Chris Fallin comments

Results 347 comments of


                                            Chris Fallin

Panics on non consecutive blocks and vregs ids

Sorry, this fell off my radar -- yes, given that a few consecutive-ID assumptions have slipped in, let's document it formally and check it. Happy to review a PR for...

Panics on non consecutive blocks and vregs ids

Looks good, thanks!

Extend Fuzzer to Check Debug Locations

@d-sonuga ah, in that case, yes the checks do look *correct* at least. I'm happy to leave this open as a draft if you'd like or we can close it,...

Support setting the module name in the name section

PRs welcome for this -- it's probably a fairly small change?

Support tail calls and bulk memory, optional function reordering pass

Thanks for this PR, @gkgoat1! I'm currently on parental leave and not writing or reviewing code, but I will take a look at this when I am back in mid-February.

Validate operator return types

These are all things that should be validated, indeed! `waffle` was codeveloped with `weval` and so has not had a separate pass of robustification or API cleanup in general --...

Support threads/atomics

Happy to take a PR if you'd like to do this!

Create relocatable wasm objects

I agree, this would be nice to support; PRs welcome!

Excessive compile time for "simple" module

The intent of `MachBuffer`'s inline `SmallVec`s was to avoid any allocations at all when on-stack, moved not at all or only once, and used for small-to-medium-sized functions, but that was...

Is the mmap memory base always guaranteed to have the same address all the time?

Regardless of underlying implementation details, what Alex is getting at (and what I would re-emphasize) is the difference between incidental details and API-contractual details. We don't *guarantee* anything about when...