[PATCH net v5] rust: net: phy: fix off-by-one bit positions in device status accessors

[PATCH net v5] rust: net: phy: fix off-by-one bit positions in device status accessors

[Chunfeng Song]

The hand-written bitfield offsets in is_link_up(), is_autoneg_enabled()
and is_autoneg_completed() were correct when the abstraction was
merged: at that time autoneg, link, and autoneg_complete were at bits
13, 14, and 15 of struct phy_device's first bitfield unit. Commit
2796ff1e3dca ("net: phy: add flag is_genphy_driven to struct phy_device")
later inserted is_genphy_driven just before autoneg, shifting the three
fields up by one, so the accessors now read:

  is_link_up()           reads bit 14 = autoneg
  is_autoneg_enabled()   reads bit 13 = is_genphy_driven
  is_autoneg_completed() reads bit 15 = link

The official ax88796b Rust driver uses all three accessors in its
link-change handling, so it inherits the bug. On genphy-driven devices
is_genphy_driven is always 1, which masks the broken
is_autoneg_enabled() check.

The ordinary bindgen accessors take &self. Calling them through
(*phydev).link() would create a shared reference to the complete
bindings::phy_device, which is not appropriate for an object wrapped in
Opaque.

Use the bindgen-generated raw accessors (link_raw(), autoneg_raw(),
and autoneg_complete_raw()) instead. They retain the bit positions and
endianness handling generated from the C layout without creating a Rust
reference to the complete phy_device. Drop the hand-written numbers
together with the TODO comment that marked them as a stopgap.

The raw accessors are only emitted by bindgen 0.71 and later, and were
added at the Rust-for-Linux project's request, so this fix can only be
backported to stable branches whose minimum bindgen version is at least
that, hence the scope on the Cc: stable line below.

Found by a static equivalence audit (C2RustDrv, a C-to-Rust driver
migration tool) that compares hand-written bitfield offsets against
the bindgen layout of struct phy_device. Verified by building the
bindings and checking the generated accessors; no runtime testing was
possible without PHY hardware.

Fixes: 2796ff1e3dca ("net: phy: add flag is_genphy_driven to struct phy_device")
Cc: stable@vger.kernel.org # Only 7.1.y and later (requires bindgen's raw pointer accessors).
Link: https://github.com/rust-lang/rust-bindgen/issues/2674
Signed-off-by: Chunfeng Song <springbreeze@stu.pku.edu.cn>
---
---
v5:
  - Move the comparison out of the `unsafe` block so that only the raw
    accessor call stays inside (Gary Guo, Miguel Ojeda).
  - Scope the `Cc: stable` backport to 7.1.y and later, since the raw
    accessors require bindgen 0.71 (Miguel Ojeda).
  - Mention the bindgen 0.71 dependency in the commit message.

v4: https://lore.kernel.org/r/20260901031824.1377498-1-springbreeze@stu.pku.edu.cn/
  - Use the bindgen-generated raw accessors to avoid creating a shared
    reference to the complete C `phy_device`, as suggested by Jakub Kicinski.
  - Update the SAFETY comments to describe pointer validity and the
    callback-context preconditions.
  - Add `#[inline]` to the three small forwarding methods.

v3: https://lore.kernel.org/r/20260825111334.795732-1-springbreeze@stu.pku.edu.cn/
  - Use the bindgen-generated accessors instead of hard-coded bit numbers.
  - Update the Fixes tag and add the net tree subject prefix.

v2: https://lore.kernel.org/r/20260824111227.742645-1-springbreeze@stu.pku.edu.cn/
  - Use bindgen-generated accessors as suggested by Andrew Lunn.
  - Explain that the offsets drifted after `is_genphy_driven` was added.

v1: https://lore.kernel.org/r/20260823103644.342849-1-springbreeze@stu.pku.edu.cn/
 rust/kernel/net/phy.rs | 44 +++++++++++++++++++++++-------------------
 1 file changed, 24 insertions(+), 20 deletions(-)

diff --git a/rust/kernel/net/phy.rs b/rust/kernel/net/phy.rs
index 956cda573ddb..256f939c5098 100644
--- a/rust/kernel/net/phy.rs
+++ b/rust/kernel/net/phy.rs
@@ -123,39 +123,43 @@ pub fn state(&self) -> DeviceState {
     /// Gets the current link state.
     ///
     /// It returns true if the link is up.
+    #[inline]
     pub fn is_link_up(&self) -> bool {
-        const LINK_IS_UP: u64 = 1;
-        // TODO: the code to access to the bit field will be replaced with automatically
-        // generated code by bindgen when it becomes possible.
-        // SAFETY: The struct invariant ensures that we may access
-        // this field without additional synchronization.
-        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
-        bit_field.get(14, 1) == LINK_IS_UP
+        let phydev = self.0.get().cast_const();
+        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
+        // `struct phy_device`, and the caller is in a context where this access
+        // is safe. The bindgen raw accessor does not create a Rust reference to
+        // the C struct.
+        let link = unsafe { bindings::phy_device::link_raw(phydev) };
+        link == 1
     }
 
