-
Notifications
You must be signed in to change notification settings - Fork 138
[WIP] ReadOnly #2866
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: Gbec926233303ce55c894f8ba65c939dd42690530
Are you sure you want to change the base?
[WIP] ReadOnly #2866
Conversation
Summary of ChangesHello @joshlf, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request introduces a new Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Code Review
This pull request introduces a new ReadOnly<T> wrapper type. While it's a work in progress, there are two critical issues with the current implementation that need to be addressed. First, ReadOnly<T> incorrectly implements DerefMut, which contradicts its purpose and allows mutation of the wrapped data. Second, it has an unsound unconditional unsafe impl of the Immutable trait, which can lead to undefined behavior by violating one of the core safety invariants of zerocopy. I've provided detailed comments and suggestions for both of these critical issues.
| // SAFETY: TODO | ||
| const _: () = unsafe { | ||
| unsafe_impl!(T: ?Sized => Immutable for ReadOnly<T>); | ||
| }; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The unconditional unsafe impl<T: ?Sized> Immutable for ReadOnly<T> is unsound. The Immutable trait is a promise that the type has no interior mutability (i.e., no UnsafeCells). Since ReadOnly<T> is a #[repr(transparent)] wrapper, it will contain an UnsafeCell if T does. For example, if T is Cell<u8>, ReadOnly<Cell<u8>> would be marked Immutable despite containing an UnsafeCell. This violates a fundamental safety invariant of zerocopy and can lead to undefined behavior, for example when zerocopy performs unsynchronized reads on types it believes to be Immutable.
If ReadOnly<T> is intended to be Immutable, it must be bounded on T: Immutable. If it's intended to be used with types that are not Immutable, then ReadOnly<T> itself cannot be Immutable.
| // SAFETY: TODO | |
| const _: () = unsafe { | |
| unsafe_impl!(T: ?Sized => Immutable for ReadOnly<T>); | |
| }; | |
| // SAFETY: `ReadOnly<T>` is a transparent wrapper over `T`. Thus, it is | |
| // `Immutable` if and only if `T` is. | |
| const _: () = unsafe { | |
| unsafe_impl!(T: ?Sized + Immutable => Immutable for ReadOnly<T>); | |
| }; |
| impl<T: ?Sized + Immutable> DerefMut for ReadOnly<T> { | ||
| #[inline(always)] | ||
| fn deref_mut(&mut self) -> &mut Self::Target { | ||
| &mut self.0 | ||
| } | ||
| } |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The ReadOnly<T> type must not implement DerefMut. The name ReadOnly strongly implies that the wrapped data cannot be mutated, but implementing DerefMut provides a mechanism to get a mutable reference to it, which completely subverts this expectation. This is a major correctness and API design issue. It allows mutating data that is supposed to be read-only, which can lead to unsoundness and violates the principle of least astonishment.
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## Gbec926233303ce55c894f8ba65c939dd42690530 #2866 +/- ##
=============================================================================
- Coverage 91.89% 91.52% -0.37%
=============================================================================
Files 20 20
Lines 5773 5796 +23
=============================================================================
Hits 5305 5305
- Misses 468 491 +23 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
gherrit-pr-id: Gbe8d7edd150d80731c79815685c596ed88460ae7
ba5b262 to
7ad85b6
Compare
Latest Update: v2 — Compare vs v1
📚 Full Patch History
Links show the diff between the row version and the column version.