Commits · 6110f2419404040da2a85b0ef1600ef657240ac6 · Kaushik Kulkarni / viennacl-dev · GitLab

Jan 16, 2013
- * Fixed issues with copy back to host when OpenCL handles are passed to CTORs... · 6110f241
  Karl Rupp authored Jan 16, 2013
  6110f241
Dec 14, 2012
- * Fixed bug in copy() to hyb_matrix as reported by Denis Demidov (thanks!) · de6376ab
  Karl Rupp authored Dec 14, 2012
  de6376ab
Dec 09, 2012
- Added SSE code contributed by Alex Christensen. Integration into host-backend is work in progress. · 7b6f92c3
  Karl Rupp authored Dec 09, 2012
  
  7b6f92c3
Dec 02, 2012
- Merge ssh://viennacl.git.sourceforge.net/gitroot/viennacl/viennacl into sf.net · 7c985934
  Karl Rupp authored Dec 02, 2012
  
  7c985934
- * Updated README · bc2a42da
  Karl Rupp authored Dec 02, 2012
  bc2a42da
- * Added more text to manual. In particular, described memory handle · 532e81ca
  Karl Rupp authored Dec 02, 2012
  532e81ca
- * Updated 'Experimental'-status in source files. · eff2ace3
  Karl Rupp authored Dec 01, 2012
  eff2ace3
Dec 01, 2012
- Further polishing: · 94900385
  Karl Rupp authored Dec 01, 2012
  94900385
- * Transition of converter from Boost.filesystem2 to Boost.filesystem3 · c1e00f54
  Karl Rupp authored Dec 01, 2012
  c1e00f54
- Release postponed by a day: · cde5bc1a
  Karl Rupp authored Nov 30, 2012
  cde5bc1a
Nov 30, 2012
- * Removed a warning regarding uninitialized members in vector_operations.hpp · c776231b
  Karl Rupp authored Nov 30, 2012
  c776231b
- * Eliminated almost all warnings with GCC at -Wextra. Only exception: SFINAE in vector-CTOR. · 60586fd8
  Karl Rupp authored Nov 30, 2012
  60586fd8
- * Fixed remaining doxygen warnings. · 1e7310c0
  Karl Rupp authored Nov 29, 2012
  1e7310c0
Nov 29, 2012
- * Updated manual · a66b8b78
  Karl Rupp authored Nov 29, 2012
  a66b8b78
- * Updated license header in source files (adding Argonne) · 356107bc
  Karl Rupp authored Nov 29, 2012
  356107bc
- * Dense matrix-vector product now accepts matrix-ranges/slices and vector-ranges/slices as well · e233b999
  Karl Rupp authored Nov 29, 2012
  e233b999
- * Renamed viennacl::linalg::single_threaded to viennacl::linalg::host_based · 9cf4f1f2
  Karl Rupp authored Nov 29, 2012
  9cf4f1f2
Nov 28, 2012
- * Added least-squares example · 11b7e84f
  Karl Rupp authored Nov 28, 2012
  11b7e84f
- * Fixed a minor bug in lu_factorize that showed up with the CUDA backend only · 8f971c47
  Karl Rupp authored Nov 28, 2012
  8f971c47
- * Reimplementation of LU factorization in viennacl/linalg/lu.hpp. Better... · e8a6e5b3
  Karl Rupp authored Nov 27, 2012
  e8a6e5b3
Nov 21, 2012
- * Reduced overhead for copying to/from ublas::compressed_matrix<> · 68ec5e72
  Karl Rupp authored Nov 21, 2012
  68ec5e72
- * Added CUDA examples/tutorials/tests to build system · ede1ed5c
  Karl Rupp authored Nov 21, 2012
  ede1ed5c
Nov 20, 2012
- * generator: Update in how opencl handles are passed to kernels · 5ac70452
  Philippe Tillet authored Nov 20, 2012
  5ac70452
Nov 18, 2012
- * Added level scheduling to ILU0. Solver cycle times look good, but setup is still quite expensive. · 730b17be
  Karl Rupp authored Nov 17, 2012
  730b17be
Nov 16, 2012
- * Added support for row-/Jacobi-preconditioner with coordinate_matrix · 013e159c
  Karl Rupp authored Nov 16, 2012
  013e159c
- * Added missing kernel initialization call to row_info() for OpenCL · 88598ec9
  Karl Rupp authored Nov 15, 2012
  88598ec9
Nov 15, 2012
- * Added element-wise operations for vectors · 119785b2
  Karl Rupp authored Nov 15, 2012
  119785b2
- Simplified implementation of inner_prod(). Might yield better performance on AMD GPUs. · 6cca4eb9
  Karl Rupp authored Nov 15, 2012
  
  6cca4eb9
- * Typesafe multi-backend transfer now working, making implementations based... · 8213cb0c
  Karl Rupp authored Nov 14, 2012
  8213cb0c
Nov 14, 2012
- Implemented support for typesafe cross-domain transfer of memory buffers. More testing required. · 1c0f1224
  Karl Rupp authored Nov 14, 2012
  
  1c0f1224
- * Pimped incomplete Cholesky factorization. Speed now comparable to ILU. · 7510910c
  Karl Rupp authored Nov 13, 2012
  7510910c
Nov 13, 2012
- Added first implementation of incomplete Cholesky preconditioner. Requires improvements. · 5118cac9
  Karl Rupp authored Nov 13, 2012
  
  5118cac9
- Added missing diagonal_assign_cpu-kernels for matrices. Initializer types now all working. · a4d0d439
  Karl Rupp authored Nov 13, 2012
  
  a4d0d439
- * Added matrix initializers (work for CPU and OpenCL, CUDA-testing required) · 32af3402
  Karl Rupp authored Nov 12, 2012
  32af3402
Nov 12, 2012
- Transfer RAM<->OpenCL<->CUDA<->RAM now implemented. · 24b5e039
  Karl Rupp authored Nov 12, 2012
  
  24b5e039
- Pimped ILU(0,T)-preconditioners, up to one order of magnitude faster with new... · 1b975b43
  Karl Rupp authored Nov 11, 2012
  1b975b43
Nov 11, 2012
- Reimplementation of ILU0 for compressed_matrix. Using the new multi-backend,... · 3a53cf71
  Karl Rupp authored Nov 11, 2012
  3a53cf71
- Sparse triangular solvers for compressed_matrix now working (CPU, OpenCL, CUDA). · 113e0c7f
  Karl Rupp authored Nov 11, 2012
  
  113e0c7f
- Sparse triangular solver for compressed_matrix now working on CPU. Moved in... · 49010db8
  Karl Rupp authored Nov 10, 2012
  49010db8
Nov 09, 2012
- Working on sparse triangular solvers. Performance still not great, but soon sufficient for ILU. · 98af43d5
  Karl Rupp authored Nov 09, 2012
  
  98af43d5