NVIDIA GPU driver infrastructure
 help / color / mirror / Atom feed
From: "Gary Guo" <gary@garyguo.net>
To: "Danilo Krummrich" <dakr@kernel.org>,
	<gregkh@linuxfoundation.org>, <rafael@kernel.org>,
	<acourbot@nvidia.com>, <aliceryhl@google.com>,
	<david.m.ertman@intel.com>, <ira.weiny@intel.com>,
	<leon@kernel.org>, <viresh.kumar@linaro.org>,
	<m.wilczynski@samsung.com>, <ukleinek@kernel.org>,
	<bhelgaas@google.com>, <kwilczynski@kernel.org>,
	<abdiel.janulgue@gmail.com>, <robin.murphy@arm.com>,
	<markus.probst@posteo.de>, <ojeda@kernel.org>, <boqun@kernel.org>,
	<gary@garyguo.net>, <bjorn3_gh@protonmail.com>,
	<lossin@kernel.org>, <a.hindborg@kernel.org>, <tmgross@umich.edu>,
	<igor.korotin@linux.dev>, <daniel.almeida@collabora.com>,
	<pcolberg@redhat.com>
Cc: <driver-core@lists.linux.dev>, <linux-kernel@vger.kernel.org>,
	<nova-gpu@lists.linux.dev>, <dri-devel@lists.freedesktop.org>,
	<linux-pm@vger.kernel.org>, <linux-pwm@vger.kernel.org>,
	<linux-pci@vger.kernel.org>, <rust-for-linux@vger.kernel.org>,
	"Eliot Courtney" <ecourtney@nvidia.com>
Subject: Re: [PATCH v5 23/24] rust: auxiliary: generalize Registration over ForLt
Date: Wed, 27 May 2026 14:55:12 +0100	[thread overview]
Message-ID: <DITIDXOEZ54S.2PU9VU9UGTSZP@garyguo.net> (raw)
In-Reply-To: <20260525202921.124698-24-dakr@kernel.org>

