profilage_code.md

----
-title: Profilage de Code
----
-
 # Code profilers
 
 In this Wiki, we will provide a methodology to install and run some code profilers. This has been tested on a specific Conda environment.
 
 The tools that were tested are proposed:
 
-- [Score-P](https://www.vi-hps.org/projects/score-p/)
-- [TAU](https://www.cs.uoregon.edu/research/tau/home.php), which allows the profiling of `for` loops
-- [Scalasca](https://www.scalasca.org/)
-- [Salt](https://github.com/ParaToolsInc/salt)
+-   [Score-P](https://www.vi-hps.org/projects/score-p/)
+-   [TAU](https://www.cs.uoregon.edu/research/tau/home.php), which allows the profiling of `for` loops
+-   [Scalasca](https://www.scalasca.org/)
+
+In order to be able to profile `for` loops, the [Salt](https://github.com/ParaToolsInc/salt) source instrumentation tool must be installed.
 
 ## Preparing the install
 
 ### Download:
 
-- Salt: `git clone --recursive https://github.com/ParaToolsInc/salt.git`
-- Score-P: https://perftools.pages.jsc.fz-juelich.de/cicd/scorep/tags/scorep-8.4/scorep-8.4.tar.gz
-- Tau: `git clone https://github.com/UO-OACISS/tau2`
-- Scalasca: https://apps.fz-juelich.de/scalasca/releases/scalasca/2.6/dist/scalasca-2.6.1.tar.gz
+-   Salt: `git clone --recursive https://github.com/ParaToolsInc/salt.git`
+-   Score-P: [https://perftools.pages.jsc.fz-juelich.de/cicd/scorep/tags/scorep-8.4/scorep-8.4.tar.gz](https://perftools.pages.jsc.fz-juelich.de/cicd/scorep/tags/scorep-8.4/scorep-8.4.tar.gz)
+-   Tau: `gh repo clone barriern/tau2`
+-   Scalasca: [https://apps.fz-juelich.de/scalasca/releases/scalasca/2.6/dist/scalasca-2.6.1.tar.gz](https://apps.fz-juelich.de/scalasca/releases/scalasca/2.6/dist/scalasca-2.6.1.tar.gz)
 
 ### Conda Environment
 
 Create the Conda environment as follows:
 
 ```
-mamba create --name profilers python=3.10 sphinx mpich-mpifort mpich-mpicc mpich-mpicxx netcdf4 six c-compiler compilers cxx-compiler clangdev llvmdev cmake
+mamba create --name profilers python=3.10 sphinx mpich-mpifort=3 mpich-mpicc=3 mpich-mpicxx=3 netcdf4 six c-compiler compilers cxx-compiler clangdev flang llvmdev cmake
 ```
 
 Activate the environment:
@@ -39,22 +36,30 @@ conda activate profilers
 Go to the `$CONDA_PREFIX/lib` folder and type:
 
 ```
-mv libbfd.a save-libbfd.a
+mkdir save
+mv libbfd.a libsframe.a save
 ```
 
-**This trick allows to avoid the use of the Conda `BFD` library and will force the use of the Score-P BFD library**
+**This trick allows to avoid the use of the Conda `BFD` library and will force the use of the Score-P and Tau BFD downloaded libraries**
 
-### Environment variables
+## Compilation
 
-Next, create an environment variable that provides the install path where profiling libraries and binaries will be stored:
+### Compilation Score-P
 
-```
-export PROFILER_PATH=/home/BARRIER/Softwares/profilers/install
-```
+Compilation of Score-P only works with GNU compilers.
 
