Re: [PATCH 4/5] perf: arm: pmu: fix reference leak on failed device registration

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Calling `platform_device_put()` on statically allocated platform devices triggers a kernel WARNING.
--

commit 3c982eab6d069fa3cf3109dda88a31b66ceb0afd
Author: Vastargazing <vebohr@gmail.com>

perf: arm: pmu: fix reference leak on failed device registration

When platform_device_register() fails in arm_acpi_register_pmu_device(),
this commit adds a call to platform_device_put() in the error branch to
prevent a reference leak.

> diff --git a/drivers/perf/arm_pmu_acpi.c b/drivers/perf/arm_pmu_acpi.c
> index e80f76d95e68..c2defbc32ad9 100644
> --- a/drivers/perf/arm_pmu_acpi.c
> +++ b/drivers/perf/arm_pmu_acpi.c
> @@ -119,8 +119,10 @@ arm_acpi_register_pmu_device(struct platform_device *pdev, u8 len,
>  
>  	pdev->resource[0].start = irq;
>  	ret = platform_device_register(pdev);
> -	if (ret)
> +	if (ret) {
> +		platform_device_put(pdev);

Because the devices passed to this function (spe_dev and trbe_dev) are
statically allocated and do not provide a .dev.release callback, will calling
platform_device_put() trigger a kernel warning?

When the reference count drops to 0, device_release() is invoked.
Since no release function is defined for these static objects, it triggers
the driver core warning about lacking a release() function, which could panic
the system if panic_on_warn is configured.

The kernel documentation's advice to use platform_device_put() applies
to dynamically allocated devices, not statically allocated global variables.

>  		acpi_unregister_gsi(gsi);
> +	}
>  
>  	return ret;
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/cover.1777889235.git.vebohr@gmail.com?part=4