N__, __t__, __n__, __e__)+
8170
8171 +ctf_user_enum_nowrite(__N__, __t__, __n__, __e__)+
8172 |
8173 Enumeration.
8174
8175 +__N__+::
8176   Name of a <<lttng-tracepoint-enum,previously defined enumeration>>.
8177
8178 +__t__+::
8179   Integer C type (`int`, `long`, `size_t`, ...).
8180
8181 +__n__+::
8182   Field name.
8183
8184 +__e__+::
8185   Argument expression.
8186
8187 |
8188 +ctf_string(__n__, __e__)+
8189
8190 +ctf_string_nowrite(__n__, __e__)+
8191
8192 +ctf_user_string(__n__, __e__)+
8193
8194 +ctf_user_string_nowrite(__n__, __e__)+
8195 |
8196 Null-terminated string; undefined behavior if +__e__+ is `NULL`.
8197
8198 +__n__+::
8199   Field name.
8200
8201 +__e__+::
8202   Argument expression.
8203
8204 |
8205 +ctf_array(__t__, __n__, __e__, __s__)+
8206
8207 +ctf_array_nowrite(__t__, __n__, __e__, __s__)+
8208
8209 +ctf_user_array(__t__, __n__, __e__, __s__)+
8210
8211 +ctf_user_array_nowrite(__t__, __n__, __e__, __s__)+
8212 |
8213 Statically-sized array of integers.
8214
8215 +__t__+::
8216   Array element C type.
8217
8218 +__n__+::
8219   Field name.
8220
8221 +__e__+::
8222   Argument expression.
8223
8224 +__s__+::
8225   Number of elements.
8226
8227 |
8228 +ctf_array_bitfield(__t__, __n__, __e__, __s__)+
8229
8230 +ctf_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
8231
8232 +ctf_user_array_bitfield(__t__, __n__, __e__, __s__)+
8233
8234 +ctf_user_array_bitfield_nowrite(__t__, __n__, __e__, __s__)+
8235 |
8236 Statically-sized array of bits.
8237
8238 The type of +__e__+ must be an integer type. +__s__+ is the number
8239 of elements of such type in +__e__+, not the number of bits.
8240
8241 +__t__+::
8242   Array element C type.
8243
8244 +__n__+::
8245   Field name.
8246
8247 +__e__+::
8248   Argument expression.
8249
8250 +__s__+::
8251   Number of elements.
8252
8253 |
8254 +ctf_array_text(__t__, __n__, __e__, __s__)+
8255
8256 +ctf_array_text_nowrite(__t__, __n__, __e__, __s__)+
8257
8258 +ctf_user_array_text(__t__, __n__, __e__, __s__)+
8259
8260 +ctf_user_array_text_nowrite(__t__, __n__, __e__, __s__)+
8261 |
8262 Statically-sized array, printed as text.
8263
8264 The string doesn't need to be null-terminated.
8265
8266 +__t__+::
8267   Array element C type (always `char`).
8268
8269 +__n__+::
8270   Field name.
8271
8272 +__e__+::
8273   Argument expression.
8274
8275 +__s__+::
8276   Number of elements.
8277
8278 |
8279 +ctf_sequence(__t__, __n__, __e__, __T__, __E__)+
8280
8281 +ctf_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
8282
8283 +ctf_user_sequence(__t__, __n__, __e__, __T__, __E__)+
8284
8285 +ctf_user_sequence_nowrite(__t__, __n__, __e__, __T__, __E__)+
8286 |
8287 Dynamically-sized array of integers.
8288
8289 The type of +__E__+ must be unsigned.
8290
8291 +__t__+::
8292   Array element C type.
8293
8294 +__n__+::
8295   Field name.
8296
8297 +__e__+::
8298   Argument expression.
8299
8300 +__T__+::
8301   Length expression C type.
8302
8303 +__E__+::
