[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <877c2zz0fh.fsf@kernel.org>
Date: Fri, 02 May 2025 14:01:22 +0200
From: Andreas Hindborg <a.hindborg@...nel.org>
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>, "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 v10 3/5] rust: Add missing SAFETY documentation for ARef
example
"Oliver Mangold" <oliver.mangold@...me> writes:
> On 250502 1241, Andreas Hindborg wrote:
>> >
>> > diff --git a/rust/kernel/types.rs b/rust/kernel/types.rs
>> > index d7fa8934c545f46a646ca900ab8957a04b0ad34d..33d2b4e4a87b991c6d934f4e8d2c6c71a15b1bcb 100644
>> > --- a/rust/kernel/types.rs
>> > +++ b/rust/kernel/types.rs
>> > @@ -498,7 +498,9 @@ pub unsafe fn from_raw(ptr: NonNull<T>) -> Self {
>> > ///
>> > /// struct Empty {}
>> > ///
>> > - /// # // SAFETY: TODO.
>> > + /// // SAFETY: The `RefCounted` implementation for `Empty` does not count references
>> > + /// // and never frees the underlying object. Thus we can act as having a
>> > + /// // refcount on the object that we pass to the newly created `ARef`.
>> > /// unsafe impl RefCounted for Empty {
>> > /// fn inc_ref(&self) {}
>> > /// unsafe fn dec_ref(_obj: NonNull<Self>) {}
>> > @@ -506,7 +508,7 @@ pub unsafe fn from_raw(ptr: NonNull<T>) -> Self {
>> > ///
>> > /// let mut data = Empty {};
>> > /// let ptr = NonNull::<Empty>::new(&mut data).unwrap();
>> > - /// # // SAFETY: TODO.
>> > + /// // SAFETY: We keep `data` around longer than the `ARef`.
>>
>> I still think this applies:
>>
>> >> How about:
>> >>
>> >> The `RefCounted` implementation for `Empty` does not count references
>> >> and never frees the underlying object. Thus we can act as having a
>> >> refcount on the object that we pass to the newly created `ARef`.
>> >>
>
> Hi Andreas,
>
> I agree. Sorry, I just messed up the fix. Your wording landed in the
> previous to-be-fixed unsafe comment, as you can see.
>
> Happens when you are too much in a hurry and didn't touch the patch for
> too long :/
>
> I will fix it in the next version.
Cool! Also check my response to v9: https://lore.kernel.org/all/87cycrz1pa.fsf@kernel.org
Best regards,
Andreas Hindborg
Powered by blists - more mailing lists