Bolt is a C++ template library optimized for GPUs. Bolt provides high-performance library implementations for common algorithms such as scan, reduce, transform, and sort.
Other
373
stars
65
forks
source link
New AMP-Functions Min Element, Max Element, Fill, Fill_n, Generate, Gene... #126
...rate_n, Copy, Copy_n and Inner_product with offset support added.
Changed the inl structures of some AMP routines with some minor corrections and Added some missing offset & UDD tests cases.
Corrected the issue for failing serial & TBB device vector offset tests for transform & count.
Added optimization changes in cl-transform, reduce & transform_reduce kernels to corresponding AMP functions.
Removed unwanted headers in test folders of AMP.