Commits · 0d1b0979f808b8425b3faa6e96a7aa26998eaf71 · Kaushik Kulkarni / viennacl-dev · GitLab

Oct 21, 2016
- CG: Fixed incorrect early convergence check for negative definite matrices. · 0d1b0979
  Karl Rupp authored Oct 21, 2016
  0d1b0979
Oct 17, 2016
- Merge pull request #206 from smarthi/sundry · 68568653
  Karl Rupp authored Oct 17, 2016
  68568653
- Fix the patch version for 1.7.1 · 2287c217
  smarthi authored Oct 16, 2016
  
  2287c217
Oct 05, 2016
- AMG: Fixed signed/unsigned comparison warning. · b09c65ad
  Karl Rupp authored Oct 05, 2016
  
  b09c65ad
Sep 22, 2016
- AMG: Fixed compilation errors and warnings for CUDA and OpenCL handles. · 6e09ddcf
  Karl Rupp authored Sep 22, 2016
  6e09ddcf
Sep 14, 2016
- Merge pull request #201 from intelfx/device-specific-forwards · a50be820
  Karl Rupp authored Sep 14, 2016
  a50be820
- Device-specific: mark functions in headers as inline · 331f7b98
  Ivan Shapovalov authored Sep 14, 2016
  
  331f7b98
Jun 14, 2016
- OpenCL: Fixed missing integer implementation of element_pow() · 0dc304c7
  Karl Rupp authored Jun 14, 2016
  0dc304c7
Apr 06, 2016
- Merge pull request #182 from ruby-accel/master · 44862d2b
  Karl Rupp authored Apr 06, 2016
  44862d2b
Mar 30, 2016
- should include Eigen/SparseCore instead of Eigen/Sparse · c72f73de
  Takashi Tamura authored Mar 30, 2016
  
  c72f73de
Mar 16, 2016
- compressed_matrix: Added option to wrap user-provided buffers plus example · 02ea0f22
  Karl Rupp authored Mar 16, 2016
  02ea0f22
Mar 03, 2016
- Merge pull request #181 from gongzg/master · cdcf992d
  Karl Rupp authored Mar 03, 2016
  cdcf992d
- Avoid incorrect flag when create ocl memory object. · 04f1063c
  Zhigang Gong authored Mar 03, 2016
  04f1063c
Feb 16, 2016
- Reordering: Fixed handling of multiple components in GPS. · b0ee36d7
  Karl Rupp authored Feb 16, 2016
  b0ee36d7
Feb 12, 2016
- SpMV: Preparations for further optimizations on CPU and Xeon Phi. · 03b423fb
  Karl Rupp authored Feb 12, 2016
  
  03b423fb
- AMG: Added preliminary implementation of classical PMIS. · 9344ae24
  Karl Rupp authored Feb 12, 2016
  9344ae24
Feb 09, 2016
- OpenMP: Added parallelization of direct AMG interpolation. · 94979fad
  Karl Rupp authored Feb 09, 2016
  94979fad
- OpenMP: Added parallel coarse point enumeration in AMG. · c095ddf5
  Karl Rupp authored Feb 09, 2016
  c095ddf5
- OpenMP: Performance optimization of amg_coarse_ag_stage1_mis2. · 70690439
  Karl Rupp authored Feb 09, 2016
  70690439
- OpenMP: Added multi-threaded sparse matrix transposition for AMG. · a9158726
  Karl Rupp authored Feb 09, 2016
  a9158726
Feb 08, 2016
- OpenMP: Added parallelization of pipelined solvers. · f96850dd
  Karl Rupp authored Feb 08, 2016
  f96850dd
Feb 04, 2016
- Merge pull request #178 from cdeterman/master · 2389cefc
  Karl Rupp authored Feb 04, 2016
  2389cefc
- return device_id as vcl_size_t · e0ff5150
  Charles Determan authored Feb 04, 2016
  
  e0ff5150
Feb 03, 2016
- functions for context and device indicies · ac45536a
  Charles Determan authored Feb 03, 2016
  
  ac45536a
Jan 31, 2016
- AMG: Fixed external linkage problems of three kernels in CUDA backend. · 15e9e434
  Karl Rupp authored Jan 31, 2016
  15e9e434
Jan 20, 2016
- DeviceDB: Chop off default-characterization in device type for lookup. · 77e9049d
  Karl Rupp authored Jan 20, 2016
  77e9049d
- Visual Studio: Fixed signed/unsigned comparison warnings. · 43567781
  Karl Rupp authored Jan 20, 2016
  
  43567781
- Clang: Fixed warnings obtained on Mac OS X Yosemite. · c87debe6
  Karl Rupp authored Jan 20, 2016
  
  c87debe6
- Mac OS: Reenabled release of memory buffers. · d254aaa4
  Karl Rupp authored Jan 20, 2016
  d254aaa4
- Version: Increased to 1.7.1. · 6cb7c249
  Karl Rupp authored Jan 20, 2016
  6cb7c249
- Copyright: Updated year span from 2010-2015 to 2010-2016. · cd4c5f40
  Karl Rupp authored Jan 20, 2016
  
  cd4c5f40
Jan 19, 2016
- changelog: Collected changes for 1.7.1 release. · 90d25adc
  Karl Rupp authored Jan 19, 2016
  
  90d25adc
- OpenCL: Removed left-over use of abs() for unsigned integers. · 08a9c71f
  Karl Rupp authored Jan 19, 2016
  08a9c71f
- OpenCL: Fixed work size restriction in sum_inner_prod kernel. · d156a78c
  Karl Rupp authored Jan 19, 2016
  d156a78c
Jan 18, 2016
- Matrix: Added arbitrary expressions for trans() · e73e96d1
  Karl Rupp authored Jan 18, 2016
  e73e96d1
- Sparse matrices: No more temporaries for y += Ax; and y -= Ax; · dc1c0413
  Karl Rupp authored Jan 18, 2016
  dc1c0413
- OpenCL: max and min on vectors now also work with work group size 1. · 26e81488
  Karl Rupp authored Jan 18, 2016
  26e81488
- OpenCL: Removed use of abs() on unsigned integers. · fa046011
  Karl Rupp authored Jan 18, 2016
  fa046011
- OpenCL: Compatibility of summation kernel for work group size 1. · 745e3400
  Karl Rupp authored Jan 18, 2016
  745e3400
Jan 17, 2016
- OpenCL: Improved compatibility of triangular solves for work size 1. · b6cf6eb2
  Karl Rupp authored Jan 17, 2016
  b6cf6eb2