scottmcm

Agreed that that's a super-weird error. @bors retry

Thanks for mentioning #102072! My [inclination](https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/Short.20term.20.23101899.20plan/near/300482983) here to to avoid further changes around `Layout` until #101899 settles a bit. Otherwise this seems perfectly reasonable to me.

It looks like the size limit will stick (https://github.com/rust-lang/rust/issues/101899#issuecomment-1290805223) so I think that part is unblocked, but https://github.com/rust-lang/rust/pull/102049 is a good callout.

Regarding naming: I'm in favour of `ilog2` because it's also an operation that would make sense to provide on floating-point numbers. Precedent: https://graphics.stanford.edu/~seander/bithacks.html#IntegerLogFloat EDIT: Also C99 https://en.cppreference.com/w/c/numeric/math and C++11 https://en.cppreference.com/w/cpp/numeric/math...

I always find it hard to judge table lookups. In microbenchmarks they often look great since cache pressure isn't a concern.

> The `Eq` derive was also [requested](https://rust-lang.zulipchat.com/#narrow/stream/219381-t-libs/topic/.60ControlFlow.3A.20Eq.60/near/303610659) by @lcnr on Zulip to allow for pattern matching. I'll cross-link https://github.com/rust-lang/rust/pull/102697#issuecomment-1274404667 here, since libs-api seems skeptical of the `CONTINUE` constant, which would...

> until `ControlFlow` itself has been stable for a bit Note that `ControlFlow` the type has been stable for about a year now: (Some of its associated stuff is still...

FCP complete; this is good to go! @bors r+

Thanks for the PR! While I'm sometimes willing to add new APIs without libs-api assent, I'm much more worried about churning signatures without contacting them to see if they like...

Thanks for the PR! While I'm sometimes willing to add new APIs without libs-api assent, I'm much more worried about churning signatures without contacting them to see if they like...