Commits · ed2d1d556719d1bb9a20e1c816a602eb1a12c026 · Kaushik Kulkarni / viennacl-dev · GitLab

Dec 15, 2013
- libviennacl: Cleanup: Applied pimpl-idiom to separate implementation from viennacl.hpp · ed2d1d55
  Karl Rupp authored Dec 15, 2013
  ed2d1d55
- libviennacl: Added extensive tests for GEMM and fixed a couple of interfacing bugs. · 1cb3a844
  Karl Rupp authored Dec 15, 2013
  
  1cb3a844
Dec 14, 2013
- libviennacl: Added GEMM for host, CUDA, and OpenCL (no tests yet). · d86cbd6e
  Karl Rupp authored Dec 14, 2013
  d86cbd6e
- OpenMP: Fixed use of unsigned integer type for outer loop in matrix-matrix products. · 7ad5cb89
  Karl Rupp authored Dec 14, 2013
  
  7ad5cb89
- SVD: Fixed flaw in test. · d3817f26
  Karl Rupp authored Dec 14, 2013
  d3817f26
- QR method: Fixed implementation and tests for non-symmetric matrices. · a0593b71
  Karl Rupp authored Dec 14, 2013
  
  a0593b71
- NMF: Added flag for controlling output to std::cout, reduced test execution time. · 815f34e7
  Karl Rupp authored Dec 14, 2013
  815f34e7
Dec 13, 2013
- Merge branch 'master' of https://github.com/albertz/viennacl-dev into albertz-master · 41412745
  Karl Rupp authored Dec 13, 2013
  41412745
- simple OpenMP implementation for matrix-matrix product · 078c83d9
  Albert Zeyer authored Dec 13, 2013
  
  078c83d9
- README: Updated version for GitHub to reflect that there is no more OpenCL... · 4db6d47d
  Karl Rupp authored Dec 13, 2013
  4db6d47d
- Tests: Switched back to building tests by default for developer version and in the source release. · 84900bf2
  Karl Rupp authored Dec 13, 2013
  84900bf2
Dec 11, 2013
- SpMdM: Added CUDA implementation for hyb_matrix, thus resolving #22 · e1617c38
  Karl Rupp authored Dec 11, 2013
  e1617c38
Dec 10, 2013
- SpMdM: Added support for hyb_matrix on CPU and OpenCL. · b4442cd3
  Karl Rupp authored Dec 10, 2013
  b4442cd3
Dec 09, 2013
- SpMdM: Added support for arbitrary layouts of dense factor and result when using CUDA. · 2b74801c
  Karl Rupp authored Dec 09, 2013
  2b74801c
- SpMdM: Added support for arbitrary layouts of dense factor and result when using OpenCL. · 2bc84594
  Karl Rupp authored Dec 09, 2013
  2bc84594
Dec 08, 2013
- SpMdM: Big steps towards supporting row-major and col-major layout for both result and factor. · 8ed0c69e
  Karl Rupp authored Dec 08, 2013
  8ed0c69e
- SpMdM: Added implementation for coordinate_matrix. · 12c1c548
  Karl Rupp authored Dec 08, 2013
  12c1c548
- libviennacl: Fixed compilation problems with CUDA and unified integer type. · c5be4ba1
  Karl Rupp authored Dec 08, 2013
  c5be4ba1
- COO OpenCL: Removed assert() in SpMV. · 18f139e1
  Karl Rupp authored Dec 08, 2013
  18f139e1
- Matrix Market: Removed status message to std::cout in reader. · 192b141f
  Karl Rupp authored Dec 08, 2013
  192b141f
- OpenCL kernels: Merged code generation for CSR times dense matrices. · 346577a9
  Karl Rupp authored Dec 08, 2013
  
  346577a9
- Spmdm: Stabilized test with respect to round-off errors. · ee720eb2
  Karl Rupp authored Dec 08, 2013
  ee720eb2
- Multithreaded: Updated tutorials to work with older Boost versions. · fd2497b6
  Karl Rupp authored Dec 08, 2013
  fd2497b6
- uBLAS: Compatibility fixes in order to compile with Boost 1.33.0 · d5575880
  Karl Rupp authored Dec 08, 2013
  
  d5575880
- CMake: Now also supporting CMake 2.6.x · 37c923f8
  Karl Rupp authored Dec 08, 2013
  37c923f8
- ViennaCLCommon.cmake: Source-package no longer builds test by default. · 9ecb61cc
  Karl Rupp authored Dec 08, 2013
  9ecb61cc
- CMake, Boost: Removed linking with filesystem, linking with system only with Boost 1.35.0 and above · 56342360
  Karl Rupp authored Dec 08, 2013
  56342360
Dec 06, 2013
- Sparse-Dense-Mult: Fixed race condition when using OpenMP. · 4f1d636c
  Karl Rupp authored Dec 06, 2013
  
  4f1d636c
Dec 05, 2013
- Spmdm test: Now testing both single and double precision. · 6854dd29
  Karl Rupp authored Dec 05, 2013
  
  6854dd29
- compressed_compressed_matrix: Fixed race condition in SpMV. · 1bf62bfd
  Karl Rupp authored Dec 05, 2013
  1bf62bfd
- coordinate_matrix: Removed left-over asserts for SpMV with strided vectors. Tests pass now. · a7ff9caf
  Karl Rupp authored Dec 05, 2013
  
  a7ff9caf
- coordinate_matrix: Fixed group boundaries for small matrices, fixed some corner cases in SpMV · deae066e
  Karl Rupp authored Dec 05, 2013
  deae066e
- Typedef'd uses of std::size_t and size_t to vcl_size_t. · 87e8c233
  Karl Rupp authored Dec 05, 2013
  87e8c233
Nov 30, 2013
- libviennacl: Fixed bug in tests (no proper dynamic allocation) · 97eb55e8
  Karl Rupp authored Nov 30, 2013
  
  97eb55e8
- Reordering: Fixed warnings on GCC · 38c061cb
  Karl Rupp authored Nov 30, 2013
  
  38c061cb
- Reordering: Improved speed of advanced Cuthill-McKee by 40%, inverted permutations · a7706964
  Karl Rupp authored Nov 30, 2013
  a7706964
- Cuthill-McKee: Tweaked implementation, resulting in a reduction of execution times by 40 percent. · a9afb67d
  Karl Rupp authored Nov 30, 2013
  a9afb67d
- libviennacl: Introduced ViennaCLInt as main integer type for OpenMP and... · 24e6faf8
  Karl Rupp authored Nov 30, 2013
  24e6faf8
Nov 28, 2013
- Merge pull request #50 from albertz/master · 87f6d2b2
  Karl Rupp authored Nov 28, 2013
  87f6d2b2
- fix a few implicit conversion warnings · 17b3541c
  Albert Zeyer authored Nov 28, 2013
  
  17b3541c