Re: [PATCH 2/2] iio: adc: ti-adc128s052: Add lower resolution devices support

[Sukrut Bellary <sbellary@baylibre.com>]

On Mon, Apr 14, 2025 at 09:40:03AM +0300, Matti Vaittinen wrote:
> On 08/04/2025 16:21, Sukrut Bellary wrote:
> > The adcxx4s communicates with a host processor via an SPI/Microwire Bus
> > interface. The device family responds with 12-bit data, of which the LSB
> > bits are transmitted by the lower resolution devices as 0.
> > The unavailable bits are 0 in LSB.
> > Shift is calculated per resolution and used in scaling and
> > raw data read.
> > 
> > Lets reuse the driver to support the family of devices with name
> > ADC<bb><c>S<sss>, where
> > * bb is the resolution in number of bits (8, 10, 12)
> > * c is the number of channels (1, 2, 4, 8)
> > * sss is the maximum conversion speed (021 for 200 kSPS, 051 for 500 kSPS
> > and 101 for 1 MSPS)
> > 
> > Complete datasheets are available at TI's website here:
> > https://www.ti.com/lit/gpn/adc<bb><c>s<sss>.pdf
> 
> I tried looking up:
> https://www.ti.com/lit/gpn/adc102s051.pdf

Sorry about that. I missed to check the link before submitting the
patch series v3.