On Mon May 25, 2026 at 9:21 PM BST, Danilo Krummrich wrote:
> Generalize Registration<T> to Registration<F: ForLt> and
> Device::registration_data<F: ForLt>() to return Pin<&F::Of<'_>>.
>
> The stored 'static lifetime is shortened to the borrow lifetime of &self
> via ForLt::cast_ref; ForLt's covariance guarantee makes this sound.
>
> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
> Reviewed-by: Eliot Courtney <ecourtney@nvidia.com>
> Signed-off-by: Danilo Krummrich <dakr@kernel.org>
> ---
>  drivers/gpu/nova-core/driver.rs       |  13 ++--
>  rust/kernel/auxiliary.rs              | 108 +++++++++++++++++++-------
>  samples/rust/rust_driver_auxiliary.rs |  19 +++--
>  3 files changed, 96 insertions(+), 44 deletions(-)
>
> diff --git a/drivers/gpu/nova-core/driver.rs b/drivers/gpu/nova-core/driver.rs
> index fa898fe5c893..d3f2245ba2e0 100644
> --- a/drivers/gpu/nova-core/driver.rs
> +++ b/drivers/gpu/nova-core/driver.rs
> @@ -3,7 +3,6 @@
>  use kernel::{
>      auxiliary,
>      device::Core,
> -    devres::Devres,
>      dma::Device,
>      dma::DmaMask,
>      pci,
> @@ -21,6 +20,7 @@
>          },
>          Arc,
>      },
> +    types::ForLt,
>  };
>  
>  use crate::gpu::Gpu;
> @@ -29,10 +29,11 @@
>  static AUXILIARY_ID_COUNTER: Atomic<u32> = Atomic::new(0);
>  
>  #[pin_data]
> -pub(crate) struct NovaCore {
> +pub(crate) struct NovaCore<'bound> {
>      #[pin]
>      pub(crate) gpu: Gpu,
> -    _reg: Devres<auxiliary::Registration<()>>,
> +    #[allow(clippy::type_complexity)]
> +    _reg: auxiliary::Registration<'bound, ForLt!(())>,
>  }
>  
>  pub(crate) struct NovaCoreDriver;
> @@ -76,13 +77,13 @@ pub(crate) struct NovaCore {
>  
>  impl pci::Driver for NovaCoreDriver {
>      type IdInfo = ();
> -    type Data<'bound> = NovaCore;
> +    type Data<'bound> = NovaCore<'bound>;
>      const ID_TABLE: pci::IdTable<Self::IdInfo> = &PCI_TABLE;
>  
>      fn probe<'bound>(
>          pdev: &'bound pci::Device<Core<'_>>,
>          _info: &'bound Self::IdInfo,
> -    ) -> impl PinInit<NovaCore, Error> + 'bound {
> +    ) -> impl PinInit<Self::Data<'bound>, Error> + 'bound {
>          pin_init::pin_init_scope(move || {
>              dev_dbg!(pdev, "Probe Nova Core GPU driver.\n");
>  
> @@ -115,7 +116,7 @@ fn probe<'bound>(
>          })
>      }
>  
> -    fn unbind<'bound>(pdev: &'bound pci::Device<Core<'_>>, this: Pin<&NovaCore>) {
> +    fn unbind<'bound>(pdev: &'bound pci::Device<Core<'_>>, this: Pin<&Self::Data<'bound>>) {
>          this.gpu.unbind(pdev.as_ref());
>      }
>  }
> diff --git a/rust/kernel/auxiliary.rs b/rust/kernel/auxiliary.rs
> index 7a1b1a7b7ca6..75ddb0220748 100644
> --- a/rust/kernel/auxiliary.rs
> +++ b/rust/kernel/auxiliary.rs
> @@ -12,7 +12,7 @@
>          RawDeviceId,
>          RawDeviceIdIndex, //
>      },
> -    devres::Devres,
> +
>      driver,
>      error::{
>          from_result,
> @@ -20,6 +20,7 @@
>      },
>      prelude::*,
>      types::{
> +        ForLt,
>          ForeignOwnable,
>          Opaque, //
>      },
> @@ -271,12 +272,16 @@ pub fn parent(&self) -> &device::Device<device::Bound> {
>  
>      /// Returns a pinned reference to the registration data set by the registering (parent) driver.
>      ///
> -    /// Returns [`EINVAL`] if `T` does not match the type used by the parent driver when calling
> +    /// `F` is the [`ForLt`](trait@ForLt) encoding of the data type. The returned
> +    /// reference has its lifetime shortened from `'static` to `&self`'s borrow lifetime via
> +    /// [`ForLt::cast_ref`].
> +    ///
> +    /// Returns [`EINVAL`] if `F` does not match the type used by the parent driver when calling
>      /// [`Registration::new()`].
>      ///
>      /// Returns [`ENOENT`] if no registration data has been set, e.g. when the device was
>      /// registered by a C driver.
> -    pub fn registration_data<T: 'static>(&self) -> Result<Pin<&T>> {
> +    pub fn registration_data<F: ForLt + 'static>(&self) -> Result<Pin<&F::Of<'_>>> {
>          // SAFETY: By the type invariant, `self.as_raw()` is a valid `struct auxiliary_device`.
>          let ptr = unsafe { (*self.as_raw()).registration_data_rust };
>          if ptr.is_null() {
> @@ -289,18 +294,23 @@ pub fn registration_data<T: 'static>(&self) -> Result<Pin<&T>> {
>  
>          // SAFETY: `ptr` is non-null and was set via `into_foreign()` in `Registration::new()`;
>          // `RegistrationData` is `#[repr(C)]` with `type_id` at offset 0, so reading a `TypeId`
> -        // at the start of the allocation is valid regardless of `T`.
> +        // at the start of the allocation is valid regardless of `F`.
>          let type_id = unsafe { ptr.cast::<TypeId>().read() };
> -        if type_id != TypeId::of::<T>() {
> +        if type_id != TypeId::of::<F>() {
>              return Err(EINVAL);
>          }
>  
> -        // SAFETY: The `TypeId` check above confirms that the stored type is `T`; `ptr` remains
> -        // valid until `Registration::drop()` calls `from_foreign()`.
> -        let wrapper = unsafe { Pin::<KBox<RegistrationData<T>>>::borrow(ptr) };
> +        // SAFETY: The `TypeId` check above confirms that the stored type matches
> +        // `F::Of<'static>`; `ptr` remains valid until `Registration::drop()` calls
> +        // `from_foreign()`.
> +        let wrapper = unsafe { Pin::<KBox<RegistrationData<F::Of<'static>>>>::borrow(ptr) };
>  
>          // SAFETY: `data` is a structurally pinned field of `RegistrationData`.
> -        Ok(unsafe { wrapper.map_unchecked(|w| &w.data) })
> +        let pinned: Pin<&F::Of<'static>> = unsafe { wrapper.map_unchecked(|w| &w.data) };