     /// Gets the current auto-negotiation configuration.
     ///
     /// It returns true if auto-negotiation is enabled.
+    #[inline]
     pub fn is_autoneg_enabled(&self) -> bool {
-        // TODO: the code to access to the bit field will be replaced with automatically
-        // generated code by bindgen when it becomes possible.
-        // SAFETY: The struct invariant ensures that we may access
-        // this field without additional synchronization.
-        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
-        bit_field.get(13, 1) == u64::from(bindings::AUTONEG_ENABLE)
+        let phydev = self.0.get().cast_const();
+        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
+        // `struct phy_device`, and the caller is in a context where this access
+        // is safe. The bindgen raw accessor does not create a Rust reference to
+        // the C struct.
+        let autoneg = unsafe { bindings::phy_device::autoneg_raw(phydev) };
+        autoneg == bindings::AUTONEG_ENABLE
     }
 
     /// Gets the current auto-negotiation state.
     ///
     /// It returns true if auto-negotiation is completed.
+    #[inline]
     pub fn is_autoneg_completed(&self) -> bool {
-        const AUTONEG_COMPLETED: u64 = 1;
-        // TODO: the code to access to the bit field will be replaced with automatically
-        // generated code by bindgen when it becomes possible.
-        // SAFETY: The struct invariant ensures that we may access
-        // this field without additional synchronization.
-        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
-        bit_field.get(15, 1) == AUTONEG_COMPLETED
+        let phydev = self.0.get().cast_const();
+        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
+        // `struct phy_device`, and the caller is in a context where this access
+        // is safe. The bindgen raw accessor does not create a Rust reference to
+        // the C struct.
+        let completed = unsafe { bindings::phy_device::autoneg_complete_raw(phydev) };
+        completed == 1
     }
 
     /// Sets the speed of the PHY.

base-commit: 2709dd5ae32f0828f386327c76bba9f39f63a1c6
-- 
2.43.0

Re: [PATCH net v5] rust: net: phy: fix off-by-one bit positions in device status accessors

[FUJITA Tomonori]

On Tue,  1 Sep 2026 14:32:51 +0000
Chunfeng Song <springbreeze@stu.pku.edu.cn> wrote:

> The hand-written bitfield offsets in is_link_up(), is_autoneg_enabled()
> and is_autoneg_completed() were correct when the abstraction was
> merged: at that time autoneg, link, and autoneg_complete were at bits
> 13, 14, and 15 of struct phy_device's first bitfield unit. Commit
> 2796ff1e3dca ("net: phy: add flag is_genphy_driven to struct phy_device")
> later inserted is_genphy_driven just before autoneg, shifting the three
> fields up by one, so the accessors now read:
> 
>   is_link_up()           reads bit 14 = autoneg
>   is_autoneg_enabled()   reads bit 13 = is_genphy_driven
>   is_autoneg_completed() reads bit 15 = link
> 
> The official ax88796b Rust driver uses all three accessors in its
> link-change handling, so it inherits the bug.

"link-change handling" is ambiguous. The driver uses all three accessors in
read_status(). It also has a link_change_notify() callback, and that one
uses none of them. Can you say read_status() instead?

> On genphy-driven devices
> is_genphy_driven is always 1, which masks the broken
> is_autoneg_enabled() check.

I don't think this applies to the ax88796b driver. phy_attach_direct()
sets is_genphy_driven only in the `if (!d->driver)` branch, the one that
falls back to the generic driver. The ax88796b driver is a real driver, so
d->driver is set and is_genphy_driven stays 0.

