Re: [PATCH v2 5/9] iio: adc: ad7173: add support for special inputs
From: Ceclan, Dumitru
Date: Thu May 16 2024 - 12:03:34 EST
On 16/05/2024 02:27, David Lechner wrote:
> On Tue, May 14, 2024 at 2:23 AM Dumitru Ceclan via B4 Relay
> <devnull+dumitru.ceclan.analog.com@xxxxxxxxxx> wrote:
>>
>> From: Dumitru Ceclan <dumitru.ceclan@xxxxxxxxxx>
>>
>> Add support for selecting REF+ and REF- inputs on all models.
>> Add support for selecting ((AVDD1 − AVSS)/5) inputs
>> on supported models.
>>
>> Signed-off-by: Dumitru Ceclan <dumitru.ceclan@xxxxxxxxxx>
>> ---
>> drivers/iio/adc/ad7173.c | 21 +++++++++++++++++++++
>> 1 file changed, 21 insertions(+)
>>
>> diff --git a/drivers/iio/adc/ad7173.c b/drivers/iio/adc/ad7173.c
>> index fb33534d63a9..1e9ba3070770 100644
>> --- a/drivers/iio/adc/ad7173.c
>> +++ b/drivers/iio/adc/ad7173.c
>> @@ -65,6 +65,10 @@
>> FIELD_PREP(AD7173_CH_SETUP_AINNEG_MASK, neg))
>> #define AD7173_AIN_TEMP_POS 17
>> #define AD7173_AIN_TEMP_NEG 18
>> +#define AD7173_AIN_COM_IN_POS 19
>> +#define AD7173_AIN_COM_IN_NEG 20
>> +#define AD7173_AIN_REF_POS 21
>> +#define AD7173_AIN_REF_NEG 22
>>
>> #define AD7172_2_ID 0x00d0
>> #define AD7175_ID 0x0cd0
>> @@ -145,6 +149,8 @@ struct ad7173_device_info {
>> unsigned int id;
>> char *name;
>> bool has_temp;
>> + /* ((AVDD1 − AVSS)/5) */
>> + bool has_common_input;
>> bool has_input_buf;
>> bool has_int_ref;
>> bool has_ref2;
>> @@ -215,6 +221,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_temp = true,
>> .has_input_buf = true,
>> .has_int_ref = true,
>> + .has_common_input = true,
>> .clock = 2 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7173_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7173_sinc5_data_rates),
>> @@ -228,6 +235,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_temp = false,
>> .has_input_buf = true,
>> .has_ref2 = true,
>> + .has_common_input = true,
>> .clock = 2 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7173_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7173_sinc5_data_rates),
>> @@ -243,6 +251,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_input_buf = true,
>> .has_int_ref = true,
>> .has_ref2 = true,
>> + .has_common_input = false,
>> .clock = 2 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7173_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7173_sinc5_data_rates),
>> @@ -257,6 +266,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_temp = true,
>> .has_input_buf = true,
>> .has_int_ref = true,
>> + .has_common_input = true,
>> .clock = 16 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7175_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7175_sinc5_data_rates),
>> @@ -271,6 +281,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_input_buf = true,
>> .has_int_ref = true,
>> .has_ref2 = true,
>> + .has_common_input = true,
>> .clock = 16 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7175_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7175_sinc5_data_rates),
>> @@ -285,6 +296,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_temp = false,
>> .has_input_buf = false,
>> .has_int_ref = true,
>> + .has_common_input = false,
>> .clock = 16 * HZ_PER_MHZ,
>> .sinc5_data_rates = ad7175_sinc5_data_rates,
>> .num_sinc5_data_rates = ARRAY_SIZE(ad7175_sinc5_data_rates),
>> @@ -298,6 +310,7 @@ static const struct ad7173_device_info ad7173_device_info[] = {
>> .has_temp = true,
>> .has_input_buf = true,
>> .has_int_ref = true,
>> + .has_common_input = true,
>> .clock = 16 * HZ_PER_MHZ,
>> .odr_start_value = AD7177_ODR_START_VALUE,
>> .sinc5_data_rates = ad7175_sinc5_data_rates,
>> @@ -920,6 +933,14 @@ static int ad7173_validate_voltage_ain_inputs(struct ad7173_state *st,
>> if (ain[i] < st->info->num_inputs)
>> continue;
>>
>> + if (ain[i] == AD7173_AIN_REF_POS || ain[i] == AD7173_AIN_REF_NEG)
>> + continue;
>> +
>> + if ((ain[i] == AD7173_AIN_COM_IN_POS ||
>> + ain[i] == AD7173_AIN_COM_IN_NEG) &&
>> + st->info->has_common_input)
>> + continue;
>> +
>
> If there is only one valid combination, it seems like these should be
> fixed channels like the temperature input rather than something coming
> from the device tree.
>
As I've said, I do not agree with forcing one channel slot to be used.
I could add a property that spawns this channel. Although as I see under,
I think I'll permit these inputs to be mixed and matched.
> It looks like on AD411x, it is the case that there is only one valid
> option for the reference input in the channel configuration. But in
> the case of AD717x since both REF+ and REF- are listed as possible
> inputs for both AINPOS0 and AINNEG0, it seems like they could be mixed
> and matched with other channels. The datasheet doesn't seem very clear
> on this though.
>
This is imposed artificially, AD411x has the same cross-point mux that
can mix and match all the inputs.
> If it is valid to combine, say AIN0 with REF+ though, then the
> validation would need to be relaxed. But I'm guessing that is not
> actually the case?
>
I think it is the case.
>> return dev_err_probe(dev, -EINVAL,
>> "Input pin number out of range for pair (%d %d).\n",
>> ain[0], ain[1]);
>>
>> --
>> 2.43.0
>>
>>