Re: [PATCH] rtc: s3c: Don't print an error on probe deferral

[Javier Martinez Canillas <javier@osg.samsung.com>]

Hello Joe,

On 03/14/2016 04:38 PM, Joe Perches wrote:
> On Mon, 2016-03-14 at 16:31 -0300, Javier Martinez Canillas wrote:
>> On 03/14/2016 04:11 PM, Joe Perches wrote:> > On Mon, 2016-03-14 at 16:05 -0300, Javier Martinez Canillas wrote:
>>>>
>>>> The clock and source clock looked up by the driver may not be available
>>>> just because the clock controller driver was not probed yet so printing
>>>> an error in this case is not correct and only adds confusion to users.
>>>>
>>>> However, knowing that a driver's probe was deferred may be useful so it
>>>> can be printed as debug information.
>>> []
>>>>
>>>> diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
>>> []
>>>>
>>>> @@ -501,18 +501,27 @@ static int s3c_rtc_probe(struct platform_device *pdev)
>>>>  
>>>>  	info->rtc_clk = devm_clk_get(&pdev->dev, "rtc");
>>>>  	if (IS_ERR(info->rtc_clk)) {
>>>> -		dev_err(&pdev->dev, "failed to find rtc clock\n");
>>>> -		return PTR_ERR(info->rtc_clk);
>>>> +		ret = PTR_ERR(info->rtc_clk);
>>>> +		if (ret != -EPROBE_DEFER)
>>>> +			dev_err(&pdev->dev, "failed to find rtc clock\n");
>>>> +		else
>>>> +			dev_dbg(&pdev->dev, "probe deferred due rtc clock\n");
>>>> +		return ret;
>>>>  	}
>>>>  	clk_prepare_enable(info->rtc_clk);
>>>>  
>>>>  	if (info->data->needs_src_clk) {
>>>>  		info->rtc_src_clk = devm_clk_get(&pdev->dev, "rtc_src");
>>>>  		if (IS_ERR(info->rtc_src_clk)) {
>>>> -			dev_err(&pdev->dev,
>>>> -				"failed to find rtc source clock\n");
>>>> +			ret = PTR_ERR(info->rtc_src_clk);
>>>> +			if (ret != -EPROBE_DEFER)
>>>> +				dev_err(&pdev->dev,
>>>> +					"failed to find rtc source clock\n");
>>>> +			else
>>>> +				dev_dbg(&pdev->dev,
>>>> +					"probe deferred due rtc source clock\n");
>>>>  			clk_disable_unprepare(info->rtc_clk);
>>>> -			return PTR_ERR(info->rtc_src_clk);
>>>> +			return ret;
>>>>  		}
>>>>  		clk_prepare_enable(info->rtc_src_clk);
>>>>  	}
>>> Maybe the debug logging messages could be object->action like:
>>>
>>> 	rtc clock probe deferred
>>> 	rtc source clock probe deferred
>>>
>> I found your suggested messages harder to read and more confusing. The
>> action that happens is a probe function deferral and that is caused by
>> a missing resource needed by the driver (clocks in this case).
>>
>> But your messages seems to imply that the probe deferred action happens
>> to a clock, it sounds like "rtc clock disabled" and that's not correct.
> 
> OK, then please change "due" to "due to" or "for" in your messages
> because they make little sense now.
>

I don't think they make little sense now since even a non-native english
speaker like me can understand it :)

But yes, it's cryptic at the very least. That's the problem with long text
and the 80 char limit to make checkpatch.pl happy. I guess I can just move
the message a little bit even if that will make to not be properly aligned.

I'll wait a couple of days to see if there's any other feedback and repost.

Best regards,
-- 
Javier Martinez Canillas
Open Source Group
Samsung Research America