Chris Fallin

Hi @theoparis -- as you can see here ``` error: Wasmtime is being compiled for an architecture that it does not support ``` Wasmtime doesn't (yet?) support `riscv32` (of any...

Indeed, it seems we skip checking operands on branch instructions: [link](https://github.com/bytecodealliance/regalloc2/blob/b778e884af221d7274639f53aa3bfd4e9428cb56/src/checker.rs#L770-L772) The stated reason there is that we don't want to check block arguments as normal uses, because we handle...

@Amanieu ah, yes, thanks for that reminder; would you be willing to separate out the checker changes in #170 from the blockparam restrictions and post them in a separate PR?

Thanks for the comment, @bwburnsides! Yes, indeed, we're aware of this trick; it's not implemented in RA2 for a combination of reasons: - There is a layered approach to performing...

@guybedford I've been poking around the test failure and I'm a little lost -- I see ``` Error running file html/webappapis/timers/evil-spec-example.any.js: setTimeout: first argument must be a function ``` and...

It appears to be an issue of memory size limits: ``` wasmtime stderr: Error: memory index 0 has a minimum page size of 165 which exceeds the limit of 160...

Actually, that appears to be an issue with an old Wasmtime ([Zulip thread](https://bytecodealliance.zulipchat.com/#narrow/stream/217126-wasmtime/topic/.E2.9C.94.20wasmtime.20serve.20wasm.20limits.20settings)); upgrading my local Wasmtime and running locally, I am unable to reproduce when running just the one...

Hi @iwanders -- thanks for the report. Actually I think the issue is a bit different, and much simpler, than this PR implies: regalloc2 requires accurate information about the range...

Doesn't that assert in `InstRange::new()` permit a range of `0..0` as it's a `

Sorry, I'm a bit confused now. You wrote in the initial report > Running clif-util test from cranelift passes on this file, that probably relabels all ids before passing them...