-Next, create some environment variables that will be used all along the compilation process:
+We will therefore need to define some environment variables:
 
 ```
+# Install folder for Score-P
+export PROFILER_PATH=/home/BARRIER/Softwares/profilers/install-scorep
+
+# Change the compiler in MPICH wrappers:
+export MPICH_CXX=g++
+export MPICH_CC=gcc
+export MPICH_FC=gfortran
+
+# Define compiler variables
 export CC=gcc
 export FC=gfortran
 export F77=gfortran
@@ -63,56 +68,72 @@ export MPICC=mpicc
 export MPICXX=mpic++
 export MPIFORT=mpifort
 export FCFLAGS="-ffree-form"
-export LD_LIBRARY_PATH=$CONDA_PREFIX/lib:$PROFILER_PATH/lib
+
+# Define library path and path
+export LD_LIBRARY_PATH=$PROFILER_PATH/lib:$CONDA_PREFIX/lib
 export PATH=$PROFILER_PATH/bin:$PATH
 export PATH=$PROFILER_PATH/x86_64/bin/:$PATH
 ```
 
-## Compilation
+When the variables are defined, go the Score-P folder, activate the `profilers` environment and type:
 
-### Compilation Salt
+```
+./configure --prefix=$PROFILER_PATH --without-shmem --with-libbfd=download --with-libunwind=download
+make
+make install
+```
+
+### Compilation Tau/Salt
 
-Type the following: 
+To allow for source instrumentation using Tau, we need to use the clang compilers. Therefore, we will need do define some variables:
 
 ```
+# Install folder for Tau
+export PROFILER_PATH=/home/BARRIER/Softwares/profilers/install-tau
+
+# Change the compiler in MPICH wrappers:
+export MPICH_CXX=clang++
+export MPICH_CC=clang
+export MPICH_FC=gfortran
+
+# Define compiler variables
 export CC=clang
+export FC=flang-new
+export F77=flang-new
 export CXX=clang++
-# configure SALT and generate build system
-cmake -Wdev -Wdeprecated -S . -B $PROFILER_PATH
-cmake --build $PROFILER_PATH --parallel # Add --verbose to debug something going wrong
-```
+export MPICC=mpicc
+export MPICXX=mpic++
+export MPIFORT=mpifort
+export FCFLAGS="-ffree-form"
 
-### Compilation PDT
+# Define library path and path
+export LD_LIBRARY_PATH=$PROFILER_PATH/lib:$CONDA_PREFIX/lib
+export PATH=$PROFILER_PATH/bin:$PATH
+export PATH=$PROFILER_PATH/x86_64/bin/:$PATH
+```
 
-Go the PDT folder type:
+When the variables are defined, go to the Tau folder and type:
 
 ```
-./configure -prefix=$PROFILER_PATH
-make
+./configure -prefix=$PROFILER_PATH/ -c++=clang++ -cc=clang -fortran=flang-new -bfd=download -unwind=download
 make install
-```
-
-### Compilation Score-P
 
-Go the Score-P folder, activate the `profilers` environment and type:
-
-```
-./configure --prefix=$PROFILER_PATH --with-mpi=mpich4 --without-shmem --with-libbfd=download --with-libunwind=download --enable-shared
-make
+./configure -prefix=$PROFILER_PATH/ -c++=clang++ -cc=clang -fortran=flang-new -bfd=download -unwind=download -mpi
 make install
 ```
 
-### Compilation TAU
+> **CAUTION** In TAU, each set of compile options generate a specific Makefile that will be used for the profiling. In the above, the generated Makefile will include PDT, Score-P and MPI. To compile a non-MPI program, re-compile TAU without the `-mpi` option
+
+### Compilation Salt
 
-In the TAU folder, activate the `profilers` environment and type
+In order to install the Salt source instrumentation tool, go to the Salt folder and type the following:
 
