Commits · d12f3be51668490a27ff638bb28c4f15a53897aa · Kaushik Kulkarni / viennacl-dev · GitLab

Aug 20, 2014
- Benchmark: Adjusted the BLAS3 benchmark size · d12f3be5
  Philippe Tillet authored Aug 21, 2014
  
  d12f3be5
- Device-specific/GEMM : Now only compile the fallback if necessary · 2bc5de5e
  Philippe Tillet authored Aug 21, 2014
  
  2bc5de5e
- Coding Style: Cleaned up viennacl/linalg/host_based/ · 278f41b9
  Karl Rupp authored Aug 21, 2014
  
  278f41b9
- Code Style: Unified the viennacl/*.hpp files · b6960e99
  Philippe Tillet authored Aug 20, 2014
  
  b6960e99
- Coding Style: Renamed ALIGNMENT to AlignmentV, and some remaining SCALARTYPE to NumericT · 8c924455
  Philippe Tillet authored Aug 20, 2014
  
  8c924455
- Coding Style: Unified the viennacl/traits/ folder · d6f2a74a
  Philippe Tillet authored Aug 20, 2014
  
  d6f2a74a
- Coding Style: Unified the viennacl/tools/ folder · e306778d
  Philippe Tillet authored Aug 20, 2014
  
  e306778d
- Coding Style: Unified the viennacl/scheduler/ folder · 22683641
  Philippe Tillet authored Aug 20, 2014
  
  22683641
- Coding Style: Unified the viennacl/ocl/ folder · d59ce6f4
  Philippe Tillet authored Aug 20, 2014
  
  d59ce6f4
- Coding Style: Unified viennacl/misc folder · 824dd297
  Philippe Tillet authored Aug 20, 2014
  
  824dd297
- Coding Style: Unified the viennacl/meta/ folder · f6502e18
  Philippe Tillet authored Aug 20, 2014
  
  f6502e18
- Coding Style: Unified the viennacl/io/ folder · eeaa672f
  Philippe Tillet authored Aug 20, 2014
  
  eeaa672f
- Coding Style: Unified viennacl/backend folder · fa3a2bd4
  Philippe Tillet authored Aug 20, 2014
  
  fa3a2bd4
- Coding Style: Unified the viennacl/device_specific/ folder · 39174a99
  Philippe Tillet authored Aug 20, 2014
  
  39174a99
- Coding style: Cleaned up viennacl/linalg/cuda folder. · 4e8b360f
  Karl Rupp authored Aug 20, 2014
  
  4e8b360f
- Coding Style: Removed space before semicolon. · 2a6cbbb5
  Karl Rupp authored Aug 20, 2014
  
  2a6cbbb5
- Coding Style: Unified space after 'if', 'for', 'while', and 'switch' · 6b8db5d9
  Karl Rupp authored Aug 20, 2014
  
  6b8db5d9
- Coding style: Replace use of "template <typename" with "template<typename" · 2bd5aa8b
  Karl Rupp authored Aug 20, 2014
  
  2bd5aa8b
- CUDA: Added missing symbolic links of new tests and benchmarks. · 8a4ba908
  Karl Rupp authored Aug 20, 2014
  
  8a4ba908
- Cleanup: Replaced tabs by two spaces in sources. · 8ce79f5b
  Karl Rupp authored Aug 20, 2014
  8ce79f5b
- FFT: Fixed compilation problems due to use of variable-length arrays (C99 extension) · 51f9dd04
  Karl Rupp authored Aug 20, 2014
  51f9dd04
Aug 19, 2014
- Device-specific : Used CRTP for cleaner templates. No more need to pass... · e290dfad
  Philippe Tillet authored Aug 19, 2014
  e290dfad
- Merge branch 'JurajKabzan-master' · 7414d54e
  Karl Rupp authored Aug 19, 2014
  
  7414d54e
- Merge branch 'master' of github.com:JurajKabzan/viennacl-dev into JurajKabzan-master · a17356ea
  Karl Rupp authored Aug 19, 2014
  a17356ea
- Benchmarks: Removed left-over inclusions of removed io.hpp · 537c09fd
  Karl Rupp authored Aug 19, 2014
  537c09fd
- NMF extended on CUDA,OpenMP. Docummentation of NMF finnished. Tests and... · 222cc426
  JurajKabzan authored Aug 19, 2014
  222cc426
- Device-Specific : Fixed vload for simd_width==1 · 295e1a1a
  Philippe Tillet authored Aug 19, 2014
  
  295e1a1a
- Device-specific/GEMM: Fixed performance regression in fallback GEMM · 02b8762e
  Philippe Tillet authored Aug 19, 2014
  
  02b8762e
Aug 18, 2014
- Device-Specific/GEMM : Fixed bug for tiny matrices · 61d6dec6
  Philippe Tillet authored Aug 19, 2014
  61d6dec6
- Device-Specific/GEMM : The fallback kernel is now similar to the optimized kernel · f757f7e6
  Philippe Tillet authored Aug 18, 2014
  f757f7e6
- scheduler: extract contexts from all supported numeric types · 41cd8fed
  Toby Smithe authored Aug 18, 2014
  
  41cd8fed
Aug 17, 2014
- Device-Specific / GEMM : Faster fallback kernel · 4381a6e9
  Philippe Tillet authored Aug 18, 2014
  
  4381a6e9
- Device-Specific / GEMV : Fixed misuse of strides · a5b54e6c
  Philippe Tillet authored Aug 17, 2014
  
  a5b54e6c
- Device-Specific : Fixed bug in fused reductions · 1c51d844
  Philippe Tillet authored Aug 17, 2014
  
  1c51d844
- Benchmarks: Refurbished the benchmark suite. · 73f46e36
  Philippe Tillet authored Aug 17, 2014
  73f46e36
- norm_frobenius: test for row vs col major · b12e4d52
  Toby Smithe authored Aug 17, 2014
  
  b12e4d52
- Fix norm_frobenius for strided or off-set matrices; note that this case incurs... · bbc626fd
  Toby Smithe authored Aug 17, 2014
  bbc626fd
- Add Frobenius norm test for matrix range / slice · 0f4094c9
  Toby Smithe authored Aug 17, 2014
  
  0f4094c9
- FFT extended on CUDA,OpenMP. Created multiple tests and tutorial. FFT... · b183be65
  JurajKabzan authored Aug 17, 2014
  b183be65
- Tests: Rewrote the GEMM test from scratch. · 6deaa42f
  Philippe Tillet authored Aug 16, 2014
  6deaa42f