oneAPI Open Source Projects
Codeplay has added NVIDIA® GPU support for DPC++ and the MKL-BLAS open source project. DPC++, an implementation of SYCL with extensions, is based on the oneAPI industry specification. Codeplay’s previous experience in developing support for NVIDIA GPUs using PTX instructions for ComputeCpp offered the experience needed to add this support. This implementation accesses the CUDA interface directly and is integrated into DPC++ as part of the LLVM compiler and the MKL-BLAS maths library project. Codeplay’s ComputeCpp compiler is also building compatibility with the oneAPI specification and can be adapted to different manufacturer processors.
The Intel® oneAPI Base Toolkit is a core set of tools and libraries for building and deploying high-performance, data-centric applications across diverse architectures. It features the Data Parallel C++ (DPC++) language, an evolution of C++ that: Allows code reuse across hardware targets—CPUs, GPUs, and FPGAs†; Permits custom tuning for individual accelerators. Domain-specific libraries and the Intel® Distribution for Python* provide drop-in acceleration across relevant architectures. Enhanced profiling, design assistance, and debug tools complete the kit.