I think `'_` is sufficient here?

> +
> +        // SAFETY: The data was pinned when stored; `cast_ref` only shortens
> +        // the lifetime, so the pinning guarantee is preserved.
> +        Ok(unsafe { Pin::new_unchecked(F::cast_ref(pinned.get_ref())) })
>      }
>  }
>  
> @@ -389,43 +399,61 @@ struct RegistrationData<T> {
>  /// This type represents the registration of a [`struct auxiliary_device`]. When its parent device
>  /// is unbound, the corresponding auxiliary device will be unregistered from the system.
>  ///
> -/// The type parameter `T` is the type of the registration data owned by the registering (parent)
> -/// driver. It can be accessed by the auxiliary driver through
> -/// [`Device::registration_data()`].
> +/// The type parameter `F` is a [`ForLt`](trait@ForLt) encoding of the registration
> +/// data type. For non-lifetime-parameterized types, use [`ForLt!(T)`](macro@ForLt).
> +/// The data can be accessed by the auxiliary driver through [`Device::registration_data()`].
>  ///
>  /// # Invariants
>  ///
>  /// `self.adev` always holds a valid pointer to an initialized and registered
>  /// [`struct auxiliary_device`] whose `registration_data_rust` field points to a
> -/// valid `Pin<KBox<RegistrationData<T>>>`.
> -pub struct Registration<T: 'static> {
> +/// valid `Pin<KBox<RegistrationData<F::Of<'static>>>>`.
> +pub struct Registration<'a, F: ForLt + 'static> {
>      adev: NonNull<bindings::auxiliary_device>,
> -    _data: PhantomData<T>,
> +    #[allow(clippy::type_complexity)]

Is this acutally needed?

> +    _phantom: PhantomData<(fn(&'a ()) -> &'a (), F)>,

Could you update this to `PhantomData<F::Of<'a>>` please? It matches better
what's is actually being stored.

'a also does not actually need to be invariant here if `F::Of` is covariant
(which currently `ForLt` guarantees). (Although, lifetimes will still be
invariant as it's used in assoc type by changing to `F::Of<'a>`). Variance of
`'a` is not what matters here; the covariance of `F::Of` is load-bearing part
for the `registration_data` method, and the !Leak requirement is the
load-bearing part for UAF protection.

Feel free to add my R-b with the change.

