From: Danilo Krummrich <dakr@kernel.org>
To: gregkh@linuxfoundation.org, rafael@kernel.org, dakr@kernel.org,
ojeda@kernel.org, boqun@kernel.org, gary@garyguo.net,
bjorn3_gh@protonmail.com, lossin@kernel.org,
a.hindborg@kernel.org, aliceryhl@google.com, tmgross@umich.edu,
acourbot@nvidia.com, ecourtney@nvidia.com,
m.wilczynski@samsung.com, david.m.ertman@intel.com,
ira.weiny@intel.com, leon@kernel.org,
daniel.almeida@collabora.com, bhelgaas@google.com,
kwilczynski@kernel.org
Cc: driver-core@lists.linux.dev, linux-kernel@vger.kernel.org,
nova-gpu@lists.linux.dev, dri-devel@lists.freedesktop.org,
linux-pwm@vger.kernel.org, rust-for-linux@vger.kernel.org
Subject: [PATCH v2 5/7] rust: devres: add DevresLt for ForLt-aware device resource access
Date: Wed, 3 Jun 2026 03:10:16 +0200 [thread overview]
Message-ID: <20260603011020.2073650-6-dakr@kernel.org> (raw)
In-Reply-To: <20260603011020.2073650-1-dakr@kernel.org>
Devres<T> stores resources as T and returns &'a T from access(). For
lifetime-parameterized types like Bar<'a, SIZE> that are transmuted to
'static for storage, this exposes the synthetic 'static lifetime to
callers -- any method on the stored type that returns a reference with
its lifetime parameter would yield a &'static reference, which is
unsound.
Add DevresLt<F: ForLt>, a thin wrapper around Devres<F::Of<'static>>
that applies ForLt's lifetime-shortening operations in all access
methods to shorten the stored 'static lifetime to the caller's borrow
lifetime.
DevresLt::new() is unsafe because the caller must guarantee that the
data remains valid for the device's full bound scope; the internal
transmute from F::Of<'a> to F::Of<'static> would otherwise allow
use-after-free.
Two access patterns are provided:
- CovariantForLt types get direct-reference accessors (access,
try_access) that return shortened references via
CovariantForLt::cast_ref.
- Plain ForLt types use closure-based accessors (access_with,
try_access_with) whose universally quantified lifetime prevents
callers from smuggling in concrete short-lived references.
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
rust/kernel/devres.rs | 110 ++++++++++++++++++++++++++++++++++++++++++
1 file changed, 110 insertions(+)
diff --git a/rust/kernel/devres.rs b/rust/kernel/devres.rs
index 82cbd8b969fb..23db91377f5e 100644
--- a/rust/kernel/devres.rs
+++ b/rust/kernel/devres.rs
@@ -24,6 +24,8 @@
Arc, //
},
types::{
+ CovariantForLt,
+ ForLt,
ForeignOwnable,
Opaque, //
},
@@ -365,6 +367,114 @@ fn drop(&mut self) {
}
}
+/// Guard returned by [`DevresLt::try_access`].
+///
+/// Dereferences to `F::Of<'a>`, shortening the lifetime of the stored data to the guard's borrow
+/// lifetime.
+pub struct DevresGuard<'a, F: CovariantForLt>(RevocableGuard<'a, F::Of<'static>>);
+
+impl<'a, F: CovariantForLt> core::ops::Deref for DevresGuard<'a, F> {
+ type Target = F::Of<'a>;
+
+ fn deref(&self) -> &Self::Target {
+ F::cast_ref(&*self.0)
+ }
+}
+
+/// Device-managed resource with [`ForLt`](trait@ForLt)-aware access.
+///
+/// `DevresLt` wraps [`Devres`] and shortens the stored `'static` lifetime to the caller's borrow
+/// lifetime in all access methods.
+///
+/// Types that implement [`trait@CovariantForLt`] get direct-reference accessors ([`Self::access`],
+/// [`Self::try_access`]). Plain [`ForLt`](trait@ForLt) types use closure-based accessors
+/// ([`Self::access_with`], [`Self::try_access_with`]).
+pub struct DevresLt<F: ForLt>(Devres<F::Of<'static>>)
+where
+ F::Of<'static>: Send;
+
+impl<F: ForLt> DevresLt<F>
+where
+ F::Of<'static>: Send,
+{
+ /// Creates a new [`DevresLt`] instance of the given `data`.
+ ///
+ /// # Safety
+ ///
+ /// The data must remain valid for the device's full bound scope. [`DevresLt`] allows
+ /// access until the device is unbound, which may outlast `'a`.
+ pub unsafe fn new<'a, E>(
+ dev: &'a Device<Bound>,
+ data: impl PinInit<F::Of<'a>, E>,
+ ) -> Result<Self>
+ where
+ Error: From<E>,
+ {
+ // SAFETY: The caller guarantees the data is valid for the device's full bound scope.
+ // Lifetimes do not affect layout, so F::Of<'a> and F::Of<'static> have identical
+ // representation; casting the slot pointer is sound.
