Gabriel Baraldi

Hmm, this seems to have made i686 very unhappy. Ok the issue is that pointer load. 32 bit has different alignment which is what is blowing this up

I don't think it will affect it too too much. Because if the rand call is super hot then things aren't going so well elsewhere :)

I don't think any benchmark is gonna show this. At least no current one. Calling the function is 5x faster that's all I can say

The difference looks to be some getindex calls that now appear in the flamegraph.

The issue is not the checkbounds btw. There's a whole getindex call that seems to be here. It might be Memory{T} bottom is 1.11  

So there are two commits that might deserve testing here. https://github.com/JuliaLang/julia/commit/9aa7980358349ee7017fa614525f571ffa92c55d as reported in https://github.com/JuliaLang/julia/issues/53158#issuecomment-1924283666 and https://github.com/JuliaLang/julia/pull/51720

The proper fix to this is to revive https://reviews.llvm.org/D136218 and make it correct

Am working on it. (it's not super trivial since it's basically a general store sinking pass). Given that, I would consider it not release blocking (So that the release gets...

I haven't yet sorry. The LLVM thing is more involved than I had imagined earlier and basically requires writing a full pass.

I think it's almost there but looking at it in my terminal it seems we are printing the SSA values in a slightly different colour  I believe it should...