After a little more effort from us, and quite much from the community, the AMD compute driver parts made it into the Mesa driver. AMD open-sourced some LLVM backend implementation, which tailed by them to work with Mesa and compute.
Thanks to that, very few things from OpenCL are compiling.
is our software tool embedded in a C++ compiler that automatically determines during compilation time which parts of the source code should be compiled to GPU (and how) while managing data transfer between CPU and GPU automatically. Our unique technology increases productivity and reduces development effort and time.
Our BRIAN QC software package helps to reduce time need and costs of discovery phase and to find the right molecule for successful nano-material and drug development projects. It uses compiler technology to generate specific integral calculation kernels for GPUs. The applied compiler enables the adaptation of field specific optimization on a chosen GPU architecture; the conventional solutions cannot compile huge code.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 767912.