Commits · bb5871e06d76eaf50087db69845aeb0a7bf7826f · Kaushik Kulkarni / viennacl-dev · GitLab

Apr 14, 2013
- Fixed issue 3608388 on sourceforge-bugtracker (copying matrix if last entry not set) · bb5871e0
  Karl Rupp authored Apr 14, 2013
  
  bb5871e0
- Improved CPU-based vector loops within the custom kernel generator · 6746adfc
  Karl Rupp authored Mar 31, 2013
  
  6746adfc
- Added improved CSR SpMV kernel for CPUs · 1f125ecc
  Karl Rupp authored Mar 30, 2013
  
  1f125ecc
- Reverted 128x128 work layout for CSR SpMV. Instead, increased search range of... · 1ceb58f2
  Karl Rupp authored Mar 29, 2013
  1ceb58f2
- Reverted to old CPU work size deduction, which is better for simple vector kernels. · 9f683d58
  Karl Rupp authored Mar 28, 2013
  9f683d58
- OpenCL work size now defaults to 128x128 on CPUs. · f9d9ddb8
  Karl Rupp authored Mar 26, 2013
  f9d9ddb8
Feb 22, 2013
- Fixed missing hint on ENABLE_PEDANTIC_FLAGS in README.md · 3f705b04
  Karl Rupp authored Feb 22, 2013
  
  3f705b04
- Added README for developer repository on github. · 7b85e311
  Karl Rupp authored Feb 22, 2013
  
  7b85e311
- Updated changelogs for 1.4.1. Added missing kernel generation file for AMD GEMM kernels. · 7dd9736a
  Karl Rupp authored Feb 21, 2013
  
  7dd9736a
- Improved detection of the Intel OpenCL SDK. · 02d72620
  Karl Rupp authored Feb 21, 2013
  
  02d72620
Feb 21, 2013
- Added (missing?) newline at end of a test file. · 87f0d7e0
  Karl Rupp authored Feb 21, 2013
  
  87f0d7e0
- Added missing preprocessor switch for old Boost.Filesystem in converter.cpp · ec4e0e93
  Karl Rupp authored Feb 21, 2013
  
  ec4e0e93
- Removed erroneous build/auxiliary/converter.cpp. · 50103ae9
  Karl Rupp authored Feb 21, 2013
  
  50103ae9
- Slightly improved Doxygen-comments. Updated manual for release. · 433a09aa
  Karl Rupp authored Feb 21, 2013
  
  433a09aa
- Benchmarks now report performance by counting additions and multiplications separately. · 439d739c
  Karl Rupp authored Feb 21, 2013
  
  439d739c
- Added support for old Boost.Filesystem (prior to 1.44) in converter script. · 3dbfeb9e
  Karl Rupp authored Feb 21, 2013
  
  3dbfeb9e
- Fixed a bug in the fast GEMM kernel for NVIDIA hardware when using strides != 1 · d34c00ba
  Karl Rupp authored Feb 21, 2013
  
  d34c00ba
- * Updated AMD GEMM kernels to include scalars alpha and beta. · f975605b
  Karl Rupp authored Feb 21, 2013
  f975605b
- Fixed remaining problems with inclusion of Philippe's GEMM kernels. Tests pass. :-) · b5d46714
  Karl Rupp authored Feb 20, 2013
  
  b5d46714
Feb 20, 2013
- Added missing kernel generator script for AMD kernels. · 0745369e
  Karl Rupp authored Feb 20, 2013
  
  0745369e
- Added fast GEMM kernels for AMD Tahiti based on input from Philippe's... · 7979fb41
  Karl Rupp authored Feb 20, 2013
  7979fb41
- Updated copyright information to "2010-2013" · a2866d1c
  Karl Rupp authored Feb 20, 2013
  
  a2866d1c
- Fixed a warning (/* in comment) in CL/cl_gl_ext.h (staying with OpenCL 1.1.... · 25856942
  Karl Rupp authored Feb 20, 2013
  25856942
- Replaced all uses of size_t by std::size_t (exception: viennacl/generator/*,... · 249ee179
  Karl Rupp authored Feb 19, 2013
  249ee179
- Ensured defined behavior when resizing an empty compressed_matrix. · 10a5f086
  Karl Rupp authored Feb 19, 2013
  
  10a5f086
Feb 19, 2013
- * Improved support for BLAS-1-type operations on matrices (e.g. B = -A). · 63931614
  Karl Rupp authored Feb 19, 2013
  63931614
Feb 18, 2013
- * Vector expressions can now be passed to inner_prod() and norm_X() directly. · 6f7f5def
  Karl Rupp authored Feb 18, 2013
  6f7f5def
Feb 13, 2013
- Better OpenMP support for vector operations. Not using OpenMP below 5000... · 576ba0ee
  Karl Rupp authored Feb 13, 2013
  576ba0ee
- Updated version number for upcoming 1.4.1 release. · aaff1b58
  Karl Rupp authored Feb 13, 2013
  
  aaff1b58
Feb 12, 2013
- Added support for Xeon Phi (using 128x128 work items). Adjusted number of work... · 8039d53f
  Karl Rupp authored Feb 12, 2013
  8039d53f
Feb 07, 2013
- Merge branch 'sf.net' of l03.iue.tuwien.ac.at:/iue/git/ViennaCL into sf.net · 1a7b51c7
  Karl Rupp authored Feb 07, 2013
  
  1a7b51c7
- Fixed assertions thrown for AMG in Debug mode (wrong matrix sizes passed over... · 48b1ead5
  Karl Rupp authored Feb 07, 2013
  48b1ead5
- Fixed assertions thrown for AMD in Debug mode. · 1b9cd8f6
  Karl Rupp authored Feb 07, 2013
  
  1b9cd8f6
Feb 04, 2013
- Replaced deprecated VIENNACL_HAVE_OPENCL with VIENNACL_WITH_OPENCL in tests/ · 2f29c388
  Karl Rupp authored Feb 04, 2013
  
  2f29c388
Feb 03, 2013
- Fixed bug in AMG: Removed unnecessary .resize() on empty operator matrices on... · 381aa697
  Karl Rupp authored Feb 03, 2013
  381aa697
Jan 30, 2013
- Fixed external linkage issues in SPAI-code. Updated erroneous test so that it won't happen again. · daa3ca64
  Karl Rupp authored Jan 30, 2013
  
  daa3ca64
Jan 16, 2013
- Added fix for segfaults on program exit when providing custom OpenCL queues... · 6dac70e5
  Karl Rupp authored Jan 16, 2013
  6dac70e5
- * 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