From: sashiko-bot@kernel.org
Subject: Re: [PATCH v5 19/24] rust: io: make IoMem and ExclusiveIoMem
 lifetime-parameterized
Reply-To: sashiko-reviews@lists.linux.dev
To: "Danilo Krummrich" <dakr@kernel.org>
Cc: linux-pci@vger.kernel.org, ojeda@kernel.org
In-Reply-To: <20260525202921.124698-20-dakr@kernel.org>
References: <20260525202921.124698-20-dakr@kernel.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Date: Mon, 25 May 2026 21:44:38 +0000
Message-Id: <20260525214438.DAC3B1F000E9@smtp.kernel.org>
Precedence: bulk

Thank you for your contribution! Sashiko AI review found 2 potential issue(=
s) to consider:

New issues:
- [High] Incorrect syntax and type mismatch in `try_pin_init!` macro leads =
to a compilation failure in `drivers/pwm/pwm_th1520.rs`.

Pre-existing issues:
- [Low] The struct construction for `ExclusiveIoMem` is missing the mandate=
d `// INVARIANT:` comment.
--

commit 67b598dd02fe735e45135c07185ac5128b22e8d0
Author: Danilo Krummrich <dakr@kernel.org>

rust: io: make IoMem and ExclusiveIoMem lifetime-parameterized

This commit adds a lifetime parameter to IoMem and ExclusiveIoMem, tying
the mapping to the device's lifetime. It mirrors the pci::Bar design and
enables drivers to hold I/O memory mappings directly in their HRT private
data.

> diff --git a/drivers/pwm/pwm_th1520.rs b/drivers/pwm/pwm_th1520.rs
> index 6c5b791f31530..48808cd80737a 100644
> --- a/drivers/pwm/pwm_th1520.rs
> +++ b/drivers/pwm/pwm_th1520.rs
[ ... ]
> @@ -352,7 +352,7 @@ fn probe<'bound>(
>              dev,
>              TH1520_MAX_PWM_NUM,
>              try_pin_init!(Th1520PwmDriverData {
> -                iomem <- request.iomap_sized::<TH1520_PWM_REG_SIZE>(),
> +                iomem <- request.iomap_sized::<TH1520_PWM_REG_SIZE>()?.i=
nto_devres(),

[Severity: High]
Does this assignment cause a type mismatch in the try_pin_init! macro?

The into_devres() method returns a Result, but the <- operator expects an
expression that implements the PinInit trait. Could this be missing a
trailing ? operator to unwrap the Result before the pin initialization?

>                  clk <- clk,
>              }),
>          )?;

> diff --git a/rust/kernel/io/mem.rs b/rust/kernel/io/mem.rs
> index 51ba347220eea..fc2a3e24f8d5d 100644
> --- a/rust/kernel/io/mem.rs
> +++ b/rust/kernel/io/mem.rs
[ ... ]
> @@ -196,26 +188,29 @@ fn ioremap(resource: &Resource) -> Result<Self> {
>              )
>              .ok_or(EBUSY)?;
> =20
> -        let iomem =3D IoMem::ioremap(resource)?;
> +        let iomem =3D IoMem::ioremap(dev, resource)?;
> =20
> -        let iomem =3D ExclusiveIoMem {
> +        Ok(ExclusiveIoMem {

[Severity: Low]
This isn't a bug introduced by this patch, but ExclusiveIoMem has an
# Invariants documentation section. According to the Rust subsystem coding
guidelines, shouldn't its construction include an // INVARIANT: comment
explaining why the invariants are satisfied?

>              iomem,
>              _region: region,
> -        };
> -
> -        Ok(iomem)
> +        })
>      }

--=20
Sashiko AI review =C2=B7 https://sashiko.dev/#/patchset/20260525202921.1246=
98-1-dakr@kernel.org?part=3D19