H.A.

Apologies if my assessment is not entirely accurate as I am not intimately familiar with all the internal intricacies of `KernelAbstractions.jl`. I am implementing reductions for a different project and...

> In this case `ndrange` should be a multiple of `groupsize` and you should perform boundschecking explicity. See how I did this with online softmax [implementation](https://github.com/pxl-th/NNop.jl/blob/3df393d76506e861643114e58d6c2189fb306143/src/softmax.jl#L19). That's a perfectly valid...