linux-kernel - Re: [PATCH v2] rust: locks: Add `get

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]

Date: Thu, 15 Feb 2024 13:50:58 -0300
From: Wedson Almeida Filho <wedsonaf@...il.com>
To: Alice Ryhl <aliceryhl@...gle.com>
Cc: mathys35.gasnier@...il.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@...sung.com>, 
	rust-for-linux@...r.kernel.org, linux-kernel@...r.kernel.org, 
	Martin Rodriguez Reboredo <yakoyoku@...il.com>
Subject: Re: [PATCH v2] rust: locks: Add `get_mut` method to `Lock`

On Mon, 12 Feb 2024 at 11:22, Alice Ryhl <aliceryhl@...gle.com> wrote:
>
> On Mon, Feb 12, 2024 at 3:13 PM Mathys-Gasnier via B4 Relay
> <devnull+mathys35.gasnier.gmail.com@...nel.org> wrote:
> > +    /// Gets the data contained in the lock
> > +    /// Having a mutable reference to the lock guarantees that no other threads have access to the lock.
> > +    /// Making it safe to get a mutable reference to the lock content.
> > +    pub fn get_mut(&mut self) -> &mut T {
> > +        self.data.get_mut()
> > +    }
>
> It's impossible to call this method. You can never have a mutable
> reference to a Linux mutex because we pin our locks. At most, you can
> have a Pin<&mut Self>.

Perhaps you meant to say that it's impossible to call this method
without unsafe blocks? From a `Pin<&mut T>`, we can call
`get_unchecked_mut` to get an `&mut T`.

This is addressing issue 924 opened by Björn some time back. The idea
here is that if there's a path where avoiding the lock/unlock calls
(which are expensive because of the memory barriers) is performance
critical, we can do it as long as we use an unsafe block.