 ```
-./configure -prefix=$PROFILER_PATH/ -pdt=$PROFILER_PATH/ -scorep=$PROFILER_PATH -mpi -c++=g++ -cc=gcc -fortran=gfortran -bfd=$PROFILER_PATH/vendor/libbfd -unwind=$PROFILER_PATH/vendor/libunwind
-make install
+# configure SALT and generate build system
+cmake -Wdev -Wdeprecated -S . -B $PROFILER_PATH -D CMAKE_C_COMPILER=clang -D CMAKE_CXX_COMPILER=clang++
+cmake --build $PROFILER_PATH --parallel # Add --verbose to debug something going wrong
 ```
 
-> **CAUTION** In TAU, each set of compile options generate a specific Makefile that will be used for the profiling. In the above, the generated Makefile will include PDT, Score-P and MPI. To compile a non-MPI program, re-compile TAU without the `-mpi` option
-
 ### Compilation Scalasca
 
 Go to your Scalasca folder and type:
@@ -151,7 +172,7 @@ In the above, the `-optKeepFiles` allows to keep intermediate files. The `-optTa
 BEGIN_EXCLUDE_LIST
 
   void quicksort(int *, int, int)
-  # The next line excludes all functions beginning with "sort_" and having 
+  # The next line excludes all functions beginning with "sort_" and having
   # arguments "int *"
   void sort_#(int *)
   void interchange(int *, int *)
@@ -188,10 +209,10 @@ Here, we specify that loops will be instrumented in all the functions of the pro
 
 Now, to compile the program, replace the standard compile commands as follows:
 
-- `gcc` -\> `tau_cc.sh`
-- `g++` -\> `tau_cxx.sh`
-- `gfortran` -\> `tau_f90.sh`
-- `f77` -\> `tau_f70.sh`
+-   `gcc` -\> `tau_cc.sh`
+-   `g++` -\> `tau_cxx.sh`
+-   `gfortran` -\> `tau_f90.sh`
+-   `f77` -\> `tau_f70.sh`
 
 For example:
 
@@ -217,13 +238,13 @@ paraprof scorep-20250310_1624_81500554863145/profile.cubex &
 
 To allow profiling with Score-P, replace the standard compile commands as follows:
 
-- `g++` -\> `scorep g++`
-- `gfortran` -\> `scorep gfortran`
-- `mpicxx` -\> `scorep mpicxx`
-- `mpif90` -\> `scorep mpif90`
-- `gcc` -\> `scorep gcc`
-- `mpicc` -\> `scorep mpicc`
-- `mpif77` -\> `scorep mpif77`
+-   `g++` -\> `scorep g++`
+-   `gfortran` -\> `scorep gfortran`
+-   `mpicxx` -\> `scorep mpicxx`
+-   `mpif90` -\> `scorep mpif90`
+-   `gcc` -\> `scorep gcc`
+-   `mpicc` -\> `scorep mpicc`
+-   `mpif77` -\> `scorep mpif77`
 
 For example
 
@@ -273,13 +294,13 @@ paraprof scorep-20250310_1624_81500554863145/profile.cubex &
 
 To profile a code with Scalasca, you need to compile the code by replacing:
 
-- `g++` -\> `skin g++`
-- `gfortran` -\> `skin gfortran`
-- `mpicxx` -\> `skin mpicxx`
-- `mpif90` -\> `skin mpif90`
-- `gcc` -\> `skin gcc`
-- `mpicc` -\> `skin mpicc`
-- `mpif77` -\> `skin mpif77`
+-   `g++` -\> `skin g++`
+-   `gfortran` -\> `skin gfortran`
+-   `mpicxx` -\> `skin mpicxx`
+-   `mpif90` -\> `skin mpif90`
+-   `gcc` -\> `skin gcc`
+-   `mpicc` -\> `skin mpicc`
+-   `mpif77` -\> `skin mpif77`
 
 Once your code has been compiled, you can execute it as follows:
 
@@ -293,9 +314,9 @@ scan test.exe
 scan -t test.exe
 ```
 
----
+------------------------------------------------------------------------
 
 Resources:
 
-- https://hpc.cea.fr/tgcc-public/en/html/toc/fulldoc/Profiling.html#
-- https://www.vi-hps.org/projects/score-p/
\ No newline at end of file
+-   https://hpc.cea.fr/tgcc-public/en/html/toc/fulldoc/Profiling.html#
+-   https://www.vi-hps.org/projects/score-p/
\ No newline at end of file