Re: [PATCH v3 2/3] iio: adc: ad7625: add driver

[Trevor Gamblin <tgamblin@baylibre.com>]

On 2024-09-04 3:58 a.m., Uwe Kleine-König wrote:
> Hi Trevor,
>
> On Tue, Aug 20, 2024 at 05:07:27PM -0400, Trevor Gamblin wrote:
>> On 2024-08-20 3:19 a.m., kernel test robot wrote:
>>> Hi Trevor,
>>>
>>> kernel test robot noticed the following build errors:
>>>
>>> [auto build test ERROR on ac6a258892793f0a255fe7084ec2b612131c67fc]
>>>
>>> url:    https://github.com/intel-lab-lkp/linux/commits/Trevor-Gamblin/dt-bindings-iio-adc-add-AD762x-AD796x-ADCs/20240819-221425
>>> base:   ac6a258892793f0a255fe7084ec2b612131c67fc
>>> patch link:    https://lore.kernel.org/r/20240819-ad7625_r1-v3-2-75d5217c76b5%40baylibre.com
>>> patch subject: [PATCH v3 2/3] iio: adc: ad7625: add driver
>>> config: alpha-randconfig-r132-20240820 (https://download.01.org/0day-ci/archive/20240820/202408201520.lFtco3eF-lkp@intel.com/config)
>>> compiler: alpha-linux-gcc (GCC) 13.3.0
>>> reproduce: (https://download.01.org/0day-ci/archive/20240820/202408201520.lFtco3eF-lkp@intel.com/reproduce)
>> Seems to be a problem with missing static inline definitions in pwm.h if
>> CONFIG_PWM isn't set. I've replied to the relevant series on the PWM mailing
>> list and will add "select PWM" to Kconfig for this driver.
> I'm not a big fan of the dummy static inlines. It seems to be a somewhat
> subjective thing, but I think that usually if a driver makes use of PWM
> functions it doesn't work at all if CONFIG_PWM=n. Does your driver work
> with CONFIG_PWM=n? If not, even if the dummy inline was there, I'd
> recommend at least a

No, it won't build without pwm_round_waveform_might_sleep() and 
pwm_set_waveform_might_sleep(), which setting CONFIG_PWM=n seems to hide.

I'll add depends on PWM to the next round of the patch series.

Thanks!

>
> 	depends on PWM || COMPILE_TEST
>
> . (This is also the implicit recommendation to use "depends" and not
> "select". Currently all drivers needing PWM use "depends" and mixing
> yields strange effects in menuconfig.)
>
> Currently there is only a single driver that uses "depends on PWM ||
> COMPILE_TEST" (i.e. SENSORS_PWM_FAN). I already considered changing that
> to plain "depends on PWM" and get rid of the dummy defines. While I
> didn't tackle that one yet, I'd like to not introduce dummys for the new
> waveform functions. So I suggest you either stick to
>
> 	depends on PWM
>
> or try to convince me that these dummys are a good idea (and then
> probably use "... || COMPILE_TEST").
>
> Best regards
> Uwe