For box expressions, use NZ drop instead of a free block

[arielb1]

This falls naturally out of making drop elaboration work with box
expressions, which is probably required for sane MIR borrow-checking.
This is a pure refactoring with no intentional functional effects.

r? @nagisa

[nagisa]

I would go ahead and remove this parameter.

[nagisa]

@bors r+

Feel free to bors r=nagisa if you fix the nit yourself.

[bors]

📌 Commit 17d2bcd has been approved by nagisa

[bors]

⌛ Testing commit 17d2bcd with merge b8266a9...

[bors]

☀️ Test successful - status-appveyor, status-travis
Approved by: nagisa
Pushing b8266a9 to master...

[oli-obk]

I think this caused let x = box 4i32; to no longer produce a StorageLive for the temporary used to assign x.

fn main() -> () {
let mut _0: (); // return pointer
scope 1 {
let _1: std::boxed::Box<i32>; // "x" in scope 1 at src/main.rs:4:9: 4:10
}
let mut _2: std::boxed::Box<i32>;
bb0: {
StorageLive(_1); // scope 0 at src/main.rs:4:9: 4:10
_2 = Box(i32); // scope 0 at src/main.rs:4:13: 4:21
(*_2) = const 4i32; // scope 0 at src/main.rs:4:17: 4:21
_1 = _2; // scope 0 at src/main.rs:4:13: 4:21
StorageDead(_2); // scope 0 at src/main.rs:4:21: 4:21
_0 = (); // scope 0 at src/main.rs:3:11: 5:2
drop(_1) -> bb1; // scope 0 at src/main.rs:5:2: 5:2
}
bb1: {
StorageDead(_1); // scope 0 at src/main.rs:5:2: 5:2
return; // scope 0 at src/main.rs:5:2: 5:2
}
}

Is this intended?

[arielb1]

Is this intended?

No.

[arielb1]

I think the difference is that we started emitting StorageDead, rather than stopping emitting StorageLive.