JF Bastien

The intent of `minNum` and `maxNum` is to follow IEEE754-2008 semantics precisely.

Right, JS adds non-determinism on top because the NaN values aren't guaranteed, but one of the goals of the SIMD spec is (IIUC?) to be shareable with WebAssembly where the...

@PeterJensen I'm not sure I understand the point about `vreinterpretq`: it seems like it's reinterpreting bits for vectors of the same bit-width, but different element width. e.g. in the webp...

> Given that we have the precedent of not being able to call functions with i64 in their signature from JS, I think it makes sense to do a similar...

This comparison, when done, should be across multiple architectures to also see which approach has better performance portability.

It would be good to quantify this: how close are we perf-wise and size-wise to: * Exceptions **off** on native platforms * Exceptions **on** but never thrown ** Exceptions **on**...

Wouldn't that remove the binary size saving?

> if we give guidelines for the producer (TBD 😉) then can the VM assume that it isn't going to have to optimize a constant 4 byte memcpy that should...

@lukewagner would you rather also have an alignment hint, so you can do fancy stuff on top?

Good question. Additionally, I'd like to understand how much of this size is metadata versus code.