8304   Length expression.
8305
8306 |
8307 +ctf_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
8308
8309 +ctf_user_sequence_hex(__t__, __n__, __e__, __T__, __E__)+
8310 |
8311 Dynamically-sized array of integers, displayed in base{nbsp}16.
8312
8313 The type of +__E__+ must be unsigned.
8314
8315 +__t__+::
8316   Array element C type.
8317
8318 +__n__+::
8319   Field name.
8320
8321 +__e__+::
8322   Argument expression.
8323
8324 +__T__+::
8325   Length expression C type.
8326
8327 +__E__+::
8328   Length expression.
8329
8330 |+ctf_sequence_network(__t__, __n__, __e__, __T__, __E__)+
8331 |
8332 Dynamically-sized array of integers in network byte order (big-endian),
8333 displayed in base{nbsp}10.
8334
8335 The type of +__E__+ must be unsigned.
8336
8337 +__t__+::
8338   Array element C type.
8339
8340 +__n__+::
8341   Field name.
8342
8343 +__e__+::
8344   Argument expression.
8345
8346 +__T__+::
8347   Length expression C type.
8348
8349 +__E__+::
8350   Length expression.
8351
8352 |
8353 +ctf_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
8354
8355 +ctf_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
8356
8357 +ctf_user_sequence_bitfield(__t__, __n__, __e__, __T__, __E__)+
8358
8359 +ctf_user_sequence_bitfield_nowrite(__t__, __n__, __e__, __T__, __E__)+
8360 |
8361 Dynamically-sized array of bits.
8362
8363 The type of +__e__+ must be an integer type. +__s__+ is the number
8364 of elements of such type in +__e__+, not the number of bits.
8365
8366 The type of +__E__+ must be unsigned.
8367
8368 +__t__+::
8369   Array element C type.
8370
8371 +__n__+::
8372   Field name.
8373
8374 +__e__+::
8375   Argument expression.
8376
8377 +__T__+::
8378   Length expression C type.
8379
8380 +__E__+::
8381   Length expression.
8382
8383 |
8384 +ctf_sequence_text(__t__, __n__, __e__, __T__, __E__)+
8385
8386 +ctf_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
8387
8388 +ctf_user_sequence_text(__t__, __n__, __e__, __T__, __E__)+
8389
8390 +ctf_user_sequence_text_nowrite(__t__, __n__, __e__, __T__, __E__)+
8391 |
8392 Dynamically-sized array, displayed as text.
8393
8394 The string doesn't need to be null-terminated.
8395
8396 The type of +__E__+ must be unsigned.
8397
8398 The behaviour is undefined if +__e__+ is `NULL`.
8399
8400 +__t__+::
8401   Sequence element C type (always `char`).
8402
8403 +__n__+::
8404   Field name.
8405
8406 +__e__+::
8407   Argument expression.
8408
8409 +__T__+::
8410   Length expression C type.
8411
8412 +__E__+::
8413   Length expression.
8414 |====
8415
8416 Use the `_user` versions when the argument expression, `e`, is
8417 a user space address. In the cases of `ctf_user_integer*()` and
8418 `ctf_user_float*()`, `&e` must be a user space address, thus `e` must
8419 be addressable.
8420
8421 The `_nowrite` versions omit themselves from the trace data, but are
8422 otherwise identical. This means LTTng won't write the `_nowrite` fields
8423 to the recorded trace. Their primary purpose is to make some of the
8424 event context available to the <<enabling-disabling-events,recording
8425 event rule filters>> without having to commit the data to
8426 <<channel,sub-buffers>>.
8427
8428
8429 [[glossary]]
8430 == Glossary
8431
8432 Terms related to LTTng and to tracing in general:
8433
8434 [[def-action]]action::
8435   The part of a <<def-trigger,trigger>> which LTTng executes when the
8436   trigger <<def-condition,condition>> is satisfied.
