linux-kernel - Re: [PATCH 2/5] rust: alloc: add Vec::pop

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: <Z90YyLPhil2vBTc_@google.com>
Date: Fri, 21 Mar 2025 07:44:08 +0000
From: Alice Ryhl <aliceryhl@...gle.com>
To: Benno Lossin <benno.lossin@...ton.me>
Cc: Danilo Krummrich <dakr@...nel.org>, rust-for-linux@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/5] rust: alloc: add Vec::pop

On Thu, Mar 20, 2025 at 10:11:09PM +0000, Benno Lossin wrote:
> On Thu Mar 20, 2025 at 2:52 PM CET, Alice Ryhl wrote:
> > This introduces a basic method that our custom Vec is missing. I expect
> > that it will be used in many places, but at the time of writing, Rust
> > Binder has six calls to Vec::pop.
> >
> > Signed-off-by: Alice Ryhl <aliceryhl@...gle.com>
> > ---
> >  rust/kernel/alloc/kvec.rs | 31 +++++++++++++++++++++++++++++++
> >  1 file changed, 31 insertions(+)
> >
> > diff --git a/rust/kernel/alloc/kvec.rs b/rust/kernel/alloc/kvec.rs
> > index 95e752ed27395fce72d372976b74fb1b0e957194..9943358c70aa63f5ad7ed9782cb8879d7a80a8fb 100644
> > --- a/rust/kernel/alloc/kvec.rs
> > +++ b/rust/kernel/alloc/kvec.rs
> > @@ -302,6 +302,37 @@ pub fn push(&mut self, v: T, flags: Flags) -> Result<(), AllocError> {
> >          Ok(())
> >      }
> >  
> > +    /// Removes the last element from a vector and returns it, or `None` if it is empty.
> > +    ///
> > +    /// # Examples
> > +    ///
> > +    /// ```
> > +    /// let mut v = KVec::new();
> > +    /// v.push(1, GFP_KERNEL)?;
> > +    /// v.push(2, GFP_KERNEL)?;
> > +    /// assert_eq!(&v, &[1, 2]);
> > +    ///
> > +    /// assert_eq!(v.pop(), Some(2));
> > +    /// assert_eq!(v.pop(), Some(1));
> > +    /// assert_eq!(v.pop(), None);
> > +    /// # Ok::<(), Error>(())
> > +    /// ```
> > +    pub fn pop(&mut self) -> Option<T> {
> > +        let Some(len_sub_1) = self.len.checked_sub(1) else {
> > +            return None;
> > +        };
> 
> Isn't it possible to do:?
>     
>     let len_sub_1 = self.len.checked_sub(1)?;

Yes, good catch.

> > +
> > +        // INVARIANT: If the first `len` elements are valid, then the first `len-1` elements are
> 
> Please add spaces around `-`.

I can do that, but does it really read better?

> > +        // valid.
> > +        self.len = len_sub_1;
> > +
> > +        // INVARIANT: This invalidates a value in this vector's allocation, but the Vec invariants
> > +        // do not require it to be valid because `self.len <= len_sub_1`.
> 
> I don't think this should be an `INVARIANT` comment. Maybe we don't even
> need it.

I can drop the INVARIANT: prefix.

Alice