Docs/2.13

The LTTng Documentation

Last update: 28 November 2023

Welcome!

Welcome to the LTTng Documentation!

The Linux Trace Toolkit: next generation is an open source software toolkit which you can use to trace the Linux kernel, user applications, and user libraries at the same time.

LTTng consists of:

  • Kernel modules to trace the Linux kernel.

  • Shared libraries to trace C/C++ user applications.

  • Java packages to trace Java applications which use java.util.logging or Apache log4j 1.2.

  • A Python package to trace Python applications which use the standard logging package.

  • A kernel module to trace shell scripts and other user applications without a dedicated instrumentation mechanism.

  • Daemons and a command-line tool, lttng, to control the LTTng tracers.

Open source documentation

Note:This is an open documentation: its source is available in a public Git repository.

Should you find any error in the content of this text, any grammatical mistake, or any dead link, we would be very grateful if you would file a GitHub issue for it or, even better, contribute a patch to this documentation by creating a pull request.

Target audience

The material of this documentation is appropriate for intermediate to advanced software developers working in a Linux environment and interested in efficient software tracing. LTTng is also worth a try for students interested in the inner mechanics of their systems.

If you do not have a programming background, you may wish to skip everything related to instrumentation, which often requires at least some programming language skills.

What’s in this documentation?

The LTTng Documentation is divided into the following sections:

  • Nuts and bolts” explains the rudiments of software tracing and the rationale behind the LTTng project.

    Skip this section if you’re familiar with software tracing and with the LTTng project.

  • Installation” describes the steps to install the LTTng packages on common Linux distributions and from their sources.

    Skip this section if you already properly installed LTTng on your target system.

  • Quick start” is a concise guide to get started quickly with LTTng kernel and user space tracing.

    We recommend this section if you’re new to LTTng or to software tracing in general.

    Skip this section if you’re not new to LTTng.

  • Core concepts” explains the concepts at the heart of LTTng.

    It’s a good idea to become familiar with the core concepts before attempting to use the toolkit.

  • Components of LTTng” describes the various components of the LTTng machinery, like the daemons, the libraries, and the command-line interface.

  • Instrumentation” shows different ways to instrument user applications and the Linux kernel for LTTng tracing.

    Instrumenting source code is essential to provide a meaningful source of events.

    Skip this section if you don’t have a programming background.

  • Tracing control” is divided into topics which demonstrate how to use the vast array of features that LTTng 2.13 offers.

  • Reference” contains API reference tables.

  • Glossary” is a specialized dictionary of terms related to LTTng or to the field of software tracing.

Convention

Function names, parameter names, variable names, command names, argument names, file system paths, file names, and other literal strings are written using a monospace typeface in this document. An italic word within such a block is a placeholder, usually described in the following sentence.

Practical tips and notes are given throughout the document using the following style:

Tip:Read the tips.

Terminal boxes are used to show command lines:

$
#
echo Command line as a regular user
echo Command line as a the `root` user

Command lines which you need to execute as a priviledged user start with the # prompt or with sudo. Other command lines start with the $ prompt.

Acknowledgements

A few people made the online LTTng Documentation possible.

Philippe Proulx wrote most of the content, created the diagrams, and formatted the document. He’s the current maintainer of the LTTng Documentation.

Daniel U. Thibault, from the DRDC, wrote “LTTng: The Linux Trace Toolkit Next Generation — A Comprehensive User’s Guide (version 2.3 edition)” which was used to complete parts of the “Core concepts” and “Components of LTTng” sections and for a few passages here and there.

The entire EfficiOS team made essential reviews of the whole document.

We sincerely thank everyone who helped enhance the quality of this documentation.

What’s new in LTTng 2.13?

LTTng 2.13 bears the name Nordicité, the product of a collaboration between Champ Libre and Boréale. This farmhouse IPA is brewed with Kveik yeast and Québec-grown barley, oats, and juniper branches. The result is a remarkable, fruity, hazy golden IPA that offers a balanced touch of resinous and woodsy bitterness.

New features and changes in LTTng 2.13:

General
User space tracing

Important:The major version part of the liblttng-ust soname is bumped, which means you must recompile your instrumented applications/libraries and tracepoint provider packages to use LTTng-UST 2.13.

This change became a necessity to clean up the library and for liblttng-ust to stop exporting private symbols.

Also, LTTng 2.13 prepends the lttng_ust_ and LTTNG_UST_ prefix to all public macro/definition/function names to offer a consistent API namespace. The LTTng 2.12 API is still available; see the “Compatibility with previous APIs” section of lttng-ust(3).

Other notable changes:

Kernel tracing
  • The preferred display base of event record integer fields which contain memory addresses is now hexadecimal instead of decimal.

  • The pid field is removed from lttng_statedump_file_descriptor event records and the file_table_address field is added.

    This new field is the address of the files_struct structure which contains the file descriptor.

    See the “statedump: introduce file_table_address” patch to learn more.

  • The flags field of syscall_entry_clone event records is now a structure containing two enumerations (exit signal and options).

    This change makes the flag values more readable and meaningful.

    See the “syscalls: Make clone()'s flags field a 2 enum struct” patch to learn more.

  • The memory footprint of the kernel tracer is improved: the latter only generates metadata for the specific system call recording event rules that you create.

Nuts and bolts

What is LTTng? As its name suggests, the Linux Trace Toolkit: next generation is a modern toolkit for tracing Linux systems and applications. So your first question might be: what is tracing?

What is tracing?

As the history of software engineering progressed and led to what we now take for granted—complex, numerous and interdependent software applications running in parallel on sophisticated operating systems like Linux—the authors of such components, software developers, began feeling a natural urge to have tools that would ensure the robustness and good performance of their masterpieces.

One major achievement in this field is, inarguably, the GNU debugger (GDB), an essential tool for developers to find and fix bugs. But even the best debugger won’t help make your software run faster, and nowadays, faster software means either more work done by the same hardware, or cheaper hardware for the same work.

A profiler is often the tool of choice to identify performance bottlenecks. Profiling is suitable to identify where performance is lost in a given piece of software. The profiler outputs a profile, a statistical summary of observed events, which you may use to discover which functions took the most time to execute. However, a profiler won’t report why some identified functions are the bottleneck. Bottlenecks might only occur when specific conditions are met, conditions that are sometimes impossible to capture by a statistical profiler, or impossible to reproduce with an application altered by the overhead of an event-based profiler. For a thorough investigation of software performance issues, a history of execution is essential, with the recorded values of variables and context fields you choose, and with as little influence as possible on the instrumented application. This is where tracing comes in handy.

Tracing is a technique used to understand what goes on in a running software system. The piece of software used for tracing is called a tracer, which is conceptually similar to a tape recorder. When recording, specific instrumentation points placed in the software source code generate events that are saved on a giant tape: a trace file. You can record user application and operating system events at the same time, opening the possibility of resolving a wide range of problems that would otherwise be extremely challenging.

Tracing is often compared to logging. However, tracers and loggers are two different tools, serving two different purposes. Tracers are designed to record much lower-level events that occur much more frequently than log messages, often in the range of thousands per second, with very little execution overhead. Logging is more appropriate for a very high-level analysis of less frequent events: user accesses, exceptional conditions (errors and warnings, for example), database transactions, instant messaging communications, and such. Simply put, logging is one of the many use cases that can be satisfied with tracing.

The list of recorded events inside a trace file can be read manually like a log file for the maximum level of detail, but it’s generally much more interesting to perform application-specific analyses to produce reduced statistics and graphs that are useful to resolve a given problem. Trace viewers and analyzers are specialized tools designed to do this.

In the end, this is what LTTng is: a powerful, open source set of tools to trace the Linux kernel and user applications at the same time. LTTng is composed of several components actively maintained and developed by its community.

Alternatives to LTTng

Excluding proprietary solutions, a few competing software tracers exist for Linux:

dtrace4linux

A port of Sun Microsystems' DTrace to Linux.

The dtrace tool interprets user scripts and is responsible for loading code into the Linux kernel for further execution and collecting the outputted data.

eBPF

A subsystem in the Linux kernel in which a virtual machine can execute programs passed from the user space to the kernel.

You can attach such programs to tracepoints and kprobes thanks to a system call, and they can output data to the user space when executed thanks to different mechanisms (pipe, VM register values, and eBPF maps, to name a few).

ftrace

The de facto function tracer of the Linux kernel.

Its user interface is a set of special files in sysfs.

perf

A performance analysis tool for Linux which supports hardware performance counters, tracepoints, as well as other counters and types of probes.

The controlling utility of perf is the perf command line/text UI tool.

strace

A command-line utility which records system calls made by a user process, as well as signal deliveries and changes of process state.

strace makes use of ptrace to fulfill its function.

sysdig

Like SystemTap, uses scripts to analyze Linux kernel events.

You write scripts, or chisels in the jargon of sysdig, in Lua and sysdig executes them while it traces the system or afterwards. The interface of sysdig is the sysdig command-line tool as well as the text UI-based csysdig tool.

SystemTap

A Linux kernel and user space tracer which uses custom user scripts to produce plain text traces.

SystemTap converts the scripts to the C language, and then compiles them as Linux kernel modules which are loaded to produce trace data. The primary user interface of SystemTap is the stap command-line tool.

The main distinctive features of LTTng is that it produces correlated kernel and user space traces, as well as doing so with the lowest overhead amongst other solutions. It produces trace files in the CTF format, a file format optimized for the production and analyses of multi-gigabyte data.

LTTng is the result of more than 10 years of active open source development by a community of passionate developers. LTTng is currently available on major desktop and server Linux distributions.

The main interface for tracing control is a single command-line tool named lttng. The latter can create several recording sessions, enable and disable recording event rules on the fly, filter events efficiently with custom user expressions, start and stop tracing, and much more. LTTng can write the traces on the file system or send them over the network, and keep them totally or partially. You can make LTTng execute user-defined actions when LTTng emits an event. You can view the traces once tracing becomes inactive or as LTTng records events.

Install LTTng now and start tracing!

Installation

LTTng is a set of software components which interact to instrument the Linux kernel and user applications, and to control tracing (start and stop recording, create recording event rules, and the rest). Those components are bundled into the following packages:

LTTng-tools

Libraries and command-line interface to control tracing.

LTTng-modules

Linux kernel modules to instrument and trace the kernel.

LTTng-UST

Libraries and Java/Python packages to instrument and trace user applications.

Most distributions mark the LTTng-modules and LTTng-UST packages as optional when installing LTTng-tools (which is always required). In the following sections, we always provide the steps to install all three, but note that:

  • You only need to install LTTng-modules if you intend to use the Linux kernel LTTng tracer.

  • You only need to install LTTng-UST if you intend to use the user space LTTng tracer.

Note:For RHEL and SLES packages, see EfficiOS Enterprise Packages.

For other distributions, build LTTng from source.

Ubuntu

LTTng 2.13 is available on Ubuntu 22.04 LTS Jammy Jellyfish, Ubuntu 23.04 Lunar Lobster, and Ubuntu 23.10 Mantic Minotaur. For previous supported releases of Ubuntu, use the LTTng Stable 2.13 PPA.

To install LTTng 2.13 on Ubuntu 22.04 LTS Jammy Jellyfish:

  1. Install the main LTTng 2.13 packages:

    #
    #
    #
    apt-get install lttng-tools
    apt-get install lttng-modules-dkms
    apt-get install liblttng-ust-dev
  2. If you need to instrument and trace Java applications, install the LTTng-UST Java agent:

    #
    apt-get install liblttng-ust-agent-java
  3. If you need to instrument and trace Python 3 applications, install the LTTng-UST Python agent:

    #
    apt-get install python3-lttngust

Ubuntu: LTTng Stable 2.13 PPA

The LTTng Stable 2.13 PPA offers the latest stable LTTng 2.13 packages for Ubuntu 18.04 LTS Bionic Beaver, Ubuntu 20.04 LTS Focal Fossa, and Ubuntu 22.04 LTS Jammy Jellyfish.

To install LTTng 2.13 from the LTTng Stable 2.13 PPA:

  1. Add the LTTng Stable 2.13 PPA repository and update the list of packages:

    #
    #
    apt-add-repository ppa:lttng/stable-2.13
    apt-get update
  2. Install the main LTTng 2.13 packages:

    #
    #
    #
    apt-get install lttng-tools
    apt-get install lttng-modules-dkms
    apt-get install liblttng-ust-dev
  3. If you need to instrument and trace Java applications, install the LTTng-UST Java agent:

    #
    apt-get install liblttng-ust-agent-java
  4. If you need to instrument and trace Python 3 applications, install the LTTng-UST Python agent:

    #
    apt-get install python3-lttngust

Debian

To install LTTng 2.13 on Debian 12 bookworm:

  1. Install the main LTTng 2.13 packages:

    #
    #
    #
    apt install lttng-modules-dkms
    apt install liblttng-ust-dev
    apt install lttng-tools
  2. If you need to instrument and trace Java applications, install the LTTng-UST Java agent:

    #
    apt install liblttng-ust-agent-java
  3. If you need to instrument and trace Python applications, install the LTTng-UST Python agent:

    #
    apt install python3-lttngust

Fedora

To install LTTng 2.13 on Fedora 37, Fedora 38, or Fedora 39:

  1. Install the LTTng-tools 2.13 and LTTng-UST 2.13 packages:

    #
    #
    yum install lttng-tools
    yum install lttng-ust
  2. Download, build, and install the latest LTTng-modules 2.13:

    $
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
    tar -xf lttng-modules-latest-2.13.tar.bz2 &&
    cd lttng-modules-2.13.* &&
    make &&
    sudo make modules_install &&
    sudo depmod -a

Java and Python application instrumentation and tracing

Important:If you need to instrument and trace Java applications on Fedora, you need to build and install LTTng-UST 2.13 from source and pass the --enable-java-agent-jul, --enable-java-agent-log4j, or --enable-java-agent-all options to the configure script, depending on which Java logging framework you use.

If you need to instrument and trace Python applications on Fedora, you need to build and install LTTng-UST 2.13 from source and pass the --enable-python-agent option to the configure script.

Arch Linux

LTTng-UST 2.13 is available in the extra repository of Arch Linux, while LTTng-tools 2.13 and LTTng-modules 2.13 are available in the AUR.

To install LTTng 2.13 on Arch Linux, using yay for the AUR packages:

  1. Install the main LTTng 2.13 packages:

    #
    $
    $
    pacman -Sy lttng-ust
    yay -Sy lttng-tools
    yay -Sy lttng-modules
  2. If you need to instrument and trace Python applications, install the LTTng-UST Python agent:

    #
    pacman -Sy python-lttngust

Alpine Linux

To install LTTng-tools 2.13 and LTTng-UST 2.13 on Alpine Linux 3.16, Alpine Linux 3.17, or Alpine Linux 3.18:

  1. Add the LTTng packages:

    #
    #
    apk add lttng-tools
    apk add lttng-ust-dev
  2. Download, build, and install the latest LTTng-modules 2.13:

    $
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget http://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
    tar -xf lttng-modules-latest-2.13.tar.bz2 &&
    cd lttng-modules-2.13.* &&
    make &&
    sudo make modules_install &&
    sudo depmod -a

Buildroot

To install LTTng 2.13 on Buildroot 2022.02, Buildroot 2022.05, Buildroot 2022.08, Buildroot 2022.11, Buildroot 2023.02, Buildroot 2023.05, or Buildroot 2023.08:

  1. Launch the Buildroot configuration tool:

    $
    make menuconfig
  2. In Kernel, check Linux kernel.

  3. In Toolchain, check Enable WCHAR support.

  4. In Target packages → Debugging, profiling and benchmark, check lttng-modules and lttng-tools.

  5. In Target packages → Libraries → Other, check lttng-libust.

OpenEmbedded and Yocto

LTTng 2.13 recipes are available in the openembedded-core layer for Yocto Project 3.3 Honister, Yocto Project 4.0 Kirkstone, Yocto Project 4.1 Langdale, Yocto Project 4.2 Mickledore, and Yocto Project 4.3 Nanbield under the following names:

  • lttng-tools

  • lttng-modules

  • lttng-ust

With BitBake, the simplest way to include LTTng recipes in your target image is to add them to IMAGE_INSTALL_append in conf/local.conf:

IMAGE_INSTALL_append = " lttng-tools lttng-modules lttng-ust"

If you use Hob:

  1. Select a machine and an image recipe.

  2. Click Edit image recipe.

  3. Under the All recipes tab, search for lttng.

  4. Check the desired LTTng recipes.

Build from source

To build and install LTTng 2.13 from source:

  1. Using the package manager of your distribution, or from source, install the following dependencies of LTTng-tools and LTTng-UST:

  2. Download, build, and install the latest LTTng-modules 2.13:

    $
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
    tar -xf lttng-modules-latest-2.13.tar.bz2 &&
    cd lttng-modules-2.13.* &&
    make &&
    sudo make modules_install &&
    sudo depmod -a
  3. Download, build, and install the latest LTTng-UST 2.13:

    $
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-ust/lttng-ust-latest-2.13.tar.bz2 &&
    tar -xf lttng-ust-latest-2.13.tar.bz2 &&
    cd lttng-ust-2.13.* &&
    ./configure &&
    make &&
    sudo make install &&
    sudo ldconfig

    Add --disable-numa to ./configure if you don’t have numactl.

    Java and Python application tracing

    Important:If you need to instrument and have LTTng trace Java applications, pass the --enable-java-agent-jul, --enable-java-agent-log4j, or --enable-java-agent-all options to the configure script, depending on which Java logging framework you use.

    If you need to instrument and have LTTng trace Python applications, pass the --enable-python-agent option to the configure script. You can set the PYTHON environment variable to the path to the Python interpreter for which to install the LTTng-UST Python agent package.

    Note:By default, LTTng-UST libraries are installed to /usr/local/lib, which is the de facto directory in which to keep self-compiled and third-party libraries.

    When linking an instrumented user application with liblttng-ust:

    • Append /usr/local/lib to the LD_LIBRARY_PATH environment variable.

    • Pass the -L/usr/local/lib and -Wl,-rpath,/usr/local/lib options to gcc(1), g++(1), or clang(1).

  4. Download, build, and install the latest LTTng-tools 2.13:

    $
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
    tar -xf lttng-tools-latest-2.13.tar.bz2 &&
    cd lttng-tools-2.13.* &&
    ./configure &&
    make &&
    sudo make install &&
    sudo ldconfig

Tip:The vlttng tool can do all the previous steps automatically for a given version of LTTng and confine the installed files to a specific directory. This can be useful to try LTTng without installing it on your system.

Linux kernel module signature

Linux kernel modules require trusted signatures in order to be loaded when any of the following is true:

  • The system boots with Secure Boot enabled.

  • The Linux kernel which boots is configured with CONFIG_MODULE_SIG_FORCE.

  • The Linux kernel boots with a command line containing module.sig_enforce=1.

Example:root user running lttng-sessiond which fails to load a required kernel module due to the signature enforcement policies.

#
 
 
 
 
lttng-sessiond
Warning: No tracing group detected
modprobe: ERROR: could not insert 'lttng_ring_buffer_client_discard': Key was rejected by service
Error: Unable to load required module lttng-ring-buffer-client-discard
Warning: No kernel tracer available

There are several methods to enroll trusted keys for signing modules that are built from source. The precise details vary from one Linux version to another, and distributions may have their own mechanisms. For example, DKMS may autogenerate a key and sign modules, but the key isn’t automatically enrolled.

See Kernel module signing facility and the documentation of your distribution to learn more about signing Linux kernel modules.

Quick start

This is a short guide to get started quickly with LTTng kernel and user space tracing.

Before you follow this guide, make sure to install LTTng.

This tutorial walks you through the steps to:

  1. Record Linux kernel events.

  2. Record the events of a user application written in C.

  3. View and analyze the recorded events.

Record Linux kernel events

Note:The following command lines start with the # prompt because you need root privileges to control the Linux kernel LTTng tracer. You can also control the kernel tracer as a regular user if your Unix user is a member of the tracing group.

  1. Create a recording session to write LTTng traces to /tmp/my-kernel-trace:

    #
    lttng create my-kernel-session --output=/tmp/my-kernel-trace
  2. List the available kernel tracepoints and system calls:

    #
    #
    lttng list --kernel
    lttng list --kernel --syscall
  3. Create recording event rules which match events having the desired names, for example the sched_switch and sched_process_fork tracepoints, and the open(2) and close(2) system calls:

    #
    #
    lttng enable-event --kernel sched_switch,sched_process_fork
    lttng enable-event --kernel --syscall open,close

    Create a recording event rule which matches all the Linux kernel tracepoint events with the --all option (recording with such a recording event rule generates a lot of data):

    #
    lttng enable-event --kernel --all
  4. Start recording:

    #
    lttng start
  5. Do some operation on your system for a few seconds. For example, load a website, or list the files of a directory.

  6. Destroy the current recording session:

    #
    lttng destroy

    The lttng-destroy(1) command doesn’t destroy the trace data; it only destroys the state of the recording session.

    The lttng-destroy(1) command also runs the lttng-stop(1) command implicitly (see “Start and stop a recording session”). You need to stop recording to make LTTng flush the remaining trace data and make the trace readable.

  7. For the sake of this example, make the recorded trace accessible to the non-root users:

    #
    chown -R $(whoami) /tmp/my-kernel-trace

See “View and analyze the recorded events” to view the recorded events.

Record user application events

This section walks you through a simple example to record the events of a Hello world program written in C.

