[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Zo6oqfFX-TNIeaIC@pollux>
Date: Wed, 10 Jul 2024 17:28:41 +0200
From: Danilo Krummrich <dakr@...hat.com>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: "Rafael J. Wysocki" <rafael@...nel.org>,
Miguel Ojeda <miguel.ojeda.sandonis@...il.com>,
Miguel Ojeda <ojeda@...nel.org>,
Alex Gaynor <alex.gaynor@...il.com>,
Wedson Almeida Filho <wedsonaf@...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>,
Alice Ryhl <aliceryhl@...gle.com>, linux-pm@...r.kernel.org,
Vincent Guittot <vincent.guittot@...aro.org>,
Stephen Boyd <sboyd@...nel.org>, Nishanth Menon <nm@...com>,
rust-for-linux@...r.kernel.org,
Manos Pitsidianakis <manos.pitsidianakis@...aro.org>,
Erik Schilling <erik.schilling@...aro.org>,
Alex Bennée <alex.bennee@...aro.org>,
Joakim Bech <joakim.bech@...aro.org>, Rob Herring <robh@...nel.org>,
linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH V3 8/8] cpufreq: Add Rust based cpufreq-dt driver
On Wed, Jul 10, 2024 at 02:26:52PM +0530, Viresh Kumar wrote:
> On 05-07-24, 13:32, Danilo Krummrich wrote:
> > On 7/3/24 09:14, Viresh Kumar wrote:
> > > + fn probe(_dev: &mut platform::Device, _id_info: Option<&Self::IdInfo>) -> Result<Self::Data> {
> > > + let drv = Arc::new(
> > > + cpufreq::Registration::<CPUFreqDTDriver>::register(
> > > + c_str!("cpufreq-dt"),
> > > + (),
> > > + cpufreq::flags::NEED_INITIAL_FREQ_CHECK | cpufreq::flags::IS_COOLING_DEV,
> > > + true,
> > > + )?,
> > > + GFP_KERNEL,
> > > + )?;
> >
> > Putting the `cpufreq::Registration` into `Arc<DeviceData>` is unsafe from a
> > lifetime point of view. Nothing prevents this `Arc` to out-live the
> > `platform::Driver`.
>
> Hmm, the platform driver layer (in Rust) should guarantee that the
> data will be freed from the driver removal path. Isn't it ?
No, the platform driver layer will only guarantee that it decreses the reference
count of the `Arc` by one, that doesn't guarantee a free. If something else
still holds a reference to the `Arc` it will keep the `Registration` alive,
unless it's wrapped by `Devres`.
>
> > Instead, you should wrap `cpufreq::Registration` into `Devres`. See
> > `drm::drv::Registration` for an example [1].
> >
> > [1] https://gitlab.freedesktop.org/drm/nova/-/blob/nova-next/rust/kernel/drm/drv.rs?ref_type=heads#L173
>
> I can convert to that too, will do it anyway.
>
> --
> viresh
>
Powered by blists - more mailing lists