Chris Fallin

I'll note that the mid-end does keep both around, rather than destructively rewriting (because egraphs!), so in the future if we have a more sophisticated cost function extractor there may...

> So basically, what I'm doing now is observe the discrepancy between the LLVM InstCombine pass and Cranelift's mid-end optimizer and then add rules to Cranelift for such missed optimization...

To the immediate question of making this rule actually fire: since we already rewrite `(iadd _ x (ineg _ y))` to `(isub _ x y)`, could you rewrite the left-hand...

> First off - there is already an effort for component record reply in progress that @cfallin is leading. You should talk to him about that effort and align with...

One additional wrinkle here that I haven't seen answered above is: how exactly can replay match up resource IDs embedded within returned composite values? To be concrete, looking at this...

To add another explicit distinction: I think the aspect of let-bindings that matters *less* (IMHO) is giving readable names to handles; that's useful if humans author this format but not...

Hi @zlfn -- unfortunately, as you've noted, regalloc2 doesn't support overlapping resources. It is theoretically possible to update it to model this by modifying the register-search logic, tracking the smallest...

> `inst1: op Def: v1i fixed(p0i), Use: v0i fixed(p1i), Clobber: p0i, Clobber: p1i, Clobber: p2i` Hi @MaxVerevkin -- this is an invalid input program: it is requesting `v0` to be...

Ah, sorry, the docs are out of date then -- this has been the interface for a while now. I'll update that as well. Ideally we would catch and return...