Re: [PATCH v2 3/9] mailbox: imx: use devm_of_platform_populate()

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] Unbalanced clk_disable_unprepare() on the error path for devm_mbox_controller_register() causes a double clock disable.
--

commit 217e9aa29dddb1723c211e93d96e446d0ea504a7
Author: Sebastian Andrzej Siewior <bigeasy@linutronix.de>

mailbox: imx: use devm_of_platform_populate()

The driver uses of_platform_populate() but does not remove the added
devices on removal, which can lead to duplicate devices upon module reload.
This commit switches to devm_of_platform_populate() to ensure populated
devices are removed when the parent device is removed.

> diff --git a/drivers/mailbox/imx-mailbox.c b/drivers/mailbox/imx-mailbox.c
> index 24cf704b4f9bd..1dda1437b87f6 100644
> --- a/drivers/mailbox/imx-mailbox.c
> +++ b/drivers/mailbox/imx-mailbox.c
> @@ -983,7 +983,7 @@ static int imx_mu_probe(struct platform_device *pdev)
>  	if (ret)
>  		goto disable_clk;

[Severity: High]
Does this error path trigger a double clock disable?

Looking at imx_mu_probe(), the clock is explicitly disabled before
devm_mbox_controller_register() is called:

	clk_disable_unprepare(priv->clk);

	ret = devm_mbox_controller_register(dev, &priv->mbox);
	if (ret)
		goto disable_clk;

If the registration fails, we jump to the disable_clk label which
unconditionally calls clk_disable_unprepare(priv->clk) a second time:

disable_clk:
	clk_disable_unprepare(priv->clk);
	return ret;

Can this imbalanced clock disable cause a framework warning and potentially
break shared clocks used by other devices?

This regression was introduced by the PM runtime refactoring earlier in the
series and remains unfixed.

>  
> -	of_platform_populate(dev->of_node, NULL, NULL, dev);
> +	devm_of_platform_populate(dev);
>  
>  	return 0;

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260603-imx_mbox_rproc-v2-0-a0059dc3b69a@linutronix.de?part=3