Commits · 946198871b4885b16ac628647077787ae53fae69 · TNL / tnl-dev · GitLab

Snippets Groups Projects

This is an archived project. Repository and other project resources are read-only.

Jul 10, 2019
- Fixed the finish method of CudaReductionKernelLauncher · 94619887
  Jakub Klinkovský authored 5 years ago
  
  94619887
Jul 09, 2019
- Fixed handling of begin and end parameters in prefix sum methods · 20c656ff
  Jakub Klinkovský authored 5 years ago
  
  20c656ff
- Throw NotImplementedError from empty prefix sum methods · 1215b84f
  Jakub Klinkovský authored 5 years ago
  
  1215b84f
- Cleaned up ArrayAssignment and VectorAssignment · abe7ae92
  Jakub Klinkovský authored 5 years ago
  
  abe7ae92
Jul 07, 2019
- Optimized CudaReductionKernel to improve compilation times · b8f60127
  Jakub Klinkovský authored 5 years ago
  
  b8f60127
- Moved getSerializationType method from Array/ArrayView to ArrayIO · f7a62b21
  Jakub Klinkovský authored 5 years ago
  
  f7a62b21
- Fixed bug in VectorOperationsCuda · aa5452fe
  Jakub Klinkovský authored 5 years ago
  
  The lambdas add1 and add2 used the Vector copy-constructor when capturing x, which made a deep copy.
  aa5452fe
Jul 04, 2019
- Rewritten parallel array operations using ParallelFor · a4ff26e4
  Jakub Klinkovský authored 5 years ago
  
  a4ff26e4
- Added missing synchronizations after CUDA kernel launches in ArrayOperations · 43c2a7e1
  Jakub Klinkovský authored 5 years ago
  
  43c2a7e1
Jun 14, 2019
- Prefix sum revision is finished. Segmented prefix sum for CUDA is not implemented yet. · 87528b48
  Tomáš Oberhuber authored 5 years ago
  
  87528b48
- Adding CUDA prefix sum unit test. · fc615e79
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  fc615e79
- Removing fetch from prefix sum which leads to templated recursive call and compiler crash. · 6f0209ca
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  Since this is not important feature I do not implement it now.
  6f0209ca
- Refactoring prefix sum. · 8923e36a
  Tomáš Oberhuber authored 5 years ago
  
  8923e36a
- Refactoring prefix sum. · 0278d7e6
  Tomáš Oberhuber authored 5 years ago
  
  0278d7e6
- Fixed CUDA prefix sum. · 2cee804b
  Tomáš Oberhuber authored 5 years ago
  
  2cee804b
- Refactoring CUDA prefix sum. · 12a7c39c
  Tomáš Oberhuber authored 5 years ago
  
  12a7c39c
- [WIP] Refactoring prefix sum in CUDA. · c1c6fc8e
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  c1c6fc8e
- Fixed segmented prefix sum on CPU. · 341117ff
  Tomáš Oberhuber authored 5 years ago
  
  341117ff
- Refactoring segmented prefix-sum. · d804c3a2
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  d804c3a2
- Refactoring prefix sum. · 127c08b4
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  127c08b4
- Refactoring prefix-sum. · aea06f06
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  aea06f06
- Fixed reduction with argument. · 9852ca1c
  Tomáš Oberhuber authored 5 years ago
  
  9852ca1c
- Implementing reduction with argument. · 61f8e317
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  61f8e317
- Optimized CUDA reduction by decreasing desired grid size. · 7a49e478
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  7a49e478
- Fixed vector addition and added vector addition benchmark to BLAS benchmark. · c81fb4cc
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  c81fb4cc
- Expression templates are working. · a06d7b8e
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  a06d7b8e
- Fixing Vector assignment. · 957e577d
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  957e577d
- Fixes of expression templates for VectorView. · 19cd813b
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  19cd813b
- Fixing expression templates for VectorView. · 0ed6586d
  Tomáš Oberhuber authored 5 years ago
  
  0ed6586d
- [WIP] Implementing expression templates for VectorView. · 75b0c9da
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  75b0c9da
- Unary expression templates seem to work. · 38ce7667
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  38ce7667
- [WIP] Implementing expression templates for static vectors. · 8e50a778
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  8e50a778
- Moving ET source code to containers. · 6abbfef6
  Tomáš Oberhuber authored 5 years ago
  
  6abbfef6
- Adding expression templates. · 34680051
  Tomáš Oberhuber authored 5 years ago and Tomáš Oberhuber committed 5 years ago
  
  34680051
Apr 19, 2019
- Refactoring ArrayIO · 67d4cba3
  Jakub Klinkovský authored 5 years ago
  
  67d4cba3
- Added NotImplementedError exception · 5dce2ff7
  Jakub Klinkovský authored 5 years ago
  
  5dce2ff7
- Fixes in ArrayOperations · b357ff4e
  Jakub Klinkovský authored 5 years ago
  
  b357ff4e
Apr 15, 2019
- Fixed use of buffer in CUDA paralell reduction. · 3f891223
  Tomáš Oberhuber authored 5 years ago
  
  3f891223
- Fixing CUDA array operations. · 58af5093
  Tomáš Oberhuber authored 5 years ago
  
  58af5093
Apr 14, 2019
- Changing numeric limits -max to lowest. · aa18c0a2
  Tomáš Oberhuber authored 5 years ago
  
  aa18c0a2