With is_genphy_driven == 0, is_autoneg_enabled() reads bit 13 as 0 and
compares it against AUTONEG_ENABLE (1), so it always returns false. This
code never runs the following code:

        if dev.is_autoneg_enabled() && dev.is_autoneg_completed() {
            dev.resolve_aneg_linkmode();
        }

> diff --git a/rust/kernel/net/phy.rs b/rust/kernel/net/phy.rs
> index 956cda573ddb..256f939c5098 100644
> --- a/rust/kernel/net/phy.rs
> +++ b/rust/kernel/net/phy.rs
> @@ -123,39 +123,43 @@ pub fn state(&self) -> DeviceState {
>      /// Gets the current link state.
>      ///
>      /// It returns true if the link is up.
> +    #[inline]
>      pub fn is_link_up(&self) -> bool {
> -        const LINK_IS_UP: u64 = 1;
> -        // TODO: the code to access to the bit field will be replaced with automatically
> -        // generated code by bindgen when it becomes possible.
> -        // SAFETY: The struct invariant ensures that we may access
> -        // this field without additional synchronization.
> -        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
> -        bit_field.get(14, 1) == LINK_IS_UP
> +        let phydev = self.0.get().cast_const();
> +        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
> +        // `struct phy_device`, and the caller is in a context where this access
> +        // is safe. The bindgen raw accessor does not create a Rust reference to
> +        // the C struct.
> +        let link = unsafe { bindings::phy_device::link_raw(phydev) };

The last sentence does not describe why bindings::phy_device::link_raw()
can be called safely. It describes the accessor itself, and the commit
message already says it.

link_raw() can be called safely here because the pointer is valid and
there is no concurrent write. The old SAFETY comment covers both.

> +        link == 1
>      }
>  
>      /// Gets the current auto-negotiation configuration.
>      ///
>      /// It returns true if auto-negotiation is enabled.
> +    #[inline]
>      pub fn is_autoneg_enabled(&self) -> bool {
> -        // TODO: the code to access to the bit field will be replaced with automatically
> -        // generated code by bindgen when it becomes possible.
> -        // SAFETY: The struct invariant ensures that we may access
> -        // this field without additional synchronization.
> -        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
> -        bit_field.get(13, 1) == u64::from(bindings::AUTONEG_ENABLE)
> +        let phydev = self.0.get().cast_const();
> +        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
> +        // `struct phy_device`, and the caller is in a context where this access
> +        // is safe. The bindgen raw accessor does not create a Rust reference to
> +        // the C struct.
> +        let autoneg = unsafe { bindings::phy_device::autoneg_raw(phydev) };
> +        autoneg == bindings::AUTONEG_ENABLE
>      }
>  
>      /// Gets the current auto-negotiation state.
>      ///
>      /// It returns true if auto-negotiation is completed.
> +    #[inline]
>      pub fn is_autoneg_completed(&self) -> bool {
> -        const AUTONEG_COMPLETED: u64 = 1;
> -        // TODO: the code to access to the bit field will be replaced with automatically
> -        // generated code by bindgen when it becomes possible.
> -        // SAFETY: The struct invariant ensures that we may access
> -        // this field without additional synchronization.
> -        let bit_field = unsafe { &(*self.0.get())._bitfield_1 };
> -        bit_field.get(15, 1) == AUTONEG_COMPLETED
> +        let phydev = self.0.get().cast_const();
> +        // SAFETY: By the type invariant of `Device`, `phydev` points to a valid
> +        // `struct phy_device`, and the caller is in a context where this access
> +        // is safe. The bindgen raw accessor does not create a Rust reference to
> +        // the C struct.
> +        let completed = unsafe { bindings::phy_device::autoneg_complete_raw(phydev) };
> +        completed == 1
>      }
>  
>      /// Sets the speed of the PHY.
> 
> base-commit: 2709dd5ae32f0828f386327c76bba9f39f63a1c6
> -- 
> 2.43.0
> 
> 

Re: [PATCH net v5] rust: net: phy: fix off-by-one bit positions in device status accessors

[Chunfeng Song]

Thanks for the review. You are right on all three points, and they are
addressed in v6:

  - The commit message now names read_status() instead of the vague
    "link-change handling".

  - The masking claim is gone, and the impact is described correctly:
    since ax88796b has its own driver, phy_attach_direct() leaves
    is_genphy_driven at 0, so the broken is_autoneg_enabled() reads bit
    13 as 0 and always returns false, and read_status() never reaches
    the resolve_aneg_linkmode() call.

  - The SAFETY comments now justify the raw accessor calls themselves:
    the pointer is valid and there is no concurrent write to the field.

The v6 patch will be posted as a separate thread.

Cheers,
Chunfeng