This is a collection of code examples focusing on porting FORTRAN codes to run DOE heterogenous architecture CPU+GPU machines. The purpose of these is to provide both learning aids for developers and OpenMP and CUDA code examples for testing vendor compilers capabilities.
While these examples have a heavy FORTRAN emphasis, some examples also include C++ usage for use in demonstrating OpenMP or CUDA with mixed C++/FORTRAN language executables.
- cuda - examples pertaining to NVIDIA CUDA
- interoperability - examples of using OpenMP and CUDA together, or different compilers
- openmp - examples using OpenMP target offloading capability to run on GPUs
- testing - a variety of cases derived from the SIERRA COE effort, including performance tests, compiler interoperabilty, compiler bugs encountered with OpenMP, etc.
RELEASE LLNL-CODE-769479