Re: [PATCH v2 2/4] rtc: s32g: add NXP S32G2/S32G3 SoC support

[Ciprian Marian Costea <ciprianmarian.costea@oss.nxp.com>]

On 10/16/2024 12:42 PM, Uwe Kleine-König wrote:

Hello Uwe,

Thanks for your review!

> Hello,
> 
> On Tue, Oct 15, 2024 at 01:51:31PM +0300, Ciprian Costea wrote:
>> +static void rtc_disable(struct rtc_priv *priv)
> 
> A very generic name for a driver specific function. I'm a big fan of
> driver specific prefixes and I wonder why this isn't called
> s34g_rtc_disable().
> 
> Also a comment about what is actually disabled here would be nice (or
> maybe a better name). I hope this doesn't stop the RTC ticking??
> 

Indeed, 's32g_rtc_disable' would be a better name. I will update in V3 
and add a comment before the function definition.

It does stop the RTC running counter because as per RTC module 
documentation from S32G2/S32G3 SoC Reference Manual, the clock source 
and divisors selection should be performed with RTC counter disabled in 
order to not cause any synchronization issues.

So following this limitation, when the clock source is switched (as it 
is the case for S32G2 and S32G3 and Suspend to RAM) the RTC counter 
should be disabled.

>> +static struct platform_driver rtc_driver = {
>> +	.driver		= {
>> +		.name			= "s32g-rtc",
>> +		.pm				= &rtc_pm_ops,
>> +		.of_match_table = rtc_dt_ids,
>> +	},
>> +	.probe		= rtc_probe,
>> +	.remove_new	= rtc_remove,
>> +};
> 
> After commit 0edb555a65d1 ("platform: Make platform_driver::remove()
> return void") .remove() is (again) the right callback to implement for
> platform drivers. Please just drop "_new".
> 

Thanks for pointing this out. I was not aware of this change. I will 
update in V3.

> Best regards
> Uwe

Best Regards,
Ciprian