From: Danilo Krummrich <dakr@kernel.org>
To: Gary Guo <gary@garyguo.net>
Cc: gregkh@linuxfoundation.org, rafael@kernel.org,
bhelgaas@google.com, ojeda@kernel.org, alex.gaynor@gmail.com,
boqun.feng@gmail.com, bjorn3_gh@protonmail.com,
benno.lossin@proton.me, tmgross@umich.edu,
a.hindborg@samsung.com, aliceryhl@google.com, airlied@gmail.com,
fujita.tomonori@gmail.com, lina@asahilina.net,
pstanner@redhat.com, ajanulgu@redhat.com, lyude@redhat.com,
robh@kernel.org, daniel.almeida@collabora.com,
saravanak@google.com, dirk.behme@de.bosch.com, j@jannau.net,
fabien.parent@linaro.org, chrisi.schrefl@gmail.com,
paulmck@kernel.org, rust-for-linux@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
devicetree@vger.kernel.org, rcu@vger.kernel.org
Subject: Re: [PATCH v7 08/16] rust: add devres abstraction
Date: Thu, 2 Jan 2025 16:59:48 +0100 [thread overview]
Message-ID: <Z3a39FM0oXHbitDz@cassiopeiae> (raw)
In-Reply-To: <Z3Zqq5qLht2j9Qqq@cassiopeiae>
On Thu, Jan 02, 2025 at 11:30:11AM +0100, Danilo Krummrich wrote:
> On Tue, Dec 24, 2024 at 09:53:23PM +0000, Gary Guo wrote:
> > On Thu, 19 Dec 2024 18:04:10 +0100
> > Danilo Krummrich <dakr@kernel.org> wrote:
> >
> > > Add a Rust abstraction for the kernel's devres (device resource
> > > management) implementation.
> > >
> > > The Devres type acts as a container to manage the lifetime and
> > > accessibility of device bound resources. Therefore it registers a
> > > devres callback and revokes access to the resource on invocation.
> > >
> > > Users of the Devres abstraction can simply free the corresponding
> > > resources in their Drop implementation, which is invoked when either the
> > > Devres instance goes out of scope or the devres callback leads to the
> > > resource being revoked, which implies a call to drop_in_place().
> > >
> > > Signed-off-by: Danilo Krummrich <dakr@kernel.org>
> > > ---
> > > MAINTAINERS | 1 +
> > > rust/helpers/device.c | 10 +++
> > > rust/helpers/helpers.c | 1 +
> > > rust/kernel/devres.rs | 178 +++++++++++++++++++++++++++++++++++++++++
> > > rust/kernel/lib.rs | 1 +
> > > 5 files changed, 191 insertions(+)
> > > create mode 100644 rust/helpers/device.c
> > > create mode 100644 rust/kernel/devres.rs
> > >
> > > <snip>
> > >
> > > +pub struct Devres<T>(Arc<DevresInner<T>>);
> > > +
> > > +impl<T> DevresInner<T> {
> > > + fn new(dev: &Device, data: T, flags: Flags) -> Result<Arc<DevresInner<T>>> {
> > > + let inner = Arc::pin_init(
> > > + pin_init!( DevresInner {
> > > + data <- Revocable::new(data),
> > > + }),
> > > + flags,
> > > + )?;
> > > +
> > > + // Convert `Arc<DevresInner>` into a raw pointer and make devres own this reference until
> > > + // `Self::devres_callback` is called.
> > > + let data = inner.clone().into_raw();
> > > +
> > > + // SAFETY: `devm_add_action` guarantees to call `Self::devres_callback` once `dev` is
> > > + // detached.
> > > + let ret = unsafe {
> > > + bindings::devm_add_action(dev.as_raw(), Some(Self::devres_callback), data as _)
> > > + };
> > > +
> > > + if ret != 0 {
> > > + // SAFETY: We just created another reference to `inner` in order to pass it to
> > > + // `bindings::devm_add_action`. If `bindings::devm_add_action` fails, we have to drop
> > > + // this reference accordingly.
> > > + let _ = unsafe { Arc::from_raw(data) };
> > > + return Err(Error::from_errno(ret));
> > > + }
> > > +
> > > + Ok(inner)
> > > + }
> > > +
> > > + #[allow(clippy::missing_safety_doc)]
> > > + unsafe extern "C" fn devres_callback(ptr: *mut kernel::ffi::c_void) {
> > > + let ptr = ptr as *mut DevresInner<T>;
> > > + // Devres owned this memory; now that we received the callback, drop the `Arc` and hence the
> > > + // reference.
