Re: [PATCH] counter: rz-mtu3-cnt: check clk_prepare_enable() return value

On Fri, 28 Aug 2026 15:11:23 +0800
"Li Youhong" <[email protected]> wrote:

> From: Li Youhong <[email protected]>
> 
> The driver ignored clk_prepare_enable() failures during probe and
> runtime resume.  Propagate the error so probe does not continue with
> the clock disabled and resume does not report success falsely.
> 
> Signed-off-by: Li Youhong <[email protected]>
> ---
>  drivers/counter/rz-mtu3-cnt.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/counter/rz-mtu3-cnt.c b/drivers/counter/rz-mtu3-cnt.c
> index 7bfb6979193c..6f746cf598c0 100644
> --- a/drivers/counter/rz-mtu3-cnt.c
> +++ b/drivers/counter/rz-mtu3-cnt.c
> @@ -817,9 +817,7 @@ static int rz_mtu3_cnt_pm_runtime_resume(struct device *dev)
>  {
>  	struct clk *const clk = dev_get_drvdata(dev);
>  
> -	clk_prepare_enable(clk);
> -
> -	return 0;
> +	return clk_prepare_enable(clk);
>  }
>  
>  static DEFINE_RUNTIME_DEV_PM_OPS(rz_mtu3_cnt_pm_ops,
> @@ -861,7 +859,10 @@ static int rz_mtu3_cnt_probe(struct platform_device *pdev)
>  
>  	mutex_init(&priv->lock);
>  	platform_set_drvdata(pdev, priv->clk);
> -	clk_prepare_enable(priv->clk);
> +	ret = clk_prepare_enable(priv->clk);
> +	if (ret)
> +		return ret;
> +
>  	pm_runtime_set_active(&pdev->dev);
>  	pm_runtime_enable(&pdev->dev);
>  	ret = devm_add_action_or_reset(&pdev->dev, rz_mtu3_cnt_pm_disable, dev);

This is fine, but any pm_runtime_get*() function called in this driver
would return a negative error code if clk_prepare_enable() fails. You
should definitely add checks whether any pm_runtime_get*() functions
succeeded, otherwise subsequent reads/writes will produce garbage.

-- 
Kind regards,
Joshua Crofts