To create the traceable user application:

  1. Create the tracepoint provider header file, which defines the tracepoints and the events they can generate:

    hello-tp.h

    #undef LTTNG_UST_TRACEPOINT_PROVIDER
    #define LTTNG_UST_TRACEPOINT_PROVIDER hello_world
    
    #undef LTTNG_UST_TRACEPOINT_INCLUDE
    #define LTTNG_UST_TRACEPOINT_INCLUDE "./hello-tp.h"
    
    #if !defined(_HELLO_TP_H) || defined(LTTNG_UST_TRACEPOINT_HEADER_MULTI_READ)
    #define _HELLO_TP_H
    
    #include <lttng/tracepoint.h>
    
    LTTNG_UST_TRACEPOINT_EVENT(
        hello_world,
        my_first_tracepoint,
        LTTNG_UST_TP_ARGS(
            int, my_integer_arg,
            char *, my_string_arg
        ),
        LTTNG_UST_TP_FIELDS(
            lttng_ust_field_string(my_string_field, my_string_arg)
            lttng_ust_field_integer(int, my_integer_field, my_integer_arg)
        )
    )
    
    #endif /* _HELLO_TP_H */
    
    #include <lttng/tracepoint-event.h>
    
  2. Create the tracepoint provider package source file:

    hello-tp.c

    #define LTTNG_UST_TRACEPOINT_CREATE_PROBES
    #define LTTNG_UST_TRACEPOINT_DEFINE
    
    #include "hello-tp.h"
    
  3. Build the tracepoint provider package:

    $
    gcc -c -I. hello-tp.c
  4. Create the Hello World application source file:

    hello.c

    #include <stdio.h>
    #include "hello-tp.h"
    
    int main(int argc, char *argv[])
    {
        unsigned int i;
    
        puts("Hello, World!\nPress Enter to continue...");
    
        /*
         * The following getchar() call only exists for the purpose of this
         * demonstration, to pause the application in order for you to have
         * time to list its tracepoints. You don't need it otherwise.
         */
        getchar();
    
        /*
         * An lttng_ust_tracepoint() call.
         *
         * Arguments, as defined in `hello-tp.h`:
         *
         * 1. Tracepoint provider name (required)
         * 2. Tracepoint name (required)
         * 3. `my_integer_arg` (first user-defined argument)
         * 4. `my_string_arg` (second user-defined argument)
         *
         * Notice the tracepoint provider and tracepoint names are
         * C identifiers, NOT strings: they're in fact parts of variables
         * that the macros in `hello-tp.h` create.
         */
        lttng_ust_tracepoint(hello_world, my_first_tracepoint, 23,
                             "hi there!");
    
        for (i = 0; i < argc; i++) {
            lttng_ust_tracepoint(hello_world, my_first_tracepoint,
                                 i, argv[i]);
        }
    
        puts("Quitting now!");
        lttng_ust_tracepoint(hello_world, my_first_tracepoint,
                             i * i, "i^2");
        return 0;
    }
    
  5. Build the application:

    $
    gcc -c hello.c
  6. Link the application with the tracepoint provider package, liblttng-ust and libdl:

    $
    gcc -o hello hello.o hello-tp.o -llttng-ust -ldl

Here’s the whole build process:

Build steps of the user space tracing tutorial.

To record the events of the user application:

  1. Run the application with a few arguments:

    $
    ./hello world and beyond

    You see:

    Hello, World!
    Press Enter to continue...
  2. Start an LTTng session daemon:

    $
    lttng-sessiond --daemonize

    Note:A session daemon might already be running, for example as a service that the service manager of your distribution started.

  3. List the available user space tracepoints:

    $
    lttng list --userspace

    You see the hello_world:my_first_tracepoint tracepoint listed under the ./hello process.

  4. Create a recording session:

    $
    lttng create my-user-space-session
  5. Create a recording event rule which matches user space tracepoint events named hello_world:my_first_tracepoint:

    $
    lttng enable-event --userspace hello_world:my_first_tracepoint
  6. Start recording:

    $
    lttng start
  7. Go back to the running hello application and press Enter.

    The program executes all lttng_ust_tracepoint() instrumentation points, emitting events as the event rule you created in step 5 matches them, and exits.

  8. Destroy the current recording session:

    $
    lttng destroy

    The lttng-destroy(1) command doesn’t destroy the trace data; it only destroys the state of the recording session.

    The lttng-destroy(1) command also runs the lttng-stop(1) command implicitly (see “Start and stop a recording session”). You need to stop recording to make LTTng flush the remaining trace data and make the trace readable.

By default, LTTng saves the traces to the $LTTNG_HOME/lttng-traces/NAME-DATE-TIME directory, where NAME is the recording session name. The LTTNG_HOME environment variable defaults to $HOME if not set.

View and analyze the recorded events

Once you have completed the Record Linux kernel events and Record user application events tutorials, you can inspect the recorded events.

There are tools you can use to read LTTng traces:

Babeltrace 2

A rich, flexible trace manipulation toolkit which includes a versatile command-line interface (babeltrace2(1)), a C library, and Python 3 bindings so that you can easily process or convert an LTTng trace with your own script.

The Babeltrace 2 project ships with a plugin (babeltrace2-plugin-ctf(7)) which supports the format of the traces which LTTng produces, CTF.

Trace Compass

A graphical user interface for viewing and analyzing any type of logs or traces, including those of LTTng.

Note:This section assumes that LTTng wrote the traces it recorded during the previous tutorials to their default location, in the $LTTNG_HOME/lttng-traces directory. The LTTNG_HOME environment variable defaults to $HOME if not set.

Use the babeltrace2 command-line tool

The simplest way to list all the recorded events of an LTTng trace is to pass its path to babeltrace2(1), without options:

$
babeltrace2 ~/lttng-traces/my-user-space-session*

The babeltrace2 command finds all traces recursively within the given path and prints all their events, sorting them chronologically.

Pipe the output of babeltrace2 into a tool like grep(1) for further filtering:

$
babeltrace2 /tmp/my-kernel-trace | grep _switch

Pipe the output of babeltrace2 into a tool like wc(1) to count the recorded events:

$
babeltrace2 /tmp/my-kernel-trace | grep _open | wc --lines

Use the Babeltrace 2 Python bindings

The text output of babeltrace2 is useful to isolate event records by simple matching using grep(1) and similar utilities. However, more elaborate filters, such as keeping only event records with a field value falling within a specific range, are not trivial to write using a shell. Moreover, reductions and even the most basic computations involving multiple event records are virtually impossible to implement.

Fortunately, Babeltrace 2 ships with Python 3 bindings which make it easy to read the event records of an LTTng trace sequentially and compute the desired information.

The following script accepts an LTTng Linux kernel trace path as its first argument and prints the short names of the top five running processes on CPU 0 during the whole trace:

top5proc.py

import bt2
import sys
import collections


def top5proc():
    # Get the trace path from the first command-line argument
    it = bt2.TraceCollectionMessageIterator(sys.argv[1])

    # This counter dictionary will hold execution times:
    #
    #     Task command name -> Total execution time (ns)
    exec_times = collections.Counter()

    # This holds the last `sched_switch` timestamp
    last_ts = None

    for msg in it:
        # We only care about event messages
        if type(msg) is not bt2._EventMessageConst:
            continue

        # Event of the event message
        event = msg.event

        # Keep only `sched_switch` events
        if event.cls.name != 'sched_switch':
            continue

        # Keep only records of events which LTTng emitted from CPU 0
        if event.packet.context_field['cpu_id'] != 0:
            continue

        # Event timestamp (ns)
        cur_ts = msg.default_clock_snapshot.ns_from_origin

        if last_ts is None:
            # Start here
            last_ts = cur_ts

        # (Short) name of the previous task command
        prev_comm = str(event.payload_field['prev_comm'])

        # Initialize an entry in our dictionary if not done yet
        if prev_comm not in exec_times:
            exec_times[prev_comm] = 0

        # Compute previous command execution time
        diff = cur_ts - last_ts

        # Update execution time of this command
        exec_times[prev_comm] += diff

        # Update last timestamp
        last_ts = cur_ts

    # Print top 5
    for name, ns in exec_times.most_common(5):
        print('{:20}{} s'.format(name, ns / 1e9))


if __name__ == '__main__':
    top5proc()

Run this script:

$
python3 top5proc.py /tmp/my-kernel-trace/kernel

Output example:

swapper/0           48.607245889 s
chromium            7.192738188 s
pavucontrol         0.709894415 s
Compositor          0.660867933 s
Xorg.bin            0.616753786 s

Note that swapper/0 is the “idle” process of CPU 0 on Linux; since we weren’t using the CPU that much when recording, its first position in the list makes sense.

Core concepts

From a user’s perspective, the LTTng system is built on a few concepts, or objects, on which the lttng command-line tool operates by sending commands to the session daemon (through liblttng-ctl).

Understanding how those objects relate to each other is key to master the toolkit.

The core concepts of LTTng are:

Note:The lttng-concepts(7) manual page also documents the core concepts of LTTng, with more links to other LTTng-tools manual pages.

Instrumentation point, event rule, and event

An instrumentation point is a point, within a piece of software, which, when executed, creates an LTTng event.

LTTng offers various types of instrumentation.

An event rule is a set of conditions to match a set of events.

When LTTng creates an event E, an event rule ER is said to match E when E satisfies all the conditions of ER. This concept is similar to a regular expression which matches a set of strings.

When an event rule matches an event, LTTng emits the event, therefore attempting to execute one or more actions.

Important:The event creation and emission processes are documentation concepts to help understand the journey from an instrumentation point to the execution of actions.

The actual creation of an event can be costly because LTTng needs to evaluate the arguments of the instrumentation point.

In practice, LTTng implements various optimizations for the Linux kernel and user space tracing domains to avoid actually creating an event when the tracer knows, thanks to properties which are independent from the event payload and current context, that it would never emit such an event. Those properties are:

In other words: if, for a given instrumentation point IP, the LTTng tracer knows that it would never emit an event, executing IP represents a simple boolean variable check and, for a Linux kernel recording event rule, a few process attribute checks.

As of LTTng 2.13, there are two places where you can find an event rule:

Recording event rule

A specific type of event rule of which the action is to record the matched event as an event record.

See “Create and enable a recording event rule” to learn more.

“Event rule matches” trigger condition (since LTTng 2.13)

When the event rule of the trigger condition matches an event, LTTng can execute user-defined actions such as sending an LTTng notification, starting a recording session, and more.

See “Add an “event rule matches” trigger to a session daemon” to learn more.

For LTTng to emit an event EE must satisfy all the basic conditions of an event rule ER, that is:

  • The instrumentation point from which LTTng creates E has a specific type.

  • A pattern matches the name of E while another pattern doesn’t.

  • The log level of the instrumentation point from which LTTng creates E is at least as severe as some value, or is exactly some value.

  • The fields of the payload of E and the current context fields satisfy a filter expression.

A recording event rule has additional, implicit conditions to satisfy.

Instrumentation point types

As of LTTng 2.13, the available instrumentation point types are, depending on the tracing domain:

Linux kernel
LTTng tracepoint

A statically defined point in the source code of the kernel image or of a kernel module using the LTTng-modules macros.

Linux kernel system call

Entry, exit, or both of a Linux kernel system call.

Linux kprobe

A single probe dynamically placed in the compiled kernel code.

When you create such an instrumentation point, you set its memory address or symbol name.

Linux user space probe

A single probe dynamically placed at the entry of a compiled user space application/library function through the kernel.

When you create such an instrumentation point, you set:

With the ELF method

Its application/library path and its symbol name.

With the USDT method

Its application/library path, its provider name, and its probe name.

“USDT” stands for SystemTap User-level Statically Defined Tracing, a DTrace-style marker.

As of LTTng 2.13, LTTng only supports USDT probes which are not reference-counted.

Linux kretprobe

Entry, exit, or both of a Linux kernel function.

When you create such an instrumentation point, you set the memory address or symbol name of its function.

User space
LTTng tracepoint

A statically defined point in the source code of a C/C++ application/library using the LTTng-UST macros.

java.util.logging, Apache log4j, and Python
Java or Python logging statement

A method call on a Java or Python logger attached to an LTTng-UST handler.

See “List the available instrumentation points” to learn how to list available Linux kernel, user space, and logging instrumentation points.

Trigger

A trigger associates a condition to one or more actions.

When the condition of a trigger is satisfied, LTTng attempts to execute its actions.

As of LTTng 2.13, the available trigger conditions and actions are:

Conditions
  • The consumed buffer size of a given recording session becomes greater than some value.

  • The buffer usage of a given channel becomes greater than some value.

  • The buffer usage of a given channel becomes less than some value.

  • There’s an ongoing recording session rotation.

  • A recording session rotation becomes completed.

  • An event rule matches an event.

Actions

A trigger belongs to a session daemon, not to a specific recording session. For a given session daemon, each Unix user has its own, private triggers. Note, however, that the root Unix user may, for the root session daemon:

  • Add a trigger as another Unix user.

  • List all the triggers, regardless of their owner.

  • Remove a trigger which belongs to another Unix user.

For a given session daemon and Unix user, a trigger has a unique name.

Recording session

A recording session (named “tracing session” prior to LTTng 2.13) is a stateful dialogue between you and a session daemon for everything related to event recording.

Everything that you do when you control LTTng tracers to record events happens within a recording session. In particular, a recording session:

  • Has its own name, unique for a given session daemon.

  • Has its own set of trace files, if any.

  • Has its own state of activity (started or stopped).

    An active recording session is an implicit recording event rule condition.

  • Has its own mode (local, network streaming, snapshot, or live).

  • Has its own channels to which are attached their own recording event rules.

  • Has its own process attribute inclusion sets.

A recording session contains channels that are members of tracing domains and contain recording event rules.

Those attributes and objects are completely isolated between different recording sessions.

A recording session is like an ATM session: the operations you do on the banking system through the ATM don’t alter the data of other users of the same system. In the case of the ATM, a session lasts as long as your bank card is inside. In the case of LTTng, a recording session lasts from the lttng-create(1) command to the lttng-destroy(1) command.

Each Unix user has its own set of recording sessions.

A recording session belongs to a session daemon. For a given session daemon, each Unix user has its own, private recording sessions. Note, however, that the root Unix user may operate on or destroy another user’s recording session.

Recording session mode

LTTng offers four recording session modes:

Local mode

Write the trace data to the local file system.

Network streaming mode

Send the trace data over the network to a listening relay daemon.

Snapshot mode

Only write the trace data to the local file system or send it to a listening relay daemon when LTTng takes a snapshot.

LTTng forces all the channels to be created to be configured to be snapshot-ready.

LTTng takes a snapshot of such a recording session when:

Live mode

Send the trace data over the network to a listening relay daemon for live reading.

An LTTng live reader (for example, babeltrace2(1)) can connect to the same relay daemon to receive trace data while the recording session is active.

Tracing domain

A tracing domain identifies a type of LTTng tracer.

A tracing domain has its own properties and features.

There are currently five available tracing domains:

  • Linux kernel

  • User space

  • java.util.logging (JUL)

  • log4j

  • Python

You must specify a tracing domain to target a type of LTTng tracer when using some lttng commands to avoid ambiguity. For example, because the Linux kernel and user space tracing domains support named tracepoints as instrumentation points, you need to specify a tracing domain when you create an event rule because both tracing domains could have tracepoints sharing the same name.

You can create channels in the Linux kernel and user space tracing domains. The other tracing domains have a single, default channel.

Channel and ring buffer

A channel is an object which is responsible for a set of ring buffers.

Each ring buffer is divided into multiple sub-buffers. When a recording event rule matches an event, LTTng can record it to one or more sub-buffers of one or more channels.

When you create a channel, you set its final attributes, that is:

A channel is always associated to a tracing domain. The java.util.logging (JUL), log4j, and Python tracing domains each have a default channel which you can’t configure.

A channel owns recording event rules.

Buffering scheme

A channel has at least one ring buffer per CPU. LTTng always records an event to the ring buffer dedicated to the CPU which emits it.

The buffering scheme of a user space channel determines what has its own set of per-CPU ring buffers:

Per-user buffering

Allocate one set of ring buffers—one per CPU—shared by all the instrumented processes of:

If your Unix user is root

Each Unix user.

Per-user buffering scheme (recording session belongs to the root Unix user).
Otherwise

Your Unix user.

Per-user buffering scheme (recording session belongs to the Bob Unix user).
Per-process buffering

Allocate one set of ring buffers—one per CPU—for each instrumented process of:

If your Unix user is root

All Unix users.

Per-process buffering scheme (recording session belongs to the root Unix user).
Otherwise

Your Unix user.

Per-process buffering scheme (recording session belongs to the Bob Unix user).

The per-process buffering scheme tends to consume more memory than the per-user option because systems generally have more instrumented processes than Unix users running instrumented processes. However, the per-process buffering scheme ensures that one process having a high event throughput won’t fill all the shared sub-buffers of the same Unix user, only its own.

The buffering scheme of a Linux kernel channel is always to allocate a single set of ring buffers for the whole system. This scheme is similar to the per-user option, but with a single, global user “running” the kernel.

Event record loss mode

When LTTng emits an event, LTTng can record it to a specific, available sub-buffer within the ring buffers of specific channels. When there’s no space left in a sub-buffer, the tracer marks it as consumable and another, available sub-buffer starts receiving the following event records. An LTTng consumer daemon eventually consumes the marked sub-buffer, which returns to the available state.

In an ideal world, sub-buffers are consumed faster than they’re filled. In the real world, however, all sub-buffers can be full at some point, leaving no space to record the following events.

In an ideal world, sub-buffers are consumed faster than they’re filled, as it’s the case in the previous animation. In the real world, however, all sub-buffers can be full at some point, leaving no space to record the following events.

By default, LTTng-modules and LTTng-UST are non-blocking tracers: when there’s no available sub-buffer to record an event, it’s acceptable to lose event records when the alternative would be to cause substantial delays in the execution of the instrumented application. LTTng privileges performance over integrity; it aims at perturbing the instrumented application as little as possible in order to make the detection of subtle race conditions and rare interrupt cascades possible.

Since LTTng 2.10, the LTTng user space tracer, LTTng-UST, supports a blocking mode. See the blocking timeout example to learn how to use the blocking mode.

When it comes to losing event records because there’s no available sub-buffer, or because the blocking timeout of the channel is reached, the event record loss mode of the channel determines what to do. The available event record loss modes are:

Discard mode

Drop the newest event records until a sub-buffer becomes available.

This is the only available mode when you specify a blocking timeout.

With this mode, LTTng increments a count of lost event records when an event record is lost and saves this count to the trace. A trace reader can use the saved discarded event record count of the trace to decide whether or not to perform some analysis even if trace data is known to be missing.

Overwrite mode

Clear the sub-buffer containing the oldest event records and start writing the newest event records there.

This mode is sometimes called flight recorder mode because it’s similar to a flight recorder: always keep a fixed amount of the latest data. It’s also similar to the roll mode of an oscilloscope.

Since LTTng 2.8, with this mode, LTTng writes to a given sub-buffer its sequence number within its data stream. With a local, network streaming, or live recording session, a trace reader can use such sequence numbers to report lost packets. A trace reader can use the saved discarded sub-buffer (packet) count of the trace to decide whether or not to perform some analysis even if trace data is known to be missing.

With this mode, LTTng doesn’t write to the trace the exact number of lost event records in the lost sub-buffers.

Which mechanism you should choose depends on your context: prioritize the newest or the oldest event records in the ring buffer?

Beware that, in overwrite mode, the tracer abandons a whole sub-buffer as soon as a there’s no space left for a new event record, whereas in discard mode, the tracer only discards the event record that doesn’t fit.

There are a few ways to decrease your probability of losing event records. The “Sub-buffer size and count” section shows how to fine-tune the sub-buffer size and count of a channel to virtually stop losing event records, though at the cost of greater memory usage.

Sub-buffer size and count

A channel has one or more ring buffer for each CPU of the target system.

See the “Buffering scheme” section to learn how many ring buffers of a given channel are dedicated to each CPU depending on its buffering scheme.

Set the size of each sub-buffer the ring buffers of a channel contain and how many there are when you create it.

Note that LTTng switching the current sub-buffer of a ring buffer (marking a full one as consumable and switching to an available one for LTTng to record the next events) introduces noticeable CPU overhead. Knowing this, the following list presents a few practical situations along with how to configure the sub-buffer size and count for them:

High event throughput

In general, prefer large sub-buffers to lower the risk of losing event records.

Having larger sub-buffers also ensures a lower sub-buffer switching frequency.

The sub-buffer count is only meaningful if you create the channel in overwrite mode: in this case, if LTTng overwrites a sub-buffer, then the other sub-buffers are left unaltered.

Low event throughput

In general, prefer smaller sub-buffers since the risk of losing event records is low.

Because LTTng emits events less frequently, the sub-buffer switching frequency should remain low and therefore the overhead of the tracer shouldn’t be a problem.

Low memory system

If your target system has a low memory limit, prefer fewer first, then smaller sub-buffers.

Even if the system is limited in memory, you want to keep the sub-buffers as large as possible to avoid a high sub-buffer switching frequency.

Note that LTTng uses CTF as its trace format, which means event record data is very compact. For example, the average LTTng kernel event record weights about 32 bytes. Therefore, a sub-buffer size of 1 MiB is considered large.

The previous scenarios highlight the major trade-off between a few large sub-buffers and more, smaller sub-buffers: sub-buffer switching frequency vs. how many event records are lost in overwrite mode. Assuming a constant event throughput and using the overwrite mode, the two following configurations have the same ring buffer total size:

Two sub-buffers of 4 MiB each

Expect a very low sub-buffer switching frequency, but if LTTng ever needs to overwrite a sub-buffer, half of the event records so far (4 MiB) are definitely lost.

Eight sub-buffers of 1 MiB each

Expect four times the tracer overhead of the configuration above, but if LTTng needs to overwrite a sub-buffer, only the eighth of event records so far (1 MiB) are definitely lost.

In discard mode, the sub-buffer count parameter is pointless: use two sub-buffers and set their size according to your requirements.

Maximum trace file size and count (trace file rotation)

By default, trace files can grow as large as needed.

Set the maximum size of each trace file that LTTng writes of a given channel when you create it.

When the size of a trace file reaches the fixed maximum size of the channel, LTTng creates another file to contain the next event records. LTTng appends a file count to each trace file name in this case.

