Expose LLVM's "or disjoint" instruction

libs
3

There is already a way to specify any assumption:

if !condition {
    unsafe { core::hint::unreachable_unchecked() };
}

The assert-unchecked crate wraps it into a macro assert_unchecked!.

So you can already write this as:

if a & b == 0 {
    a | b
} else {
   unsafe { unreachable_unchecked() }
}

Similarly, you can already write:

unsafe { a.checked_mul(b).unwrap_unchecked() }

so it's strange to me that additional functions like unchecked_mul are being added to u32 for this specific operation.

The rationale for adding unchecked_mul is that the above doesn't optimize equally well, but wouldn't a better solution be to improve the optimizer rather than add redundant functions?