Re: [PATCH v2 4/4] cdx: use generic driver_override infrastructure

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] Stale kernel-doc entry for removed `@driver_override` member in `struct cdx_device`.
--

commit c9a2a632f33a0544d4ef0145d2020070218d6838
Author: Runyu Xiao <runyu.xiao@seu.edu.cn>

cdx: use generic driver_override infrastructure

This commit switches the CDX subsystem to use the generic driver-core
driver_override infrastructure. It removes the private driver_override
storage and uses device_match_driver_override() to prevent races during
bus matching.

Link: https://lore.kernel.org/driver-core/DGRGTIRHA62X.3RY09D9SOK77P@kernel.org/

[ ... ]

> diff --git a/include/linux/cdx/cdx_bus.h b/include/linux/cdx/cdx_bus.h
> index b1ba97f6c9ad6..f1a107b232daa 100644
> --- a/include/linux/cdx/cdx_bus.h
> +++ b/include/linux/cdx/cdx_bus.h
> @@ -165,7 +165,6 @@ struct cdx_device {
>  	bool enabled;
>  	u32 msi_dev_id;
>  	u32 num_msi;
> -	const char *driver_override;

[Severity: Low]
Does this leave a stale kernel-doc entry for @driver_override in the
struct cdx_device documentation block?

The struct documentation still references the removed field:

 * @driver_override: driver name to force a match; do not set directly,
 *                   because core frees it; use driver_set_override() to
 *                   set or clear it.

This might trigger an excess kerneldoc parameter warning during the kernel
build (with W=1).

>  	struct mutex irqchip_lock;
>  	bool msi_write_pending;
>  };

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260604035239.1711889-1-runyu.xiao@seu.edu.cn?part=4