ambiguous_wide_pointer_comparisons: bad suggestion

[matthiaskrgr]

Code

//@ check-pass

use std::cmp::PartialEq;
use std::rc::Rc;
use std::sync::Arc;

struct A;
struct B;

trait T {}
impl T for A {}
impl T for B {}

fn main() {
    fn cmp<T: ?Sized>(a: *mut T, b: *mut T) -> bool {
        let _ = a == b;
        //~^ WARN ambiguous wide pointer comparison

        panic!();
    }
}

Current output

warning: unused import: `std::cmp::PartialEq`
 --> src/main.rs:3:5
  |
3 | use std::cmp::PartialEq;
  |     ^^^^^^^^^^^^^^^^^^^
  |
  = note: `#[warn(unused_imports)]` on by default

warning: unused import: `std::rc::Rc`
 --> src/main.rs:4:5
  |
4 | use std::rc::Rc;
  |     ^^^^^^^^^^^

warning: unused import: `std::sync::Arc`
 --> src/main.rs:5:5
  |
5 | use std::sync::Arc;
  |     ^^^^^^^^^^^^^^

warning: struct `A` is never constructed
 --> src/main.rs:7:8
  |
7 | struct A;
  |        ^
  |
  = note: `#[warn(dead_code)]` on by default

warning: struct `B` is never constructed
 --> src/main.rs:8:8
  |
8 | struct B;
  |        ^

warning: trait `T` is never used
  --> src/main.rs:10:7
   |
10 | trait T {}
   |       ^

warning: function `cmp` is never used
  --> src/main.rs:15:8
   |
15 |     fn cmp<T: ?Sized>(a: *mut T, b: *mut T) -> bool {
   |        ^^^

warning: ambiguous wide pointer comparison, the comparison includes metadata which may not be expected
  --> src/main.rs:16:17
   |
16 |         let _ = a == b;
   |                 ^^^^^^
   |
   = note: `#[warn(ambiguous_wide_pointer_comparisons)]` on by default
help: use `std::ptr::addr_eq` or untyped pointers to only compare their addresses
   |
16 |         let _ = std::ptr::addr_eq(a, b);
   |                 ++++++++++++++++++ ~  +
help: use explicit `std::ptr::eq` method to compare metadata and addresses
   |
16 |         let _ = std::ptr::eq(a, b);
   |                 +++++++++++++ ~  +

Desired output

No response

Rationale and extra context

The lint is marked cargo-fix able but applying the suggestion fails:

after fixes were automatically applied the compiler reported errors within these files:

  * src/main.rs

This likely indicates a bug in either rustc or cargo itself,
and we would appreciate a bug report! You're likely to see
a number of compiler warnings after this message which cargo
attempted to fix but failed. If you could open an issue at
https://github.com/rust-lang/rust/issues
quoting the full output of this command we'd be very appreciative!
Note that you may be able to make some more progress in the near-term
fixing code with the `--broken-code` flag

The following errors were reported:
error[E0061]: this function takes 2 arguments but 1 argument was supplied
    --> src/main.rs:16:17
     |
16   |         let _ = std::ptr::eq(std::ptr::addr_eq(a, b));
     |                 ^^^^^^^^^^^^------------------------- an argument of type `*const _` is missing
     |
note: expected `*const _`, found `bool`
    --> src/main.rs:16:30
     |
16   |         let _ = std::ptr::eq(std::ptr::addr_eq(a, b));
     |                              ^^^^^^^^^^^^^^^^^^^^^^^
     = note: expected raw pointer `*const _`
                       found type `bool`
note: function defined here
    --> /home/matthias/.rustup/toolchains/nightly-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/ptr/mod.rs:1932:8
     |
1932 | pub fn eq<T: ?Sized>(a: *const T, b: *const T) -> bool {
     |        ^^
help: provide the argument
     |
16   |         let _ = std::ptr::eq(/* a */, /* b */);
     |                             ~~~~~~~~~~~~~~~~~~

error: aborting due to 1 previous error

For more information about this error, try `rustc --explain E0061`.
Original diagnostics will follow.

warning: `wiiide` (bin "wiiide" test) generated 7 warnings (7 duplicates)

Other cases

No response

Rust Version

rustc 1.78.0-nightly (3246e7951 2024-02-19)
binary: rustc
commit-hash: 3246e79513cb89ddbfc0f21cb5a877e5b321dcc5
commit-date: 2024-02-19
host: x86_64-unknown-linux-gnu
release: 1.78.0-nightly
LLVM version: 18.1.0

Anything else?

No response

[matthiaskrgr]

Same problem with

pub fn main() {
   pub fn cmp<T: ?Sized>(a: *mut T, b: *mut T) -> bool {
        let _ = a.eq(&b);
        //~^ WARN ambiguous wide pointer comparison

        todo!();
    }
}

warning: ambiguous wide pointer comparison, the comparison includes metadata which may not be expected
 --> src/main.rs:3:17
  |
3 |         let _ = a.eq(&b);
  |                 ^^^^^^^^
  |
  = note: `#[warn(ambiguous_wide_pointer_comparisons)]` on by default
help: use `std::ptr::addr_eq` or untyped pointers to only compare their addresses
  |
3 |         let _ = std::ptr::addr_eq(a, b);
  |                 ++++++++++++++++++ ~  ~
help: use explicit `std::ptr::eq` method to compare metadata and addresses
  |
3 |         let _ = std::ptr::eq(a, b);
  |                 +++++++++++++ ~  ~

warning: `wiiide` (bin "wiiide" test) generated 1 warning (run `cargo fix --bin "wiiide" --tests` to apply 1 suggestion)
warning: failed to automatically apply fixes suggested by rustc to crate `wiiide`

after fixes were automatically applied the compiler reported errors within these files:

  * src/main.rs

This likely indicates a bug in either rustc or cargo itself,
and we would appreciate a bug report! You're likely to see
a number of compiler warnings after this message which cargo
attempted to fix but failed. If you could open an issue at
https://github.com/rust-lang/rust/issues
quoting the full output of this command we'd be very appreciative!
Note that you may be able to make some more progress in the near-term
fixing code with the `--broken-code` flag

The following errors were reported:
error: mismatched closing delimiter: `}`
 --> src/main.rs:3:29
  |
3 |         let _ = std::ptr::eq(std::ptr::addr_eq(a, b);
  |                             ^ unclosed delimiter
...
7 |     }
  |     ^ mismatched closing delimiter

error: aborting due to 1 previous error

Original diagnostics will follow.

[jieyouxu]

@rustbot claim

[Urgau]

@matthiaskrgr @jieyouxu I think this is a known issue with cargo fix with applying all suggestions even if they are multually exclusive. I'm pretty sure there's nothing to do in the lint, it correctly suggests the two (valid) suggestions, only one of them should be applied.

[matthiaskrgr]

Well we can either remove one of the suggestions or make the lint MaybeIncorrect or something?

[Urgau]

I think this is #53934.

Well we can either remove one of the suggestions or make the lint MaybeIncorrect or something?

We can't remove one of the suggestions, they are both necessary. We can however make them MaybeIncorrect for the time being.

[jieyouxu]

I think this is #53934.

That is indeed unfortunate.