Add missing mut to pin.rs docs

[justanotheranonymoususer]

Per my understanding, needed for mut access next line.

[rustbot]

r? @scottmcm

rustbot has assigned @scottmcm.
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

[scottmcm]

Punting on pin,
@rustbot reroll

[joboet]

Sure, though this isn't intended to compile anyway...
@bors r+ rollup

[rust-bors]

📌 Commit 39b08b6 has been approved by joboet

It is now in the queue for this repository.

[justanotheranonymoususer]

this isn't intended to compile anyway...

Only because get_pin_mut doesn't exist. It compiles without that line.

Also, I figured the argument to exploit_ref_cell should be mut.

This compiles:

use std::cell::RefCell;
use std::pin::Pin;

trait RefCellPinExt<'a, T> {
    fn get_pin_mut(self) -> Pin<&'a mut T>;
}

impl<'a, T> RefCellPinExt<'a, T> for Pin<&'a mut RefCell<T>> {
    fn get_pin_mut(self) -> Pin<&'a mut T> {
        // SAFETY: That's actually unsafe. Don't do this in real code!
        unsafe { self.map_unchecked_mut(|rc| rc.get_mut()) }
    }
}

fn exploit_ref_cell<T>(mut rc: Pin<&mut RefCell<T>>) {
    // Here we get pinned access to the `T`.
    let _: Pin<&mut T> = rc.as_mut().get_pin_mut();

    // And here we have `&mut T` to the same data.
    let shared: &RefCell<T> = rc.into_ref().get_ref();
    let mut borrow = shared.borrow_mut();
    let content = &mut *borrow;
}

https://play.rust-lang.org/?version=stable&mode=debug&edition=2024&gist=e54a0a6082785b63359650b128f15071

Maybe we better replace it with this version? It's longer, but the compiler can check that it compiles. It's incorrect and unsafe, but having it compile makes sure the example is not bogus.

[rust-bors]

⚠️ A new commit 2a68b2ec8a6735435fe91af086b0f32042540151 was pushed to the branch, the PR will need to be re-approved.

[justanotheranonymoususer]

Updated to code that compiles, LMK WDYT

[justanotheranonymoususer]

@joboet ping

[joboet]

Updated to code that compiles, LMK WDYT

I don't think that's a good idea. The example not compiling makes it even clearer that this is unsound. Could you please revert to the previous version?

@joboet _pin_g

😄. Sorry, I filter my review queue by S-waiting-on-review, so your PR didn't appear. You can use @rustbot ready to change the label.

[justanotheranonymoususer]

@rustbot ready

[joboet]

Thanks! Just one thing before I approve this again: Could you squash the commits into one, please?

[justanotheranonymoususer]

Not via github, I'm afraid. Can you use squash merge? Or is it an absolute requirement for me to clone the repo?

[joboet]

We use a custom merge bot that isn't able to do squash merges unfortunately. I've performed the squash for you.

[justanotheranonymoususer]

@rustbot ready

[joboet]

@bors r+ rollup

[rust-bors]

📌 Commit 672828e has been approved by joboet

It is now in the queue for this repository.