> > > + // SAFETY: Safe, since we leaked an `Arc` reference to devm_add_action() in
> > > + // `DevresInner::new`.
> > > + let inner = unsafe { Arc::from_raw(ptr) };
> > > +
> > > + inner.data.revoke();
> > > + }
> > > +}
> > > +
> > > +impl<T> Devres<T> {
> > > + /// Creates a new [`Devres`] instance of the given `data`. The `data` encapsulated within the
> > > + /// returned `Devres` instance' `data` will be revoked once the device is detached.
> > > + pub fn new(dev: &Device, data: T, flags: Flags) -> Result<Self> {
> > > + let inner = DevresInner::new(dev, data, flags)?;
> > > +
> > > + Ok(Devres(inner))
> > > + }
> > > +
> > > + /// Same as [`Devres::new`], but does not return a `Devres` instance. Instead the given `data`
> > > + /// is owned by devres and will be revoked / dropped, once the device is detached.
> > > + pub fn new_foreign_owned(dev: &Device, data: T, flags: Flags) -> Result {
> > > + let _ = DevresInner::new(dev, data, flags)?;
> > > +
> > > + Ok(())
> > > + }
> > > +}
> > > +
> > > +impl<T> Deref for Devres<T> {
> > > + type Target = Revocable<T>;
> > > +
> > > + fn deref(&self) -> &Self::Target {
> > > + &self.0.data
> > > + }
> > > +}
> > > +
> > > +impl<T> Drop for Devres<T> {
> > > + fn drop(&mut self) {
> > > + // Revoke the data, such that it gets dropped already and the actual resource is freed.
> > > + //
> > > + // `DevresInner` has to stay alive until the devres callback has been called. This is
> > > + // necessary since we don't know when `Devres` is dropped and calling
> > > + // `devm_remove_action()` instead could race with `devres_release_all()`.
> >
> > IIUC, the outcome of that race is the `WARN` if
> > devres_release_all takes the spinlock first and has already remvoed the
> > action?
>
> Yes, this was one issue. But I think there was another when you have a class
> `Registration` that is owned by a `Devres`, which holds private data that is
> encapsulated in a `Devres` too.
>
> I have this case in Nova where the DRM device' private data holds the PCI bar,
> and the DRM device registration has a reference of the corresponding DRM device.
>
> But maybe this also was something else. I will double check and if I can confirm
> that the WARN_ON() in devm_remove_action() is the only issue, we can certainly
> change this.
Ok, I double checked and this should indeed be the only issue.
I can reproduce the race and can confirm that a devm_remove_action_nowarn()
works, as long as we make it return the integer that indicates whether the
action has been removed already, such that we can handle it from the Rust side
accordingly.
Generally, I don't think it's a big deal, drivers usually hold the resources
they request anyways until remove(). But it's still an improvement, so I'll send
a patch for that.
>
> >
> > Could you do a custom devres_release here that mimick
> > `devm_remove_action` but omit the `WARN`? This way it allows the memory
> > behind DevresInner to be freed early without keeping it allocated until
> > the end of device lifetime.
> >
> > > + //
> > > + // SAFETY: When `drop` runs, it's guaranteed that nobody is accessing the revocable data
> > > + // anymore, hence it is safe not to wait for the grace period to finish.
