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

[Trevor Gamblin <tgamblin@baylibre.com>]

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.
>
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202408201520.lFtco3eF-lkp@intel.com/
>
> All errors (new ones prefixed by >>):
>
>     drivers/iio/adc/ad7625.c: In function 'ad7625_set_sampling_freq':
>>> drivers/iio/adc/ad7625.c:191:15: error: implicit declaration of function 'pwm_round_waveform_might_sleep' [-Werror=implicit-function-declaration]
>       191 |         ret = pwm_round_waveform_might_sleep(st->cnv_pwm, &cnv_wf);
>           |               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>     drivers/iio/adc/ad7625.c: In function 'ad7625_buffer_preenable':
>>> drivers/iio/adc/ad7625.c:420:15: error: implicit declaration of function 'pwm_set_waveform_might_sleep' [-Werror=implicit-function-declaration]
>       420 |         ret = pwm_set_waveform_might_sleep(st->cnv_pwm, &st->cnv_wf, false);
>           |               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
>     cc1: some warnings being treated as errors
>
>
> vim +/pwm_round_waveform_might_sleep +191 drivers/iio/adc/ad7625.c
>
>     175	
>     176	static int ad7625_set_sampling_freq(struct ad7625_state *st, int freq)
>     177	{
>     178		u64 target;
>     179		struct pwm_waveform clk_gate_wf = { }, cnv_wf = { };
>     180		int ret;
>     181	
>     182		target = DIV_ROUND_UP_ULL(NSEC_PER_SEC, freq);
>     183		cnv_wf.period_length_ns = clamp(target, 100, 10 * KILO);
>     184	
>     185		/*
>     186		 * Use the maximum conversion time t_CNVH from the datasheet as
>     187		 * the duty_cycle for ref_clk, cnv, and clk_gate
>     188		 */
>     189		cnv_wf.duty_length_ns = st->info->timing_spec->conv_high_ns;
>     190	
>   > 191		ret = pwm_round_waveform_might_sleep(st->cnv_pwm, &cnv_wf);
>     192		if (ret)
>     193			return ret;
>     194	
>     195		/*
>     196		 * Set up the burst signal for transferring data. period and
>     197		 * offset should mirror the CNV signal
>     198		 */
>     199		clk_gate_wf.period_length_ns = cnv_wf.period_length_ns;
>     200	
>     201		clk_gate_wf.duty_length_ns = DIV_ROUND_UP_ULL((u64)NSEC_PER_SEC *
>     202			st->info->chan_spec.scan_type.realbits,
>     203			st->ref_clk_rate_hz);
>     204	
>     205		/* max t_MSB from datasheet */
>     206		clk_gate_wf.duty_offset_ns = st->info->timing_spec->conv_msb_ns;
>     207	
>     208		ret = pwm_round_waveform_might_sleep(st->clk_gate_pwm, &clk_gate_wf);
>     209		if (ret)
>     210			return ret;
>     211	
>     212		st->cnv_wf = cnv_wf;
>     213		st->clk_gate_wf = clk_gate_wf;
>     214	
>     215		/* TODO: Add a rounding API for PWMs that can simplify this */
>     216		target = DIV_ROUND_CLOSEST_ULL(st->ref_clk_rate_hz, freq);
>     217		st->sampling_freq_hz = DIV_ROUND_CLOSEST_ULL(st->ref_clk_rate_hz,
>     218							     target);
>     219	
>     220		return 0;
>     221	}
>     222	
>