Commits · dafcb6f48e98db44d1ea65e3155db44677fa38d6 · TNL / tnl-dev

Mar 19, 2016
- Added --size-step-factor parameter to tnl-cuda-benchmarks · dafcb6f4
  Jakub Klinkovský authored 9 years ago
  
  dafcb6f4
- Added parser of command line parameters to tnl-cuda-benchmarks · 49f4459a
  Jakub Klinkovský authored 9 years ago
  
  49f4459a
- Added benchmark for tnlChunkedEllpackMatrix · ae90399d
  Jakub Klinkovský authored 9 years ago
  
  ae90399d
- Benchmarks: include error messages in the log file · 929e524c
  Jakub Klinkovský authored 9 years ago
  
  929e524c
- Benchmarks: fixed description · 41284115
  Jakub Klinkovský authored 9 years ago
  
  41284115
Jan 02, 2016
- Benchmarks: rewritten objectively, with support for output into a log file · 77e7a205
  Jakub Klinkovský authored 9 years ago
  
  77e7a205
Dec 31, 2015
- Benchmarks: fixing copy-paste error · 88ac33c7
  Jakub Klinkovský authored 9 years ago
  
  88ac33c7
- Benchmarks: refactoring, generalization for other operations · 7d4d00e7
  Jakub Klinkovský authored 9 years ago
  
  7d4d00e7
- Benchmarks: simplified initialization of timers, removed obsolete TODO notes · 55003e09
  Jakub Klinkovský authored 9 years ago
  
  55003e09
Dec 30, 2015
- Benchmarks: add a case with explicit cudaThreadsSynchronize() call · 5b36e3c1
  Jakub Klinkovský authored 9 years ago
  
  5b36e3c1
- Splitting benchmarks of vector operations into separate header · 742581b2
  Jakub Klinkovský authored 9 years ago
  
  742581b2
- Splitting main benchmark functions into separate header · da855c19
  Jakub Klinkovský authored 9 years ago
  
  da855c19
- Benchmarks: adding TODO notes, small fixes · a09ffbcd
  Jakub Klinkovský authored 9 years ago
  
  a09ffbcd
- Refactoring benchmarks using more lambda functions · 5e5d94ce
  Jakub Klinkovský authored 9 years ago
  
  5e5d94ce
- Used lambda functions to slightly generalize benchmarks · 8bb94c12
  Jakub Klinkovský authored 9 years ago
  
  8bb94c12
- added TODO note · 2771e103
  Jakub Klinkovský authored 9 years ago
  
  2771e103
- Generalized SpMV benchmarks on CUDA for more matrix formats · a7482bfe
  Jakub Klinkovský authored 9 years ago
  
  a7482bfe
- Reorganized output of CUDA benchmarks · 3de4fa58
  Jakub Klinkovský authored 9 years ago
  
  3de4fa58
- Optimized error checking in tnl-cuda-benchmarks · 75742f7c
  Jakub Klinkovský authored 9 years ago
  
  75742f7c
- Fixed order of fields printed to stdout in the CUDA benchmark · 4bfcb781
  Jakub Klinkovský authored 9 years ago
  
  4bfcb781
Dec 13, 2015
- Fixing a bug in host and device arrays comparison. · 31ee230c
  Tomáš Oberhuber authored 9 years ago
  
  31ee230c
Dec 12, 2015
- Tuning Ellpack formats. · 994ee56d
  Tomáš Oberhuber authored 9 years ago
  
  994ee56d
- Tuning Ellpack format. · 670f1892
  Tomáš Oberhuber authored 9 years ago
  
  670f1892
Dec 11, 2015
- Fixing bug in tnlVector::operator !=. · c2f4c4ab
  Tomáš Oberhuber authored 9 years ago
  
  c2f4c4ab
Dec 10, 2015
- Optimizing CUDA L2 norm. · 15dcbd43
  Tomáš Oberhuber authored 9 years ago
  
  15dcbd43
- Adding banchmark for CUDA lp norm. · 7aa09ad1
  Tomáš Oberhuber authored 9 years ago
  
  7aa09ad1
- Optimized error checking in tnl-cuda-benchmarks · 03905634
  Jakub Klinkovský authored 9 years ago
  
  03905634
- Fixed calculation of bandwidth (lpNorm is unary operation). · a19fd6cf
  Jakub Klinkovský authored 9 years ago
  
  a19fd6cf
Dec 09, 2015
- Added benchmarks for lpNorm to tnl-cuda-benchmarks · fb1474d3
  Jakub Klinkovský authored 9 years ago
  
  fb1474d3
- Fixed order of fields printed to stdout in the CUDA benchmark · d2514e28
  Jakub Klinkovský authored 9 years ago
  
  d2514e28
Jun 18, 2015
- Optimizing CUDA parallel reduction. · e36a896a
  Tomáš Oberhuber authored 9 years ago
  
  e36a896a
Jun 17, 2015
- Optimizing parallel reduction in CUDA. · 87a42926
  Tomáš Oberhuber authored 9 years ago
  
  87a42926
Jun 14, 2015
- Implementing CUDA benchmark. · 3aa06ead
  Tomáš Oberhuber authored 9 years ago
  
  3aa06ead
May 28, 2015
- Basic CUDA benchmark was added. · 0031f809
  Tomáš Oberhuber authored 9 years ago
  
  0031f809
May 03, 2015

A keyword __cuda_callable__ was introduced. · 07179b5a

Min. interations feature was fixed in the GMRES method.
Timers for explicit and semi-implicit timestepping were added.

07179b5a

Mar 24, 2013
- Refactoring the reduction on the CUDA device. · c52cb76b
  Tomáš Oberhuber authored 12 years ago
  
  Explicit instantiation of template CUDA reduction.
  c52cb76b
Jul 09, 2012

TNL is now build as one library, linking is easier. · 56541aee

Tomáš Oberhuber authored 12 years ago

Adding new class tnlArrayManager for managing arrays - allocation/deallocation, load and save functions etc.
tnlVector is derived from tnlArrayManager.

56541aee

Jun 27, 2012
- Renaming tnlVector to tnlTuple and tnlLongVector to tnlVector. · eb5820e0
  Tomáš Oberhuber authored 12 years ago
  
  eb5820e0
Jun 14, 2011
- Converting project to CMake. · 1fab565d
  Tomáš Oberhuber authored 13 years ago
  
  1fab565d
May 11, 2011
- Rewriting SPMV benchmarks. · 5ac31abf
  Tomáš Oberhuber authored 13 years ago
  
  5ac31abf