Best,
Gary

>  }
>  
> -impl<T: Send + Sync + 'static> Registration<T> {
> +impl<'a, F: ForLt> Registration<'a, F>
> +where
> +    for<'b> F::Of<'b>: Send + Sync,
> +{
>      /// Create and register a new auxiliary device with the given registration data.
>      ///
>      /// The `data` is owned by the registration and can be accessed through the auxiliary device
>      /// via [`Device::registration_data()`].
> -    pub fn new<E>(
> -        parent: &device::Device<device::Bound>,
> +    ///
> +    /// # Safety
> +    ///
> +    /// The caller must not `mem::forget()` the returned [`Registration`] or otherwise prevent its
> +    /// [`Drop`] implementation from running, since the registration data may contain borrowed
> +    /// references that become invalid after `'a` ends.
> +    ///
> +    /// If the registration data is `'static`, use the safe [`Registration::new()`] instead.
> +    pub unsafe fn new_with_lt<E>(
> +        parent: &'a device::Device<device::Bound>,
>          name: &CStr,
>          id: u32,
>          modname: &CStr,
> -        data: impl PinInit<T, E>,
> -    ) -> Result<Devres<Self>>
> +        data: impl PinInit<F::Of<'a>, E>,
> +    ) -> Result<Self>
>      where
>          Error: From<E>,
>      {
>          let data = KBox::pin_init::<Error>(
>              try_pin_init!(RegistrationData {
> -                type_id: TypeId::of::<T>(),
> +                type_id: TypeId::of::<F>(),
>                  data <- data,
>              }),
>              GFP_KERNEL,
>          )?;
>  
> +        // SAFETY: `'a` is invariant (via `Registration`'s `PhantomData`). Lifetimes do not
> +        // affect layout, so RegistrationData<F::Of<'a>> and RegistrationData<F::Of<'static>>
> +        // have identical representation.
> +        let data: Pin<KBox<RegistrationData<F::Of<'static>>>> =
> +            unsafe { core::mem::transmute(data) };
> +
>          let boxed: KBox<Opaque<bindings::auxiliary_device>> = KBox::zeroed(GFP_KERNEL)?;
>          let adev = boxed.get();
>  
> @@ -455,7 +483,9 @@ pub fn new<E>(
>          if ret != 0 {
>              // SAFETY: `registration_data` was set above via `into_foreign()`.
>              drop(unsafe {
> -                Pin::<KBox<RegistrationData<T>>>::from_foreign((*adev).registration_data_rust)
> +                Pin::<KBox<RegistrationData<F::Of<'static>>>>::from_foreign(
> +                    (*adev).registration_data_rust,
> +                )If it's un
>              });
>  
>              // SAFETY: `adev` is guaranteed to be a valid pointer to a
> @@ -467,18 +497,36 @@ pub fn new<E>(
>  
>          // INVARIANT: The device will remain registered until `auxiliary_device_delete()` is
>          // called, which happens in `Self::drop()`.
> -        let reg = Self {
> +        Ok(Self {
>              // SAFETY: `adev` is guaranteed to be non-null, since the `KBox` was allocated
>              // successfully.
>              adev: unsafe { NonNull::new_unchecked(adev) },
> -            _data: PhantomData,
> -        };
> +            _phantom: PhantomData,
> +        })
> +    }
>  
> -        Devres::new::<core::convert::Infallible>(parent, reg)
> +    /// Create and register a new auxiliary device with `'static` registration data.
> +    ///
> +    /// Safe variant of [`Registration::new_with_lt()`] for registration data that does not contain
> +    /// borrowed references.
> +    pub fn new<E>(
> +        parent: &'a device::Device<device::Bound>,
> +        name: &CStr,
> +        id: u32,
> +        modname: &CStr,
> +        data: impl PinInit<F::Of<'a>, E>,
> +    ) -> Result<Self>
> +    where
> +        F::Of<'a>: 'static,
> +        Error: From<E>,
> +    {
> +        // SAFETY: `F::Of<'a>: 'static` guarantees the data contains no borrowed references,
> +        // so forgetting the `Registration` cannot cause use-after-free.
> +        unsafe { Self::new_with_lt(parent, name, id, modname, data) }
>      }
>  }


