The goal of this project is to serve as a collection of free OpenCL algorithms with a bias towards embedded systems. The license used is BSD.
Each algorithm shall be provided with a documentation explaining how it was parallelized. For instance, in case of a vector summation, the documentation could explain that thanks to the associativity of the addition operator the computation can be organized as a tree.
Algorithms shall be listed here with a short description if needed. Each algorithms shall come with the full implementation : Haskell + OpenCL kernels. The Haskell one being the reference.
The design document is giving more details about the test framework which is using Haskell and QuickCheck.