https://github.com/fsquillace/cusp

Improved version of CUSP Parallel Algorithms for Sparse Matrix
https://github.com/fsquillace/cusp

Last synced: 23 days ago
JSON representation

Improved version of CUSP Parallel Algorithms for Sparse Matrix

• Host: GitHub
• URL: https://github.com/fsquillace/cusp
• Owner: fsquillace
• License: apache-2.0
• Created: 2012-03-31T11:00:21.000Z (almost 13 years ago)
• Default Branch: master
• Last Pushed: 2012-04-22T16:54:15.000Z (over 12 years ago)
• Last Synced: 2024-11-01T22:07:02.300Z (2 months ago)
• Language: C
• Homepage:
• Size: 1.59 MB
• Stars: 3
• Watchers: 2
• Forks: 1
• Open Issues: 2
• Metadata Files:
  • Readme: README
  • Changelog: CHANGELOG
  • License: LICENSE

Awesome Lists containing this project

README

        
Refer to the QuickStartGuide for detailed installation and usage instructions:

http://code.google.com/p/cusp-library/wiki/QuickStartGuide

Instructions for the Impatient

------------------------------

    0) Verify that CUDA 4.x is installed

        $ nvcc --version

        nvcc: NVIDIA (R) Cuda compiler driver

        Copyright (c) 2005-2011 NVIDIA Corporation

        Built on Fri_Jan_13_01:52:47_PST_2012

        Cuda compilation tools, release 4.1, V0.2.1221

    1) Download Cusp v0.3.x

        http://cusp-library.googlecode.com/files/cusp-v0.3.0.zip

    

    2) Extract contents of ZIP files to the CUDA include directory:

        /usr/local/cuda/include/ on a Linux and Mac OSX

        C:\CUDA\include\ on a Windows system

    

    3) Save the following program to a file named version.cu

        #include 

        #include 

        #include 

        int main(void)

        {

            std::cout << "The following libraries were found:" << std::endl;

            std::cout << "    CUDA   v" << (CUDA_VERSION / 1000) << "." <<

                                           (CUDA_VERSION % 1000) / 10 << std::endl;

            std::cout << "    Thrust v" << THRUST_MAJOR_VERSION << "." << 

                                           THRUST_MINOR_VERSION << "." << 

                                           THRUST_SUBMINOR_VERSION << std::endl;

            std::cout << "    Cusp   v" << CUSP_MAJOR_VERSION << "." << 

                                           CUSP_MINOR_VERSION << "." << 

                                           CUSP_SUBMINOR_VERSION << std::endl;

            return 0;

        }

    4) Compile version.cu with nvcc to verify installation

        $ nvcc version.cu -o version

        $ ./version

        The following libraries were found:

            CUDA   v4.1

            Thrust v1.5.1

            Cusp   v0.3.0