8437
8438 Babeltrace::
8439   The https://diamon.org/babeltrace[Babeltrace] project, which includes:
8440 +
8441 * The
8442   https://babeltrace.org/docs/v2.0/man1/babeltrace2.1/[cmd:babeltrace2]
8443   command-line interface.
8444 * The libbabeltrace2 library which offers a
8445   https://babeltrace.org/docs/v2.0/libbabeltrace2/[C API].
8446 * https://babeltrace.org/docs/v2.0/python/bt2/[Python{nbsp}3 bindings].
8447 * Plugins.
8448
8449 [[def-buffering-scheme]]<<channel-buffering-schemes,buffering scheme>>::
8450   A layout of <<def-sub-buffer,sub-buffers>> applied to a given channel.
8451
8452 [[def-channel]]<<channel,channel>>::
8453   An entity which is responsible for a set of
8454   <<def-ring-buffer,ring buffers>>.
8455 +
8456 <<def-recording-event-rule,Recording event rules>> are always attached
8457 to a specific channel.
8458
8459 clock::
8460   A source of time for a <<def-tracer,tracer>>.
8461
8462 [[def-condition]]condition::
8463   The part of a <<def-trigger,trigger>> which must be satisfied for
8464   LTTng to attempt to execute the trigger <<def-action,actions>>.
8465
8466 [[def-consumer-daemon]]<<lttng-consumerd,consumer daemon>>::
8467   A program which is responsible for consuming the full
8468   <<def-sub-buffer,sub-buffers>> and write them to a file system or
8469   send them over the network.
8470
8471 [[def-current-trace-chunk]]current trace chunk::
8472   A <<def-trace-chunk,trace chunk>> which includes the current content
8473   of all the <<def-sub-buffer,sub-buffers>> of the
8474   <<def-tracing-session,recording session>> and the stream files
8475   produced since the latest event amongst:
8476 +
8477 * The creation of the recording session.
8478 * The last <<def-tracing-session-rotation,recording session rotation>>, if
8479   any.
8480
8481 <<channel-overwrite-mode-vs-discard-mode,discard mode>>::
8482   The <<def-event-record-loss-mode,event record loss mode>> in which
8483   the <<def-tracer,tracer>> _discards_ new <<def-event-record,event
8484   records>> when there's no <<def-sub-buffer,sub-buffer>> space left to
8485   store them.
8486
8487 [[def-event]]event::
8488   The execution of an <<def-instrumentation-point,instrumentation
8489   point>>, like a <<def-tracepoint,tracepoint>> that you manually place
8490   in some source code, or a Linux kprobe.
8491 +
8492 When an instrumentation point is executed, LTTng creates an event.
8493 +
8494 When an <<def-event-rule,event rule>> matches the event,
8495 <<def-lttng,LTTng>> executes some action, for example:
8496 +
8497 * Record its payload to a <<def-sub-buffer,sub-buffer>> as an
8498   <<def-event-record,event record>>.
8499 * Attempt to execute the user-defined actions of a
8500   <<def-trigger,trigger>> with an
8501   <<add-event-rule-matches-trigger,``event rule matches''>> condition.
8502
8503 [[def-event-name]]event name::
8504   The name of an <<def-event,event>>, which is also the name of the
8505   <<def-event-record,event record>>.
8506 +
8507 This is also called the _instrumentation point name_.
8508
8509 [[def-event-record]]event record::
8510   A record (binary serialization), in a <<def-trace,trace>>, of the
8511   payload of an <<def-event,event>>.
8512 +
8513 The payload of an event record has zero or more _fields_.
8514
8515 [[def-event-record-loss-mode]]<<channel-overwrite-mode-vs-discard-mode,event record loss mode>>::
8516   The mechanism by which event records of a given
8517   <<def-channel,channel>> are lost (not recorded) when there's no
8518   <<def-sub-buffer,sub-buffer>> space left to store them.
