How to build

From Medusa: Coordinate Free Mehless Method implementation
Revision as of 12:47, 31 May 2018 by Jureslak (talk | contribs) (Installation)

Jump to: navigation, search

Installation

To make this work from plain Ubuntu installation, run

sudo apt-get install cmake doxygen graphviz libboost-dev libhdf5-serial-dev
git clone https://gitlab.com/e62Lab/e62numcodes.git
cd e62numcodes
./run_tests.sh

which installs dependencies, clones the repository, goes into the root folder of the repository and runs tests. This will check the configuration, notify you of potentially missing dependencies, build and run all tests, check code style and docs. If this works, you are ready to go! Otherwise, install any missing packages and if it still fails, raise an issue!

Building

List of dependencies:

Out of source builds are preferred. Run

mkdir -p build
cd build
cmake ..
make

Note that you only have to run cmake once, after that only make is sufficient.

Binaries are placed into bin folder. Test can be run all at once via make run_all_tests or individually via eg. make basisfunc_run_tests.

Linear Algebra

We use Eigen as our matrix library. See here for use reference and documentation. For a quick transition from Matlab see here.

Drawing

Some tests include drawing. We are using SMFL library, which can be installed on most linux systems easly as sudo apt-get install libsfml-dev or sudo pacman -S sfml. After the installation uncomment a test case in domain_draw_test.cpp and run make test_domain_draw to see the visual effect.

Binaries using SFML require additional linker flags -lsfml-graphics -lsfml-window -lsfml-system, but the makefile should take care of that for you.

HDF5

In order to use IO you need hdf5 library. You can install it easily using the command sudo apt-get install libhdf5- dev or sudo pacman -S hdf5-cpp-fortran.

Ubuntu places hdf5 libs in a werid folder /usr/lib/x86_64-linux-gnu/hdf5/serial/. If you get error similar to -lhdf5 not found and you have hdf5 installed, you might have to link the libraries into a discoverable place, like /usr/lib/ or add the above directory to the linker path. If using cmake, you can add the following line to your CMakeLists.txt

 link_directories(/usr/lib/x86_64-linux-gnu/hdf5/serial/)

Using Intel Math Kernel Library (MKL)

Eigen has great support for MKL all you have to do is define a EIGEN_USE_MKL_ALL macro before any includes. You can see further instructions on their website.

Besides setting #define EIGEN_USE_MKL_ALL in your code, some linker and compilation fixes are needed. You have to set MKL and MKLROOT variables in cmake. You can define the variable MKLROOT as a system variable (using export) which is enough. You can also define it manually when calling cmake. If it is not set in either way it will default to "/opt/intel/compilers_and_libraries_2017.2.174/linux/mkl".

cmake .. -DMKL=ON -DMKLROOT=/opt/intel/compilers_and_libraries_2016.1.150/linux/mkl

Your target has to be linked with some MKL libraries so make sure to add the following link to your cmake file.

target_link_libraries(target ${LMKL})

Building on Mac OSX

This method was tested on El Capitan. Linking the OpenMP library is still not functioning as intended.

First install all dependencies from homebrew

brew install llvm cmake homebrew/science/hdf5 SFML

Now you can clone and build the project with CLang using the following commands

git clone https://gitlab.com/e62Lab/e62numcodes.git
cd e62numcodes
mkdir build
cd build
cmake .. -DCMAKE_C_COMPILER=/usr/local/opt/llvm/bin/clang -DCMAKE_CXX_COMPILER=/usr/local/opt/llvm/bin/clang++
make domain_run_tests

Using Intel C/C++ Compiler

In order to use Intel's compiler you have to first set the CXX and CC bash variables. Before calling cmake for the first time you have to export the following:

export CXX="icpc"
export CC="icc"

or you can define the compiler when first calling cmake like so:

cmake .. -DCMAKE_C_COMPILER=$(which icc) -DCMAKE_CXX_COMPILER=$(which icpc)


You can also compile it directly for Intel® Xeon Phi™ Coprocessor. You do this by adding -Dmmic=ON flag to the cmake command:

cmake .. -Dmmic=ON -DCMAKE_C_COMPILER=$(which icc) -DCMAKE_CXX_COMPILER=$(which icpc)

Note: All features that depend on system third-party libraries are not available on MIC (Many Integrated Core). This includes: