linux-kernel - Re: [PATCH 07/10] rust: Add arrayvec

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: <Z-VjRmiWDj6teoFL@cassiopeiae>
Date: Thu, 27 Mar 2025 15:40:06 +0100
From: Danilo Krummrich <dakr@...nel.org>
To: Remo Senekowitsch <remo@...nzli.dev>
Cc: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
	Daniel Scally <djrscally@...il.com>,
	Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
	Sakari Ailus <sakari.ailus@...ux.intel.com>,
	Rob Herring <robh@...nel.org>, Dirk Behme <dirk.behme@...bosch.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	"Rafael J. Wysocki" <rafael@...nel.org>,
	Saravana Kannan <saravanak@...gle.com>,
	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>,
	linux-kernel@...r.kernel.org, linux-acpi@...r.kernel.org,
	devicetree@...r.kernel.org, rust-for-linux@...r.kernel.org
Subject: Re: [PATCH 07/10] rust: Add arrayvec

On Wed, Mar 26, 2025 at 06:13:46PM +0100, Remo Senekowitsch wrote:
> This patch is basically a proof of concept intendend to gather feedback
> about how to do this properly. Normally I would want to use the crate
> from crates.io[1], but that's not an option in the kernel. We could also
> vendor the entire source code of arrayvec. I'm not sure if people will
> be happy with that.

Do we really need this? The only user in this series I could spot was
property_get_reference_args(). And I think that with a proper abstraction of
struct fwnode_reference_args we could avoid to copy memory at all.

If it turns out we actually need something like this, I'd prefer to move it to
rust/kernel/alloc/ and see if it's worth to derive a common trait that maybe can
share a few things between ArrayVec and Vec.

> 
> [1] https://crates.io/crates/arrayvec
> 
> Signed-off-by: Remo Senekowitsch <remo@...nzli.dev>
> ---
>  rust/kernel/arrayvec.rs | 81 +++++++++++++++++++++++++++++++++++++++++
>  rust/kernel/lib.rs      |  1 +
>  2 files changed, 82 insertions(+)
>  create mode 100644 rust/kernel/arrayvec.rs
> 
> diff --git a/rust/kernel/arrayvec.rs b/rust/kernel/arrayvec.rs
> new file mode 100644
> index 000000000..041e7dcce
> --- /dev/null
> +++ b/rust/kernel/arrayvec.rs
> @@ -0,0 +1,81 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +//! Provides [ArrayVec], a stack-allocated vector with static capacity.
> +
> +use core::mem::MaybeUninit;
> +
> +/// A stack-allocated vector with statically fixed capacity.
> +///
> +/// This can be useful to avoid heap allocation and still ensure safety where a
> +/// small but dynamic number of elements is needed.
> +///
> +/// For example, consider a function that returns a variable number of values,
> +/// but no more than 8. In C, one might achieve this by passing a pointer to
> +/// a stack-allocated array as an out-parameter and making the function return
> +/// the actual number of elements. This is not safe, because nothing prevents
> +/// the caller from reading elements from the array that weren't actually
> +/// initialized by the function. `ArrayVec` solves this problem, users are
> +/// prevented from accessing uninitialized elements.
> +///
> +/// This basically exists already (in a much more mature form) on crates.io:
> +/// <https://crates.io/crates/arrayvec>
> +#[derive(Debug)]
> +pub struct ArrayVec<const N: usize, T> {
> +    array: [core::mem::MaybeUninit<T>; N],
> +    len: usize,
> +}
> +
> +impl<const N: usize, T> ArrayVec<N, T> {
> +    /// Adds a new element to the end of the vector.
> +    ///
> +    /// # Panics
> +    ///
> +    /// Panics if the vector is already full.
> +    pub fn push(&mut self, elem: T) {
> +        if self.len == N {
> +            panic!("OOM")

Please do not panic, this should return a Result instead.