8519
8520 [[def-event-rule]]<<event-rule,event rule>>::
8521   Set of conditions which an <<def-event,event>> must satisfy
8522   for LTTng to execute some action.
8523 +
8524 An event rule is said to _match_ events, like a
8525 https://en.wikipedia.org/wiki/Regular_expression[regular expression]
8526 matches strings.
8527 +
8528 A <<def-recording-event-rule,recording event rule>> is a specific type
8529 of event rule of which the action is to <<def-record,record>> the event
8530 to a <<def-sub-buffer,sub-buffer>>.
8531
8532 [[def-incl-set]]inclusion set::
8533   In the <<pid-tracking,process attribute inclusion set>> context: a
8534   set of <<def-proc-attr,process attributes>> of a given type.
8535
8536 <<instrumenting,instrumentation>>::
8537   The use of <<def-lttng,LTTng>> probes to make a kernel or
8538   <<def-user-application,user application>> traceable.
8539
8540 [[def-instrumentation-point]]instrumentation point::
8541   A point in the execution path of a kernel or
8542   <<def-user-application,user application>> which, when executed,
8543   create an <<def-event,event>>.
8544
8545 instrumentation point name::
8546   See _<<def-event-name,event name>>_.
8547
8548 `java.util.logging`::
8549   The
8550   https://docs.oracle.com/javase/7/docs/api/java/util/logging/package-summary.html[core logging facilities]
8551   of the Java platform.
8552
8553 log4j::
8554   A https://logging.apache.org/log4j/1.2/[logging library] for Java
8555   developed by the Apache Software Foundation.
8556
8557 log level::
8558   Level of severity of a log statement or user space
8559   <<def-instrumentation-point,instrumentation point>>.
8560
8561 [[def-lttng]]LTTng::
8562   The _Linux Trace Toolkit: next generation_ project.
8563
8564 <<lttng-cli,cmd:lttng>>::
8565   A command-line tool provided by the <<def-lttng-tools,LTTng-tools>>
8566   project which you can use to send and receive control messages to and
8567   from a <<def-session-daemon,session daemon>>.
8568
8569 LTTng analyses::
8570   The https://github.com/lttng/lttng-analyses[LTTng analyses] project,
8571   which is a set of analyzing programs that you can use to obtain a
8572   higher level view of an <<def-lttng,LTTng>> <<def-trace,trace>>.
8573
8574 cmd:lttng-consumerd::
8575   The name of the <<def-consumer-daemon,consumer daemon>> program.
8576
8577 cmd:lttng-crash::
8578   A utility provided by the <<def-lttng-tools,LTTng-tools>> project
8579   which can convert <<def-ring-buffer,ring buffer>> files (usually
8580   <<persistent-memory-file-systems,saved on a persistent memory file
8581   system>>) to <<def-trace,trace>> files.
8582 +
8583 See man:lttng-crash(1).
8584
8585 LTTng Documentation::
8586   This document.
8587
8588 <<lttng-live,LTTng live>>::
8589   A communication protocol between the <<lttng-relayd,relay daemon>> and
8590   live readers which makes it possible to show or analyze
8591   <<def-event-record,event records>> ``live'', as they're received by
8592   the <<def-relay-daemon,relay daemon>>.
8593
8594 <<lttng-modules,LTTng-modules>>::
8595   The https://github.com/lttng/lttng-modules[LTTng-modules] project,
8596   which contains the Linux kernel modules to make the Linux kernel
8597   <<def-instrumentation-point,instrumentation points>> available for
8598   <<def-lttng,LTTng>> tracing.
8599
8600 cmd:lttng-relayd::
8601   The name of the <<def-relay-daemon,relay daemon>> program.
8602
8603 cmd:lttng-sessiond::
8604   The name of the <<def-session-daemon,session daemon>> program.
8605
8606 [[def-lttng-tools]]LTTng-tools::
8607   The https://github.com/lttng/lttng-tools[LTTng-tools] project, which
8608   contains the various programs and libraries used to
8609   <<controlling-tracing,control tracing>>.
