Implement feature common to all `AbstractGPUArrays` through KernelAbstractions.jl

Specific Domains GPU
,
1

Hello,

Let’s imagine I have a very common function

LinearAlgebra.kron(a::VT, b::VT) where VT<:AbstractGPUArray

which can support any AbstractGPUArray since I’m using KernelAbstractions.jl. Where should I implement such feature? Is there a general place? Or should I implement it on every platform as CUDA.jl, Metal.jl etc?

2

Hi. You can do this in GPUArrays.jl package. Every GPU package has it as a dependency

3

Thank you. Just made a PR in GPUArrays.jl