Add a conversion from &mut T to &mut UnsafeCell<T>

[joseph-gio]

Provides a safe way of downgrading an exclusive reference into an alias-able &UnsafeCell<T> reference.

I would appreciate guidance on what exactly I should put for the stability attribute.

ACP: rust-lang/libs-team#198.

[rustbot]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @joshtriplett (or someone else) soon.

Please see the contribution instructions for more information.

[rustbot]

Hey! It looks like you've submitted a new PR for the library teams!

If this PR contains changes to any rust-lang/rust public library APIs then please comment with @rustbot label +T-libs-api -T-libs to tag it appropriately. If this PR contains changes to any unstable APIs please edit the PR description to add a link to the relevant API Change Proposal or create one if you haven't already. If you're unsure where your change falls no worries, just leave it as is and the reviewer will take a look and make a decision to forward on if necessary.

Examples of T-libs-api changes:

• Stabilizing library features
• Introducing insta-stable changes such as new implementations of existing stable traits on existing stable types
• Introducing new or changing existing unstable library APIs (excluding permanently unstable features / features without a tracking issue)
• Changing public documentation in ways that create new stability guarantees
• Changing observable runtime behavior of library APIs

[joseph-gio]

@rustbot label +T-libs-api -T-libs

[joseph-gio]

@rustbot label -T-libs-api +T-libs

[clubby789]

This is a libs-api change:

Introducing new or changing existing unstable library APIs

You'll also need to make an ACP and mark this as S-waiting-on-ACP once you've created it

[joseph-gio]

I see, thank you.

@rustbot label +T-libs-api +S-waiting-on-ACP -T-libs

[KamilaBorowska]

I'm not sure if that should return &mut UnsafeCell<T> when Cell::from_mut returns &Cell<T>. That said, I'm not seeing obvious safety problems, so it may as well be fine, and &mut UnsafeCell<T> is more useful as it has get_mut method.

It's worth noting that we already guarantee converting from &mut T to &UnsafeCell<T> to be possible, https://doc.rust-lang.org/stable/std/cell/struct.UnsafeCell.html has the following function in the examples:

fn get_shared<T>(ptr: &mut T) -> &UnsafeCell<T> {
  let t = ptr as *mut T as *const UnsafeCell<T>;
  // SAFETY: `T` and `UnsafeCell<T>` have the same memory layout
  unsafe { &*t }
}

[clarfonthey]

I'm not sure if that should return &mut UnsafeCell<T> when Cell::from_mut returns &Cell<T>. That said, I'm not seeing obvious safety problems, so it may as well be fine, and &mut UnsafeCell<T> is more useful as it has get_mut method.

Note that Cell<T> also has a get_mut method, so, I would say that this argument doesn't apply here.

[joseph-gio]

I think it was probably a mistake for Cell::from_mut to return &Cell -- &mut Cell is more useful and I'm not aware of any issues introduced by returning it.

[anden3]

@JoJoJet Just want to ping you as part of the triage procedure to let you know that CI fails for this PR :)

[joseph-gio]

@anden3 Yes, thank you. Like I said, I'll need a bit of guidance, as I haven't contributed to this repo before :).

I believe that I need to add a feature flag for this and gate the new function behind it, but I wasn't able to locate the guidelines for this type of thing.

[anden3]

I believe that I need to add a feature flag for this and gate the new function behind it, but I wasn't able to locate the guidelines for this type of thing.

Would this help?
https://rustc-dev-guide.rust-lang.org/feature-gates.html

[joseph-gio]

Yes that helps, thank you.

[Dylan-DPC]

@JoJoJet this needs an ACP, have you created one already for it? If so can you post it here or preferably add it to the title of the pr? thanks

[joseph-gio]

The ACP is at rust-lang/libs-team#198. What do you want me to add to the title?

[Dylan-DPC]

The ACP link but that's fine i'll do it :P

[joseph-gio]

@rustbot label -S-waiting-on-author

[joshtriplett]

Please add a tracking issue for this. Once that's done, r=me (you can write @ bors r = joshtriplett without spaces).

@bors delegate+

[bors]

✌️ @JoJoJet can now approve this pull request

[lukas-code]

This isn't necessary for library features, the feature name will picked up from the #[unstable] attribute directly.

[joseph-gio]

I see, thank you.

[joseph-gio]

@bors r=joshtriplett

[bors]

📌 Commit 810a87bb92347101cf6230d944993a19983544cb has been approved by joshtriplett

It is now in the queue for this repository.

[klensy]

0de00d1 https://rustc-dev-guide.rust-lang.org/git.html#no-merge-policy. 10 commits for 22 lines == squash commits.

