| layout | page |
|---|---|
| title | About |
| permalink | /about/ |
AClang is a research project conducted at the University of Campinas (IC-UNICAMP) in Brazil. It is based on the open-source LLVM/Clang compiler framework and implements the OpenMP 4.0 Accelerator Programming Model. ACLang automatically converts OpenMP annotated program regions into OpenCL/SPIR kernels, while providing a set of polyhedral based optimizations like tiling and vectorization. OpenCL kernels resulting from ACLang can be executed on any OpenCL/SPIR compatible acceleration device, not only GPUs, but also FPGA accelerators like those found in the Intel HARP architecture.