linux-kernel - Re: [PATCH v8 5/6] rust: ww_mutex: add Mutex, AcquireCtx and MutexGuard

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] [day] [month] [year] [list]

Message-Id: <55E30790-0370-4910-B8E9-C007052BF285@collabora.com>
Date: Wed, 17 Dec 2025 05:54:46 -0300
From: Daniel Almeida <daniel.almeida@...labora.com>
To: Onur Özkan <work@...rozkan.dev>
Cc: Alice Ryhl <aliceryhl@...gle.com>,
 rust-for-linux@...r.kernel.org,
 lossin@...nel.org,
 lyude@...hat.com,
 ojeda@...nel.org,
 alex.gaynor@...il.com,
 boqun.feng@...il.com,
 gary@...yguo.net,
 a.hindborg@...nel.org,
 tmgross@...ch.edu,
 dakr@...nel.org,
 peterz@...radead.org,
 mingo@...hat.com,
 will@...nel.org,
 longman@...hat.com,
 felipe_life@...e.com,
 daniel@...lak.dev,
 thomas.hellstrom@...ux.intel.com,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8 5/6] rust: ww_mutex: add Mutex, AcquireCtx and
 MutexGuard



> On 15 Dec 2025, at 06:10, Onur Özkan <work@...rozkan.dev> wrote:
> 
> On Thu, 4 Dec 2025 09:07:18 +0000
> Alice Ryhl <aliceryhl@...gle.com> wrote:
> 
>> On Wed, Dec 03, 2025 at 02:23:14PM -0300, Daniel Almeida wrote:
>>> 
>>> 
>>>> On 3 Dec 2025, at 10:26, Alice Ryhl <aliceryhl@...gle.com> wrote:
>>>> 
>>>> On Mon, Dec 01, 2025 at 01:28:54PM +0300, Onur Özkan wrote:
>>>>> Yeah :(. We could get rid of them easily by keeping the class
>>>>> that was passed to the constructor functions but that becomes a
>>>>> problem for the from_raw implementations.
>>>>> 
>>>>> I think the best solution would be to expose ww_class type from
>>>>> ww_acquire_ctx and ww_mutex unconditionally (right now it
>>>>> depends on DEBUG_WW_MUTEXES). That way we can just access the
>>>>> class and verify that the mutex and acquire_ctx classes match.
>>>>> 
>>>>> What do you think? I can submit a patch for the C-side
>>>>> implementation. It should be straightforward and shouldn't have
>>>>> any runtime impact.
>>>> 
>>>> I think there is a better solution. We can create a different
>>>> type for every single class, like how
>>>> rust/kernel/sync/lock/global.rs creates a different type for
>>>> every single mutex. Then, you know that the classes are the same
>>>> since the class is part of the type.
>>> 
>>> I don’t think this would work with the from_raw() functions. What
>>> class would you assign then? I think this is precisely what sparked
>>> the current solution.
>> 
>> There can be a way to create a type for a C-defined class, and
>> from_raw() can require that you don't use the same Rust type for
>> different C classes.
>> 
> 
> Do you think this is a better alternative? IMO it doesn't seem worth
> it for what it's doing. Current approach adds less complexity and is
> easier to maintain. It's not just helping from_raw functions, the class
> validation is being much simpler without having to deal with storing
> class references or creating new types.
> 
> I am holding off the next version because we don't have a clear
> consensus on this.
> 
> - Onur
> 
>> Alice


I am frankly not sure whether what Alice suggested is an improvement either.
Having the class stored seems like a much simpler solution to the problem.

> Then when you call from_raw(), you call
> 
> Mutex::<T, MY_C_CLASS>::from_raw(ptr_to_mutex)

That’s what we are trying to avoid in the first place. It does not really
matter that from_raw() is unsafe, because this is not really about UB, but
rather about trying to avoid broken code if possible.

IMHO, if we can check this, then I propose that we do. Hence why we are storing
the class.


— Daniel