> > 
> > Tested only with ti-adc102s051 on BegalePlay SBC.
> > https://www.beagleboard.org/boards/beagleplay
> > 
> > Co-developed-by: Nishanth Menon <nm@ti.com>
> > Signed-off-by: Nishanth Menon <nm@ti.com>
> > Signed-off-by: Sukrut Bellary <sbellary@baylibre.com>
> > ---
> > Changes in v3:
> >          - used be16_to_cpu() for the endian conversion.
> >          - used config index enum while setting up the adc128_config[]
> > 
> > - Link to v2:
> >          https://lore.kernel.org/lkml/20231022031203.632153-1-sukrut.bellary@linux.com/
> > 
> > Changes in v2:
> >          - Arranged of_device_id and spi_device_id in numeric order.
> >          - Used enum to index into adc128_config.
> >          - Reorder adc128_config in alphabetical.
> >          - Include channel resolution information.
> >          - Shift is calculated per resolution and used in scaling and
> >          raw data read.
> > 
> > - Link to v1: https://lore.kernel.org/all/20220701042919.18180-1-nm@ti.com/
> > ---
> >   drivers/iio/adc/ti-adc128s052.c | 149 ++++++++++++++++++++++++--------
> >   1 file changed, 112 insertions(+), 37 deletions(-)
> > 
> 
> Hi dee Ho,
> 
> Thanks for improving this! It's always nice to be able to support more
> devices with small(ish) changes!
> 
> This looks good to me. I will take another, hopefully more in-depth look at
> the rebased version when available though.
> 
> I have just one comment for now, but it's not strictly related to this
> change. If you wish to go the extra mile, then I'd appreciated it. If not,
> then it can be re-worked later. Anyways, please, see below.
> 
> > diff --git a/drivers/iio/adc/ti-adc128s052.c b/drivers/iio/adc/ti-adc128s052.c
> > index a456ea78462f..d4b76fd85abd 100644
> > --- a/drivers/iio/adc/ti-adc128s052.c
> > +++ b/drivers/iio/adc/ti-adc128s052.c
> > @@ -7,6 +7,22 @@
> >    * https://www.ti.com/lit/ds/symlink/adc128s052.pdf
> >    * https://www.ti.com/lit/ds/symlink/adc122s021.pdf
> >    * https://www.ti.com/lit/ds/symlink/adc124s021.pdf
> > + *
> > + * The adcxx4s communicates with a host processor via an SPI/Microwire Bus
> > + * interface. This driver supports the whole family of devices with a name
> > + * ADC<bb><c>S<sss>, where
> > + * bb is the resolution in number of bits (8, 10, 12)
> > + * c is the number of channels (1, 2, 4, 8)
> > + * sss is the maximum conversion speed (021 for 200 kSPS, 051 for 500 kSPS
> > + * and 101 for 1 MSPS)
> > + *
> > + * Complete datasheets are available at TI's website here:
> > + *   https://www.ti.com/lit/gpn/adc<bb><c>s<sss>.pdf
> > + *
> > + * 8, 10, and 12 bits converters send 12-bit data with
> > + * unavailable bits set to 0 in LSB.
> > + * Shift is calculated per resolution and used in scaling and
> > + * raw data read.
> >    */
> >   #include <linux/err.h>
> > @@ -53,7 +69,7 @@ static int adc128_adc_conversion(struct adc128 *adc, u8 channel)
> >   	if (ret < 0)
> >   		return ret;
> > -	return ((adc->buffer[0] << 8 | adc->buffer[1]) & 0xFFF);
> > +	return be16_to_cpu(*((__be16 *)adc->buffer));
> >   }
> >   static int adc128_read_raw(struct iio_dev *indio_dev,
> > @@ -70,7 +86,8 @@ static int adc128_read_raw(struct iio_dev *indio_dev,
> >   		if (ret < 0)
> >   			return ret;
> > -		*val = ret;
> > +		*val = (ret >> channel->scan_type.shift) &
> > +			GENMASK(channel->scan_type.realbits - 1, 0);
> >   		return IIO_VAL_INT;
> >   	case IIO_CHAN_INFO_SCALE:
> > @@ -80,7 +97,7 @@ static int adc128_read_raw(struct iio_dev *indio_dev,
> >   			return ret;
> >   		*val = ret / 1000;
> > -		*val2 = 12;
> > +		*val2 = channel->scan_type.realbits;
> >   		return IIO_VAL_FRACTIONAL_LOG2;
> >   	default:
> > @@ -89,24 +106,34 @@ static int adc128_read_raw(struct iio_dev *indio_dev,
> >   }
> > -#define ADC128_VOLTAGE_CHANNEL(num)	\
> > -	{ \
> > -		.type = IIO_VOLTAGE, \
> > -		.indexed = 1, \
> > -		.channel = (num), \
> > -		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW), \
> > -		.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE) \
> > +#define _ADC128_VOLTAGE_CHANNEL(num, real_bits, store_bits)		\
> > +	{								\
> > +		.type = IIO_VOLTAGE,					\
> > +		.indexed = 1,						\
> > +		.channel = (num),					\
> > +		.info_mask_separate = BIT(IIO_CHAN_INFO_RAW),		\
> > +		.info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),	\
> > +		.scan_index = (num),					\
> > +		.scan_type = {						\
> > +			.sign = 'u',					\
> > +			.realbits = (real_bits),			\
> > +			.storagebits = (store_bits),			\
> > +			.shift = (12 - real_bits),			\
> > +		},							\
> >   	}
> > -static const struct iio_chan_spec adc128s052_channels[] = {
> > -	ADC128_VOLTAGE_CHANNEL(0),
> > -	ADC128_VOLTAGE_CHANNEL(1),
> > -	ADC128_VOLTAGE_CHANNEL(2),
> > -	ADC128_VOLTAGE_CHANNEL(3),
> > -	ADC128_VOLTAGE_CHANNEL(4),
> > -	ADC128_VOLTAGE_CHANNEL(5),
