Josh Stone

That's interesting -- maybe `Result::from_residual` could also have `#[track_caller]`? But that may bloat a lot of callers in cases that won't ever use the data.

@tmccombs It gives you internal iteration for more complicated iterators like `Chain`, rather than calling the outermost `next()` each time. But while I think that's a useful transformation under user...

Yes -- #69 fixed the tests, but I haven't bothered publishing that yet since it doesn't affect "normal" users. 😉

I imagine it's fine, but it does seem more like a marketing/trademark question for the foundation.

Since there's already `fn cis(T) -> Complex`, maybe this could be `fn into_cis(self) -> Self`? Is there any naming precedent from other languages for this form?

@a-samea that's a strange example -- $cis(1) \approx 0.54 + 0.84i$, so it would have to round or truncate badly for `Complex`.

Can you share your benchmark code somewhere? I'd like to compare results with various Intel and AMD machines, and also when compared to `-Ctarget-cpu=native`. I'm not sure how much the...

By x32, I mean the odd target of `x86_64-unknown-linux-gnux32`, which is using a 64-bit CPU with 32-bit pointers. In Rust that looks like `cfg(all(target_arch="x86_64", target_pointer_width="32"))`. See https://en.wikipedia.org/wiki/X32_ABI Thanks for your...

AFAICS, we *already* get the same kind of unrolled optimization as your non-asm version, only it chooses 4 at a time instead of your 5, and when I use `-Ctarget-cpu=native`...

It's possible to add, and you can emulate this now with `T::min_value() < T::zero()`, which is probably how we would write a default implementation in the trait. Since the `to_*`...