RE: [PATCH 1/3] rtc: da9052: convert to devm_rtc_allocate_device

[Adam Thomson]

On 06 March 2020 07:36, Alexandre Belloni wrote:

> This allows further improvement of the driver.
> 
> Signed-off-by: Alexandre Belloni <alexandre.belloni@xxxxxxxxxxx>
> ---
>  drivers/rtc/rtc-da9052.c | 8 ++++----
>  1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/rtc/rtc-da9052.c b/drivers/rtc/rtc-da9052.c
> index 204eb7cf1aa4..e159c5911122 100644
> --- a/drivers/rtc/rtc-da9052.c
> +++ b/drivers/rtc/rtc-da9052.c
> @@ -298,12 +298,12 @@ static int da9052_rtc_probe(struct platform_device
> *pdev)
>  		rtc_err(rtc, "Failed to disable TICKS: %d\n", ret);
> 
>  	device_init_wakeup(&pdev->dev, true);
> -	rtc->rtc = devm_rtc_device_register(&pdev->dev, pdev->name,
> -				       &da9052_rtc_ops, THIS_MODULE);
> -
> +	rtc->rtc = devm_rtc_allocate_device(&pdev->dev);
>  	if (IS_ERR(rtc->rtc))
>  		return PTR_ERR(rtc->rtc);
> 
> +	rtc->rtc->ops = &da9052_rtc_ops;
> +
>  	ret = da9052_request_irq(rtc->da9052, DA9052_IRQ_ALARM, "ALM",
>  				da9052_rtc_irq, rtc);
>  	if (ret != 0) {
> @@ -311,7 +311,7 @@ static int da9052_rtc_probe(struct platform_device
> *pdev)
>  		return ret;
>  	}
> 
> -	return 0;
> +	return rtc_register_device(rtc->rtc);

It seems with this change there's no tidy up around the IRQ.
da9052_request_irq() isn't calling a devm_* function so I'm not sure how the
IRQ is being freed on failure. In actual fact it looks like there's no _remove()
function for this driver which, although that's a separate topic.

>  }
> 
>  static struct platform_driver da9052_rtc_driver = {
> --
> 2.24.1