  reply	other threads:[~2026-05-27 13:55 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-25 20:20 [PATCH v5 00/24] rust: device: Higher-Ranked Lifetime Types for device drivers Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 01/24] rust: pci: use 'static lifetime for PCI BAR resource names Danilo Krummrich
2026-05-26  0:38   ` Eliot Courtney
2026-05-26  2:22   ` Alexandre Courbot
2026-05-27 11:58   ` Gary Guo
2026-05-25 20:20 ` [PATCH v5 02/24] rust: alloc: remove `'static` bound on `ForeignOwnable` Danilo Krummrich
2026-05-26 18:21   ` Miguel Ojeda
2026-05-25 20:20 ` [PATCH v5 03/24] rust: driver: move 'static bounds to constructor Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 04/24] rust: driver: decouple driver private data from driver type Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 05/24] rust: driver core: drop drvdata before devres release Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 06/24] rust: pci: implement Sync for Device<Bound> Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 07/24] rust: platform: " Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 08/24] rust: auxiliary: " Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 09/24] rust: usb: " Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 10/24] rust: device: " Danilo Krummrich
2026-05-25 20:20 ` [PATCH v5 11/24] rust: device: make Core and CoreInternal lifetime-parameterized Danilo Krummrich
2026-05-27 13:13   ` Gary Guo
2026-05-25 20:20 ` [PATCH v5 12/24] rust: pci: make Driver trait lifetime-parameterized Danilo Krummrich
2026-05-27 13:21   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 13/24] rust: platform: " Danilo Krummrich
2026-05-27 13:22   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 14/24] rust: auxiliary: " Danilo Krummrich
2026-05-27 13:22   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 15/24] rust: usb: " Danilo Krummrich
2026-05-27 13:22   ` Gary Guo
2026-05-27 13:38   ` Daniel Almeida
2026-05-25 20:21 ` [PATCH v5 16/24] rust: i2c: " Danilo Krummrich
2026-05-27 13:23   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 17/24] rust: driver: update module documentation for GAT-based Data type Danilo Krummrich
2026-05-27 13:24   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 18/24] rust: pci: make Bar lifetime-parameterized Danilo Krummrich
2026-05-27 13:30   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 19/24] rust: io: make IoMem and ExclusiveIoMem lifetime-parameterized Danilo Krummrich
2026-05-27 13:31   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 20/24] samples: rust: rust_driver_pci: use HRT lifetime for Bar Danilo Krummrich
2026-05-27 13:41   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 21/24] gpu: nova-core: separate driver type from driver data Danilo Krummrich
2026-05-27 13:39   ` Gary Guo
2026-05-25 20:21 ` [PATCH v5 22/24] rust: types: add `ForLt` trait for higher-ranked lifetime support Danilo Krummrich
2026-05-26  5:44   ` Alexandre Courbot
2026-05-26 18:49   ` Miguel Ojeda
2026-05-25 20:21 ` [PATCH v5 23/24] rust: auxiliary: generalize Registration over ForLt Danilo Krummrich
2026-05-27 13:55   ` Gary Guo [this message]
2026-05-25 20:21 ` [PATCH v5 24/24] samples: rust: rust_driver_auxiliary: showcase lifetime-bound registration data Danilo Krummrich
2026-05-27 21:24 ` [PATCH v5 00/24] rust: device: Higher-Ranked Lifetime Types for device drivers 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=DITIDXOEZ54S.2PU9VU9UGTSZP@garyguo.net \
    --to=gary@garyguo.net \
    --cc=a.hindborg@kernel.org \
    --cc=abdiel.janulgue@gmail.com \
    --cc=acourbot@nvidia.com \
    --cc=aliceryhl@google.com \
    --cc=bhelgaas@google.com \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun@kernel.org \
    --cc=dakr@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=gregkh@linuxfoundation.org \
    --cc=igor.korotin@linux.dev \
    --cc=ira.weiny@intel.com \
    --cc=kwilczynski@kernel.org \
    --cc=leon@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=linux-pwm@vger.kernel.org \
    --cc=lossin@kernel.org \
    --cc=m.wilczynski@samsung.com \
    --cc=markus.probst@posteo.de \
    --cc=nova-gpu@lists.linux.dev \
    --cc=ojeda@kernel.org \
    --cc=pcolberg@redhat.com \
    --cc=rafael@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=tmgross@umich.edu \
    --cc=ukleinek@kernel.org \
    --cc=viresh.kumar@linaro.org \
    /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