+ let data = unsafe {
+ pin_init::pin_init_from_closure::<F::Of<'static>, E>(move |slot| {
+ data.__pinned_init(slot.cast())
+ })
+ };
+
+ Ok(Self(Devres::new(dev, data)?))
+ }
+
+ /// Return a reference of the [`Device`] this [`DevresLt`] instance has been created with.
+ pub fn device(&self) -> &Device {
+ self.0.device()
+ }
+
+ /// Obtain `&F::Of<'_>`, bypassing the [`Revocable`], through a closure.
+ ///
+ /// This method works like [`DevresLt::access`](DevresLt::access) but accepts any
+ /// [`trait@ForLt`] type, not just [`trait@CovariantForLt`].
+ pub fn access_with<R, G>(&self, dev: &Device<Bound>, f: G) -> Result<R>
+ where
+ G: for<'a> FnOnce(&'a F::Of<'a>) -> R,
+ {
+ self.0.access(dev).map(|data| {
+ // SAFETY: The closure's HRTB `for<'a>` prevents the caller from smuggling in
+ // references with a concrete short lifetime, making the round-trip from `'static`
+ // sound regardless of variance.
+ f(unsafe { F::cast_ref_unchecked(data) })
+ })
+ }
+
+ /// [`DevresLt`] accessor for [`Revocable::try_access_with`].
+ pub fn try_access_with<R, G>(&self, f: G) -> Option<R>
+ where
+ G: for<'a> FnOnce(&'a F::Of<'a>) -> R,
+ {
+ self.0.data().try_access_with(|data| {
+ // SAFETY: The closure's HRTB `for<'a>` prevents the caller from smuggling in
+ // references with a concrete short lifetime, making the round-trip from `'static`
+ // sound regardless of variance.
+ f(unsafe { F::cast_ref_unchecked(data) })
+ })
+ }
+}
+
+impl<F: CovariantForLt> DevresLt<F>
+where
+ F::Of<'static>: Send,
+{
+ /// Obtain `&'a F::Of<'a>`, bypassing the [`Revocable`].
+ ///
+ /// This method works like [`Devres::access`], but shortens the returned reference's lifetime
+ /// from `'static` to `'a` via [`CovariantForLt::cast_ref`].
+ pub fn access<'a>(&'a self, dev: &'a Device<Bound>) -> Result<&'a F::Of<'a>> {
+ self.0.access(dev).map(F::cast_ref)
+ }
+
+ /// [`DevresLt`] accessor for [`Revocable::try_access`].
+ pub fn try_access(&self) -> Option<DevresGuard<'_, F>> {
+ self.0.data().try_access().map(DevresGuard)
+ }
+}
+
/// Consume `data` and [`Drop::drop`] `data` once `dev` is unbound.
fn register_foreign<P>(dev: &Device<Bound>, data: P) -> Result
where
--
2.54.0
next prev parent reply other threads:[~2026-06-03 1:10 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-03 1:10 [PATCH v2 0/7] ForLt/CovariantForLt split, auxiliary closure API and DevresLt Danilo Krummrich
2026-06-03 1:10 ` [PATCH v2 1/7] rust: types: rename ForLt to CovariantForLt Danilo Krummrich
2026-06-03 11:59 ` Gary Guo
2026-06-03 1:10 ` [PATCH v2 2/7] rust: types: introduce ForLt base trait for CovariantForLt Danilo Krummrich
2026-06-03 12:04 ` Gary Guo
2026-06-03 1:10 ` [PATCH v2 3/7] rust: auxiliary: add registration_data_with() for ForLt types Danilo Krummrich
2026-06-03 12:05 ` Gary Guo
2026-06-03 1:10 ` [PATCH v2 4/7] rust: auxiliary: sample: demonstrate ForLt with invariant Mutex type Danilo Krummrich
2026-06-03 1:10 ` Danilo Krummrich [this message]
2026-06-03 1:10 ` [PATCH v2 6/7] rust: pci: return DevresLt from Bar::into_devres() Danilo Krummrich
2026-06-03 1:10 ` [PATCH v2 7/7] rust: io: mem: return DevresLt from IoMem/ExclusiveIoMem::into_devres() Danilo Krummrich
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=20260603011020.2073650-6-dakr@kernel.org \
--to=dakr@kernel.org \
--cc=a.hindborg@kernel.org \
--cc=acourbot@nvidia.com \
--cc=aliceryhl@google.com \
--cc=bhelgaas@google.com \
--cc=bjorn3_gh@protonmail.com \
--cc=boqun@kernel.org \
--cc=daniel.almeida@collabora.com \
--cc=david.m.ertman@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=driver-core@lists.linux.dev \
--cc=ecourtney@nvidia.com \
--cc=gary@garyguo.net \
--cc=gregkh@linuxfoundation.org \
--cc=ira.weiny@intel.com \
--cc=kwilczynski@kernel.org \
--cc=leon@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pwm@vger.kernel.org \
--cc=lossin@kernel.org \
--cc=m.wilczynski@samsung.com \
--cc=nova-gpu@lists.linux.dev \
--cc=ojeda@kernel.org \
--cc=rafael@kernel.org \
--cc=rust-for-linux@vger.kernel.org \
--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