Giter VIP home page Giter VIP logo

sjunwl / suitesparse-metis-for-windows Goto Github PK

View Code? Open in Web Editor NEW

This project forked from jlblancoc/suitesparse-metis-for-windows

0.0 0.0 0.0 60.21 MB

CMake scripts for painless usage of SuiteSparse+METIS from Visual Studio and the rest of Windows/Linux/OSX IDEs supported by CMake

License: BSD 3-Clause "New" or "Revised" License

C 69.76% Python 0.06% C++ 8.64% CMake 0.38% Makefile 1.41% Fortran 0.68% TeX 3.82% MATLAB 11.78% Shell 0.06% Awk 0.01% Cuda 0.32% Assembly 0.01% Java 1.06% HTML 0.39% Perl 0.04% Ruby 0.20% Objective-C 1.33% D 0.02% Batchfile 0.01% Roff 0.04%

suitesparse-metis-for-windows's Introduction

Build Status Grunt status

CMake scripts for painless usage of Tim Davis' SuiteSparse (CHOLMOD,UMFPACK,AMD,LDL,SPQR,...) and METIS from Visual Studio and the rest of Windows/Linux/OSX IDEs supported by CMake. The project includes precompiled BLAS/LAPACK DLLs for easy use with Visual C++. Licensed under BSD 3-Clause License.

The goal is using one single CMake code to build against SuiteSparse in standard Linux package systems (e.g. libsuitesparse-dev) and in manual compilations under Windows.

Credits: Jose Luis Blanco (Universidad de Almeria); Jerome Esnault (INRIA); @NeroBurner

logo

1. Instructions

  • (1) Install CMake.

  • (2) Only for Linux/Mac: Install LAPACK & BLAS. In Debian/Ubuntu: sudo apt-get install liblapack-dev libblas-dev

  • (3) Clone or download this project (latest release) and extract it somewhere in your disk, say SP_ROOT.

    • (OPTIONAL) CMake will download SuiteSparse sources automatically for you (skip to step 4), but you may do it manually if preferred:
      • Populate the directories within SP_ROOT with the original sources from each project:
        • SuiteSparse:
          • Download SuiteSparse-X.Y.Z.tar.gz from Tim Davis' webpage.
          • Extract it.
          • Merge (e.g. copy and paste from Windows Explorer) the tree SuiteSparse/* into SP_ROOT/SuiteSparse/*.
          • Make sure of looking for patches in the original webpage and apply them to prevent build errors.
        • METIS: (Optional, only if need METIS for partitioning)
          • Download metis-X.Y.Z.tar.gz.
          • Extract it.
          • Merge the tree metis-X.Y.Z/* into SP_ROOT/metis/*.
          • Add the command cmake_policy(SET CMP0022 NEW) right after the line project(METIS) in metis/CMakeLists.txt.
  • (4) Run CMake (cmake-gui), then:

    • Set the "Source code" directory to SP_ROOT
    • Set the "Build" directory to any empty directory, typically SP_ROOT/build
    • Press "Configure", change anything (if needed)
    • Important: I recommend setting the CMAKE_INSTALL_PREFIX to some other directory different than "Program Files" or "/usr/local" so the INSTALL command does not require Admin privileges. By default it will point to SP_ROOT/build/install.
    • If you have an error like: "Cannot find source file: GKlib/conf/check_thread_storage.c", then manually adjust GKLIB_PATH to the correct path SP_ROOT/metis/GKlib.
    • HAVE_COMPLEX is OFF by default to avoid errors related to complex numbers in some compilers.
    • Press "Generate".
  • (5) Compile and install:

    • In Visual Studio, open SuiteSparseProject.sln and build the INSTALL project in Debug and Release. You may get hundreds of warnings, but it's ok.
    • In Unix: Just execute make install or sudo make install if you did set the install prefix to /usr/*
  • (6) Notice that a file SuiteSparseConfig.cmake should be located in your install directory. It will be required for your programs to correctly build and link against SuiteSparse.

  • (7) Only for Windows: You will have to append CMAKE_INSTALL_PREFIX\lib*\lapack_blas_windows\ and CMAKE_INSTALL_PREFIX\lib* to the environment variable PATH before executing any program, for Windows to localize the required BLAS/Fortran libraries (.DLLs).

2. Test program

Example CMake programs are provided for testing, based on Tim Davis' code in his manual:

An example to test CUDA support can be found here.

3. Integration in your code (unique code for Windows/Linux)

  • Add a block like this to your CMake code (see complete example), and set the CMake variable SuiteSparse_DIR to SP_INSTALL_DIR/lib/cmake/suitesparse-5.4.0/ or the equivalent place where SuiteSparse-config.cmake was installed.

    # ------------------------------------------------------------------
    # Detect SuiteSparse libraries:
    # If not found automatically, set SuiteSparse_DIR in CMake to the
    # directory where SuiteSparse-config.cmake was installed.
    # ------------------------------------------------------------------
    find_package(SuiteSparse CONFIG REQUIRED)
    target_link_libraries(MY_PROGRAM PRIVATE ${SuiteSparse_LIBRARIES})
    
    # or directly add only the libs you need:
    target_link_libraries(MY_PROGRAM PRIVATE
      	SuiteSparse::suitesparseconfig
      	SuiteSparse::amd
      	SuiteSparse::btf
      	SuiteSparse::camd
      	SuiteSparse::ccolamd
      	SuiteSparse::colamd
      	SuiteSparse::cholmod
      	SuiteSparse::cxsparse
      	SuiteSparse::klu
      	SuiteSparse::ldl
      	SuiteSparse::umfpack
      	SuiteSparse::spqr
      	SuiteSparse::metis
      )
    
  • In Windows, if you have a CMake error like:

    Could not find a package configuration file provided by "LAPACK" with any
     of the following names:
    
       LAPACKConfig.cmake
       lapack-config.cmake
    

    set the CMake variable LAPACK_DIR to SP_ROOT/lapack_windows/x64/ (or x32 for 32bit builds).

4. Why did you create this project?

Porting SuiteSparse to CMake wasn't trivial because this package makes extensive usage of a one-source-multiple-objects philosophy by compiling the same sources with different compiler flags, and this ain't directly possible to CMake's design.

My workaround to this limitation includes auxiliary Python scripts and dummy source files, so the whole thing became large enough to be worth publishing online so many others may benefit.

suitesparse-metis-for-windows's People

Contributors

jlblancoc avatar neroburner avatar riegl-gc avatar xoviat avatar dtmoodie avatar nightlark avatar awf avatar andr3wmac avatar kinddragon avatar simbaforrest avatar saedrna avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.