If you set the trace file size attribute when you create a channel, the maximum number of trace files that LTTng creates is unlimited by default. To limit them, set a maximum number of trace files. When the number of trace files reaches the fixed maximum count of the channel, LTTng overwrites the oldest trace file. This mechanism is called trace file rotation.

Important:Even if you don’t limit the trace file count, always assume that LTTng manages all the trace files of the recording session.

In other words, there’s no safe way to know if LTTng still holds a given trace file open with the trace file rotation feature.

The only way to obtain an unmanaged, self-contained LTTng trace before you destroy the recording session is with the recording session rotation feature, which is available since LTTng 2.11.

Timers

Each channel can have up to three optional timers:

Switch timer

When this timer expires, a sub-buffer switch happens: for each ring buffer of the channel, LTTng marks the current sub-buffer as consumable and switches to an available one to record the next events.

A switch timer is useful to ensure that LTTng consumes and commits trace data to trace files or to a distant relay daemon periodically in case of a low event throughput.

Such a timer is also convenient when you use large sub-buffers to cope with a sporadic high event throughput, even if the throughput is otherwise low.

Set the period of the switch timer of a channel when you create it with the --switch-timer option.

Read timer

When this timer expires, LTTng checks for full, consumable sub-buffers.

By default, the LTTng tracers use an asynchronous message mechanism to signal a full sub-buffer so that a consumer daemon can consume it.

When such messages must be avoided, for example in real-time applications, use this timer instead.

Set the period of the read timer of a channel when you create it with the --read-timer option.

Monitor timer

When this timer expires, the consumer daemon samples some channel statistics to evaluate the following trigger conditions:

  1. The consumed buffer size of a given recording session becomes greater than some value.

  2. The buffer usage of a given channel becomes greater than some value.

  3. The buffer usage of a given channel becomes less than some value.

If you disable the monitor timer of a channel C:

  • The consumed buffer size value of the recording session of C could be wrong for trigger condition type 1: the consumed buffer size of C won’t be part of the grand total.

  • The buffer usage trigger conditions (types 2 and 3) for C will never be satisfied.

Set the period of the monitor timer of a channel when you create it with the --monitor-timer option.

Recording event rule and event record

A recording event rule is a specific type of event rule of which the action is to serialize and record the matched event as an event record.

Set the explicit conditions of a recording event rule when you create it. A recording event rule also has the following implicit conditions:

  • The recording event rule itself is enabled.

    A recording event rule is enabled on creation.

  • The channel to which the recording event rule is attached is enabled.

    A channel is enabled on creation.

  • The recording session of the recording event rule is active (started).

    A recording session is inactive (stopped) on creation.

  • The process for which LTTng creates an event to match is allowed to record events.

    All processes are allowed to record events on recording session creation.

You always attach a recording event rule to a channel, which belongs to a recording session, when you create it.

When a recording event rule ER matches an event E, LTTng attempts to serialize and record E to one of the available sub-buffers of the channel to which E is attached.

When multiple matching recording event rules are attached to the same channel, LTTng attempts to serialize and record the matched event once. In the following example, the second recording event rule is redundant when both are enabled:

$
$
lttng enable-event --userspace hello:world
lttng enable-event --userspace hello:world --loglevel=INFO
Logical path from an instrumentation point to an event record.

As of LTTng 2.13, you cannot remove a recording event rule: it exists as long as its recording session exists.

Components of LTTng

The second T in LTTng stands for toolkit: it would be wrong to call LTTng a simple tool since it’s composed of multiple interacting components.

This section describes those components, explains their respective roles, and shows how they connect together to form the LTTng ecosystem.

The following diagram shows how the most important components of LTTng interact with user applications, the Linux kernel, and you:

Control and trace data paths between LTTng components.

The LTTng project integrates:

LTTng-tools

Libraries and command-line interface to control recording sessions:

LTTng-UST

Libraries and Java/Python packages to instrument and trace user applications:

LTTng-modules

Linux kernel modules to instrument and trace the kernel:

  • LTTng kernel tracer module.

  • Recording ring buffer kernel modules.

  • Probe kernel modules.

  • LTTng logger kernel module.

Tracing control command-line interface

The lttng(1) command-line tool is the standard user interface to control LTTng recording sessions.

The lttng tool is part of LTTng-tools.

The lttng tool is linked with liblttng-ctl to communicate with one or more session daemons behind the scenes.

The lttng tool has a Git-like interface:

$
lttng [GENERAL OPTIONS] <COMMAND> [COMMAND OPTIONS]

The “Tracing control” section explores the available features of LTTng through its lttng tool.

Tracing control library

The tracing control library.

The LTTng control library, liblttng-ctl, is used to communicate with a session daemon using a C API that hides the underlying details of the protocol.

liblttng-ctl is part of LTTng-tools.

The lttng command-line tool is linked with liblttng-ctl.

Use liblttng-ctl in C or C++ source code by including its “master” header:

#include <lttng/lttng.h>

As of LTTng 2.13, the best available developer documentation for liblttng-ctl is its installed header files. Functions and structures are documented with header comments.

User space tracing library

The user space tracing library.

The user space tracing library, liblttng-ust (see lttng-ust(3)), is the LTTng user space tracer.

liblttng-ust receives commands from a session daemon, for example to allow specific instrumentation points to emit LTTng events, and writes event records to ring buffers shared with a consumer daemon.

liblttng-ust is part of LTTng-UST.

liblttng-ust can also send asynchronous messages to the session daemon when it emits an event. This supports the “event rule matches” trigger condition feature (see “Add an “event rule matches” trigger to a session daemon”).

Public C header files are installed beside liblttng-ust to instrument any C or C++ application.

LTTng-UST agents, which are regular Java and Python packages, use their own tracepoint provider package which is linked with liblttng-ust.

An application or library doesn’t have to initialize liblttng-ust manually: its constructor does the necessary tasks to register the application to a session daemon. The initialization phase also configures instrumentation points depending on the event rules that you already created.

User space tracing agents

The user space tracing agents.

The LTTng-UST Java and Python agents are regular Java and Python packages which add LTTng tracing capabilities to the native logging frameworks.

The LTTng-UST agents are part of LTTng-UST.

In the case of Java, the java.util.logging core logging facilities and Apache log4j 1.2 are supported. Note that Apache Log4j 2 isn’t supported.

In the case of Python, the standard logging package is supported. Both Python 2 and Python 3 modules can import the LTTng-UST Python agent package.

The applications using the LTTng-UST agents are in the java.util.logging (JUL), log4j, and Python tracing domains.

Both agents use the same mechanism to convert log statements to LTTng events. When an agent initializes, it creates a log handler that attaches to the root logger. The agent also registers to a session daemon. When the user application executes a log statement, the root logger passes it to the log handler of the agent. The custom log handler of the agent calls a native function in a tracepoint provider package shared library linked with liblttng-ust, passing the formatted log message and other fields, like its logger name and its log level. This native function contains a user space instrumentation point, therefore tracing the log statement.

The log level condition of a recording event rule is considered when tracing a Java or a Python application, and it’s compatible with the standard java.util.logging, log4j, and Python log levels.

LTTng kernel modules

The LTTng kernel modules.

The LTTng kernel modules are a set of Linux kernel modules which implement the kernel tracer of the LTTng project.

The LTTng kernel modules are part of LTTng-modules.

The LTTng kernel modules include:

  • A set of probe modules.

    Each module attaches to a specific subsystem of the Linux kernel using its tracepoint instrument points.

    There are also modules to attach to the entry and return points of the Linux system call functions.

  • Ring buffer modules.

    A ring buffer implementation is provided as kernel modules. The LTTng kernel tracer writes to ring buffers; a consumer daemon reads from ring buffers.

  • The LTTng kernel tracer module.

  • The LTTng logger module.

    The LTTng logger module implements the special /proc/lttng-logger (and /dev/lttng-logger, since LTTng 2.11) files so that any executable can generate LTTng events by opening those files and writing to them.

The LTTng kernel tracer can also send asynchronous messages to the session daemon when it emits an event. This supports the “event rule matches” trigger condition feature (see “Add an “event rule matches” trigger to a session daemon”).

Generally, you don’t have to load the LTTng kernel modules manually (using modprobe(8), for example): a root session daemon loads the necessary modules when starting. If you have extra probe modules, you can specify to load them to the session daemon on the command line (see the --extra-kmod-probes option). See also Linux kernel module signature.

The LTTng kernel modules are installed in /usr/lib/modules/release/extra by default, where release is the kernel release (output of uname --kernel-release).

Session daemon

The session daemon.

The session daemon, lttng-sessiond(8), is a daemon which:

The session daemon is part of LTTng-tools.

The session daemon sends control requests to and receives control responses from:

  • The user space tracing library.

    Any instance of the user space tracing library first registers to a session daemon. Then, the session daemon can send requests to this instance, such as:

    • Get the list of tracepoints.

    • Share a recording event rule so that the user space tracing library can decide whether or not a given tracepoint can emit events. Amongst the possible conditions of a recording event rule is a filter expression which liblttng-ust evaluates before it emits an event.

    • Share channel attributes and ring buffer locations.

    The session daemon and the user space tracing library use a Unix domain socket to communicate.

  • The user space tracing agents.

    Any instance of a user space tracing agent first registers to a session daemon. Then, the session daemon can send requests to this instance, such as:

    • Get the list of loggers.

    • Enable or disable a specific logger.

    The session daemon and the user space tracing agent use a TCP connection to communicate.

  • The LTTng kernel tracer.

  • The consumer daemon.

    The session daemon sends requests to the consumer daemon to instruct it where to send the trace data streams, amongst other information.

  • The relay daemon.

The session daemon receives commands from the tracing control library.

The session daemon can receive asynchronous messages from the user space and kernel tracers when they emit events. This supports the “event rule matches” trigger condition feature (see “Add an “event rule matches” trigger to a session daemon”).

The root session daemon loads the appropriate LTTng kernel modules on startup. It also spawns one or more consumer daemons as soon as you create a recording event rule.

The session daemon doesn’t send and receive trace data: this is the role of the consumer daemon and relay daemon. It does, however, generate the CTF metadata stream.

Each Unix user can have its own session daemon instance. The recording sessions which different session daemons manage are completely independent.

The root user’s session daemon is the only one which is allowed to control the LTTng kernel tracer, and its spawned consumer daemon is the only one which is allowed to consume trace data from the LTTng kernel tracer. Note, however, that any Unix user which is a member of the tracing group is allowed to create channels in the Linux kernel tracing domain, and therefore to use the Linux kernel LTTng tracer.

The lttng command-line tool automatically starts a session daemon when using its create command if none is currently running. You can also start the session daemon manually.

Consumer daemon

The consumer daemon.

The consumer daemon, lttng-consumerd, is a daemon which shares ring buffers with user applications or with the LTTng kernel modules to collect trace data and send it to some location (file system or to a relay daemon over the network).

The consumer daemon is part of LTTng-tools.

You don’t start a consumer daemon manually: a consumer daemon is always spawned by a session daemon as soon as you create a recording event rule, that is, before you start recording. When you kill its owner session daemon, the consumer daemon also exits because it’s the child process of the session daemon. Command-line options of lttng-sessiond(8) target the consumer daemon process.

There are up to two running consumer daemons per Unix user, whereas only one session daemon can run per user. This is because each process can be either 32-bit or 64-bit: if the target system runs a mixture of 32-bit and 64-bit processes, it’s more efficient to have separate corresponding 32-bit and 64-bit consumer daemons. The root user is an exception: it can have up to three running consumer daemons: 32-bit and 64-bit instances for its user applications, and one more reserved for collecting kernel trace data.

Relay daemon

The relay daemon.

The relay daemon, lttng-relayd(8), is a daemon acting as a bridge between remote session and consumer daemons, local trace files, and a remote live trace reader.

The relay daemon is part of LTTng-tools.

The main purpose of the relay daemon is to implement a receiver of trace data over the network. This is useful when the target system doesn’t have much file system space to write trace files locally.

The relay daemon is also a server to which a live trace reader can connect. The live trace reader sends requests to the relay daemon to receive trace data as the target system records events. The communication protocol is named LTTng live; it’s used over TCP connections.

Note that you can start the relay daemon on the target system directly. This is the setup of choice when the use case is to view/analyze events as the target system records them without the need of a remote system.

Instrumentation

There are many examples of tracing and monitoring in our everyday life:

  • You have access to real-time and historical weather reports and forecasts thanks to weather stations installed around the country.

  • You know your heart is safe thanks to an electrocardiogram.

  • You make sure not to drive your car too fast and to have enough fuel to reach your destination thanks to gauges visible on your dashboard.

All the previous examples have something in common: they rely on instruments. Without the electrodes attached to the surface of your body skin, cardiac monitoring is futile.

LTTng, as a tracer, is no different from those real life examples. If you’re about to trace a software system or, in other words, record its history of execution, you better have instrumentation points in the subject you’re tracing, that is, the actual software system.

Various ways were developed to instrument a piece of software for LTTng tracing. The most straightforward one is to manually place static instrumentation points, called tracepoints, in the source code of the application. The Linux kernel tracing domain also makes it possible to dynamically add instrumentation points.

If you’re only interested in tracing the Linux kernel, your instrumentation needs are probably already covered by the built-in Linux kernel instrumentation points of LTTng. You may also wish to have LTTng trace a user application which is already instrumented for LTTng tracing. In such cases, skip this whole section and read the topics of the “Tracing control” section.

Many methods are available to instrument a piece of software for LTTng tracing:

Instrument a C/C++ user application

The high level procedure to instrument a C or C++ user application with the LTTng user space tracing library, liblttng-ust, is:

  1. Create the source files of a tracepoint provider package.

  2. Add tracepoints to the source code of the application.

  3. Build and link a tracepoint provider package and the user application.

If you need quick, printf(3)-like instrumentation, skip those steps and use lttng_ust_tracef() or lttng_ust_tracelog() instead.

Important:You need to install LTTng-UST to instrument a user application with liblttng-ust.

Create the source files of a tracepoint provider package

A tracepoint provider is a set of compiled functions which provide tracepoints to an application, the type of instrumentation point which LTTng-UST provides.

Those functions can make LTTng emit events with user-defined fields and serialize those events as event records to one or more LTTng-UST channel sub-buffers. The lttng_ust_tracepoint() macro, which you insert in the source code of a user application, calls those functions.

A tracepoint provider package is an object file (.o) or a shared library (.so) which contains one or more tracepoint providers. Its source files are:

A tracepoint provider package is dynamically linked with liblttng-ust, the LTTng user space tracer, at run time.

User application linked with liblttng-ust and containing a tracepoint provider.

Note:If you need quick, printf(3)-like instrumentation, skip creating and using a tracepoint provider and use lttng_ust_tracef() or lttng_ust_tracelog() instead.

Create a tracepoint provider header file template

A tracepoint provider header file contains the tracepoint definitions of a tracepoint provider.

To create a tracepoint provider header file:

  1. Start from this template:

    Tracepoint provider header file template (.h file extension).

    #undef LTTNG_UST_TRACEPOINT_PROVIDER
    #define LTTNG_UST_TRACEPOINT_PROVIDER provider_name
    
    #undef LTTNG_UST_TRACEPOINT_INCLUDE
    #define LTTNG_UST_TRACEPOINT_INCLUDE "./tp.h"
    
    #if !defined(_TP_H) || defined(LTTNG_UST_TRACEPOINT_HEADER_MULTI_READ)
    #define _TP_H
    
    #include <lttng/tracepoint.h>
    
    /*
     * Use LTTNG_UST_TRACEPOINT_EVENT(), LTTNG_UST_TRACEPOINT_EVENT_CLASS(),
     * LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(), and
     * LTTNG_UST_TRACEPOINT_LOGLEVEL() here.
     */
    
    #endif /* _TP_H */
    
    #include <lttng/tracepoint-event.h>
    

    Replace:

    • provider_name with the name of your tracepoint provider.

    • "tp.h" with the name of your tracepoint provider header file.

  2. Below the #include <lttng/tracepoint.h> line, put your tracepoint definitions.

Your tracepoint provider name must be unique amongst all the possible tracepoint provider names used on the same target system. We suggest to include the name of your project or company in the name, for example, org_lttng_my_project_tpp.

Create a tracepoint definition

A tracepoint definition defines, for a given tracepoint:

  • Its input arguments.

    They’re the macro parameters that the lttng_ust_tracepoint() macro accepts for this particular tracepoint in the source code of the user application.

  • Its output event fields.

    They’re the sources of event fields that form the payload of any event that the execution of the lttng_ust_tracepoint() macro emits for this particular tracepoint.

Create a tracepoint definition with the LTTNG_UST_TRACEPOINT_EVENT() macro below the #include <lttng/tracepoint.h> line in the tracepoint provider header file template.

The syntax of the LTTNG_UST_TRACEPOINT_EVENT() macro is:

LTTNG_UST_TRACEPOINT_EVENT() macro syntax.

LTTNG_UST_TRACEPOINT_EVENT(
    /* Tracepoint provider name */
    provider_name,

    /* Tracepoint name */
    tracepoint_name,

    /* Input arguments */
    LTTNG_UST_TP_ARGS(
        arguments
    ),

    /* Output event fields */
    LTTNG_UST_TP_FIELDS(
        fields
    )
)

Replace:

  • provider_name with your tracepoint provider name.

  • tracepoint_name with your tracepoint name.

  • arguments with the input arguments.

  • fields with the output event field definitions.

The full name of this tracepoint is provider_name:tracepoint_name.

Event name length limitation

Important:The concatenation of the tracepoint provider name and the tracepoint name must not exceed 254 characters. If it does, the instrumented application compiles and runs, but LTTng throws multiple warnings and you could experience serious issues.

The syntax of the LTTNG_UST_TP_ARGS() macro is:

LTTNG_UST_TP_ARGS() macro syntax.

LTTNG_UST_TP_ARGS(
    type, arg_name
)

Replace:

  • type with the C type of the argument.

  • arg_name with the argument name.

You can repeat type and arg_name up to 10 times to have more than one argument.

Example:LTTNG_UST_TP_ARGS() usage with three arguments.

LTTNG_UST_TP_ARGS(
    int, count,
    float, ratio,
    const char*, query
)

The LTTNG_UST_TP_ARGS() and LTTNG_UST_TP_ARGS(void) forms are valid to create a tracepoint definition with no input arguments.

The LTTNG_UST_TP_FIELDS() macro contains a list of lttng_ust_field_*() macros. Each lttng_ust_field_*() macro defines one event field. See lttng-ust(3) for a complete description of the available lttng_ust_field_*() macros. A lttng_ust_field_*() macro specifies the type, size, and byte order of one event field.

Each lttng_ust_field_*() macro takes an argument expression parameter. This is a C expression that the tracer evaluates at the lttng_ust_tracepoint() macro site in the source code of the application. This expression provides the source of data of a field. The argument expression can include input argument names listed in the LTTNG_UST_TP_ARGS() macro.

Each lttng_ust_field_*() macro also takes a field name parameter. Field names must be unique within a given tracepoint definition.

Here’s a complete tracepoint definition example:

Example:Tracepoint definition.

The following tracepoint definition defines a tracepoint which takes three input arguments and has four output event fields.

#include "my-custom-structure.h"

LTTNG_UST_TRACEPOINT_EVENT(
    my_provider,
    my_tracepoint,
    LTTNG_UST_TP_ARGS(
        const struct my_custom_structure *, my_custom_structure,
        float, ratio,
        const char *, query
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_string(query_field, query)
        lttng_ust_field_float(double, ratio_field, ratio)
        lttng_ust_field_integer(int, recv_size,
                                my_custom_structure->recv_size)
        lttng_ust_field_integer(int, send_size,
                                my_custom_structure->send_size)
    )
)

Refer to this tracepoint definition with the lttng_ust_tracepoint() macro in the source code of your application like this:

lttng_ust_tracepoint(my_provider, my_tracepoint,
                     my_structure, some_ratio, the_query);

Note:The LTTng-UST tracer only evaluates the arguments of a tracepoint at run time when such a tracepoint could emit an event. See this note to learn more.

Use a tracepoint class

A tracepoint class is a class of tracepoints which share the same output event field definitions. A tracepoint instance is one instance of such a defined tracepoint class, with its own tracepoint name.

The LTTNG_UST_TRACEPOINT_EVENT() macro is actually a shorthand which defines both a tracepoint class and a tracepoint instance at the same time.

When you build a tracepoint provider package, the C or C++ compiler creates one serialization function for each tracepoint class. A serialization function is responsible for serializing the event fields of a tracepoint to a sub-buffer when recording.

For various performance reasons, when your situation requires multiple tracepoint definitions with different names, but with the same event fields, we recommend that you manually create a tracepoint class and instantiate as many tracepoint instances as needed. One positive effect of such a design, amongst other advantages, is that all tracepoint instances of the same tracepoint class reuse the same serialization function, thus reducing cache pollution.

Example:Use a tracepoint class and tracepoint instances.

Consider the following three tracepoint definitions:

LTTNG_UST_TRACEPOINT_EVENT(
    my_app,
    get_account,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, userid, userid)
        lttng_ust_field_integer(size_t, len, len)
    )
)

LTTNG_UST_TRACEPOINT_EVENT(
    my_app,
    get_settings,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, userid, userid)
        lttng_ust_field_integer(size_t, len, len)
    )
)

LTTNG_UST_TRACEPOINT_EVENT(
    my_app,
    get_transaction,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, userid, userid)
        lttng_ust_field_integer(size_t, len, len)
    )
)

In this case, we create three tracepoint classes, with one implicit tracepoint instance for each of them: get_account, get_settings, and get_transaction. However, they all share the same event field names and types. Hence three identical, yet independent serialization functions are created when you build the tracepoint provider package.

