Cross-architecture parallel algorithms for Julia's GPU backends, from a unified KernelAbstractions.jl codebase. Targets Intel oneAPI, AMD ROCm, Apple Metal, Nvidia CUDA.
MIT License
Bot releases are visible (Hide)
Merged pull requests:
Closed issues:
GPUArraysCore
? (#1)sortperm!
on AMDGPU (#2)Published by anicusan about 1 month ago
First release of AcceleratedKernels.jl, for archiving purposes supporting the "AcceleratedKernels.jl: Cross-Architecture Parallel Algorithms from a Unified, Transpiled Codebase" paper.