> > > + unsafe { self.revoke_nosync() };
> > > + }
> > > +}
> > > diff --git a/rust/kernel/lib.rs b/rust/kernel/lib.rs
> > > index 6c836ab73771..2b61bf99d1ee 100644
> > > --- a/rust/kernel/lib.rs
> > > +++ b/rust/kernel/lib.rs
> > > @@ -41,6 +41,7 @@
> > > pub mod cred;
> > > pub mod device;
> > > pub mod device_id;
> > > +pub mod devres;
> > > pub mod driver;
> > > pub mod error;
> > > #[cfg(CONFIG_RUST_FW_LOADER_ABSTRACTIONS)]
> >
next prev parent reply other threads:[~2025-01-02 15:59 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-19 17:04 [PATCH v7 00/16] Device / Driver PCI / Platform Rust abstractions Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 01/16] rust: module: add trait `ModuleMetadata` Danilo Krummrich
2024-12-24 20:51 ` Gary Guo
2024-12-19 17:04 ` [PATCH v7 02/16] rust: implement generic driver registration Danilo Krummrich
2024-12-24 19:58 ` Gary Guo
2025-01-02 9:34 ` Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 03/16] rust: implement `IdArray`, `IdTable` and `RawDeviceId` Danilo Krummrich
2024-12-24 20:10 ` Gary Guo
2025-01-02 9:36 ` Danilo Krummrich
2025-03-15 16:52 ` Tamir Duberstein
2025-03-15 16:59 ` Danilo Krummrich
2025-03-15 17:31 ` Tamir Duberstein
2024-12-19 17:04 ` [PATCH v7 04/16] rust: add rcu abstraction Danilo Krummrich
2024-12-24 20:54 ` Gary Guo
2025-01-02 9:44 ` Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 05/16] rust: types: add `Opaque::pin_init` Danilo Krummrich
2024-12-24 20:21 ` Gary Guo
2024-12-19 17:04 ` [PATCH v7 06/16] rust: add `Revocable` type Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 07/16] rust: add `io::{Io, IoRaw}` base types Danilo Krummrich
2025-01-16 10:31 ` Fiona Behrens
2025-01-20 12:54 ` Danilo Krummrich
2025-02-21 1:20 ` Alistair Popple
2025-02-21 3:58 ` Miguel Ojeda
2025-02-25 5:50 ` Alistair Popple
2025-02-25 11:04 ` Danilo Krummrich
2025-02-27 0:25 ` Alistair Popple
2025-02-27 10:01 ` Danilo Krummrich
2025-02-28 5:29 ` Alistair Popple
2025-02-28 10:12 ` Danilo Krummrich
2025-02-28 10:22 ` Miguel Ojeda
2025-02-27 16:06 ` Miguel Ojeda
2024-12-19 17:04 ` [PATCH v7 08/16] rust: add devres abstraction Danilo Krummrich
2024-12-24 21:53 ` Gary Guo
2025-01-02 10:30 ` Danilo Krummrich
2025-01-02 15:59 ` Danilo Krummrich [this message]
2025-01-14 16:33 ` Danilo Krummrich
2025-01-15 6:41 ` Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 09/16] rust: pci: add basic PCI device / driver abstractions Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 10/16] rust: pci: implement I/O mappable `pci::Bar` Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 11/16] samples: rust: add Rust PCI sample driver Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 12/16] rust: of: add `of::DeviceId` abstraction Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 13/16] rust: driver: implement `Adapter` Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 14/16] rust: platform: add basic platform device / driver abstractions Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 15/16] samples: rust: add Rust platform sample driver Danilo Krummrich
2024-12-19 17:04 ` [PATCH v7 16/16] MAINTAINERS: add Danilo to DRIVER CORE Danilo Krummrich
2024-12-20 7:24 ` [PATCH v7 00/16] Device / Driver PCI / Platform Rust abstractions Dirk Behme
2024-12-20 16:44 ` Greg KH
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Z3a39FM0oXHbitDz@cassiopeiae \
--to=dakr@kernel.org \
--cc=a.hindborg@samsung.com \
--cc=airlied@gmail.com \
--cc=ajanulgu@redhat.com \
--cc=alex.gaynor@gmail.com \
--cc=aliceryhl@google.com \
--cc=benno.lossin@proton.me \
--cc=bhelgaas@google.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun.feng@gmail.com \
--cc=chrisi.schrefl@gmail.com \
--cc=daniel.almeida@collabora.com \
--cc=devicetree@vger.kernel.org \
--cc=dirk.behme@de.bosch.com \
--cc=fabien.parent@linaro.org \
--cc=fujita.tomonori@gmail.com \
--cc=gary@garyguo.net \
--cc=gregkh@linuxfoundation.org \
--cc=j@jannau.net \
--cc=lina@asahilina.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=lyude@redhat.com \
--cc=ojeda@kernel.org \
--cc=paulmck@kernel.org \
--cc=pstanner@redhat.com \
--cc=rafael@kernel.org \
--cc=rcu@vger.kernel.org \
--cc=robh@kernel.org \
--cc=rust-for-linux@vger.kernel.org \
--cc=saravanak@google.com \
--cc=tmgross@umich.edu \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).