Thomas Lively

Full-sized values based on the type of the array makes sense. So an i8 array would have an entry for each source byte and an i32 array would have an...

@askeksa-google, do I remember correctly that at one point you had performance data showing the benefit of initializing these constants in globals as well? Generally having these as constant instructions...

I think for the data count section we can frame it as allowing the existing data count section (with the existing section id) to appear in one of multiple possible...

Thanks for the write up! It should be fairly simple to measure the code size benefit of switching from two indices to one index on accessor instructions. Since the precise...

Sorry @RossTate, I stated my reasoning imprecisely and too strongly. I agree that this is orthogonal to the type system question and I agree this is worth considering for the...

Is anyone interested in investigating this option further? If not, we can close this issue.

> Let's suppose ref A has the property that the type of its n-th field is uniquely/deterministically determined to be tA, and as such struct.get accepts it. Let's suppose an...

Yes, I am asking for a hypothetical Post-MVP extension here :) I don't think we should spend a lot of time trying to predict whether we will or will not...

Thanks for the examples, @RossTate! My curiosity is satisfied. It sounds good to me to pull this in once(/if) we resolve other discussions about it with the knowledge that it...

We decided on the EH proposal that traps would not be handled as exceptions, so EH won't give you any extra ability to recover from a trap, whether due to...