diff options
| author | Cedric Nugteren <web@cedricnugteren.nl> | 2018-01-08 21:07:01 +0100 |
|---|---|---|
| committer | Cedric Nugteren <web@cedricnugteren.nl> | 2018-01-08 21:07:01 +0100 |
| commit | 99a4df88a6d808ea77c9116ce63621503c00b57a (patch) | |
| tree | d8f8bc1b3884c0340df9f6d95b4837ed3dff8deb /CHANGELOG | |
| parent | 13f0f6fc6e612a5f77c6fd78b983f1b2bb8e36b6 (diff) | |
Implemented the in-direct version of the strided-batched GEMM kernel
Diffstat (limited to 'CHANGELOG')
| -rw-r--r-- | CHANGELOG | 3 |
1 files changed, 3 insertions, 0 deletions
@@ -10,6 +10,9 @@ Development (next version) - Improved compilation time by splitting the tuning database into multiple compilation units - Various minor fixes and enhancements - Added tuned parameters for various devices (see README) +- Added a strided-batched (not part of the BLAS standard) routine, faster but less generic compared + to the existing xGEMMBATCHED routines: + * SGEMMSTRIDEDBATCHED/DGEMMSTRIDEDBATCHED/CGEMMSTRIDEDBATCHED/ZGEMMSTRIDEDBATCHED/HGEMMSTRIDEDBATCHED Version 1.2.0 - Fixed a bug in the TRSM/TRSV routines due to missing synchronisations after GEMM/GEMV calls |