About the toolkit

The artdaq toolkit is a data-acquisition framework designed for high-energy physics experiments. It provides a flexible, reliable backbone for data transfers and has several locations where users can perform custom analysis tasks using the art framework.

The artdaq suite consists of the following packages:

trace: High-performance message logging
artdaq-core: Data formats used by the artdaq toolkit
artdaq-utilities: Online tools, primarily metrics reporting
artdaq-mfextensions: Extensions to the MessageFacility product which are useful in DAQ context
artdaq: Application and data transfer framework
artdaq-core-demo: Data formats used by the artdaq demonstration system
artdaq-demo: "User" implementations for the artdaq demonstration system
artdaq-daqinterface: Command line run control and example configurations
artdaq-database: Bindings for MongoDB or local "filesystemdb" configuration databases
artdaq-epics-plugin: Metric endpoint for the EPICS control system

About this package

This package consists of the TRACE high-performance message logging package. It can be built and used independently from the rest of the artdaq suite.

To get started, see the quick-start.txt file.

Please note that using a version of gcc that supports the c11 and/or c++11 standard is desirable.

RPM packaging provided by Pat Riehecky on the Scientific Linux team.

Public readonly GIT access via: git clone https://github.com/art-daq/trace

tar files with binaries at: https://scisoft.fnal.gov/scisoft/packages/TRACE/

Public svn export svn export http://cdcvs.fnal.gov/subversion/trace-svn/trunk/ trace

Authorized read-write SVN access via: svn co svn+ssh://p-trace@cdcvs.fnal.gov:/cvs/projects/trace-svn/trunk trace

BUILDING:

make:

cd trace
make OUT=$PWD 
PATH=$PWD/*/bin:$PATH
. script/trace_functions.sh
# if UPS environment, 2 lines above can be replaced by: setup -r$PWD-z$PWD TRACE 

cmake

cd trace
# next 4 to git :) cetmodules; used in building examples
git clone git@github.com:FNALssi/cetmodules -b3.27.03 # use tag closest to cmake version
mkdir cetmodules/build;cd $_
cmake .. -DCMAKE_INSTALL_PREFIX=$PWD && make install && export CMAKE_PREFIX_PATH=$PWD
cd ../..
mkdir build; cd build
cmake .. -DCMAKE_INSTALL_PREFIX=$PWD
make install
echo ":$PATH:" | grep :$PWD/bin: || PATH="$PWD/bin:$PATH"
export PYTHONPATH=$PWD/python LD_LIBRARY_PATH=`echo $PWD/lib*`
. etc/profile.d/trace_functions.sh
tcntl TRACE INFO hello
python -c 'import TRACE;TRACE.INFO("hello")'

For a system install, use either -DCMAKE_INSTALL_PREFIX=/usr or -DCMAKE_INSTALL_PREFIX=/. Most install destinations follow GNUInstallDirs and land under /usr either way (for example, headers in /usr/include/TRACE, binaries in /usr/bin, docs in /usr/share, and source/support trees in /usr/share/TRACE by default).

Two destinations are intentionally special:

trace_functions.sh is installed to /etc/profile.d for system integration (for -DCMAKE_INSTALL_PREFIX=/usr, also set -DCMAKE_INSTALL_SYSCONFDIR=/etc if you need /etc/profile.d).
If -DWANT_KMOD=ON, TRACE.ko installs to /lib/modules/... for system prefixes (/ and /usr). For non-system prefixes, it installs under <prefix>/lib/modules/....

To skip source/support tree installation entirely, configure with -DTRACE_INSTALL_SOURCE_TREE=OFF.

spack - work-in-progress

wget https://raw.githubusercontent.com/art-daq/artdaq_demo/refs/heads/develop/tools/setup_spack_build_system_v0.28.sh
echo "c70e6c68ec1f7fddbf4afdcd5b24a3b1d9bbb660  setup_spack_build_system_v0.28.sh" | sha1sum -c -
 
source setup_spack_build_system_v0.28.sh
# Note that install_spack_build_system sources setup-env.sh
install_spack_build_system $PWD spack 0 # Installs Spack into the spack/ subdirectory
spack find --format "{name}@{version}%{compiler}/{hash} {arch}={platform}-{os}-{target}" trace
spack install --reuse trace@develop ~mf # Don't install MessageFacility dependency
 
spack load trace@develop