Chris Fallin

> The main thing that I'm worried about is a situation like: [sub rsp once, many stores with large offsets] Fortunately we seem to [decrement rsp and then store in...

(And likewise the [dynamic loop version](https://github.com/bytecodealliance/wasmtime/blob/17146ab212d1968c6f466bb97e13d42e3a7adfaa/cranelift/codegen/src/isa/x64/inst/emit.rs#L1518-L1539) seems to have the same property)

You would know better than me but that sounds plausible, given your work on helper calls in the linked PR!

@alexcrichton thanks a bunch for pushing this further! I agree it's best to do this only if it actually matters. In my own work I can at least test your...

Yep, I missed that entirely in this iteration of the patch -- sorry about that! Just updated. I ended up wiring the limit through to a new config option (with...

The merge queue did its job (nice!) and found a conflict with #8515 which merged after this PR's CI ran. I pushed a fix which is fairly small but @fitzgen...

(or @jameysharp feel free to review last commit too as it interacts with your change in #8515!)

This would be really neat! I've definitely thought that the backend's lowering rules serve as the best sort of cost function actually (some thoughts in my [slides](https://cfallin.org/pubs/egraphs2023_aegraphs_slides.pdf) slide 185 --...

Hmm -- I took a look at this and was able to reproduce locally. I'm not sure where the `mach bootstrap` error is coming from though. The version of SpiderMonkey...

My [comment in #146](https://github.com/bytecodealliance/regalloc2/pull/146#issuecomment-1608610742) should probably have gone here; I think there's a way to do this more generally, by doing splits in a more general way (in the middle...