Chris Elrod

It doesn't handle `samples::Vector{Vector{Float64}}` at the moment. It probably shouldn't be too hard to add support. The second example works because it pulls `samples[i]` out of the loop, and `samples[i]`...

I will keep it open for future reference.

Not a great suggest, I know, but perhaps you can define ```julia using VectorizationBase, Static VectorizationBase.num_cores() = static(24) ``` Older versions of CPUSummary used Hwloc.jl, but `using Hwloc` crashes Julia...

A little odd that removing the line `c += d*(exp(-d/β + lu[I] + lv[J]))` makes the error go away. The problematic lines are: ```julia t = (I-J).I d = (t[1]^2+t[2]^2)...

Yeah, I should add `ifelse` to the known reductions. Reductions have to be known to correctly initialize `0`s for separate accumulators. To SIMD a reduction, you need to do multiple...

`begin` would throw that error too. The expression parser should replace those with `firstindex`/`lastindex`.

Trying to decide the best way to handle this. LoopVectorization's parsing currently doesn't keep track of the index with respect to the parent array, but with respect to the stridedpointer...

Seems like it may be useful for internals documentation? "Users" probably don't need to know sausage manufacturing. I'd like to encourage contributors eventually / make things easier to hack on,...

I mean that I can't follow the `Makie` example [here](https://juliapolyhedra.github.io/Polyhedra.jl/latest/plot/#D-plotting-with-Plots-2), as `Makie` is not supported on Linux with AMD GPUs and throws an error on loading. However, `using AbstractPlotting, WGLMakie,...

Any plans to support this? And/or abstracted information such as "number of vector registers", and SIMD-width for different types? E.g., to know that AVX512 has 32 registers vs 16 for...