8610
8611 [[def-lttng-ust]]<<lttng-ust,LTTng-UST>>::
8612   The https://github.com/lttng/lttng-ust[LTTng-UST] project, which
8613   contains libraries to instrument
8614   <<def-user-application,user applications>>.
8615
8616 <<lttng-ust-agents,LTTng-UST Java agent>>::
8617   A Java package provided by the <<def-lttng-ust,LTTng-UST>> project to
8618   allow the LTTng instrumentation of `java.util.logging` and Apache
8619   log4j{nbsp}1.2 logging statements.
8620
8621 <<lttng-ust-agents,LTTng-UST Python agent>>::
8622   A Python package provided by the <<def-lttng-ust,LTTng-UST>> project
8623   to allow the <<def-lttng,LTTng>> instrumentation of Python logging
8624   statements.
8625
8626 <<channel-overwrite-mode-vs-discard-mode,overwrite mode>>::
8627   The <<def-event-record-loss-mode,event record loss mode>> in which new
8628   <<def-event-record,event records>> _overwrite_ older event records
8629   when there's no <<def-sub-buffer,sub-buffer>> space left to store
8630   them.
8631
8632 <<channel-buffering-schemes,per-process buffering>>::
8633   A <<def-buffering-scheme,buffering scheme>> in which each instrumented
8634   process has its own <<def-sub-buffer,sub-buffers>> for a given user
8635   space <<def-channel,channel>>.
8636
8637 <<channel-buffering-schemes,per-user buffering>>::
8638   A <<def-buffering-scheme,buffering scheme>> in which all the processes
8639   of a Unix user share the same <<def-sub-buffer,sub-buffers>> for a
8640   given user space <<def-channel,channel>>.
8641
8642 [[def-proc-attr]]process attribute::
8643   In the <<pid-tracking,process attribute inclusion set>> context:
8644 +
8645 * A process ID.
8646 * A virtual process ID.
8647 * A Unix user ID.
8648 * A virtual Unix user ID.
8649 * A Unix group ID.
8650 * A virtual Unix group ID.
8651
8652 record (_noun_)::
8653   See <<def-event-record,_event record_>>.
8654
8655 [[def-record]]record (_verb_)::
8656   Serialize the binary payload of an <<def-event,event>> to a
8657   <<def-sub-buffer,sub-buffer>>.
8658
8659 [[def-recording-event-rule]]<<event,recording event rule>>::
8660   Specific type of <<def-event-rule,event rule>> of which the action is
8661   to <<def-record,record>> the matched event to a
8662   <<def-sub-buffer,sub-buffer>>.
8663
8664 [[def-tracing-session]][[def-recording-session]]<<tracing-session,recording session>>::
8665   A stateful dialogue between you and a <<lttng-sessiond,session daemon>>.
8666
8667 [[def-tracing-session-rotation]]<<session-rotation,recording session rotation>>::
8668   The action of archiving the
8669   <<def-current-trace-chunk,current trace chunk>> of a
8670   <<def-tracing-session,recording session>>.
8671
8672 [[def-relay-daemon]]<<lttng-relayd,relay daemon>>::
8673   A process which is responsible for receiving the <<def-trace,trace>>
8674   data which a distant <<def-consumer-daemon,consumer daemon>> sends.
8675
8676 [[def-ring-buffer]]ring buffer::
8677   A set of <<def-sub-buffer,sub-buffers>>.
8678
8679 rotation::
8680   See _<<def-tracing-session-rotation,recording session rotation>>_.
8681
8682 [[def-session-daemon]]<<lttng-sessiond,session daemon>>::
8683   A process which receives control commands from you and orchestrates
8684   the <<def-tracer,tracers>> and various <<def-lttng,LTTng>> daemons.