A better design choice is to define a single tracepoint class and three tracepoint instances:

/* The tracepoint class */
LTTNG_UST_TRACEPOINT_EVENT_CLASS(
    /* Tracepoint class provider name */
    my_app,

    /* Tracepoint class name */
    my_class,

    /* Input arguments */
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    ),

    /* Output event fields */
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, userid, userid)
        lttng_ust_field_integer(size_t, len, len)
    )
)

/* The tracepoint instances */
LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
    /* Tracepoint class provider name */
    my_app,

    /* Tracepoint class name */
    my_class,

    /* Instance provider name */
    my_app,

    /* Tracepoint name */
    get_account,

    /* Input arguments */
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    )
)
LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
    my_app,
    my_class,
    get_settings,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    )
)
LTTNG_UST_TRACEPOINT_EVENT_INSTANCE(
    my_app,
    my_class,
    get_transaction,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    )
)

The tracepoint class and instance provider names must be the same if the LTTNG_UST_TRACEPOINT_EVENT_CLASS() and LTTNG_UST_TRACEPOINT_EVENT_INSTANCE() expansions are part of the same translation unit. See lttng-ust(3) to learn more.

Assign a log level to a tracepoint definition

Assign a log level to a tracepoint definition with the LTTNG_UST_TRACEPOINT_LOGLEVEL() macro.

Assigning different levels of severity to tracepoint definitions can be useful: when you create a recording event rule, you can target tracepoints having a log level at least as severe as a specific value.

The concept of LTTng-UST log levels is similar to the levels found in typical logging frameworks:

  • In a logging framework, the log level is given by the function or method name you use at the log statement site: debug(), info(), warn(), error(), and so on.

  • In LTTng-UST, you statically assign the log level to a tracepoint definition; any lttng_ust_tracepoint() macro invocation which refers to this definition has this log level.

You must use LTTNG_UST_TRACEPOINT_LOGLEVEL() after the LTTNG_UST_TRACEPOINT_EVENT() or LTTNG_UST_TRACEPOINT_INSTANCE() macro for a given tracepoint.

The syntax of the LTTNG_UST_TRACEPOINT_LOGLEVEL() macro is:

LTTNG_UST_TRACEPOINT_LOGLEVEL() macro syntax.

LTTNG_UST_TRACEPOINT_LOGLEVEL(provider_name, tracepoint_name, log_level)

Replace:

  • provider_name with the tracepoint provider name.

  • tracepoint_name with the tracepoint name.

  • log_level with the log level to assign to the tracepoint definition named tracepoint_name in the provider_name tracepoint provider.

    See lttng-ust(3) for a list of available log level names.

Example:Assign the LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_UNIT log level to a tracepoint definition.

/* Tracepoint definition */
LTTNG_UST_TRACEPOINT_EVENT(
    my_app,
    get_transaction,
    LTTNG_UST_TP_ARGS(
        int, userid,
        size_t, len
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, userid, userid)
        lttng_ust_field_integer(size_t, len, len)
    )
)

/* Log level assignment */
LTTNG_UST_TRACEPOINT_LOGLEVEL(my_app, get_transaction,
                              LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_UNIT)
Create a tracepoint provider package source file

A tracepoint provider package source file is a C source file which includes a tracepoint provider header file to expand its macros into event serialization and other functions.

Use the following tracepoint provider package source file template:

Tracepoint provider package source file template.

#define LTTNG_UST_TRACEPOINT_CREATE_PROBES

#include "tp.h"

Replace tp.h with the name of your tracepoint provider header file name. You may also include more than one tracepoint provider header file here to create a tracepoint provider package holding more than one tracepoint providers.

Add tracepoints to the source code of an application

Once you create a tracepoint provider header file, use the lttng_ust_tracepoint() macro in the source code of your application to insert the tracepoints that this header defines.

The lttng_ust_tracepoint() macro takes at least two parameters: the tracepoint provider name and the tracepoint name. The corresponding tracepoint definition defines the other parameters.

Example:lttng_ust_tracepoint() usage.

The following tracepoint definition defines a tracepoint which takes two input arguments and has two output event fields.

Tracepoint provider header file.

#include "my-custom-structure.h"

LTTNG_UST_TRACEPOINT_EVENT(
    my_provider,
    my_tracepoint,
    LTTNG_UST_TP_ARGS(
        int, argc,
        const char *, cmd_name
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_string(cmd_name, cmd_name)
        lttng_ust_field_integer(int, number_of_args, argc)
    )
)

Refer to this tracepoint definition with the lttng_ust_tracepoint() macro in the source code of your application like this:

Application source file.

#include "tp.h"

int main(int argc, char* argv[])
{
    lttng_ust_tracepoint(my_provider, my_tracepoint, argc, argv[0]);
    return 0;
}

Note how the source code of the application includes the tracepoint provider header file containing the tracepoint definitions to use, tp.h.

Example:lttng_ust_tracepoint() usage with a complex tracepoint definition.

Consider this complex tracepoint definition, where multiple event fields refer to the same input arguments in their argument expression parameter:

Tracepoint provider header file.

/* For `struct stat` */
#include <sys/types.h>
#include <sys/stat.h>
#include <unistd.h>

LTTNG_UST_TRACEPOINT_EVENT(
    my_provider,
    my_tracepoint,
    LTTNG_UST_TP_ARGS(
        int, my_int_arg,
        char *, my_str_arg,
        struct stat *, st
    ),
    LTTNG_UST_TP_FIELDS(
        lttng_ust_field_integer(int, my_constant_field, 23 + 17)
        lttng_ust_field_integer(int, my_int_arg_field, my_int_arg)
        lttng_ust_field_integer(int, my_int_arg_field2,
                                my_int_arg * my_int_arg)
        lttng_ust_field_integer(int, sum4_field,
                                my_str_arg[0] + my_str_arg[1] +
                                my_str_arg[2] + my_str_arg[3])
        lttng_ust_field_string(my_str_arg_field, my_str_arg)
        lttng_ust_field_integer_hex(off_t, size_field, st->st_size)
        lttng_ust_field_float(double, size_dbl_field, (double) st->st_size)
        lttng_ust_field_sequence_text(char, half_my_str_arg_field,
                                      my_str_arg, size_t,
                                      strlen(my_str_arg) / 2)
    )
)

Refer to this tracepoint definition with the lttng_ust_tracepoint() macro in the source code of your application like this:

Application source file.

#define LTTNG_UST_TRACEPOINT_DEFINE
#include "tp.h"

int main(void)
{
    struct stat s;

    stat("/etc/fstab", &s);
    lttng_ust_tracepoint(my_provider, my_tracepoint, 23,
                         "Hello, World!", &s);

    return 0;
}

If you look at the event record that LTTng writes when recording this program, assuming the file size of /etc/fstab is 301 bytes, it should look like this:

Event record fields

Field name Field value

my_constant_field

40

my_int_arg_field

23

my_int_arg_field2

529

sum4_field

389

my_str_arg_field

Hello, World!

size_field

0x12d

size_dbl_field

301.0

half_my_str_arg_field

Hello,

Sometimes, the arguments you pass to lttng_ust_tracepoint() are expensive to evaluate—they use the call stack, for example. To avoid this computation when LTTng wouldn’t emit any event anyway, use the lttng_ust_tracepoint_enabled() and lttng_ust_do_tracepoint() macros.

The syntax of the lttng_ust_tracepoint_enabled() and lttng_ust_do_tracepoint() macros is:

lttng_ust_tracepoint_enabled() and lttng_ust_do_tracepoint() macros syntax.

lttng_ust_tracepoint_enabled(provider_name, tracepoint_name)

lttng_ust_do_tracepoint(provider_name, tracepoint_name, ...)

Replace:

  • provider_name with the tracepoint provider name.

  • tracepoint_name with the tracepoint name.

lttng_ust_tracepoint_enabled() returns a non-zero value if executing the tracepoint named tracepoint_name from the provider named provider_name could make LTTng emit an event, depending on the payload of said event.

lttng_ust_do_tracepoint() is like lttng_ust_tracepoint(), except that it doesn’t check what lttng_ust_tracepoint_enabled() checks. Using lttng_ust_tracepoint() with lttng_ust_tracepoint_enabled() is dangerous because lttng_ust_tracepoint() also contains the lttng_ust_tracepoint_enabled() check; therefore, a race condition is possible in this situation:

Possible race condition when using lttng_ust_tracepoint_enabled() with lttng_ust_tracepoint().

if (lttng_ust_tracepoint_enabled(my_provider, my_tracepoint)) {
    stuff = prepare_stuff();
}

lttng_ust_tracepoint(my_provider, my_tracepoint, stuff);

If lttng_ust_tracepoint_enabled() is false, but would be true after the conditional block, then stuff isn’t prepared: the emitted event will either contain wrong data, or the whole application could crash (with a segmentation fault, for example).

Note:Neither lttng_ust_tracepoint_enabled() nor lttng_ust_do_tracepoint() have an STAP_PROBEV() call. If you need it, you must emit this call yourself.

Build and link a tracepoint provider package and an application

Once you have one or more tracepoint provider header files and a tracepoint provider package source file, create the tracepoint provider package by compiling its source file. From here, multiple build and run scenarios are possible. The following table shows common application and library configurations along with the required command lines to achieve them.

In the following diagrams, we use the following file names:

app

Executable application.

app.o

Application object file.

tpp.o

Tracepoint provider package object file.

tpp.a

Tracepoint provider package archive file.

libtpp.so

Tracepoint provider package shared object file.

emon.o

User library object file.

libemon.so

User library shared object file.

We use the following symbols in the diagrams of table below:

Symbols used in the build scenario diagrams.

We assume that . is part of the LD_LIBRARY_PATH environment variable in the following instructions.

Common tracepoint provider package scenarios.

Scenario Instructions

The instrumented application is statically linked with the tracepoint provider package object.

To build the tracepoint provider package object file:

To build the instrumented application:

  1. In app.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
    gcc -o app app.o tpp.o -llttng-ust -ldl

To run the instrumented application:

  • Start the application:

    $
    ./app

The instrumented application is statically linked with the tracepoint provider package archive file.

To create the tracepoint provider package archive file:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -c tpp.c
  2. Create the tracepoint provider package archive file:

    $
    ar rcs tpp.a tpp.o

To build the instrumented application:

  1. In app.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
    gcc -o app app.o tpp.a -llttng-ust -ldl

To run the instrumented application:

  • Start the application:

    $
    ./app

The instrumented application is linked with the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented application:

  1. In app.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
    gcc -o app app.o -ldl -L. -ltpp

To run the instrumented application:

  • Start the application:

    $
    ./app

The tracepoint provider package shared object is preloaded before the instrumented application starts.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented application:

  1. In app.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
    gcc -o app app.o -ldl

To run the instrumented application with tracing support:

  • Preload the tracepoint provider package shared object and start the application:

    $
    LD_PRELOAD=./libtpp.so ./app

To run the instrumented application without tracing support:

  • Start the application:

    $
    ./app

The instrumented application dynamically loads the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented application:

  1. In app.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
    gcc -o app app.o -ldl

To run the instrumented application:

  • Start the application:

    $
    ./app

The application is linked with the instrumented user library.

The instrumented user library is statically linked with the tracepoint provider package object file.

To build the tracepoint provider package object file:

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o tpp.o -llttng-ust -ldl

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -L. -lemon

To run the application:

  • Start the application:

    $
    ./app

The application is linked with the instrumented user library.

The instrumented user library is linked with the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl -L. -ltpp

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -L. -lemon

To run the application:

  • Start the application:

    $
    ./app

The tracepoint provider package shared object is preloaded before the application starts.

The application is linked with the instrumented user library.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -L. -lemon

To run the application with tracing support:

  • Preload the tracepoint provider package shared object and start the application:

    $
    LD_PRELOAD=./libtpp.so ./app

To run the application without tracing support:

  • Start the application:

    $
    ./app

The application is linked with the instrumented user library.

The instrumented user library dynamically loads the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -L. -lemon

To run the application:

  • Start the application:

    $
    ./app

The application dynamically loads the instrumented user library.

The instrumented user library is linked with the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl -L. -ltpp

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -ldl -L. -lemon

To run the application:

  • Start the application:

    $
    ./app

The application dynamically loads the instrumented user library.

The instrumented user library dynamically loads the tracepoint provider package shared object.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -ldl -L. -lemon

To run the application:

  • Start the application:

    $
    ./app

The tracepoint provider package shared object is preloaded before the application starts.

The application dynamically loads the instrumented user library.

To build the tracepoint provider package shared object:

  1. Compile the tracepoint provider package source file:

    $
    gcc -I. -fpic -c tpp.c
  2. Build the tracepoint provider package shared object:

    $
    gcc -shared -o libtpp.so tpp.o -llttng-ust -ldl

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following lines:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    #define LTTNG_UST_TRACEPOINT_PROBE_DYNAMIC_LINKAGE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o -ldl

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o -L. -lemon

To run the application with tracing support:

  • Preload the tracepoint provider package shared object and start the application:

    $
    LD_PRELOAD=./libtpp.so ./app

To run the application without tracing support:

  • Start the application:

    $
    ./app

The application is statically linked with the tracepoint provider package object file.

The application is linked with the instrumented user library.

To build the tracepoint provider package object file:

To build the instrumented user library:

  1. In emon.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  3. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o

To build the application:

  1. Compile the application source file:

    $
    gcc -c app.c
  2. Build the application:

    $
    gcc -o app app.o tpp.o -llttng-ust -ldl -L. -lemon

To run the instrumented application:

  • Start the application:

    $
    ./app

The application is statically linked with the tracepoint provider package object file.

The application dynamically loads the instrumented user library.

To build the tracepoint provider package object file:

To build the application:

  1. In app.c, before including tpp.h, add the following line:

    #define LTTNG_UST_TRACEPOINT_DEFINE
    
  2. Compile the application source file:

    $
    gcc -c app.c
  3. Build the application:

    $
     
    gcc -Wl,--export-dynamic -o app app.o tpp.o \
        -llttng-ust -ldl

    The --export-dynamic option passed to the linker is necessary for the dynamically loaded library to “see” the tracepoint symbols defined in the application.

To build the instrumented user library:

  1. Compile the user library source file:

    $
    gcc -I. -fpic -c emon.c
  2. Build the user library shared object:

    $
    gcc -shared -o libemon.so emon.o

To run the application:

  • Start the application:

    $
    ./app
Use LTTng-UST with daemons

If your instrumented application calls fork(2), clone(2), or BSD’s rfork(2), without a following exec(3)-family system call, you must preload the liblttng-ust-fork.so shared object when you start the application.

$
LD_PRELOAD=liblttng-ust-fork.so ./my-app

If your tracepoint provider package is a shared library which you also preload, you must put both shared objects in LD_PRELOAD:

$
LD_PRELOAD=liblttng-ust-fork.so:/path/to/tp.so ./my-app
Use LTTng-UST with applications which close file descriptors that don’t belong to them
Since 2.9

If your instrumented application closes one or more file descriptors which it did not open itself, you must preload the liblttng-ust-fd.so shared object when you start the application:

$
LD_PRELOAD=liblttng-ust-fd.so ./my-app

Typical use cases include closing all the file descriptors after fork(2) or rfork(2) and buggy applications doing “double closes”.

Use pkg-config

On some distributions, LTTng-UST ships with a pkg-config metadata file. If this is your case, then use pkg-config to build an application on the command line:

$
gcc -o my-app my-app.o tp.o $(pkg-config --cflags --libs lttng-ust)
Build a 32-bit instrumented application for a 64-bit target system

In order to trace a 32-bit application running on a 64-bit system, LTTng must use a dedicated 32-bit consumer daemon.

The following steps show how to build and install a 32-bit consumer daemon, which is not part of the default 64-bit LTTng build, how to build and install the 32-bit LTTng-UST libraries, and how to build and link an instrumented 32-bit application in that context.

To build a 32-bit instrumented application for a 64-bit target system, assuming you have a fresh target system with no installed Userspace RCU or LTTng packages:

  1. Download, build, and install a 32-bit version of Userspace RCU:

    $
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/urcu/userspace-rcu-latest-0.13.tar.bz2 &&
    tar -xf userspace-rcu-latest-0.13.tar.bz2 &&
    cd userspace-rcu-0.13.* &&
    ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 &&
    make &&
    sudo make install &&
    sudo ldconfig
  2. Using the package manager of your distribution, or from source, install the 32-bit versions of the following dependencies of LTTng-tools and LTTng-UST:

  3. Download, build, and install a 32-bit version of the latest LTTng-UST 2.13:

    $
     
     
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-ust/lttng-ust-latest-2.13.tar.bz2 &&
    tar -xf lttng-ust-latest-2.13.tar.bz2 &&
    cd lttng-ust-2.13.* &&
    ./configure --libdir=/usr/local/lib32 \
                CFLAGS=-m32 CXXFLAGS=-m32 \
                LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' &&
    make &&
    sudo make install &&
    sudo ldconfig

    Add --disable-numa to ./configure if you don’t have numactl.

    Note:Depending on your distribution, 32-bit libraries could be installed at a different location than /usr/lib32. For example, Debian is known to install some 32-bit libraries in /usr/lib/i386-linux-gnu.

    In this case, make sure to set LDFLAGS to all the relevant 32-bit library paths, for example:

    $
    LDFLAGS='-L/usr/lib/i386-linux-gnu -L/usr/lib32'
  4. Download the latest LTTng-tools 2.13, build, and install the 32-bit consumer daemon:

    $
     
     
     
     
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
    tar -xf lttng-tools-latest-2.13.tar.bz2 &&
    cd lttng-tools-2.13.* &&
    ./configure --libdir=/usr/local/lib32 CFLAGS=-m32 CXXFLAGS=-m32 \
                LDFLAGS='-L/usr/local/lib32 -L/usr/lib32' \
                --disable-bin-lttng --disable-bin-lttng-crash \
                --disable-bin-lttng-relayd --disable-bin-lttng-sessiond &&
    make &&
    cd src/bin/lttng-consumerd &&
    sudo make install &&
    sudo ldconfig
  5. From your distribution or from source, install the 64-bit versions of LTTng-UST and Userspace RCU.

  6. Download, build, and install the 64-bit version of the latest LTTng-tools 2.13:

    $
     
     
     
     
     
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-tools/lttng-tools-latest-2.13.tar.bz2 &&
    tar -xf lttng-tools-latest-2.13.tar.bz2 &&
    cd lttng-tools-2.13.* &&
    ./configure --with-consumerd32-libdir=/usr/local/lib32 \
                --with-consumerd32-bin=/usr/local/lib32/lttng/libexec/lttng-consumerd &&
    make &&
    sudo make install &&
    sudo ldconfig
  7. Pass the following options to gcc(1), g++(1), or clang(1) when linking your 32-bit application:

    -m32 -L/usr/lib32 -L/usr/local/lib32 \
    -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32

    For example, let’s rebuild the quick start example in “Record user application events” as an instrumented 32-bit application:

    $
    $
    $
     
     
     
    gcc -m32 -c -I. hello-tp.c
    gcc -m32 -c hello.c
    gcc -m32 -o hello hello.o hello-tp.o \
        -L/usr/lib32 -L/usr/local/lib32 \
        -Wl,-rpath,/usr/lib32,-rpath,/usr/local/lib32 \
        -llttng-ust -ldl

No special action is required to execute the 32-bit application and for LTTng to trace it: use the command-line lttng(1) tool as usual.

Use lttng_ust_tracef()

Since 2.5

lttng_ust_tracef(3) is a small LTTng-UST API designed for quick, printf(3)-like instrumentation without the burden of creating and building a tracepoint provider package.

To use lttng_ust_tracef() in your application:

  1. In the C or C++ source files where you need to use lttng_ust_tracef(), include <lttng/tracef.h>:

    #include <lttng/tracef.h>
    
  2. In the source code of the application, use lttng_ust_tracef() like you would use printf(3):

        /* ... */
    
        lttng_ust_tracef("my message: %d (%s)", my_integer, my_string);
    
        /* ... */
    
  3. Link your application with liblttng-ust:

    $
    gcc -o app app.c -llttng-ust

To record the events that lttng_ust_tracef() calls emit:

Limitations of lttng_ust_tracef()

Important:The lttng_ust_tracef() utility function was developed to make user space tracing super simple, albeit with notable disadvantages compared to user-defined tracepoints:

  • All the created events have the same tracepoint provider and tracepoint names, respectively lttng_ust_tracef and event.

  • There’s no static type checking.

  • The only event record field you actually get, named msg, is a string potentially containing the values you passed to lttng_ust_tracef() using your own format string. This also means that you can’t filter events with a custom expression at run time because there are no isolated fields.

  • Since lttng_ust_tracef() uses the vasprintf(3) function of the C standard library behind the scenes to format the strings at run time, its expected performance is lower than with user-defined tracepoints, which don’t require a conversion to a string.

Taking this into consideration, lttng_ust_tracef() is useful for some quick prototyping and debugging, but you shouldn’t consider it for any permanent and serious applicative instrumentation.

Use lttng_ust_tracelog()

Since 2.7

The tracelog(3) API is very similar to lttng_ust_tracef(), with the difference that it accepts an additional log level parameter.

The goal of lttng_ust_tracelog() is to ease the migration from logging to tracing.

To use lttng_ust_tracelog() in your application:

  1. In the C or C++ source files where you need to use tracelog(), include <lttng/tracelog.h>:

    #include <lttng/tracelog.h>
    
  2. In the source code of the application, use lttng_ust_tracelog() like you would use printf(3), except for the first parameter which is the log level:

        /* ... */
    
        tracelog(LTTNG_UST_TRACEPOINT_LOGLEVEL_WARNING,
                "my message: %d (%s)", my_integer, my_string);
    
        /* ... */
    

    See lttng-ust(3) for a list of available log level names.