[Noratrieb]

Yeah, we have a no merge policy and also there are indeed a lot of commits. Could you squash them together? If you need help, the linked page above should contain instructions.
@bors r-
nice catch @klensy

[rustbot]

Some changes occurred in src/tools/cargo

cc @ehuss

[rust-log-analyzer]

The job mingw-check failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

   Compiling clap v4.2.4
   Compiling clap_complete v4.2.2
    Finished dev [unoptimized] target(s) in 17.08s
##[endgroup]
error: current package believes it's in a workspace when it's not:
workspace: /checkout/Cargo.toml


this may be fixable by adding `src/tools/cargo` to the `workspace.members` array of the manifest located at: /checkout/Cargo.toml
Alternatively, to keep it out of the workspace, add the package to the `workspace.exclude` array, or add an empty `[workspace]` table to the package's manifest.
thread 'main' panicked at 'command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "metadata" "--format-version" "1" "--no-deps" "--manifest-path" "/checkout/src/tools/cargo/Cargo.toml"
expected success, got: exit status: 101', metadata.rs:83:31
Build completed unsuccessfully in 0:00:28
make: *** [Makefile:58: prepare] Error 1
Command failed. Attempt 2/5:
##[group]Building bootstrap
##[group]Building bootstrap
    Finished dev [unoptimized] target(s) in 0.05s
##[endgroup]
error: current package believes it's in a workspace when it's not:
workspace: /checkout/Cargo.toml


this may be fixable by adding `src/tools/cargo` to the `workspace.members` array of the manifest located at: /checkout/Cargo.toml
Alternatively, to keep it out of the workspace, add the package to the `workspace.exclude` array, or add an empty `[workspace]` table to the package's manifest.
thread 'main' panicked at 'command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "metadata" "--format-version" "1" "--no-deps" "--manifest-path" "/checkout/src/tools/cargo/Cargo.toml"
expected success, got: exit status: 101', metadata.rs:83:31
Build completed unsuccessfully in 0:00:00
make: *** [Makefile:58: prepare] Error 1
Command failed. Attempt 3/5:
##[group]Building bootstrap
##[group]Building bootstrap
    Finished dev [unoptimized] target(s) in 0.04s
##[endgroup]
error: current package believes it's in a workspace when it's not:
workspace: /checkout/Cargo.toml


this may be fixable by adding `src/tools/cargo` to the `workspace.members` array of the manifest located at: /checkout/Cargo.toml
Alternatively, to keep it out of the workspace, add the package to the `workspace.exclude` array, or add an empty `[workspace]` table to the package's manifest.
thread 'main' panicked at 'command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "metadata" "--format-version" "1" "--no-deps" "--manifest-path" "/checkout/src/tools/cargo/Cargo.toml"
expected success, got: exit status: 101', metadata.rs:83:31
Build completed unsuccessfully in 0:00:00
make: *** [Makefile:58: prepare] Error 1
Command failed. Attempt 4/5:
##[group]Building bootstrap
##[group]Building bootstrap
    Finished dev [unoptimized] target(s) in 0.04s
##[endgroup]
error: current package believes it's in a workspace when it's not:
workspace: /checkout/Cargo.toml


this may be fixable by adding `src/tools/cargo` to the `workspace.members` array of the manifest located at: /checkout/Cargo.toml
Alternatively, to keep it out of the workspace, add the package to the `workspace.exclude` array, or add an empty `[workspace]` table to the package's manifest.
thread 'main' panicked at 'command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "metadata" "--format-version" "1" "--no-deps" "--manifest-path" "/checkout/src/tools/cargo/Cargo.toml"
expected success, got: exit status: 101', metadata.rs:83:31
Build completed unsuccessfully in 0:00:00
make: *** [Makefile:58: prepare] Error 1
Command failed. Attempt 5/5:
##[group]Building bootstrap
##[group]Building bootstrap
    Finished dev [unoptimized] target(s) in 0.04s
##[endgroup]
error: current package believes it's in a workspace when it's not:
workspace: /checkout/Cargo.toml


this may be fixable by adding `src/tools/cargo` to the `workspace.members` array of the manifest located at: /checkout/Cargo.toml
Alternatively, to keep it out of the workspace, add the package to the `workspace.exclude` array, or add an empty `[workspace]` table to the package's manifest.
thread 'main' panicked at 'command did not execute successfully: "/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo" "metadata" "--format-version" "1" "--no-deps" "--manifest-path" "/checkout/src/tools/cargo/Cargo.toml"
expected success, got: exit status: 101', metadata.rs:83:31
Build completed unsuccessfully in 0:00:00
make: *** [Makefile:58: prepare] Error 1
The command has failed after 5 attempts.