linux-kernel - Re: [PATCH v5 4/4] rust: adding OwnableRefCounted and SimpleOwnableRefCounted

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CANiq72mYfhuRWkjomb1vOMMPOaxvdS6qjfVLAwxUw6ecdqyh2A@mail.gmail.com>
Date: Fri, 7 Mar 2025 14:16:32 +0100
From: Miguel Ojeda <miguel.ojeda.sandonis@...il.com>
To: Oliver Mangold <oliver.mangold@...me>
Cc: Miguel Ojeda <ojeda@...nel.org>, Alex Gaynor <alex.gaynor@...il.com>, 
	Boqun Feng <boqun.feng@...il.com>, Gary Guo <gary@...yguo.net>, 
	Björn Roy Baron <bjorn3_gh@...tonmail.com>, 
	Benno Lossin <benno.lossin@...ton.me>, Andreas Hindborg <a.hindborg@...nel.org>, 
	Alice Ryhl <aliceryhl@...gle.com>, Trevor Gross <tmgross@...ch.edu>, 
	Asahi Lina <lina@...hilina.net>, rust-for-linux@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5 4/4] rust: adding OwnableRefCounted and SimpleOwnableRefCounted

Hi Oliver,

Some general style nits for this and other series you may send in the
future (not a review). Please note that most may apply several times.

On Fri, Mar 7, 2025 at 11:04 AM Oliver Mangold <oliver.mangold@...me> wrote:
>
> Types implementing one of these traits
> can safely convert between an ARef<T> and an Owned<T>.

The wrapping is strange here, and it also happens in your code
comments. Please use the same width as the rest of the code in a file
etc.

> +/// - The same safety requirements as for [`Ownable`] and [`RefCounted`] apply.

"same" sounds like no extra requirements -- what about something like:

    The safety requirements from both [.....

I wonder if we should expand/inline them, even if they come from the
supertraits.

> +/// - the uniqueness invariant of [`Owned`] is upheld until dropped.

Please use uppercase to start sentences.

> +///         // Use a KBox to handle the actual allocation

Please use Markdown for comments too, and period at the end (also for
"SAFETY: ..." comments).

> +/// // SAFETY: we implement the trait correctly by ensuring

There is no need to say "we implement the trait correctly", i.e. the
`SAFETY` tag is enough to introduce the comment; the same way we don't
say "SAFETY: the following unsafe block is correct because ..." etc.

> +///     }
> +///     fn is_unique(&self) -> bool {

Newline between items.

> +/// let foo = Foo::new().unwrap();

In general, we try to avoid showing patterns that people should avoid
when writing actual code (at least in non-hidden code) -- could we
avoid the `unwrap()`?

> +// TODO: enable this when compiler supports it (>=1.85)
> +// #[diagnostic::do_not_recommend]

I think (untested) we could conditionally enable this already and
remove the `TODO` with something like:

    config RUSTC_HAS_DO_NOT_RECOMMEND
        def_bool RUSTC_VERSION >= 108500

    #[cfg_attr(CONFIG_RUSTC_HAS_DO_NOT_RECOMMEND, diagnostic::do_not_recommend)]

Thanks!

Cheers,
Miguel