  3. Link your application with liblttng-ust:

    $
    gcc -o app app.c -llttng-ust

To record the events that lttng_ust_tracelog() calls emit with a log level at least as severe as a specific log level:

  • Create a recording event rule which matches user space tracepoint events named lttng_ust_tracelog:* and with some minimum level of severity:

    $
     
    lttng enable-event --userspace 'lttng_ust_tracelog:*' \
                       --loglevel=WARNING

To record the events that lttng_ust_tracelog() calls emit with a specific log level:

  • Create a recording event rule which matches tracepoint events named lttng_ust_tracelog:* and with a specific log level:

    $
     
    lttng enable-event --userspace 'lttng_ust_tracelog:*' \
                       --loglevel-only=INFO

Load a prebuilt user space tracing helper

The LTTng-UST package provides a few helpers in the form of preloadable shared objects which automatically instrument system functions and calls.

The helper shared objects are normally found in /usr/lib. If you built LTTng-UST from source, they’re probably located in /usr/local/lib.

The installed user space tracing helpers in LTTng-UST 2.13 are:

liblttng-ust-libc-wrapper.so
liblttng-ust-pthread-wrapper.so

C standard library memory and POSIX threads function tracing.

liblttng-ust-cyg-profile.so
liblttng-ust-cyg-profile-fast.so

Function entry and exit tracing.

liblttng-ust-dl.so

Dynamic linker tracing.

To use a user space tracing helper with any user application:

  • Preload the helper shared object when you start the application:

    $
    LD_PRELOAD=liblttng-ust-libc-wrapper.so my-app

    You can preload more than one helper:

    $
    LD_PRELOAD=liblttng-ust-libc-wrapper.so:liblttng-ust-dl.so my-app

Instrument C standard library memory and POSIX threads functions

Since 2.3

The liblttng-ust-libc-wrapper.so and liblttng-ust-pthread-wrapper.so helpers add instrumentation to some C standard library and POSIX threads functions.

Functions instrumented by preloading liblttng-ust-libc-wrapper.so.

TP provider name TP name Instrumented function

lttng_ust_libc

malloc

malloc(3)

calloc

calloc(3)

realloc

realloc(3)

free

free(3)

memalign

memalign(3)

posix_memalign

posix_memalign(3)

Functions instrumented by preloading liblttng-ust-pthread-wrapper.so.

TP provider name TP name Instrumented function

lttng_ust_pthread

pthread_mutex_lock_req

pthread_mutex_lock(3p) (request time)

pthread_mutex_lock_acq

pthread_mutex_lock(3p) (acquire time)

pthread_mutex_trylock

pthread_mutex_trylock(3p)

pthread_mutex_unlock

pthread_mutex_unlock(3p)

When you preload the shared object, it replaces the functions listed in the previous tables by wrappers which contain tracepoints and call the replaced functions.

Instrument function entry and exit

The liblttng-ust-cyg-profile*.so helpers can add instrumentation to the entry and exit points of functions.

gcc(1) and clang(1) have an option named -finstrument-functions which generates instrumentation calls for entry and exit to functions. The LTTng-UST function tracing helpers, liblttng-ust-cyg-profile.so and liblttng-ust-cyg-profile-fast.so, take advantage of this feature to add tracepoints to the two generated functions (which contain cyg_profile in their names, hence the name of the helper).

To use the LTTng-UST function tracing helper, the source files to instrument must be built using the -finstrument-functions compiler flag.

There are two versions of the LTTng-UST function tracing helper:

  • liblttng-ust-cyg-profile-fast.so is a lightweight variant that you should only use when it can be guaranteed that the complete event stream is recorded without any lost event record. Any kind of duplicate information is left out.

    Assuming no event record is lost, having only the function addresses on entry is enough to create a call graph, since an event record always contains the ID of the CPU that generated it.

    Use a tool like addr2line(1) to convert function addresses back to source file names and line numbers.

  • liblttng-ust-cyg-profile.so is a more robust variant which also works in use cases where event records might get discarded or not recorded from application startup. In these cases, the trace analyzer needs more information to be able to reconstruct the program flow.

See lttng-ust-cyg-profile(3) to learn more about the instrumentation points of this helper.

All the tracepoints that this helper provides have the log level LTTNG_UST_TRACEPOINT_LOGLEVEL_DEBUG_FUNCTION (see lttng-ust(3)).

Tip:It’s sometimes a good idea to limit the number of source files that you compile with the -finstrument-functions option to prevent LTTng from writing an excessive amount of trace data at run time. When using gcc(1), use the -finstrument-functions-exclude-function-list option to avoid instrument entries and exits of specific function names.

Instrument the dynamic linker

Since 2.4

The liblttng-ust-dl.so helper adds instrumentation to the dlopen(3) and dlclose(3) function calls.

See lttng-ust-dl(3) to learn more about the instrumentation points of this helper.

Instrument a Java application

Since 2.4

You can instrument any Java application which uses one of the following logging frameworks:

LTTng-UST Java agent imported by a Java application.

Note that the methods described below are new in LTTng 2.8. Previous LTTng versions use another technique.

Note:We use OpenJDK 8 for development and continuous integration, thus this version is directly supported. However, the LTTng-UST Java agent is also tested with OpenJDK 7.

Use the LTTng-UST Java agent for java.util.logging

Since 2.8

To use the LTTng-UST Java agent in a Java application which uses java.util.logging (JUL):

  1. In the source code of the Java application, import the LTTng-UST log handler package for java.util.logging:

    import org.lttng.ust.agent.jul.LttngLogHandler;
    
  2. Create an LTTng-UST java.util.logging log handler:

    Handler lttngUstLogHandler = new LttngLogHandler();
    
  3. Add this handler to the java.util.logging loggers which should emit LTTng events:

    Logger myLogger = Logger.getLogger("some-logger");
    
    myLogger.addHandler(lttngUstLogHandler);
    
  4. Use java.util.logging log statements and configuration as usual. The loggers with an attached LTTng-UST log handler can emit LTTng events.

  5. Before exiting the application, remove the LTTng-UST log handler from the loggers attached to it and call its close() method:

    myLogger.removeHandler(lttngUstLogHandler);
    lttngUstLogHandler.close();
    

    This isn’t strictly necessary, but it’s recommended for a clean disposal of the resources of the handler.

  6. Include the common and JUL-specific JAR files of the LTTng-UST Java agent, lttng-ust-agent-common.jar and lttng-ust-agent-jul.jar, in the class path when you build the Java application.

    The JAR files are typically located in /usr/share/java.

    Important:The LTTng-UST Java agent must be installed for the logging framework your application uses.

Example:Use the LTTng-UST Java agent for java.util.logging.

Test.java

import java.io.IOException;
import java.util.logging.Handler;
import java.util.logging.Logger;
import org.lttng.ust.agent.jul.LttngLogHandler;

public class Test
{
    private static final int answer = 42;

    public static void main(String[] argv) throws Exception
    {
        // Create a logger
        Logger logger = Logger.getLogger("jello");

        // Create an LTTng-UST log handler
        Handler lttngUstLogHandler = new LttngLogHandler();

        // Add the LTTng-UST log handler to our logger
        logger.addHandler(lttngUstLogHandler);

        // Log at will!
        logger.info("some info");
        logger.warning("some warning");
        Thread.sleep(500);
        logger.finer("finer information; the answer is " + answer);
        Thread.sleep(123);
        logger.severe("error!");

        // Not mandatory, but cleaner
        logger.removeHandler(lttngUstLogHandler);
        lttngUstLogHandler.close();
    }
}

Build this example:

$
javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java

Create a recording session, create a recording event rule matching JUL events named jello, and start recording:

$
$
$
lttng create
lttng enable-event --jul jello
lttng start

Run the compiled class:

$
java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test

Stop recording and inspect the recorded events:

$
$
lttng stop
lttng view

In the resulting trace, an event record which a Java application using java.util.logging generated is named lttng_jul:event and has the following fields:

msg

Log record message.

logger_name

Logger name.

class_name

Name of the class in which the log statement was executed.

method_name

Name of the method in which the log statement was executed.

long_millis

Logging time (timestamp in milliseconds).

int_loglevel

Log level integer value.

int_threadid

ID of the thread in which the log statement was executed.

Use the --loglevel or --loglevel-only option of the lttng-enable-event(1) command to target a range of java.util.logging log levels or a specific java.util.logging log level.

Use the LTTng-UST Java agent for Apache log4j

Since 2.8

To use the LTTng-UST Java agent in a Java application which uses Apache log4j 1.2:

  1. In the source code of the Java application, import the LTTng-UST log appender package for Apache log4j:

    import org.lttng.ust.agent.log4j.LttngLogAppender;
    
  2. Create an LTTng-UST log4j log appender:

    Appender lttngUstLogAppender = new LttngLogAppender();
    
  3. Add this appender to the log4j loggers which should emit LTTng events:

    Logger myLogger = Logger.getLogger("some-logger");
    
    myLogger.addAppender(lttngUstLogAppender);
    
  4. Use Apache log4j log statements and configuration as usual. The loggers with an attached LTTng-UST log appender can emit LTTng events.

  5. Before exiting the application, remove the LTTng-UST log appender from the loggers attached to it and call its close() method:

    myLogger.removeAppender(lttngUstLogAppender);
    lttngUstLogAppender.close();
    

    This isn’t strictly necessary, but it’s recommended for a clean disposal of the resources of the appender.

  6. Include the common and log4j-specific JAR files of the LTTng-UST Java agent, lttng-ust-agent-common.jar and lttng-ust-agent-log4j.jar, in the class path when you build the Java application.

    The JAR files are typically located in /usr/share/java.

    Important:The LTTng-UST Java agent must be installed for the logging framework your application uses.

Example:Use the LTTng-UST Java agent for Apache log4j.

Test.java

import org.apache.log4j.Appender;
import org.apache.log4j.Logger;
import org.lttng.ust.agent.log4j.LttngLogAppender;

public class Test
{
    private static final int answer = 42;

    public static void main(String[] argv) throws Exception
    {
        // Create a logger
        Logger logger = Logger.getLogger("jello");

        // Create an LTTng-UST log appender
        Appender lttngUstLogAppender = new LttngLogAppender();

        // Add the LTTng-UST log appender to our logger
        logger.addAppender(lttngUstLogAppender);

        // Log at will!
        logger.info("some info");
        logger.warn("some warning");
        Thread.sleep(500);
        logger.debug("debug information; the answer is " + answer);
        Thread.sleep(123);
        logger.fatal("error!");

        // Not mandatory, but cleaner
        logger.removeAppender(lttngUstLogAppender);
        lttngUstLogAppender.close();
    }
}

Build this example ($LOG4JPATH is the path to the Apache log4j JAR file):

$
javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH Test.java

Create a recording session, create a recording event rule matching log4j events named jello, and start recording:

$
$
$
lttng create
lttng enable-event --log4j jello
lttng start

Run the compiled class:

$
java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-log4j.jar:$LOG4JPATH:. Test

Stop recording and inspect the recorded events:

$
$
lttng stop
lttng view

In the resulting trace, an event record which a Java application using log4j generated is named lttng_log4j:event and has the following fields:

msg

Log record message.

logger_name

Logger name.

class_name

Name of the class in which the log statement was executed.

method_name

Name of the method in which the log statement was executed.

filename

Name of the file in which the executed log statement is located.

line_number

Line number at which the log statement was executed.

timestamp

Logging timestamp.

int_loglevel

Log level integer value.

thread_name

Name of the Java thread in which the log statement was executed.

Use the --loglevel or --loglevel-only option of the lttng-enable-event(1) command to target a range of Apache log4j log levels or a specific log4j log level.

Provide application-specific context fields in a Java application

Since 2.8

A Java application-specific context field is a piece of state which the Java application provides. You can add such a context field to be recorded, using the lttng-add-context(1) command, to each event record which the log statements of this application produce.

For example, a given object might have a current request ID variable. You can create a context information retriever for this object and assign a name to this current request ID. You can then, using the lttng-add-context(1) command, add this context field by name so that LTTng writes it to the event records of a given java.util.logging or log4j channel.

To provide application-specific context fields in a Java application:

  1. In the source code of the Java application, import the LTTng-UST Java agent context classes and interfaces:

    import org.lttng.ust.agent.context.ContextInfoManager;
    import org.lttng.ust.agent.context.IContextInfoRetriever;
    
  2. Create a context information retriever class, that is, a class which implements the IContextInfoRetriever interface:

    class MyContextInfoRetriever implements IContextInfoRetriever
    {
        @Override
        public Object retrieveContextInfo(String key)
        {
            if (key.equals("intCtx")) {
                return (short) 17;
            } else if (key.equals("strContext")) {
                return "context value!";
            } else {
                return null;
            }
        }
    }
    

    This retrieveContextInfo() method is the only member of the IContextInfoRetriever interface. Its role is to return the current value of a state by name to create a context field. The names of the context fields and which state variables they return depends on your specific scenario.

    All primitive types and objects are supported as context fields. When retrieveContextInfo() returns an object, the context field serializer calls its toString() method to add a string field to event records. The method can also return null, which means that no context field is available for the required name.

  3. Register an instance of your context information retriever class to the context information manager singleton:

    IContextInfoRetriever cir = new MyContextInfoRetriever();
    ContextInfoManager cim = ContextInfoManager.getInstance();
    cim.registerContextInfoRetriever("retrieverName", cir);
    
  4. Before exiting the application, remove your context information retriever from the context information manager singleton:

    ContextInfoManager cim = ContextInfoManager.getInstance();
    cim.unregisterContextInfoRetriever("retrieverName");
    

    This isn’t strictly necessary, but it’s recommended for a clean disposal of some resources of the manager.

  5. Build your Java application with LTTng-UST Java agent support as usual, following the procedure for either the java.util.logging or Apache log4j framework.

Example:Provide application-specific context fields in a Java application.

Test.java

import java.util.logging.Handler;
import java.util.logging.Logger;
import org.lttng.ust.agent.jul.LttngLogHandler;
import org.lttng.ust.agent.context.ContextInfoManager;
import org.lttng.ust.agent.context.IContextInfoRetriever;

public class Test
{
    // Our context information retriever class
    private static class MyContextInfoRetriever
    implements IContextInfoRetriever
    {
        @Override
        public Object retrieveContextInfo(String key) {
            if (key.equals("intCtx")) {
                return (short) 17;
            } else if (key.equals("strContext")) {
                return "context value!";
            } else {
                return null;
            }
        }
    }

    private static final int answer = 42;

    public static void main(String args[]) throws Exception
    {
        // Get the context information manager instance
        ContextInfoManager cim = ContextInfoManager.getInstance();

        // Create and register our context information retriever
        IContextInfoRetriever cir = new MyContextInfoRetriever();
        cim.registerContextInfoRetriever("myRetriever", cir);

        // Create a logger
        Logger logger = Logger.getLogger("jello");

        // Create an LTTng-UST log handler
        Handler lttngUstLogHandler = new LttngLogHandler();

        // Add the LTTng-UST log handler to our logger
        logger.addHandler(lttngUstLogHandler);

        // Log at will!
        logger.info("some info");
        logger.warning("some warning");
        Thread.sleep(500);
        logger.finer("finer information; the answer is " + answer);
        Thread.sleep(123);
        logger.severe("error!");

        // Not mandatory, but cleaner
        logger.removeHandler(lttngUstLogHandler);
        lttngUstLogHandler.close();
        cim.unregisterContextInfoRetriever("myRetriever");
    }
}

Build this example:

$
javac -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar Test.java

Create a recording session and create a recording event rule matching java.util.logging events named jello:

$
$
lttng create
lttng enable-event --jul jello

Add the application-specific context fields to be recorded to the event records of the java.util.logging channel:

$
$
lttng add-context --jul --type='$app.myRetriever:intCtx'
lttng add-context --jul --type='$app.myRetriever:strContext'

Start recording:

$
lttng start

Run the compiled class:

$
java -cp /usr/share/java/jarpath/lttng-ust-agent-common.jar:/usr/share/java/jarpath/lttng-ust-agent-jul.jar:. Test

Stop recording and inspect the recorded events:

$
$
lttng stop
lttng view

Instrument a Python application

Since 2.7

You can instrument a Python 2 or Python 3 application which uses the standard logging package.

Each log statement creates an LTTng event once the application module imports the LTTng-UST Python agent package.

A Python application importing the LTTng-UST Python agent.

To use the LTTng-UST Python agent:

  1. In the source code of the Python application, import the LTTng-UST Python agent:

    import lttngust
    

    The LTTng-UST Python agent automatically adds its logging handler to the root logger at import time.

    A log statement that the application executes before this import doesn’t create an LTTng event.

    Important:The LTTng-UST Python agent must be installed.

  2. Use log statements and logging configuration as usual. Since the LTTng-UST Python agent adds a handler to the root logger, any log statement from any logger can emit an LTTng event.

Example:Use the LTTng-UST Python agent.

test.py

import lttngust
import logging
import time


def example():
    logging.basicConfig()
    logger = logging.getLogger('my-logger')

    while True:
        logger.debug('debug message')
        logger.info('info message')
        logger.warn('warn message')
        logger.error('error message')
        logger.critical('critical message')
        time.sleep(1)


if __name__ == '__main__':
    example()

Note:logging.basicConfig(), which adds to the root logger a basic logging handler which prints to the standard error stream, isn’t strictly required for LTTng-UST tracing to work, but in versions of Python preceding 3.2, you could see a warning message which indicates that no handler exists for the logger my-logger.

Create a recording session, create a recording event rule matching Python logging events named my-logger, and start recording:

$
$
$
lttng create
lttng enable-event --python my-logger
lttng start

Run the Python script:

$
python test.py

Stop recording and inspect the recorded events:

$
$
lttng stop
lttng view

In the resulting trace, an event record which a Python application generated is named lttng_python:event and has the following fields:

asctime

Logging time (string).

msg

Log record message.

logger_name

Logger name.

funcName

Name of the function in which the log statement was executed.

lineno

Line number at which the log statement was executed.

int_loglevel

Log level integer value.

thread

ID of the Python thread in which the log statement was executed.

threadName

Name of the Python thread in which the log statement was executed.

Use the --loglevel or --loglevel-only option of the lttng-enable-event(1) command to target a range of Python log levels or a specific Python log level.

When an application imports the LTTng-UST Python agent, the agent tries to register to a session daemon. Note that you must start the session daemon before you run the Python application. If a session daemon is found, the agent tries to register to it during five seconds, after which the application continues without LTTng tracing support. Override this timeout value with the LTTNG_UST_PYTHON_REGISTER_TIMEOUT environment variable (milliseconds).

If the session daemon stops while a Python application with an imported LTTng-UST Python agent runs, the agent retries to connect and to register to a session daemon every three seconds. Override this delay with the LTTNG_UST_PYTHON_REGISTER_RETRY_DELAY environment variable.

Use the LTTng logger

Since 2.5

The lttng-tracer Linux kernel module, part of LTTng-modules, creates the special LTTng logger files /proc/lttng-logger and /dev/lttng-logger (since LTTng 2.11) when it’s loaded. Any application can write text data to any of those files to create one or more LTTng events.

An application writes to the LTTng logger file to create one or more LTTng events.

The LTTng logger is the quickest method—not the most efficient, however—to add instrumentation to an application. It’s designed mostly to instrument shell scripts:

$
echo "Some message, some $variable" > /dev/lttng-logger

Any event that the LTTng logger creates is named lttng_logger and belongs to the Linux kernel tracing domain. However, unlike other instrumentation points in the kernel tracing domain, any Unix user can create a recording event rule which matches events named lttng_logger, not only the root user or users in the tracing group.

To use the LTTng logger:

  • From any application, write text data to the /dev/lttng-logger file.

The msg field of lttng_logger event records contains the recorded message.

Note:The maximum message length of an LTTng logger event is 1024 bytes. Writing more than this makes the LTTng logger emit more than one event to contain the remaining data.

You shouldn’t use the LTTng logger to trace a user application which you can instrument in a more efficient way, namely:

Example:Use the LTTng logger.

test.bash

echo 'Hello, World!' > /dev/lttng-logger
sleep 2
df --human-readable --print-type / > /dev/lttng-logger

Create a recording session, create a recording event rule matching Linux kernel tracepoint events named lttng_logger, and start recording:

$
$
$
lttng create
lttng enable-event --kernel lttng_logger
lttng start

Run the Bash script:

$
bash test.bash

Stop recording and inspect the recorded events:

$
$
lttng stop
lttng view

Instrument a Linux kernel image or module

Note:This section shows how to add instrumentation points to the Linux kernel. The subsystems of the kernel are already thoroughly instrumented at strategic points for LTTng when you install the LTTng-modules package.

Add an LTTng layer to an existing ftrace tracepoint

This section shows how to add an LTTng layer to existing ftrace instrumentation using the TRACE_EVENT() API.

This section doesn’t document the TRACE_EVENT() macro. Read the following articles to learn more about this API:

The following procedure assumes that your ftrace tracepoints are correctly defined in their own header and that they’re created in one source file using the CREATE_TRACE_POINTS definition.

To add an LTTng layer over an existing ftrace tracepoint:

  1. Make sure the following kernel configuration options are enabled:

    • CONFIG_MODULES

    • CONFIG_KALLSYMS

    • CONFIG_HIGH_RES_TIMERS

    • CONFIG_TRACEPOINTS

  2. Build the Linux source tree with your custom ftrace tracepoints.

  3. Boot the resulting Linux image on your target system.

    Confirm that the tracepoints exist by looking for their names in the /sys/kernel/debug/tracing/events/subsys directory, where subsys is your subsystem name.

  4. Get a copy of the latest LTTng-modules 2.13:

    $
     
     
     
    cd $(mktemp -d) &&
    wget https://lttng.org/files/lttng-modules/lttng-modules-latest-2.13.tar.bz2 &&
    tar -xf lttng-modules-latest-2.13.tar.bz2 &&
    cd lttng-modules-2.13.*
  5. In instrumentation/events/lttng-module, relative to the root of the LTTng-modules source tree, create a header file named subsys.h for your custom subsystem subsys and write your LTTng-modules tracepoint definitions using the LTTng-modules macros in it.