8685
8686 <<taking-a-snapshot,snapshot>>::
8687   A copy of the current data of all the <<def-sub-buffer,sub-buffers>>
8688   of a given <<def-tracing-session,recording session>>, saved as
8689   <<def-trace,trace>> files.
8690
8691 [[def-sub-buffer]]sub-buffer::
8692   One part of an <<def-lttng,LTTng>> <<def-ring-buffer,ring buffer>>
8693   which contains <<def-event-record,event records>>.
8694
8695 timestamp::
8696   The time information attached to an <<def-event,event>> when LTTng
8697   creates it.
8698
8699 [[def-trace]]trace (_noun_)::
8700   A set of:
8701 +
8702 * One https://diamon.org/ctf/[CTF] metadata stream file.
8703 * One or more CTF data stream files which are the concatenations of one
8704   or more flushed <<def-sub-buffer,sub-buffers>>.
8705
8706 [[def-trace-verb]]trace (_verb_)::
8707   From the perspective of a <<def-tracer,tracer>>: attempt to execute
8708   one or more actions when emitting an <<def-event,event>> in an
8709   application or in a system.
8710
8711 [[def-trace-chunk]]trace chunk::
8712   A self-contained <<def-trace,trace>> which is part of a
8713   <<def-tracing-session,recording session>>. Each
8714   <<def-tracing-session-rotation, recording session rotation>> produces a
8715   <<def-trace-chunk-archive,trace chunk archive>>.
8716
8717 [[def-trace-chunk-archive]]trace chunk archive::
8718   The result of a <<def-tracing-session-rotation, recording session
8719   rotation>>.
8720 +
8721 <<def-lttng,LTTng>> doesn't manage any trace chunk archive, even if its
8722 containing <<def-tracing-session,recording session>> is still active: you
8723 are free to read it, modify it, move it, or remove it.
8724
8725 Trace Compass::
8726   The http://tracecompass.org[Trace Compass] project and application.
8727
8728 [[def-tracepoint]]tracepoint::
8729   An instrumentation point using the tracepoint mechanism of the Linux
8730   kernel or of <<def-lttng-ust,LTTng-UST>>.
8731
8732 tracepoint definition::
8733   The definition of a single <<def-tracepoint,tracepoint>>.
8734
8735 tracepoint name::
8736   The name of a <<def-tracepoint,tracepoint>>.
8737
8738 [[def-tracepoint-provider]]tracepoint provider::
8739   A set of functions providing <<def-tracepoint,tracepoints>> to an
8740   instrumented <<def-user-application,user application>>.
8741 +
8742 Not to be confused with a <<def-tracepoint-provider-package,tracepoint
8743 provider package>>: many tracepoint providers can exist within a
8744 tracepoint provider package.
8745
8746 [[def-tracepoint-provider-package]]tracepoint provider package::
8747   One or more <<def-tracepoint-provider,tracepoint providers>> compiled
8748   as an https://en.wikipedia.org/wiki/Object_file[object file] or as a
8749   link:https://en.wikipedia.org/wiki/Library_(computing)#Shared_libraries[shared
8750   library].
8751
8752 [[def-tracer]]tracer::
8753   A piece of software which executes some action when it emits
8754   an <<def-event,event>>, like <<def-record,record>> it to some
8755   buffer.
8756
8757 <<domain,tracing domain>>::
8758   A type of LTTng <<def-tracer,tracer>>.
8759
8760 <<tracing-group,tracing group>>::
8761   The Unix group which a Unix user can be part of to be allowed to
8762   control the Linux kernel LTTng <<def-tracer,tracer>>.
8763
8764 [[def-trigger]]<<trigger,trigger>>::
8765   A <<def-condition,condition>>-<<def-action,actions>> pair; when the
8766   condition of a trigger is satisfied, LTTng attempts to execute its
8767   actions.
8768
8769 [[def-user-application]]user application::
8770   An application (program or library) running in user space, as opposed
8771   to a Linux kernel module, for example.