linux-kernel - Re: [PATCH v8 4/9] rust: acpi: add `acpi::DeviceId` abstraction

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: <aF2RNVJVIHyKuead@cassiopeiae>
Date: Thu, 26 Jun 2025 20:28:05 +0200
From: Danilo Krummrich <dakr@...nel.org>
To: Igor Korotin <igor.korotin.linux@...il.com>
Cc: "Rafael J . Wysocki" <rafael@...nel.org>,
	Alex Gaynor <alex.gaynor@...il.com>,
	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	Miguel Ojeda <ojeda@...nel.org>, Rob Herring <robh@...nel.org>,
	Saravana Kannan <saravanak@...gle.com>,
	Alex Hung <alex.hung@....com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Jakub Kicinski <kuba@...nel.org>,
	Jonathan Cameron <Jonathan.Cameron@...wei.com>,
	Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
	Mauro Carvalho Chehab <mchehab+huawei@...nel.org>,
	Remo Senekowitsch <remo@...nzli.dev>,
	Tamir Duberstein <tamird@...il.com>,
	Viresh Kumar <viresh.kumar@...aro.org>,
	Wedson Almeida Filho <wedsonaf@...il.com>,
	Xiangfei Ding <dingxiangfei2009@...il.com>,
	devicetree@...r.kernel.org, linux-acpi@...r.kernel.org,
	linux-kernel@...r.kernel.org, rust-for-linux@...r.kernel.org,
	Björn Roy Baron <bjorn3_gh@...tonmail.com>,
	Alice Ryhl <aliceryhl@...gle.com>,
	Andreas Hindborg <a.hindborg@...nel.org>,
	Benno Lossin <lossin@...nel.org>, Boqun Feng <boqun.feng@...il.com>,
	Gary Guo <gary@...yguo.net>, Len Brown <lenb@...nel.org>,
	Trevor Gross <tmgross@...ch.edu>
Subject: Re: [PATCH v8 4/9] rust: acpi: add `acpi::DeviceId` abstraction

On Thu, Jun 26, 2025 at 06:40:06PM +0100, Igor Korotin wrote:
> 
> 
> On 6/26/25 16:25, Danilo Krummrich wrote:
> > On 6/20/25 5:24 PM, Igor Korotin wrote:
> >> +impl DeviceId {
> >> +    const ACPI_ID_LEN: usize = 16;
> >> +
> >> +    /// Create a new device id from an ACPI 'id' string.
> >> +    pub const fn new<const N: usize>(id: &[u8; N]) -> Self {
> > 
> > Didn't notice before, but why was this silently changed from &CStr to
> > &[u8; N]
> > from v6 to v7?
> > 
> >> +        build_assert!(N <= Self::ACPI_ID_LEN, "ID exceeds 16 bytes");
> >> +        // Replace with `bindings::acpi_device_id::default()` once
> >> stabilized for `const`.
> >> +        // SAFETY: FFI type is valid to be zero-initialized.
> >> +        let mut acpi: bindings::acpi_device_id = unsafe
> >> { core::mem::zeroed() };
> >> +        let mut i = 0;
> >> +        while i < N {
> >> +            acpi.id[i] = id[i];
> >> +            i += 1;
> >> +        }
> >> +
> >> +        Self(acpi)
> >> +    }
> >> +}
> 
> In v6 I was asked to change assert! (runtime) to build_assert! (build time)
> It was as follows:
> 
> > +    pub const fn new(id: &'static CStr) -> Self {
> > +        assert!(id.len() <= Self::ACPI_ID_LEN, "ID exceeds 16 bytes");
> 
> but id.len() breaks const context and so build_assert! triggers
> assertion.

It does indeed, but I'm not sure why it does...

> If I needed to explicitly describe change from CStr to
> [u8;20], then it's my bad.

Yes, that's usually better. Otherwise reviewers might skip the changed part. In
this case I think it actually introduced a bug:

Checking for N <= Self::ACPI_ID_LEN it can happen that acpi_device_id::id is not
NULL terminated anymore, whereas before this was ensured by
CStr::as_bytes_with_nul(). See also [1].

I think we can easily fix this by just checking N < Self::ACPI_ID_LEN.

However, I'd still like to check why build_assert!() does not work with
CStr::len() in this case first.

[1] https://elixir.bootlin.com/linux/v6.15.3/source/drivers/acpi/bus.c#L899