    Start with this template:

    instrumentation/events/lttng-module/my_subsys.h

    #undef TRACE_SYSTEM
    #define TRACE_SYSTEM my_subsys
    
    #if !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ)
    #define _LTTNG_MY_SUBSYS_H
    
    #include "../../../probes/lttng-tracepoint-event.h"
    #include <linux/tracepoint.h>
    
    LTTNG_TRACEPOINT_EVENT(
        /*
         * Format is identical to the TRACE_EVENT() version for the three
         * following macro parameters:
         */
        my_subsys_my_event,
        TP_PROTO(int my_int, const char *my_string),
        TP_ARGS(my_int, my_string),
    
        /* LTTng-modules specific macros */
        TP_FIELDS(
            ctf_integer(int, my_int_field, my_int)
            ctf_string(my_bar_field, my_bar)
        )
    )
    
    #endif /* !defined(_LTTNG_MY_SUBSYS_H) || defined(TRACE_HEADER_MULTI_READ) */
    
    #include "../../../probes/define_trace.h"
    

    The entries in the TP_FIELDS() section are the list of fields for the LTTng tracepoint. This is similar to the TP_STRUCT__entry() part of the TRACE_EVENT() ftrace macro.

    See “Tracepoint fields macros” for a complete description of the available ctf_*() macros.

  6. Create the kernel module C source file of the LTTng-modules probe, probes/lttng-probe-subsys.c, where subsys is your subsystem name:

    probes/lttng-probe-my-subsys.c

    #include <linux/module.h>
    #include "../lttng-tracer.h"
    
    /*
     * Build-time verification of mismatch between mainline
     * TRACE_EVENT() arguments and the LTTng-modules adaptation
     * layer LTTNG_TRACEPOINT_EVENT() arguments.
     */
    #include <trace/events/my_subsys.h>
    
    /* Create LTTng tracepoint probes */
    #define LTTNG_PACKAGE_BUILD
    #define CREATE_TRACE_POINTS
    #define TRACE_INCLUDE_PATH ../instrumentation/events/lttng-module
    
    #include "../instrumentation/events/lttng-module/my_subsys.h"
    
    MODULE_LICENSE("GPL and additional rights");
    MODULE_AUTHOR("Your name <your-email>");
    MODULE_DESCRIPTION("LTTng my_subsys probes");
    MODULE_VERSION(__stringify(LTTNG_MODULES_MAJOR_VERSION) "."
        __stringify(LTTNG_MODULES_MINOR_VERSION) "."
        __stringify(LTTNG_MODULES_PATCHLEVEL_VERSION)
        LTTNG_MODULES_EXTRAVERSION);
    
  7. Edit probes/KBuild and add your new kernel module object next to the existing ones:

    probes/KBuild

    # ...
    
    obj-m += lttng-probe-module.o
    obj-m += lttng-probe-power.o
    
    obj-m += lttng-probe-my-subsys.o
    
    # ...
    
  8. Build and install the LTTng kernel modules:

    $
    #
    make KERNELDIR=/path/to/linux
    make modules_install && depmod -a

    Replace /path/to/linux with the path to the Linux source tree where you defined and used tracepoints with the TRACE_EVENT() ftrace macro.

Note that you can also use the LTTNG_TRACEPOINT_EVENT_CODE() macro instead of LTTNG_TRACEPOINT_EVENT() to use custom local variables and C code that need to be executed before LTTng records the event fields.

The best way to learn how to use the previous LTTng-modules macros is to inspect the existing LTTng-modules tracepoint definitions in the instrumentation/events/lttng-module header files. Compare them with the Linux kernel mainline versions in the include/trace/events directory of the Linux source tree.

Use custom C code to access the data for tracepoint fields
Since 2.7

Although we recommended to always use the LTTNG_TRACEPOINT_EVENT() macro to describe the arguments and fields of an LTTng-modules tracepoint when possible, sometimes you need a more complex process to access the data that the tracer records as event record fields. In other words, you need local variables and multiple C statements instead of simple argument-based expressions that you pass to the ctf_*() macros of TP_FIELDS().

Use the LTTNG_TRACEPOINT_EVENT_CODE() macro instead of LTTNG_TRACEPOINT_EVENT() to declare custom local variables and define a block of C code to be executed before LTTng records the fields. The structure of this macro is:

LTTNG_TRACEPOINT_EVENT_CODE() macro syntax.

LTTNG_TRACEPOINT_EVENT_CODE(
    /*
     * Format identical to the LTTNG_TRACEPOINT_EVENT()
     * version for the following three macro parameters:
     */
    my_subsys_my_event,
    TP_PROTO(int my_int, const char *my_string),
    TP_ARGS(my_int, my_string),

    /* Declarations of custom local variables */
    TP_locvar(
        int a = 0;
        unsigned long b = 0;
        const char *name = "(undefined)";
        struct my_struct *my_struct;
    ),

    /*
     * Custom code which uses both tracepoint arguments
     * (in TP_ARGS()) and local variables (in TP_locvar()).
     *
     * Local variables are actually members of a structure pointed
     * to by the special variable tp_locvar.
     */
    TP_code(
        if (my_int) {
            tp_locvar->a = my_int + 17;
            tp_locvar->my_struct = get_my_struct_at(tp_locvar->a);
            tp_locvar->b = my_struct_compute_b(tp_locvar->my_struct);
            tp_locvar->name = my_struct_get_name(tp_locvar->my_struct);
            put_my_struct(tp_locvar->my_struct);

            if (tp_locvar->b) {
                tp_locvar->a = 1;
            }
        }
    ),

    /*
     * Format identical to the LTTNG_TRACEPOINT_EVENT()
     * version for this, except that tp_locvar members can be
     * used in the argument expression parameters of
     * the ctf_*() macros.
     */
    TP_FIELDS(
        ctf_integer(unsigned long, my_struct_b, tp_locvar->b)
        ctf_integer(int, my_struct_a, tp_locvar->a)
        ctf_string(my_string_field, my_string)
        ctf_string(my_struct_name, tp_locvar->name)
    )
)

Important:The C code defined in TP_code() must not have any side effects when executed. In particular, the code must not allocate memory or get resources without deallocating this memory or putting those resources afterwards.

Load and unload a custom probe kernel module

You must load a created LTTng-modules probe kernel module in the kernel before it can emit LTTng events.

To load the default probe kernel modules and a custom probe kernel module:

  • Use the --extra-kmod-probes option to give extra probe modules to load when starting a root session daemon:

    Example:Load the my_subsys, usb, and the default probe modules.

    #
    lttng-sessiond --extra-kmod-probes=my_subsys,usb

    You only need to pass the subsystem name, not the whole kernel module name.

To load only a given custom probe kernel module:

  • Use the --kmod-probes option to give the probe modules to load when starting a root session daemon:

    Example:Load only the my_subsys and usb probe modules.

    #
    lttng-sessiond --kmod-probes=my_subsys,usb

To confirm that a probe module is loaded:

  • Use lsmod(8):

    $
    lsmod | grep lttng_probe_usb

To unload the loaded probe modules:

  • Kill the session daemon with SIGTERM:

    #
    pkill lttng-sessiond

    You can also use the --remove option of modprobe(8) if the session daemon terminates abnormally.

Tracing control

Once an application or a Linux kernel is instrumented for LTTng tracing, you can trace it.

In the LTTng context, tracing means making sure that LTTng attempts to execute some action(s) when a CPU executes an instrumentation point.

This section is divided in topics on how to use the various components of LTTng, in particular the lttng command-line tool, to control the LTTng daemons and tracers.

Note:In the following subsections, we refer to an lttng(1) command using its man page name. For example, instead of “Run the create command to”, we write “Run the lttng-create(1) command to”.

Start a session daemon

In some situations, you need to run a session daemon (lttng-sessiond(8)) before you can use the lttng(1) command-line tool.

You will see the following error when you run a command while no session daemon is running:

Error: No session daemon is available

The only command that automatically runs a session daemon is lttng-create(1), which you use to create a recording session. While this could be your most used first operation, sometimes it’s not. Some examples are:

All the examples above don’t require a recording session to operate on.

Each Unix user can have its own running session daemon to use the user space LTTng tracer. The session daemon that the root user starts is the only one allowed to control the LTTng kernel tracer. Members of the Unix tracing group may connect to and control the root session daemon, even for user space tracing. See the “Session daemon connection” section of lttng(1) to learn more about the Unix tracing group.

To start a user session daemon:

To start the root session daemon:

In both cases, remove the --daemonize option to start the session daemon in foreground.

To stop a session daemon, kill its process (see kill(1)) with the standard TERM signal.

Note that some Linux distributions could manage the LTTng session daemon as a service. In this case, we suggest that you use the service manager to start, restart, and stop session daemons.

Create and destroy a recording session

Many LTTng control operations happen in the scope of a recording session, which is the dialogue between the session daemon and you for everything related to event recording.

To create a recording session with a generated name:

The name of the created recording session is auto followed by the creation date.

To create a recording session with a specific name:

  • Use the optional argument of the lttng-create(1) command:

    $
    lttng create SESSION

    Replace SESSION with your specific recording session name.

In local mode, LTTng writes the traces of a recording session to the $LTTNG_HOME/lttng-traces/NAME-DATE-TIME directory by default, where NAME is the name of the recording session. Note that the LTTNG_HOME environment variable defaults to $HOME if not set.

To output LTTng traces to a non-default location:

You may create as many recording sessions as you wish.

To list all the existing recording sessions for your Unix user, or for all users if your Unix user is root:

When you create a recording session, the lttng-create(1) command sets it as the current recording session. The following lttng(1) commands operate on the current recording session when you don’t specify one:

To change the current recording session:

  • Use the lttng-set-session(1) command:

    $
    lttng set-session SESSION

    Replace SESSION with the name of the new current recording session.

When you’re done recording in a given recording session, destroy it. This operation frees the resources taken by the recording session to destroy; it doesn’t destroy the trace data that LTTng wrote for this recording session (see “Clear a recording session” for one way to do this).

To destroy the current recording session:

The lttng-destroy(1) command also runs the lttng-stop(1) command implicitly (see “Start and stop a recording session”). You need to stop recording to make LTTng flush the remaining trace data and make the trace readable.

List the available instrumentation points

The session daemon can query the running instrumented user applications and the Linux kernel to get a list of available instrumentation points:

  • LTTng tracepoints and system calls for the Linux kernel tracing domain.

  • LTTng tracepoints for the user space tracing domain.

To list the available instrumentation points:

  1. Make sure there’s a running session daemon to which your Unix user can connect.

  2. Use the lttng-list(1) command with the option of the requested tracing domain amongst:

    --kernel

    Linux kernel tracepoints.

    Your Unix user must be root, or it must be a member of the Unix tracing group.

    --kernel with --syscall

    Linux kernel system calls.

    Your Unix user must be root, or it must be a member of the Unix tracing group.

    --userspace

    User space tracepoints.

    --jul

    java.util.logging loggers.

    --log4j

    Apache log4j loggers.

    --python

    Python loggers.

Example:List the available user space tracepoints.

$
lttng list --userspace

Example:List the available Linux kernel system calls.

$
lttng list --kernel --syscall

Create and enable a recording event rule

Once you create a recording session, you can create recording event rules with the lttng-enable-event(1) command.

The lttng-enable-event(1) command always attaches an event rule to a channel on creation. The command can create a default channel, named channel0, for you. The lttng-enable-event(1) command reuses the default channel each time you run it for the same tracing domain and session.

A recording event rule is always enabled at creation time.

The following examples show how to combine the command-line arguments of the lttng-enable-event(1) command to create simple to more complex recording event rules within the current recording session.

Example:Create a recording event rule matching specific Linux kernel tracepoint events (default channel).

#
lttng enable-event --kernel sched_switch

Example:Create a recording event rule matching Linux kernel system call events with four specific names (default channel).

#
lttng enable-event --kernel --syscall open,write,read,close

Example:Create recording event rules matching tracepoint events which satisfy a filter expressions (default channel).

#
lttng enable-event --kernel sched_switch --filter='prev_comm == "bash"'
#
 
lttng enable-event --kernel --all \
                   --filter='$ctx.tid == 1988 || $ctx.tid == 1534'
$
 
lttng enable-event --jul my_logger \
                   --filter='$app.retriever:cur_msg_id > 3'

Important:Make sure to always single-quote the filter string when you run lttng(1) from a shell.

See also “Allow specific processes to record events” which offers another, more efficient filtering mechanism for process ID, user ID, and group ID attributes.

Example:Create a recording event rule matching any user space event from the my_app tracepoint provider and with a log level range (default channel).

$
lttng enable-event --userspace my_app:'*' --loglevel=INFO

Important:Make sure to always single-quote the wildcard character when you run lttng(1) from a shell.

Example:Create a recording event rule matching user space events named specifically, but with name exclusions (default channel).

$
 
lttng enable-event --userspace my_app:'*' \
                   --exclude=my_app:set_user,my_app:handle_sig

Example:Create a recording event rule matching any Apache log4j event with a specific log level (default channel).

$
lttng enable-event --log4j --all --loglevel-only=WARN

Example:Create a recording event rule, attached to a specific channel, and matching user space tracepoint events named my_app:my_tracepoint.

$
 
lttng enable-event --userspace my_app:my_tracepoint \
                   --channel=my-channel

Example:Create a recording event rule matching user space probe events for the malloc function entry in /usr/lib/libc.so.6:

#
 
 
lttng enable-event --kernel \
                   --userspace-probe=/usr/lib/libc.so.6:malloc \
                   libc_malloc

Example:Create a recording event rule matching user space probe events for the server/accept_request USDT probe in /usr/bin/serv:

#
 
 
lttng enable-event --kernel \
                   --userspace-probe=sdt:serv:server:accept_request \
                   server_accept_request

The recording event rules of a given channel form a whitelist: as soon as an event rule matches an event, LTTng emits it once and therefore can record it. For example, the following rules both match user space tracepoint events named my_app:my_tracepoint with an INFO log level:

$
$
 
lttng enable-event --userspace my_app:my_tracepoint
lttng enable-event --userspace my_app:my_tracepoint \
                   --loglevel=INFO

The second recording event rule is redundant: the first one includes the second one.

Disable a recording event rule

To disable a recording event rule that you created previously, use the lttng-disable-event(1) command.

lttng-disable-event(1) can only find recording event rules to disable by their instrumentation point type and event name conditions. Therefore, you cannot disable recording event rules having a specific instrumentation point log level condition, for example.

LTTng doesn’t emit (and, therefore, won’t record) an event which only disabled recording event rules match.

Example:Disable event rules matching Python logging events from the my-logger logger (default channel, current recording session).

$
lttng disable-event --python my-logger

Example:Disable event rules matching all java.util.logging events (default channel, recording session my-session).

$
lttng disable-event --jul --session=my-session '*'

Example:Disable all the Linux kernel recording event rules (channel my-chan, current recording session).

The --all-events option isn’t, like the --all option of the lttng-enable-event(1) command, an alias for the event name globbing pattern *: it disables all the recording event rules of a given channel.

#
lttng disable-event --kernel --channel=my-chan --all-events

Note:You can’t remove a recording event rule once you create it.

Get the status of a recording session

To get the status of the current recording session, that is, its parameters, its channels, recording event rules, and their attributes:

To get the status of any recording session:

  • Use the lttng-list(1) command with the name of the recording session:

    $
    lttng list SESSION

    Replace SESSION with the recording session name.

Start and stop a recording session

Once you create a recording session and create one or more recording event rules, you can start and stop the tracers for this recording session.

To start the current recording session:

LTTng is flexible: you can launch user applications before or after you start the tracers. An LTTng tracer only records an event if a recording event rule matches it, which means the tracer is active.

The start-session trigger action can also start a recording session.

To stop the current recording session:

Important:You need to stop recording to make LTTng flush the remaining trace data and make the trace readable. Note that the lttng-destroy(1) command (see “Create and destroy a recording session”) also runs the lttng-stop(1) command implicitly.

The stop-session trigger action can also stop a recording session.

Clear a recording session

Since 2.12

You might need to remove all the current tracing data of one or more recording sessions between multiple attempts to reproduce a problem without interrupting the LTTng recording activity.

To clear the tracing data of the current recording session:

To clear the tracing data of all the recording sessions:

  • Use the lttng clear command with its --all option:

    $
    lttng clear --all

Create a channel

Once you create a recording session, you can create a channel with the lttng-enable-channel(1) command.

Note that LTTng can automatically create a default channel when you create a recording event rule. Therefore, you only need to create a channel when you need non-default attributes.

Specify each non-default channel attribute with a command-line option when you run the lttng-enable-channel(1) command.

You can only create a custom channel in the Linux kernel and user space tracing domains: the Java/Python logging tracing domains have their own default channel which LTTng automatically creates when you create a recording event rule.

Important:As of LTTng 2.13, you may not perform the following operations with the lttng-enable-channel(1) command:

  • Change an attribute of an existing channel.

  • Enable a disabled channel once its recording session has been active at least once.

  • Create a channel once its recording session has been active at least once.

  • Create a user space channel with a given buffering scheme and create a second user space channel with a different buffering scheme in the same recording session.

The following examples show how to combine the command-line options of the lttng-enable-channel(1) command to create simple to more complex channels within the current recording session.

Example:Create a Linux kernel channel with default attributes.

#
lttng enable-channel --kernel my-channel

Example:Create a user space channel with four sub-buffers or 1 MiB each, per CPU, per instrumented process.

$
 
lttng enable-channel --userspace --num-subbuf=4 --subbuf-size=1M \
                     --buffers-pid my-channel

Example:Create a default user space channel with an infinite blocking timeout.

Create a recording session, create the channel, create a recording event rule, and start recording:

$
$
$
$
lttng create
lttng enable-channel --userspace --blocking-timeout=inf blocking-chan
lttng enable-event --userspace --channel=blocking-chan --all
lttng start

Run an application instrumented with LTTng-UST tracepoints and allow it to block:

$
LTTNG_UST_ALLOW_BLOCKING=1 my-app

Example:Create a Linux kernel channel which rotates eight trace files of 4 MiB each for each stream.

#
 
lttng enable-channel --kernel --tracefile-count=8 \
                     --tracefile-size=4194304 my-channel

Example:Create a user space channel in overwrite (or “flight recorder”) mode.

$
lttng enable-channel --userspace --overwrite my-channel

Example:Create the same recording event rule attached to two different channels.

$
$
lttng enable-event --userspace --channel=my-channel app:tp
lttng enable-event --userspace --channel=other-channel app:tp

When a CPU executes the app:tp user space tracepoint, the two recording event rules above match the created event, making LTTng emit the event. Because the recording event rules are not attached to the same channel, LTTng records the event twice.

Disable a channel

To disable a specific channel that you created previously, use the lttng-disable-channel(1) command.

Example:Disable a specific Linux kernel channel (current recording session).

#
lttng disable-channel --kernel my-channel

An enabled channel is an implicit recording event rule condition.

Note:As of LTTng 2.13, you may not enable a disabled channel once its recording session has been started at least once.

Add context fields to be recorded to the event records of a channel

Event record fields in trace files provide important information about previously emitted events, but sometimes some external context may help you solve a problem faster.

Examples of context fields are:

  • The process ID, thread ID, process name, and process priority of the thread from which LTTng emits the event.

  • The hostname of the system on which LTTng emits the event.

  • The Linux kernel and user call stacks (since LTTng 2.11).

  • The current values of many possible performance counters using perf, for example:

    • CPU cycles, stalled cycles, idle cycles, and the other cycle types.

    • Cache misses.

    • Branch instructions, misses, and loads.

    • CPU faults.

  • Any state defined at the application level (supported for the java.util.logging and Apache log4j tracing domains).

To get the full list of available context fields:

Example:Add context fields to be recorded to the event records of all the channels of the current recording session.

The following command line adds the virtual process identifier and the per-thread CPU cycles count fields to all the user space channels of the current recording session.

$
lttng add-context --userspace --type=vpid --type=perf:thread:cpu-cycles

Example:Add performance counter context fields by raw ID

See lttng-add-context(1) for the exact format of the context field type, which is partly compatible with the format used in perf-record(1).

#
#
lttng add-context --userspace --type=perf:thread:raw:r0110:test
lttng add-context --kernel --type=perf:cpu:raw:r0013c:x86unhalted

Example:Add context fields to be recorded to the event records of a specific channel.

The following command line adds the thread identifier and user call stack context fields to the Linux kernel channel named my-channel of the current recording session.

#
 
lttng add-context --kernel --channel=my-channel \
                  --type=tid --type=callstack-user

Example:Add an application-specific context field to be recorded to the event records of a specific channel.

The following command line makes sure LTTng writes the cur_msg_id context field of the retriever context retriever to all the Java logging event records of the channel named my-channel:

#
 
lttng add-context --kernel --channel=my-channel \
                  --type='$app:retriever:cur_msg_id'

Important:Make sure to always single-quote the $ character when you run lttng-add-context(1) from a shell.

Note:You can’t undo what the lttng-add-context(1) command does.

Allow specific processes to record events

Since 2.7

It’s often useful to only allow processes with specific attributes to record events. For example, you may wish to record all the system calls which a given process makes (à la strace(1)).

The lttng-track(1) and lttng-untrack(1) commands serve this purpose. Both commands operate on inclusion sets of process attributes. The available process attribute types are:

Linux kernel tracing domain
  • Process ID (PID).

  • Virtual process ID (VPID).

    This is the PID as seen by the application.

  • Unix user ID (UID).

  • Virtual Unix user ID (VUID).

    This is the UID as seen by the application.

  • Unix group ID (GID).

  • Virtual Unix group ID (VGID).

    This is the GID as seen by the application.

User space tracing domain
  • VPID

  • VUID

  • VGID

A recording session has nine process attribute inclusion sets: six for the Linux kernel tracing domain and three for the user space tracing domain.

