This repository contains examples that demonstrate how to use the CUDA backend in SYCL.
The examples are built and test in Linux with GCC 7.4, NVCC 10.1 and the experimental support for CUDA in the DPC++ SYCL implementation.
CUDA is a registered trademark of NVIDIA Corporation SYCL is a trademark of the Khronos Group Inc.
These examples are intended to be used with this docker image. It provides all the examples, libraries and the required environment variables.
NVIDIA Container Toolkit must be installed to run the image.
A useful guide for setting up docker and the NVIDIA Container Toolkit can be found here.
Once docker and the NVIDIA Container Toolkit are installed, we can create a new container and run the examples witin it.
$ sudo docker run --gpus all -it ruyman/dpcpp_cuda_examplesOnce inside the docker image, navigate to /home/examples/ to find a local clone of this repo. Make sure to pull the latest changes:
$ cd /home/examples/SYCL-For-CUDA-Examples
$ git pullRefer to each example and/or exercise for detailed instructions on how to run it.
This trivial example can be used to compare a simple vector addition in CUDA to an equivalent implementation in SYCL for CUDA. The aim of the example is also to highlight how to build an application with SYCL for CUDA using DPC++ support, for which an example CMakefile is provided.
This demonstrates an example of how to call a SYCL function from a CUDA fortran code.
This example shows how to integrate MPI calls within the SYCL DAG using Host Tasks for integration.
This demonstrates using SYCL's host_task for CUDA interoperability, calling CUBLAS's SGEMM routine for matrix multiplication.
This example combines the MPI and SGEMM Interop examples to distribute a matrix multiplication problem between MPI ranks.
Kokkos is a middle-layer for scientific computing which features a SYCL backend. This example shows a small Kokkos test case (vector-matrix-vector multiplication), adapted from a test case in the Kokkos repo; there is no SYCL code in the example, but it includes scripts to build Kokkos with SYCL support.
This example is slightly different - it benchmarks a series of hashing algorithms.