Issues

• Device-aware Auto() tuning: reduce default items_per_thread leaves ~1.8x on AMD

  JuliaGPU/AcceleratedKernels.jl#150 · Filter by author shreyas-omkar opened on Oct 7, 2026

• Int32 index-type specialization: no runtime gain (memory-bound)

  JuliaGPU/AcceleratedKernels.jl#149 · Filter by author shreyas-omkar opened on Oct 7, 2026

• Operations on a BitArray fail without an explicit backend

  JuliaGPU/AcceleratedKernels.jl#143 · Filter by author maleadt opened on Oct 1, 2026
  1

• Multi-source mapreduce with a closure capturing a device array fails to compile on OpenCL

  JuliaGPU/AcceleratedKernels.jl#142 · Filter by author maleadt opened on Oct 1, 2026

• Performance of small and whole-array reductions, and of Cartesian findall

  JuliaGPU/AcceleratedKernels.jl#135 · Filter by author maleadt opened on Sep 24, 2026

• Typos in Section I (Introduction) of AcceleratedKernels.jl preprint

  JuliaGPU/AcceleratedKernels.jl#89 · Filter by author DiegoLopezInc opened on Jun 23, 2026

• Statistics kernels?

  JuliaGPU/AcceleratedKernels.jl#62 · Filter by author Moelf opened on Aug 10, 2025
  3

• Use matrix input in searchsortedfirst

  JuliaGPU/AcceleratedKernels.jl#61 · Filter by author mtagliazucchi opened on Aug 7, 2025

• Use heuristic to determine block size instead of fixed value

  JuliaGPU/AcceleratedKernels.jl#60 · Filter by author christiangnrd opened on Aug 5, 2025

• Performance comparison against KernelAbstractions.jl

  JuliaGPU/AcceleratedKernels.jl#50 · Filter by author mtagliazucchi opened on Jul 5, 2025
  3

• Benchmarks

  JuliaGPU/AcceleratedKernels.jl#43 · Filter by author christiangnrd opened on May 28, 2025
  2

• Faster Reductions using Coalesced Reads

  JuliaGPU/AcceleratedKernels.jl#12 · Filter by author THargreaves opened on Dec 18, 2024
  2