For a given recording session, a process P is allowed to record LTTng events for a given tracing domain D if all the attributes of P are part of the inclusion sets of D.

Whether a process is allowed or not to record LTTng events is an implicit condition of all recording event rules. Therefore, if LTTng creates an event E for a given process, but this process may not record events, then no recording event rule matches E, which means LTTng won’t emit and record E.

When you create a recording session, all its process attribute inclusion sets contain all the possible values. In other words, all processes are allowed to record events.

Add values to an inclusion set with the lttng-track(1) command and remove values with the lttng-untrack(1) command.

Note:The process attribute values are numeric.

Should a process with a given ID (part of an inclusion set), for example, exit, and then a new process be given this same ID, then the latter would also be allowed to record events.

With the lttng-track(1) command, you can add Unix user and group names to the user and group inclusion sets: the session daemon finds the corresponding UID, VUID, GID, or VGID once on addition to the inclusion set. This means that if you rename the user or group after you run the lttng-track(1) command, its user/group ID remains part of the inclusion sets.

Example:Allow processes to record events based on their virtual process ID (VPID).

For the sake of the following example, assume the target system has 16 possible VPIDs.

When you create a recording session, the user space VPID inclusion set contains all the possible VPIDs:

The VPID inclusion set is full.

When the inclusion set is full and you run the lttng-track(1) command to specify some VPIDs, LTTng:

  1. Clears the inclusion set.

  2. Adds the specific VPIDs to the inclusion set.

After:

$
lttng track --userspace --vpid=3,4,7,10,13

the VPID inclusion set is:

The VPID inclusion set contains the VPIDs 3, 4, 7, 10, and 13.

Add more VPIDs to the inclusion set afterwards:

$
lttng track --userspace --vpid=1,15,16

The result is:

VPIDs 1, 15, and 16 are added to the inclusion set.

The lttng-untrack(1) command removes entries from process attribute inclusion sets. Given the previous example, the following command:

$
lttng untrack --userspace --vpid=3,7,10,13

leads to this VPID inclusion set:

VPIDs 3, 7, 10, and 13 are removed from the inclusion set.

You can make the VPID inclusion set full again with the --all option:

$
lttng track --userspace --vpid --all

The result is, again:

The VPID inclusion set is full.

Example:Allow specific processes to record events based on their user ID (UID).

A typical use case with process attribute inclusion sets is to start with an empty inclusion set, then start the tracers, and finally add values manually while the tracers are active.

Use the --all option of the lttng-untrack(1) command to clear the inclusion set after you create a recording session, for example (with UIDs):

#
lttng untrack --kernel --uid --all

gives:

The UID inclusion set is empty.

If the LTTng tracer runs with this inclusion set configuration, it records no events within the current recording session because no processes is allowed to do so. Use the lttng-track(1) command as usual to add specific values to the UID inclusion set when you need to, for example:

#
lttng track --kernel --uid=http,11

Result:

UIDs 6 (http) and 11 are part of the UID inclusion set.

Save and load recording session configurations

Since 2.5

Configuring a recording session can be long. Some of the tasks involved are:

If you use LTTng to solve real world problems, chances are you have to record events using the same recording session setup over and over, modifying a few variables each time in your instrumented program or environment.

To avoid constant recording session reconfiguration, the lttng(1) command-line tool can save and load recording session configurations to/from XML files.

To save a given recording session configuration:

  • Use the lttng-save(1) command:

    $
    lttng save SESSION

    Replace SESSION with the name of the recording session to save.

LTTng saves recording session configurations to $LTTNG_HOME/.lttng/sessions by default. Note that the LTTNG_HOME environment variable defaults to $HOME if not set. See lttng-save(1) to learn more about the recording session configuration output path.

LTTng saves all configuration parameters, for example:

  • The recording session name.

  • The trace data output path.

  • The channels, with their state and all their attributes.

  • The context fields you added to channels.

  • The recording event rules with their state and conditions.

To load a recording session:

  • Use the lttng-load(1) command:

    $
    lttng load SESSION

    Replace SESSION with the name of the recording session to load.

When LTTng loads a configuration, it restores your saved recording session as if you just configured it manually.

You can also save and load many sessions at a time; see lttng-save(1) and lttng-load(1) to learn more.

Send trace data over the network

LTTng can send the recorded trace data of a recording session to a remote system over the network instead of writing it to the local file system.

To send the trace data over the network:

  1. On the remote system (which can also be the target system), start an LTTng relay daemon (lttng-relayd(8)):

    $
    lttng-relayd
  2. On the target system, create a recording session configured to send trace data over the network:

    $
    lttng create my-session --set-url=net://remote-system

    Replace remote-system with the host name or IP address of the remote system. See lttng-create(1) for the exact URL format.

  3. On the target system, use the lttng(1) command-line tool as usual.

    When recording is active, the consumer daemon of the target sends the contents of sub-buffers to the remote relay daemon instead of flushing them to the local file system. The relay daemon writes the received packets to its local file system.

See the “Output directory” section of lttng-relayd(8) to learn where a relay daemon writes its received trace data.

View events as LTTng records them (LTTng live)

Since 2.4

LTTng live is a network protocol implemented by the relay daemon (lttng-relayd(8)) to allow compatible trace readers to display or analyze event records as LTTng records events on the target system while recording is active.

The relay daemon creates a tee: it forwards the trace data to both the local file system and to connected live readers:

The relay daemon creates a tee, forwarding the trace data to both trace files and a connected live reader.

To use LTTng live:

  1. On the target system, create a recording session in live mode:

    $
    lttng create my-session --live

    This operation spawns a local relay daemon.

  2. Start the live reader and configure it to connect to the relay daemon.

    For example, with babeltrace2(1):

    $
    babeltrace2 net://localhost/host/HOSTNAME/my-session

    Replace HOSTNAME with the host name of the target system.

  3. Configure the recording session as usual with the lttng(1) command-line tool, and start recording.

List the available live recording sessions with babeltrace2(1):

$
babeltrace2 net://localhost

You can start the relay daemon on another system. In this case, you need to specify the URL of the relay daemon when you create the recording session with the --set-url option of the lttng-create(1) command. You also need to replace localhost in the procedure above with the host name of the system on which the relay daemon runs.

Take a snapshot of the current sub-buffers of a recording session

Since 2.3

The normal behavior of LTTng is to append full sub-buffers to growing trace data files. This is ideal to keep a full history of the events which the target system emitted, but it can represent too much data in some situations.

For example, you may wish to have LTTng record your application continuously until some critical situation happens, in which case you only need the latest few recorded events to perform the desired analysis, not multi-gigabyte trace files.

With the lttng-snapshot(1) command, you can take a snapshot of the current sub-buffers of a given recording session. LTTng can write the snapshot to the local file system or send it over the network.

A snapshot is a copy of the current sub-buffers, which LTTng does not clear after the operation.

The snapshot feature of LTTng is similar to how a flight recorder or the “roll” mode of an oscilloscope work.

Tip:If you wish to create unmanaged, self-contained, non-overlapping trace chunk archives instead of a simple copy of the current sub-buffers, see the recording session rotation feature (available since LTTng 2.11).

To take a snapshot of the current recording session:

  1. Create a recording session in snapshot mode:

    $
    lttng create my-session --snapshot

    The event record loss mode of channels created in this mode is automatically set to overwrite.

  2. Configure the recording session as usual with the lttng(1) command-line tool, and start recording.

  3. Optional: When you need to take a snapshot, stop recording.

    You can take a snapshot when the tracers are active, but if you stop them first, you’re guaranteed that the trace data in the sub-buffers doesn’t change before you actually take the snapshot.

  4. Take a snapshot:

    $
    lttng snapshot record --name=my-first-snapshot

    LTTng writes the current sub-buffers of all the channels of the current recording session to trace files on the local file system. Those trace files have my-first-snapshot in their name.

There’s no difference between the format of a normal trace file and the format of a snapshot: LTTng trace readers also support LTTng snapshots.

By default, LTTng writes snapshot files to the path shown by

$
lttng snapshot list-output

You can change this path or decide to send snapshots over the network using either:

  1. An output path or URL that you specify when you create the recording session.

  2. A snapshot output path or URL that you add using the add-output action of the lttng-snapshot(1) command.

  3. An output path or URL that you provide directly to the record action of the lttng-snapshot(1) command.

Method 3 overrides method 2, which overrides method 1. When you specify a URL, a relay daemon must listen on a remote system (see “Send trace data over the network”).

The snapshot-session trigger action can also take a recording session snapshot.

Archive the current trace chunk (rotate a recording session)

Since 2.11

The snapshot user guide shows how to dump the current sub-buffers of a recording session to the file system or send them over the network. When you take a snapshot, LTTng doesn’t clear the ring buffers of the recording session: if you take another snapshot immediately after, both snapshots could contain overlapping trace data.

Inspired by log rotation, recording session rotation is a feature which appends the content of the ring buffers to what’s already on the file system or sent over the network since the creation of the recording session or since the last rotation, and then clears those ring buffers to avoid trace data overlaps.

What LTTng is about to write when performing a recording session rotation is called the current trace chunk. When LTTng writes or sends over the network this current trace chunk, it becomes a trace chunk archive. Therefore, a recording session rotation operation archives the current trace chunk.

A recording session rotation operation archives the current trace chunk.

A trace chunk archive is a self-contained LTTng trace which LTTng doesn’t manage anymore: you can read it, modify it, move it, or remove it.

As of LTTng 2.13, there are three methods to perform a recording session rotation:

To perform an immediate rotation of the current recording session:

  1. Create a recording session in local mode or network streaming mode (only those two recording session modes support recording session rotation):

    #
    lttng create my-session
  2. Create one or more recording event rules and start recording:

    #
    #
    lttng enable-event --kernel sched_'*'
    lttng start
  3. When needed, immediately rotate the current recording session:

    #
    lttng rotate

    The lttng-rotate(1) command prints the path to the created trace chunk archive. See its manual page to learn about the format of trace chunk archive directory names.

    Perform other immediate rotations while the recording session is active. It’s guaranteed that all the trace chunk archives don’t contain overlapping trace data. You can also perform an immediate rotation once you have stopped the recording session.

  4. When you’re done recording, destroy the current recording session:

    #
    lttng destroy

    The recording session destruction operation creates one last trace chunk archive from the current trace chunk.

A recording session rotation schedule is a planned rotation which LTTng performs automatically based on one of the following conditions:

  • A timer with a configured period expires.

  • The total size of the flushed part of the current trace chunk becomes greater than or equal to a configured value.

To schedule a rotation of the current recording session, set a rotation schedule:

  1. Create a recording session in local mode or network streaming mode (only those two creation modes support recording session rotation):

    #
    lttng create my-session
  2. Create one or more recording event rules:

    #
    lttng enable-event --kernel sched_'*'
  3. Set a recording session rotation schedule:

    #
    lttng enable-rotation --timer=10s

    In this example, we set a rotation schedule so that LTTng performs a recording session rotation every ten seconds.

    See lttng-enable-rotation(1) to learn more about other ways to set a rotation schedule.

  4. Start recording:

    #
    lttng start

    LTTng performs recording session rotations automatically while the recording session is active thanks to the rotation schedule.

  5. When you’re done recording, destroy the current recording session:

    #
    lttng destroy

    The recording session destruction operation creates one last trace chunk archive from the current trace chunk.

Unset a recording session rotation schedule with the lttng-disable-rotation(1) command.

Add an “event rule matches” trigger to a session daemon

Since 2.13

With the lttng-add-trigger(1) command, you can add a trigger to a session daemon.

A trigger associates an LTTng tracing condition to one or more actions: when the condition is satisfied, LTTng attempts to execute the actions.

A trigger doesn’t need any recording session to exist: it belongs to a session daemon.

As of LTTng 2.13, many condition types are available through the liblttng-ctl C API, but the lttng-add-trigger(1) command only accepts the “event rule matches” condition.

An “event rule matches” condition is satisfied when its event rule matches an event.

Unlike a recording event rule, the event rule of an “event rule matches” trigger condition has no implicit conditions, that is:

Both the lttng-add-trigger(1) and lttng-enable-event(1) commands accept command-line arguments to specify an event rule. That being said, the former is a more recent command and therefore follows the common event rule specification format (see lttng-event-rule(7)).

Example:Start a recording session when an event rule matches.

This example shows how to add the following trigger to the root session daemon:

Condition

An event rule matches a Linux kernel system call event of which the name starts with exec and */ls matches the filename payload field.

With such an event rule, LTTng emits an event when the ls program starts.

Action

Start the recording session named pitou.

To add such a trigger to the root session daemon:

  1. If there’s no currently running LTTng root session daemon, start one:

    #
    lttng-sessiond --daemonize
  2. Create a recording session named pitou and create a recording event rule matching all the system call events:

    #
    #
    lttng create pitou
    lttng enable-event --kernel --syscall --all
  3. Add the trigger to the root session daemon:

    #
     
     
     
    lttng add-trigger --condition=event-rule-matches \
                      --type=syscall --name='exec*' \
                      --filter='filename == "*/ls"' \
                      --action=start-session pitou

    Confirm that the trigger exists with the lttng-list-triggers(1) command:

    #
    lttng list-triggers
  4. Make sure the pitou recording session is still inactive (stopped):

    #
    lttng list pitou

    The first line should be something like:

    Recording session pitou: [inactive]

Run the ls program to fire the LTTng trigger above:

$
ls ~

At this point, the pitou recording session should be active (started). Confirm this with the lttng-list(1) command again:

#
lttng list pitou

The first line should now look like:

Recording session pitou: [active]

This line confirms that the LTTng trigger you added fired, therefore starting the pitou recording session.

Example:Send a notification to a user application when an event rule matches.

This example shows how to add the following trigger to the root session daemon:

Condition

An event rule matches a Linux kernel tracepoint event named sched_switch and of which the value of the next_comm payload field is bash.

With such an event rule, LTTng emits an event when Linux gives access to the processor to a process named bash.

Action

Send an LTTng notification to a user application.

Moreover, we’ll specify a capture descriptor with the event-rule-matches trigger condition so that the user application can get the value of a specific sched_switch event payload field.

First, write and build the user application:

  1. Create the C source file of the application:

    notif-app.c

    #include <stdlib.h>
    #include <stdio.h>
    #include <stdbool.h>
    #include <assert.h>
    #include <string.h>
    #include <lttng/lttng.h>
    
    /*
     * Subscribes to notifications, through the notification channel
     * `notification_channel`, which match the condition of the trigger
     * named `trigger_name`.
     *
     * Returns `true` on success.
     */
    static bool subscribe(struct lttng_notification_channel *notification_channel,
                          const char *trigger_name)
    {
        const struct lttng_condition *condition = NULL;
        struct lttng_triggers *triggers = NULL;
        unsigned int trigger_count;
        unsigned int i;
        enum lttng_error_code error_code;
        enum lttng_trigger_status trigger_status;
        bool ret = false;
    
        /* Get all LTTng triggers */
        error_code = lttng_list_triggers(&triggers);
        assert(error_code == LTTNG_OK);
    
        /* Get the number of triggers */
        trigger_status = lttng_triggers_get_count(triggers, &trigger_count);
        assert(trigger_status == LTTNG_TRIGGER_STATUS_OK);
    
        /* Find the trigger named `trigger_name` */
        for (i = 0; i < trigger_count; i++) {
            const struct lttng_trigger *trigger;
            const char *this_trigger_name;
    
            trigger = lttng_triggers_get_at_index(triggers, i);
            trigger_status = lttng_trigger_get_name(trigger, &this_trigger_name);
            assert(trigger_status == LTTNG_TRIGGER_STATUS_OK);
    
            if (strcmp(this_trigger_name, trigger_name) == 0) {
                /* Trigger found: subscribe with its condition */
                enum lttng_notification_channel_status notification_channel_status;
    
                notification_channel_status = lttng_notification_channel_subscribe(
                    notification_channel,
                    lttng_trigger_get_const_condition(trigger));
                assert(notification_channel_status ==
                       LTTNG_NOTIFICATION_CHANNEL_STATUS_OK);
                ret = true;
                break;
            }
        }
    
        lttng_triggers_destroy(triggers);
        return ret;
    }
    
    /*
     * Handles the evaluation `evaluation` of a single notification.
     */
    static void handle_evaluation(const struct lttng_evaluation *evaluation)
    {
        enum lttng_evaluation_status evaluation_status;
        const struct lttng_event_field_value *array_field_value;
        const struct lttng_event_field_value *string_field_value;
        enum lttng_event_field_value_status event_field_value_status;
        const char *string_field_string_value;
    
        /* Get the value of the first captured (string) field */
        evaluation_status = lttng_evaluation_event_rule_matches_get_captured_values(
            evaluation, &array_field_value);
        assert(evaluation_status == LTTNG_EVALUATION_STATUS_OK);
        event_field_value_status =
            lttng_event_field_value_array_get_element_at_index(
                array_field_value, 0, &string_field_value);
        assert(event_field_value_status == LTTNG_EVENT_FIELD_VALUE_STATUS_OK);
        assert(lttng_event_field_value_get_type(string_field_value) ==
               LTTNG_EVENT_FIELD_VALUE_TYPE_STRING);
        event_field_value_status = lttng_event_field_value_string_get_value(
                string_field_value, &string_field_string_value);
        assert(event_field_value_status == LTTNG_EVENT_FIELD_VALUE_STATUS_OK);
    
        /* Print the string value of the field */
        puts(string_field_string_value);
    }
    
    int main(int argc, char *argv[])
    {
        int exit_status = EXIT_SUCCESS;
        struct lttng_notification_channel *notification_channel;
        enum lttng_notification_channel_status notification_channel_status;
        const struct lttng_condition *condition;
        const char *trigger_name;
        bool subscribe_res;
    
        assert(argc >= 2);
        trigger_name = argv[1];
    
        /*
         * Create a notification channel.
         *
         * A notification channel connects the user application to the LTTng
         * session daemon.
         *
         * You can use this notification channel to listen to various types
         * of notifications.
         */
        notification_channel = lttng_notification_channel_create(
            lttng_session_daemon_notification_endpoint);
        assert(notification_channel);
    
        /*
         * Subscribe to notifications which match the condition of the
         * trigger named `trigger_name`.
         */
        if (!subscribe(notification_channel, trigger_name)) {
            fprintf(stderr,
                    "Error: Failed to subscribe to notifications (trigger `%s`).\n",
                    trigger_name);
            exit_status = EXIT_FAILURE;
            goto end;
        }
    
        /*
         * Notification loop.
         *
         * Put this in a dedicated thread to avoid blocking the main thread.
         */
        while (true) {
            struct lttng_notification *notification;
            enum lttng_notification_channel_status status;
            const struct lttng_evaluation *notification_evaluation;
    
            /* Receive the next notification */
            status = lttng_notification_channel_get_next_notification(
                notification_channel, &notification);
    
            switch (status) {
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
                break;
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
                /*
                 * The session daemon can drop notifications if a receiving
                 * application doesn't consume the notifications fast
                 * enough.
                 */
                continue;
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
                /*
                 * The session daemon closed the notification channel.
                 *
                 * This is typically caused by a session daemon shutting
                 * down.
                 */
                goto end;
            default:
                /* Unhandled conditions or errors */
                exit_status = EXIT_FAILURE;
                goto end;
            }
    
            /*
             * Handle the condition evaluation.
             *
             * A notification provides, amongst other things:
             *
             * * The condition that caused LTTng to send this notification.
             *
             * * The condition evaluation, which provides more specific
             *   information on the evaluation of the condition.
             */
            handle_evaluation(lttng_notification_get_evaluation(notification));
    
            /* Destroy the notification object */
            lttng_notification_destroy(notification);
        }
    
    end:
        lttng_notification_channel_destroy(notification_channel);
        return exit_status;
    }
    

    This application prints the first captured string field value of the condition evaluation of each LTTng notification it receives.

  2. Build the notif-app application, using pkg-config to provide the right compiler and linker flags:

    $
    gcc -o notif-app notif-app.c $(pkg-config --cflags --libs lttng-ctl)

Now, to add the trigger to the root session daemon:

  1. If there’s no currently running LTTng root session daemon, start one:

    #
    lttng-sessiond --daemonize
  2. Add the trigger, naming it sched-switch-notif, to the root session daemon:

    #
     
     
     
     
    lttng add-trigger --name=sched-switch-notif \
                      --condition=event-rule-matches \
                      --type=kernel --name=sched_switch \
                      --filter='next_comm == "bash"' --capture=prev_comm \
                      --action=notify

    Confirm that the sched-switch-notif trigger exists with the lttng-list-triggers(1) command:

    #
    lttng list-triggers

Run the notif-app application, passing the name of the trigger of which to watch the notifications:

#
./notif-app sched-switch-notif

Now, in an interactive Bash, type a few keys to fire the sched-switch-notif trigger. Watch the notif-app application print the previous process names.

Use the machine interface

Since 2.6

With any command of the lttng(1) command-line tool, set the --mi option to xml (before the command name) to get an XML machine interface output, for example:

$
lttng --mi=xml list my-session

A schema definition (XSD) is available to ease the integration with external tools as much as possible.

Regenerate the metadata of an LTTng trace

Since 2.8

An LTTng trace, which is a CTF trace, has both data stream files and a metadata stream file. This metadata file contains, amongst other things, information about the offset of the clock sources which LTTng uses to assign timestamps to event records when recording.

If, once a recording session is started, a major NTP correction happens, the clock offset of the trace also needs to be updated. Use the metadata item of the lttng-regenerate(1) command to do so.

The main use case of this command is to allow a system to boot with an incorrect wall time and have LTTng trace it before its wall time is corrected. Once the system is known to be in a state where its wall time is correct, you can run lttng regenerate metadata.

To regenerate the metadata stream files of the current recording session:

Regenerate the state dump event records of a recording session

Since 2.9

The LTTng kernel and user space tracers generate state dump event records when the application starts or when you start a recording session.

