r? @lcnr

rustbot has assigned @lcnr.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Some changes occurred to the CTFE machinery

cc @rust-lang/wg-const-eval

@bors r=compiler-errors

📌 Commit a61cd86 has been approved by compiler-errors

It is now in the queue for this repository.

Uuh, I'm seeing a bunch of new false positives for the clippy::missing-const-for-fn lint popping up, presumably due to this change.

For example

//@ run-pass
use std::ops::Add;

fn f<T: Add>(a: T, b: T) -> <T as Add>::Output {
    a + b
}

fn main() {
    println!("a + b is {}", f::<f32>(100f32, 200f32));
}

now lints

warning: this could be a `const fn`
 --> ./tests/ui/issues/issue-22258.rs:4:1
  |
4 | / fn f<T: Add>(a: T, b: T) -> <T as Add>::Output {
5 | |     a + b
6 | | }
  | |_^
  |
  = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#missing_const_for_fn
  = note: requested on the command line with `-W clippy::missing-const-for-fn`
help: make the function `const`
  |
4 | const fn f<T: Add>(a: T, b: T) -> <T as Add>::Output {
  | +++++

which is not correct

yes, the clippy lint needs to look at const stability of traits. The add trait specifically is unstably const