[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20251125215919.6af03a2e@nimda.home>
Date: Tue, 25 Nov 2025 21:59:19 +0300
From: Onur Özkan <work@...rozkan.dev>
To: Daniel Almeida <daniel.almeida@...labora.com>
Cc: 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, aliceryhl@...gle.com,
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, bjorn3_gh@...tonmail.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 4/6] rust: ww_mutex: add Mutex, AcquireCtx and
MutexGuard
On Tue, 25 Nov 2025 15:32:08 -0300
Daniel Almeida <daniel.almeida@...labora.com> wrote:
[...]
> > + /// Similar to `lock`, but can be interrupted by signals.
> > + ///
> > + /// # Safety
> > + ///
> > + /// The given `mutex` must be created with the [`Class`] that
> > was used
> > + /// to initialize this [`AcquireCtx`].
> > + pub unsafe fn lock_interruptible<'a, T>(
> > + &'a self,
> > + mutex: &'a Mutex<'a, T>,
> > + ) -> Result<MutexGuard<'a, T>> {
> > + // SAFETY: By the safety contract, `mutex` belongs to the
> > same `Class`
> > + // as `self` does.
> > + unsafe { lock_common(mutex, Some(self),
> > LockKind::Interruptible) }
> > + }
> > +
> > + /// Locks the given mutex on this [`AcquireCtx`] using the
> > slow path.
> > + ///
> > + /// This function should be used when `lock` fails (typically
> > due to a potential deadlock).
> > + ///
> > + /// # Safety
> > + ///
> > + /// The given `mutex` must be created with the [`Class`] that
> > was used
> > + /// to initialize this [`AcquireCtx`].
>
> I’m assuming we either can’t or otherwise don’t want to store this in
> the Mutex itself?
>
> IOW: the class is ’static IIUC, so it’s always safe to store a
> reference anyways, at the cost of providing the storage.
>
We used to do this but I had to revert them when I implemented
`from_raw` functions. I wrote about it on [1], let me know what you
think about the solution idea I wrote there.
[1]: https://lore.kernel.org/all/20251124184928.30b8bbaf@nimda/
> > + pub unsafe fn lock_slow<'a, T>(&'a self, mutex: &'a Mutex<'a,
> > T>) -> Result<MutexGuard<'a, T>> {
> > + // SAFETY: By the safety contract, `mutex` belongs to the
> > same `Class`
> > + // as `self` does.
> > + unsafe { lock_common(mutex, Some(self), LockKind::Slow) }
> > + }
> > +
> > + /// Similar to `lock_slow`, but can be interrupted by signals.
> > + ///
> > + /// # Safety
> > + ///
> > + /// The given `mutex` must be created with the [`Class`] that
> > was used
> > + /// to initialize this [`AcquireCtx`].
> > + pub unsafe fn lock_slow_interruptible<'a, T>(
> > + &'a self,
> > + mutex: &'a Mutex<'a, T>,
> > + ) -> Result<MutexGuard<'a, T>> {
> > + // SAFETY: By the safety contract, `mutex` belongs to the
> > same `Class`
> > + // as `self` does.
> > + unsafe { lock_common(mutex, Some(self),
> > LockKind::SlowInterruptible) }
> > + }
> > +
> > + /// Tries to lock the mutex on this [`AcquireCtx`] without
> > blocking.
> > + ///
> > + /// Unlike `lock`, no deadlock handling is performed.
>
> You can use [`lock`] here for better docs.
>
I will update it as [`Self::lock`] ([`lock`] will not work). There are
some other lines where I need to do the same thing.
> > + ///
> > + /// # Safety
> > + ///
> > + /// The given `mutex` must be created with the [`Class`] that
[...]
BR,
Onur
Powered by blists - more mailing lists