> > -	ADC128_VOLTAGE_CHANNEL(6),
> > -	ADC128_VOLTAGE_CHANNEL(7),
> > +#define ADC082_VOLTAGE_CHANNEL(num) _ADC128_VOLTAGE_CHANNEL(num, 8, 16)
> > +#define ADC102_VOLTAGE_CHANNEL(num) _ADC128_VOLTAGE_CHANNEL(num, 10, 16)
> > +#define ADC128_VOLTAGE_CHANNEL(num) _ADC128_VOLTAGE_CHANNEL(num, 12, 16)
> > +
> > +static const struct iio_chan_spec adc082s021_channels[] = {
> > +	ADC082_VOLTAGE_CHANNEL(0),
> > +	ADC082_VOLTAGE_CHANNEL(1),
> > +};
> > +
> > +static const struct iio_chan_spec adc102s021_channels[] = {
> > +	ADC102_VOLTAGE_CHANNEL(0),
> > +	ADC102_VOLTAGE_CHANNEL(1),
> >   };
> >   static const struct iio_chan_spec adc122s021_channels[] = {
> > @@ -121,10 +148,46 @@ static const struct iio_chan_spec adc124s021_channels[] = {
> >   	ADC128_VOLTAGE_CHANNEL(3),
> >   };
> > +static const struct iio_chan_spec adc128s052_channels[] = {
> > +	ADC128_VOLTAGE_CHANNEL(0),
> > +	ADC128_VOLTAGE_CHANNEL(1),
> > +	ADC128_VOLTAGE_CHANNEL(2),
> > +	ADC128_VOLTAGE_CHANNEL(3),
> > +	ADC128_VOLTAGE_CHANNEL(4),
> > +	ADC128_VOLTAGE_CHANNEL(5),
> > +	ADC128_VOLTAGE_CHANNEL(6),
> > +	ADC128_VOLTAGE_CHANNEL(7),
> > +};
> > +
> > +enum adc128_configuration_index {
> > +	ADC128_CONFIG_INDEX_082S,
> > +	ADC128_CONFIG_INDEX_102S,
> > +	ADC128_CONFIG_INDEX_122S,
> > +	ADC128_CONFIG_INDEX_124S,
> > +	ADC128_CONFIG_INDEX_128S,
> > +};
> 
> I like the fact you added these indexes as it makes this a lot clearer.
> But...
> 
> > +
> >   static const struct adc128_configuration adc128_config[] = {
> > -	{ adc128s052_channels, ARRAY_SIZE(adc128s052_channels) },
> > -	{ adc122s021_channels, ARRAY_SIZE(adc122s021_channels) },
> > -	{ adc124s021_channels, ARRAY_SIZE(adc124s021_channels) },
> > +	[ADC128_CONFIG_INDEX_082S] = {
> > +		.channels = adc082s021_channels,
> > +		.num_channels = ARRAY_SIZE(adc082s021_channels)
> > +	},
> > +	[ADC128_CONFIG_INDEX_102S] = {
> > +		.channels = adc102s021_channels,
> > +		.num_channels = ARRAY_SIZE(adc102s021_channels)
> > +	},
> > +	[ADC128_CONFIG_INDEX_122S] = {
> > +		.channels = adc122s021_channels,
> > +		.num_channels = ARRAY_SIZE(adc122s021_channels)
> > +	},
> > +	[ADC128_CONFIG_INDEX_124S] = {
> > +		.channels = adc124s021_channels,
> > +		.num_channels = ARRAY_SIZE(adc124s021_channels)
> > +	},
> > +	[ADC128_CONFIG_INDEX_128S] = {
> > +		.channels = adc128s052_channels,
> > +		.num_channels = ARRAY_SIZE(adc128s052_channels)
> > +	},
> >   };
> 
> ... I don't really love this array. I believe the code would be clearer if
> this array was changed to individual structs because ...
> 
> >   static const struct iio_info adc128_info = {
> > @@ -177,31 +240,43 @@ static int adc128_probe(struct spi_device *spi)
> >   }
> >   static const struct of_device_id adc128_of_match[] = {
> > -	{ .compatible = "ti,adc128s052", .data = &adc128_config[0] },
> > -	{ .compatible = "ti,adc122s021", .data = &adc128_config[1] },
> > -	{ .compatible = "ti,adc122s051", .data = &adc128_config[1] },
> > -	{ .compatible = "ti,adc122s101", .data = &adc128_config[1] },
> > -	{ .compatible = "ti,adc124s021", .data = &adc128_config[2] },
> > -	{ .compatible = "ti,adc124s051", .data = &adc128_config[2] },
> > -	{ .compatible = "ti,adc124s101", .data = &adc128_config[2] },
> > +	{ .compatible = "ti,adc082s021", .data = &adc128_config[ADC128_CONFIG_INDEX_082S] },
> > +	{ .compatible = "ti,adc082s051", .data = &adc128_config[ADC128_CONFIG_INDEX_082S] },
> > +	{ .compatible = "ti,adc082s101", .data = &adc128_config[ADC128_CONFIG_INDEX_082S] },
> > +	{ .compatible = "ti,adc102s021", .data = &adc128_config[ADC128_CONFIG_INDEX_102S] },
> > +	{ .compatible = "ti,adc102s051", .data = &adc128_config[ADC128_CONFIG_INDEX_102S] },
> > +	{ .compatible = "ti,adc102s101", .data = &adc128_config[ADC128_CONFIG_INDEX_102S] },
> > +	{ .compatible = "ti,adc122s021", .data = &adc128_config[ADC128_CONFIG_INDEX_122S] },
> > +	{ .compatible = "ti,adc122s051", .data = &adc128_config[ADC128_CONFIG_INDEX_122S] },
> > +	{ .compatible = "ti,adc122s101", .data = &adc128_config[ADC128_CONFIG_INDEX_122S] },
> > +	{ .compatible = "ti,adc124s021", .data = &adc128_config[ADC128_CONFIG_INDEX_124S] },
> > +	{ .compatible = "ti,adc124s051", .data = &adc128_config[ADC128_CONFIG_INDEX_124S] },
> > +	{ .compatible = "ti,adc124s101", .data = &adc128_config[ADC128_CONFIG_INDEX_124S] },
> > +	{ .compatible = "ti,adc128s052", .data = &adc128_config[ADC128_CONFIG_INDEX_128S] },
> 
> ... here we could then directly refer to individual structs. That way we
> would not need to define the names for the array indexes (for clarity), or
> look up the individual array members based on magic numbers.

Thanks for the review.
yes, I will take care of this in v4.

> >   	{ /* sentinel */ },
> >   };
> >   MODULE_DEVICE_TABLE(of, adc128_of_match);
> 
> Yours,
> 	-- Matti