Ben L. Titzer

I was reasoning that `br_on_not_i31_to_i32_s/u` would leave an `i32` on the stack, which, if simplified to only be unsigned, would need only a single extension. Likewise with the other branches;...

While I agree with what @rossberg wrote above w.r.t. to implementing polymorphism (though Virgil does whole-program static specialization and does not have polymorphic recursion or first-class polymorphism), the stronger argument...

> However, we'll run into this with all post gc-mvp features, and the solution is for runtimes to keep implementing these features before full standardization and making them available for...

I feel uncomfortable reasoning in detail about all future runtime systems at this point, so I don't accept the premise to this proposal. There haven't yet been any example systems...

> Also, my sentiment of "improving what we have" or more generally "working together instead of against each other" applies to other similar discussions as well, as anything else is...

It's worth pointing out that in many languages, particularly Java and C#, arrays are not growable. In order to protect themselves from their arrays' lengths being modified by outside code...

It's also worth pointing out that arrays in JavaScript are pretty complicated. One can delete elements from the middle array (where subsequent accesses poke through to the prototype chain), and...

It's a difficult situation and I don't envy you :-) Arrays in JavaScript are complicated enough that it might be easier to use JS arrays directly and "proxy" them to...

We also emit an alignment check for atomic accesses. I think we may be stuck with the explicit branch because of the constraints we have, at least if I understand...

I see that, but that might cause other problems, since now the input is 32 user-controlled bits, shifted left by 1, 2, or 3 bits. Should the (implicit) bounds check...