An analysis can use the state dump event records to set an initial state before it builds the rest of the state from the subsequent event records. Trace Compass is a notable example of an application which uses the state dump of an LTTng trace.

When you take a snapshot, it’s possible that the state dump event records aren’t included in the snapshot trace files because they were recorded to a sub-buffer that has been consumed or overwritten already.

Use the statedump item of the lttng-regenerate(1) command to emit and record the state dump events again.

To regenerate the state dump of the current recording session, provided you created it in snapshot mode, before you take a snapshot:

  1. Use the statedump item of the lttng-regenerate(1) command:

    $
    lttng regenerate statedump
  2. Stop the recording session:

    $
    lttng stop
  3. Take a snapshot:

    $
    lttng snapshot record --name=my-snapshot

Depending on the event throughput, you should run steps 1 and 2 as closely as possible.

Note:To record the state dump events, you need to create recording event rules which enable them:

  • The names of LTTng-UST state dump tracepoints start with lttng_ust_statedump:.

  • The names of LTTng-modules state dump tracepoints start with lttng_statedump_.

Record trace data on persistent memory file systems

Since 2.7

Non-volatile random-access memory (NVRAM) is random-access memory that retains its information when power is turned off (non-volatile). Systems with such memory can store data structures in RAM and retrieve them after a reboot, without flushing to typical storage.

Linux supports NVRAM file systems thanks to either DAX + pmem (requires Linux 4.1+) or PRAMFS (requires Linux < 4).

This section doesn’t describe how to operate such file systems; we assume that you have a working persistent memory file system.

When you create a recording session, you can specify the path of the shared memory holding the sub-buffers. If you specify a location on an NVRAM file system, then you can retrieve the latest recorded trace data when the system reboots after a crash.

To record trace data on a persistent memory file system and retrieve the trace data after a system crash:

  1. Create a recording session with a sub-buffer shared memory path located on an NVRAM file system:

    $
    lttng create my-session --shm-path=/path/to/shm/on/nvram
  2. Configure the recording session as usual with the lttng(1) command-line tool, and start recording.

  3. After a system crash, use the lttng-crash(1) command-line tool to read the trace data recorded on the NVRAM file system:

    $
    lttng-crash /path/to/shm/on/nvram

The binary layout of the ring buffer files isn’t exactly the same as the trace files layout. This is why you need to use lttng-crash(1) instead of some standard LTTng trace reader.

To convert the ring buffer files to LTTng trace files:

Get notified when the buffer usage of a channel is too high or too low

Since 2.10

With the notification and trigger C API of liblttng-ctl, LTTng can notify your user application when the buffer usage of one or more channels becomes too low or too high.

Use this API and enable or disable recording event rules while a recording session is active to avoid discarded event records, for example.

Example:Send a notification to a user application when the buffer usage of an LTTng channel is too high.

In this example, we create and build an application which gets notified when the buffer usage of a specific LTTng channel is higher than 75 %.

We only print that it’s the case in this example, but we could as well use the liblttng-ctl C API to disable recording event rules when this happens, for example.

  1. Create the C source file of the application:

    notif-app.c

    #include <stdlib.h>
    #include <stdio.h>
    #include <assert.h>
    #include <lttng/lttng.h>
    
    int main(int argc, char *argv[])
    {
        int exit_status = EXIT_SUCCESS;
        struct lttng_notification_channel *notification_channel;
        struct lttng_condition *condition;
        struct lttng_action *action;
        struct lttng_trigger *trigger;
        const char *recording_session_name;
        const char *channel_name;
    
        assert(argc >= 3);
        recording_session_name = argv[1];
        channel_name = argv[2];
    
        /*
         * Create a notification channel.
         *
         * A notification channel connects the user application to the LTTng
         * session daemon.
         *
         * You can use this notification channel to listen to various types
         * of notifications.
         */
        notification_channel = lttng_notification_channel_create(
            lttng_session_daemon_notification_endpoint);
    
        /*
         * Create a "buffer usage becomes greater than" condition.
         *
         * In this case, the condition is satisfied when the buffer usage
         * becomes greater than or equal to 75 %.
         *
         * We create the condition for a specific recording session name,
         * channel name, and for the user space tracing domain.
         *
         * The following condition types also exist:
         *
         * * The buffer usage of a channel becomes less than a given value.
         *
         * * The consumed data size of a recording session becomes greater
         *   than a given value.
         *
         * * A recording session rotation becomes ongoing.
         *
         * * A recording session rotation becomes completed.
         *
         * * A given event rule matches an event.
         */
        condition = lttng_condition_buffer_usage_high_create();
        lttng_condition_buffer_usage_set_threshold_ratio(condition, .75);
        lttng_condition_buffer_usage_set_session_name(condition,
                                                      recording_session_name);
        lttng_condition_buffer_usage_set_channel_name(condition,
                                                      channel_name);
        lttng_condition_buffer_usage_set_domain_type(condition,
                                                     LTTNG_DOMAIN_UST);
    
        /*
         * Create an action (receive a notification) to execute when the
         * condition created above is satisfied.
         */
        action = lttng_action_notify_create();
    
        /*
         * Create a trigger.
         *
         * A trigger associates a condition to an action: LTTng executes
         * the action when the condition is satisfied.
         */
        trigger = lttng_trigger_create(condition, action);
    
        /* Register the trigger to the LTTng session daemon. */
        lttng_register_trigger(trigger);
    
        /*
         * Now that we have registered a trigger, LTTng will send a
         * notification every time its condition is met through a
         * notification channel.
         *
         * To receive this notification, we must subscribe to notifications
         * which match the same condition.
         */
        lttng_notification_channel_subscribe(notification_channel,
                                             condition);
    
        /*
         * Notification loop.
         *
         * Put this in a dedicated thread to avoid blocking the main thread.
         */
        for (;;) {
            struct lttng_notification *notification;
            enum lttng_notification_channel_status status;
            const struct lttng_evaluation *notification_evaluation;
            const struct lttng_condition *notification_condition;
            double buffer_usage;
    
            /* Receive the next notification. */
            status = lttng_notification_channel_get_next_notification(
                notification_channel, &notification);
    
            switch (status) {
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_OK:
                break;
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_NOTIFICATIONS_DROPPED:
                /*
                 * The session daemon can drop notifications if a monitoring
                 * application isn't consuming the notifications fast
                 * enough.
                 */
                continue;
            case LTTNG_NOTIFICATION_CHANNEL_STATUS_CLOSED:
                /*
                 * The session daemon closed the notification channel.
                 *
                 * This is typically caused by a session daemon shutting
                 * down.
                 */
                goto end;
            default:
                /* Unhandled conditions or errors. */
                exit_status = EXIT_FAILURE;
                goto end;
            }
    
            /*
             * A notification provides, amongst other things:
             *
             * * The condition that caused LTTng to send this notification.
             *
             * * The condition evaluation, which provides more specific
             *   information on the evaluation of the condition.
             *
             * The condition evaluation provides the buffer usage
             * value at the moment the condition was satisfied.
             */
            notification_condition = lttng_notification_get_condition(
                notification);
            notification_evaluation = lttng_notification_get_evaluation(
                notification);
    
            /* We're subscribed to only one condition. */
            assert(lttng_condition_get_type(notification_condition) ==
                LTTNG_CONDITION_TYPE_BUFFER_USAGE_HIGH);
    
            /*
             * Get the exact sampled buffer usage from the condition
             * evaluation.
             */
            lttng_evaluation_buffer_usage_get_usage_ratio(
                notification_evaluation, &buffer_usage);
    
            /*
             * At this point, instead of printing a message, we could do
             * something to reduce the buffer usage of the channel, like
             * disable specific events, for example.
             */
            printf("Buffer usage is %f %% in recording session \"%s\", "
                   "user space channel \"%s\".\n", buffer_usage * 100,
                   recording_session_name, channel_name);
    
            /* Destroy the notification object. */
            lttng_notification_destroy(notification);
        }
    
    end:
        lttng_action_destroy(action);
        lttng_condition_destroy(condition);
        lttng_trigger_destroy(trigger);
        lttng_notification_channel_destroy(notification_channel);
        return exit_status;
    }
    
  2. Build the notif-app application, linking it with liblttng-ctl:

    $
    gcc -o notif-app notif-app.c $(pkg-config --cflags --libs lttng-ctl)
  3. Create a recording session, create a recording event rule matching all the user space tracepoint events, and start recording:

    $
    $
    $
    lttng create my-session
    lttng enable-event --userspace --all
    lttng start

    If you create the channel manually with the lttng-enable-channel(1) command, you can set its monitor timer to control how frequently LTTng samples the current values of the channel properties to evaluate user conditions.

  4. Run the notif-app application.

    This program accepts the recording session and user space channel names as its two first arguments. The channel which LTTng automatically creates with the lttng-enable-event(1) command above is named channel0:

    $
    ./notif-app my-session channel0
  5. In another terminal, run an application with a very high event throughput so that the 75 % buffer usage condition is reached.

    In the first terminal, the application should print lines like this:

    Buffer usage is 81.45197 % in recording session "my-session", user space
    channel "channel0".

    If you don’t see anything, try to make the threshold of the condition in notif-app.c lower (0.1 %, for example), and then rebuild the notif-app application (step 2) and run it again (step 4).

Reference

LTTng-modules

LTTNG_TRACEPOINT_ENUM() usage

Since 2.9

Use the LTTNG_TRACEPOINT_ENUM() macro to define an enumeration:

LTTNG_TRACEPOINT_ENUM(name, TP_ENUM_VALUES(entries))

Replace:

  • name with the name of the enumeration (C identifier, unique amongst all the defined enumerations).

  • entries with a list of enumeration entries.

The available enumeration entry macros are:

ctf_enum_value(name, value)

Entry named name mapped to the integral value value.

ctf_enum_range(name, begin, end)

Entry named name mapped to the range of integral values between begin (included) and end (included).

ctf_enum_auto(name)

Entry named name mapped to the integral value following the last mapping value.

The last value of a ctf_enum_value() entry is its value parameter.

The last value of a ctf_enum_range() entry is its end parameter.

If ctf_enum_auto() is the first entry in the list, its integral value is 0.

Use the ctf_enum() field definition macro to use a defined enumeration as a tracepoint field.

Example:Define an enumeration with LTTNG_TRACEPOINT_ENUM().

LTTNG_TRACEPOINT_ENUM(
    my_enum,
    TP_ENUM_VALUES(
        ctf_enum_auto("AUTO: EXPECT 0")
        ctf_enum_value("VALUE: 23", 23)
        ctf_enum_value("VALUE: 27", 27)
        ctf_enum_auto("AUTO: EXPECT 28")
        ctf_enum_range("RANGE: 101 TO 303", 101, 303)
        ctf_enum_auto("AUTO: EXPECT 304")
    )
)

Tracepoint fields macros (for TP_FIELDS())

Since 2.7

The available macros to define tracepoint fields, which must be listed within TP_FIELDS() in LTTNG_TRACEPOINT_EVENT(), are:

Available macros to define LTTng-modules tracepoint fields

Macro Description and parameters

ctf_integer(t, n, e)

ctf_integer_nowrite(t, n, e)

ctf_user_integer(t, n, e)

ctf_user_integer_nowrite(t, n, e)

Standard integer, displayed in base 10.

t

Integer C type (int, long, size_t, …).

n

Field name.

e

Argument expression.

ctf_integer_hex(t, n, e)

ctf_user_integer_hex(t, n, e)

Standard integer, displayed in base 16.

t

Integer C type.

n

Field name.

e

Argument expression.

ctf_integer_oct(t, n, e)

Standard integer, displayed in base 8.

t

Integer C type.

n

Field name.

e

Argument expression.

ctf_integer_network(t, n, e)

ctf_user_integer_network(t, n, e)

Integer in network byte order (big-endian), displayed in base 10.

t

Integer C type.

n

Field name.

e

Argument expression.

ctf_integer_network_hex(t, n, e)

ctf_user_integer_network_hex(t, n, e)

Integer in network byte order, displayed in base 16.

t

Integer C type.

n

Field name.

e

Argument expression.

ctf_enum(N, t, n, e)

ctf_enum_nowrite(N, t, n, e)

ctf_user_enum(N, t, n, e)

ctf_user_enum_nowrite(N, t, n, e)

Enumeration.

N

Name of a previously defined enumeration.

t

Integer C type (int, long, size_t, …).

n

Field name.

e

Argument expression.

ctf_string(n, e)

ctf_string_nowrite(n, e)

ctf_user_string(n, e)

ctf_user_string_nowrite(n, e)

Null-terminated string; undefined behavior if e is NULL.

n

Field name.

e

Argument expression.

ctf_array(t, n, e, s)

ctf_array_nowrite(t, n, e, s)

ctf_user_array(t, n, e, s)

ctf_user_array_nowrite(t, n, e, s)

Statically-sized array of integers.

t

Array element C type.

n

Field name.

e

Argument expression.

s

Number of elements.

ctf_array_bitfield(t, n, e, s)

ctf_array_bitfield_nowrite(t, n, e, s)

ctf_user_array_bitfield(t, n, e, s)

ctf_user_array_bitfield_nowrite(t, n, e, s)

Statically-sized array of bits.

The type of e must be an integer type. s is the number of elements of such type in e, not the number of bits.

t

Array element C type.

n

Field name.

e

Argument expression.

s

Number of elements.

ctf_array_text(t, n, e, s)

ctf_array_text_nowrite(t, n, e, s)

ctf_user_array_text(t, n, e, s)

ctf_user_array_text_nowrite(t, n, e, s)

Statically-sized array, printed as text.

The string doesn’t need to be null-terminated.

t

Array element C type (always char).

n

Field name.

e

Argument expression.

s

Number of elements.

ctf_sequence(t, n, e, T, E)

ctf_sequence_nowrite(t, n, e, T, E)

ctf_user_sequence(t, n, e, T, E)

ctf_user_sequence_nowrite(t, n, e, T, E)

Dynamically-sized array of integers.

The type of E must be unsigned.

t

Array element C type.

n

Field name.

e

Argument expression.

T

Length expression C type.

E

Length expression.

ctf_sequence_hex(t, n, e, T, E)

ctf_user_sequence_hex(t, n, e, T, E)

Dynamically-sized array of integers, displayed in base 16.

The type of E must be unsigned.

t

Array element C type.

n

Field name.

e

Argument expression.

T

Length expression C type.

E

Length expression.

ctf_sequence_network(t, n, e, T, E)

Dynamically-sized array of integers in network byte order (big-endian), displayed in base 10.

The type of E must be unsigned.

t

Array element C type.

n

Field name.

e

Argument expression.

T

Length expression C type.

E

Length expression.

ctf_sequence_bitfield(t, n, e, T, E)

ctf_sequence_bitfield_nowrite(t, n, e, T, E)

ctf_user_sequence_bitfield(t, n, e, T, E)

ctf_user_sequence_bitfield_nowrite(t, n, e, T, E)

Dynamically-sized array of bits.

The type of e must be an integer type. s is the number of elements of such type in e, not the number of bits.

The type of E must be unsigned.

t

Array element C type.

n

Field name.

e

Argument expression.

T

Length expression C type.

E

Length expression.

ctf_sequence_text(t, n, e, T, E)

ctf_sequence_text_nowrite(t, n, e, T, E)

ctf_user_sequence_text(t, n, e, T, E)

ctf_user_sequence_text_nowrite(t, n, e, T, E)

Dynamically-sized array, displayed as text.

The string doesn’t need to be null-terminated.

The type of E must be unsigned.

The behaviour is undefined if e is NULL.

t

Sequence element C type (always char).

n

Field name.

e

Argument expression.

T

Length expression C type.

E

Length expression.

Use the _user versions when the argument expression, e, is a user space address. In the cases of ctf_user_integer*() and ctf_user_float*(), &e must be a user space address, thus e must be addressable.

The _nowrite versions omit themselves from the trace data, but are otherwise identical. This means LTTng won’t write the _nowrite fields to the recorded trace. Their primary purpose is to make some of the event context available to the recording event rule filters without having to commit the data to sub-buffers.

Glossary

Terms related to LTTng and to tracing in general:

action

The part of a trigger which LTTng executes when the trigger condition is satisfied.

Babeltrace

The Babeltrace project, which includes:

buffering scheme

A layout of sub-buffers applied to a given channel.

channel

An entity which is responsible for a set of ring buffers.

Recording event rules are always attached to a specific channel.

clock

A source of time for a tracer.

condition

The part of a trigger which must be satisfied for LTTng to attempt to execute the trigger actions.

consumer daemon

A program which is responsible for consuming the full sub-buffers and write them to a file system or send them over the network.

current trace chunk

A trace chunk which includes the current content of all the sub-buffers of the recording session and the stream files produced since the latest event amongst:

discard mode

The event record loss mode in which the tracer discards new event records when there’s no sub-buffer space left to store them.

event

The execution of an instrumentation point, like a tracepoint that you manually place in some source code, or a Linux kprobe.

When an instrumentation point is executed, LTTng creates an event.

When an event rule matches the event, LTTng executes some action, for example:

event name

The name of an event, which is also the name of the event record.

This is also called the instrumentation point name.

event record

A record (binary serialization), in a trace, of the payload of an event.

The payload of an event record has zero or more fields.

event record loss mode

The mechanism by which event records of a given channel are lost (not recorded) when there’s no sub-buffer space left to store them.

event rule

Set of conditions which an event must satisfy for LTTng to execute some action.

An event rule is said to match events, like a regular expression matches strings.

A recording event rule is a specific type of event rule of which the action is to record the event to a sub-buffer.

inclusion set

In the process attribute inclusion set context: a set of process attributes of a given type.

instrumentation

The use of LTTng probes to make a kernel or user application traceable.

instrumentation point

A point in the execution path of a kernel or user application which, when executed, create an event.

instrumentation point name

See event name.

java.util.logging

The core logging facilities of the Java platform.

log4j

A logging library for Java developed by the Apache Software Foundation.

log level

Level of severity of a log statement or user space instrumentation point.

LTTng

The Linux Trace Toolkit: next generation project.

lttng

A command-line tool provided by the LTTng-tools project which you can use to send and receive control messages to and from a session daemon.

lttng-consumerd

The name of the consumer daemon program.

lttng-crash

A utility provided by the LTTng-tools project which can convert ring buffer files (usually saved on a persistent memory file system) to trace files.

See lttng-crash(1).

LTTng Documentation

This document.

LTTng live

A communication protocol between the relay daemon and live readers which makes it possible to show or analyze event records “live”, as they’re received by the relay daemon.

LTTng-modules

The LTTng-modules project, which contains the Linux kernel modules to make the Linux kernel instrumentation points available for LTTng tracing.

lttng-relayd

The name of the relay daemon program.

lttng-sessiond

The name of the session daemon program.

LTTng-tools

The LTTng-tools project, which contains the various programs and libraries used to control tracing.

LTTng-UST

The LTTng-UST project, which contains libraries to instrument user applications.

LTTng-UST Java agent

A Java package provided by the LTTng-UST project to allow the LTTng instrumentation of java.util.logging and Apache log4j 1.2 logging statements.

LTTng-UST Python agent

A Python package provided by the LTTng-UST project to allow the LTTng instrumentation of Python logging statements.

overwrite mode

The event record loss mode in which new event records overwrite older event records when there’s no sub-buffer space left to store them.

per-process buffering

A buffering scheme in which each instrumented process has its own sub-buffers for a given user space channel.

per-user buffering

A buffering scheme in which all the processes of a Unix user share the same sub-buffers for a given user space channel.

process attribute

In the process attribute inclusion set context:

  • A process ID.

  • A virtual process ID.

  • A Unix user ID.

  • A virtual Unix user ID.

  • A Unix group ID.

  • A virtual Unix group ID.

record (noun)

See event record.

record (verb)

Serialize the binary payload of an event to a sub-buffer.

recording event rule

Specific type of event rule of which the action is to record the matched event to a sub-buffer.

recording session

A stateful dialogue between you and a session daemon.

recording session rotation

The action of archiving the current trace chunk of a recording session.

relay daemon

A process which is responsible for receiving the trace data which a distant consumer daemon sends.

ring buffer

A set of sub-buffers.

rotation

See recording session rotation.

session daemon

A process which receives control commands from you and orchestrates the tracers and various LTTng daemons.

snapshot

A copy of the current data of all the sub-buffers of a given recording session, saved as trace files.

sub-buffer

One part of an LTTng ring buffer which contains event records.

timestamp

The time information attached to an event when LTTng creates it.

trace (noun)

A set of:

  • One CTF metadata stream file.

  • One or more CTF data stream files which are the concatenations of one or more flushed sub-buffers.

trace (verb)

From the perspective of a tracer: attempt to execute one or more actions when emitting an event in an application or in a system.

trace chunk

A self-contained trace which is part of a recording session. Each recording session rotation produces a trace chunk archive.

trace chunk archive

The result of a recording session rotation.

LTTng doesn’t manage any trace chunk archive, even if its containing recording session is still active: you are free to read it, modify it, move it, or remove it.

Trace Compass

The Trace Compass project and application.

tracepoint

An instrumentation point using the tracepoint mechanism of the Linux kernel or of LTTng-UST.

tracepoint definition

The definition of a single tracepoint.

tracepoint name

The name of a tracepoint.

tracepoint provider

A set of functions providing tracepoints to an instrumented user application.

Not to be confused with a tracepoint provider package: many tracepoint providers can exist within a tracepoint provider package.

tracepoint provider package

One or more tracepoint providers compiled as an object file or as a shared library.

tracer

A piece of software which executes some action when it emits an event, like record it to some buffer.

tracing domain

A type of LTTng tracer.

tracing group

The Unix group which a Unix user can be part of to be allowed to control the Linux kernel LTTng tracer.

trigger

A condition-actions pair; when the condition of a trigger is satisfied, LTTng attempts to execute its actions.

user application

An application (program or library) running in user space, as opposed to a Linux kernel module, for example.