GPU optimized implementation of Radix Sort using OpenCL.
C++20-enabled compilers are supported, e.g.:
- GCC 14
- Clang 16
- Visual Studio 2019
- cmake 4.0+
Every OpenCL 1.2 compliant driver should be supported. For NVIDIA devices, install CUDA drivers.
Libraries and tests can be built as follows:
git clone github.com/gyatskov/radix-sort
cd radix-sort
mkdir build
cmake -H. -B build
cmake --build buildTests and RadixSort.cl kernel will be installed to build/tests.
Run
ctest --test-dir build/tests --output-on-failure
The implementation is based on papers referenced in doc.pdf