Re: [PATCH] Add support for the Texas Intruments ADS868x ADC.

[Jonathan Cameron <jic23@kernel.org>]

On 10/09/15 08:22, Sean Nyekjær wrote:
> Hi Jonathan
> 
> Regarding ads868x_show_scales.
> Yes it is register values, the challenge here is the [±1.25*vref, +2.5*vref] and [±0.625*VREF, +1.25*VREF] results in the same scaling values. How can we control this?
Compute the actual scale and offsets and export them via *_available.
Then a write to either will change to relevant register setting.

So as per the ABI thought obviously it can be a bit fiddly to compute
the values!

Jonathan
> 
> Sean
> 
> On 2015-09-09 21:33, Jonathan Cameron wrote:
>> On 07/09/15 10:34, Sean Nyekjaer wrote:
>> A comment here describing the device and the functionality supported by
>> the driver would be good.
>>
>> Also, there are device tree bindings in here.  They should have
>> accompanying documentation (cc'd to the device tree list).
>>
>> Various comments inline. Nothing particularly major, though be
>> careful to obey the ABI and not use magic values in attributes.
>>
>> Jonathan
>>> Signed-off-by: Sean Nyekjaer <sean.nyekjaer@prevas.dk>
>>> Reviewed-by: Martin Hundebøll <martin.hundeboll@prevas.dk>
>>> ---
>>>   drivers/iio/adc/Kconfig      |  10 ++
>>>   drivers/iio/adc/Makefile     |   1 +
>>>   drivers/iio/adc/ti-ads868x.c | 398 +++++++++++++++++++++++++++++++++++++++++++
>>>   3 files changed, 409 insertions(+)
>>>   create mode 100644 drivers/iio/adc/ti-ads868x.c
>>>
>>> diff --git a/drivers/iio/adc/Kconfig b/drivers/iio/adc/Kconfig
>>> index deea62c..39924d5 100644
>>> --- a/drivers/iio/adc/Kconfig
>>> +++ b/drivers/iio/adc/Kconfig
>>> @@ -322,6 +322,16 @@ config TI_ADC128S052
>>>         This driver can also be built as a module. If so, the module will be
>>>         called ti-adc128s052.
>>>
>>> +config TI_ADS868X
>>> +    tristate "Texas Instruments ADS8684/8"
>>> +    depends on SPI && OF
>>> +    help
>>> +      If you say yes here you get support for Texas Instruments ADS8684 and
>>> +      and ADS8688 ADC chips
>>> +
>>> +      This driver can also be built as a module. If so, the module will be
>>> +      called ti-ads868x.
>>> +
>>>   config TI_AM335X_ADC
>>>       tristate "TI's AM335X ADC driver"
>>>       depends on MFD_TI_AM335X_TSCADC
>>> diff --git a/drivers/iio/adc/Makefile b/drivers/iio/adc/Makefile
>>> index fa5723a..75170d2 100644
>>> --- a/drivers/iio/adc/Makefile
>>> +++ b/drivers/iio/adc/Makefile
>>> @@ -31,6 +31,7 @@ obj-$(CONFIG_QCOM_SPMI_VADC) += qcom-spmi-vadc.o
>>>   obj-$(CONFIG_ROCKCHIP_SARADC) += rockchip_saradc.o
>>>   obj-$(CONFIG_TI_ADC081C) += ti-adc081c.o
>>>   obj-$(CONFIG_TI_ADC128S052) += ti-adc128s052.o
>>> +obj-$(CONFIG_TI_ADS868X) += ti-ads868x.o
>>>   obj-$(CONFIG_TI_AM335X_ADC) += ti_am335x_adc.o
>>>   obj-$(CONFIG_TWL4030_MADC) += twl4030-madc.o
>>>   obj-$(CONFIG_TWL6030_GPADC) += twl6030-gpadc.o
>>> diff --git a/drivers/iio/adc/ti-ads868x.c b/drivers/iio/adc/ti-ads868x.c
>>> new file mode 100644
>>> index 0000000..368e22c
>>> --- /dev/null
>>> +++ b/drivers/iio/adc/ti-ads868x.c
>>> @@ -0,0 +1,398 @@
>>> +/*
>>> + * Copyright (C) 2015 Prevas A/S
>>> + *
>>> + * This program is free software; you can redistribute it and/or modify
>>> + * it under the terms of the GNU General Public License version 2 as
>>> + * published by the Free Software Foundation.
>>> + */
>>> +
>>> +#include <linux/device.h>
>>> +#include <linux/kernel.h>
>>> +#include <linux/slab.h>
>>> +#include <linux/sysfs.h>
>>> +#include <linux/spi/spi.h>
>>> +#include <linux/regulator/consumer.h>
>>> +#include <linux/err.h>
>>> +#include <linux/delay.h>
>> Not used that I can immediately spot.
>>
>>> +#include <linux/module.h>
>>> +#include <linux/interrupt.h>
>> not used
>>
>>> +#include <linux/of.h>
>>> +
>>> +#include <linux/iio/iio.h>
>>> +#include <linux/iio/sysfs.h>
>>> +#include <linux/iio/buffer.h>
>> Not currently used.
>>> +
>>> +#define ADS868X_CMD_REG(x)        (x << 8)
>>> +#define    ADS868X_CMD_REG_NOOP        0x00
>>> +#define ADS868X_CMD_REG_RST        0x85
>>> +#define ADS868X_CMD_REG_MAN_CH(chan)    (0xC0 | (4 * chan))
>>> +#define ADS868X_CMD_DONT_CARE_BITS    16
>>> +
>>> +#define ADS868X_PROG_REG(x)        (x << 9)
>>> +#define ADS868X_PROG_REG_RANGE_CH(chan)    (0x05 + chan)
>>> +#define ADS868X_PROG_WR_BIT        (1 << 8)
>> BIT Macro preferred.
>>> +#define ADS868X_PROG_DONT_CARE_BITS    8
>>> +
>>> +#define ADS868X_VREF_MV            4096
>>> +
>>> +struct ads868x_chip_info {
>>> +    unsigned int id;
>>> +    const struct iio_chan_spec *channels;
>>> +    unsigned int num_channels;
>>> +    unsigned int flags;
>>> +    const struct iio_info *iio_info;
>>> +};
>>> +
>>> +struct ads868x_state {
>>> +    const struct ads868x_chip_info    *chip_info;
>>> +    struct spi_device        *spi;
>>> +    struct regulator        *reg;
>>> +    unsigned long            vref_mv;
>>> +    char                range[8];
>>> +    union {
>>> +        __be32 d32;
>>> +        u8 d8[4];
>>> +    } data[2] ____cacheline_aligned;
>>> +};
>>> +
>>> +enum ads868x_id {
>>> +    ID_ADS8684,
>>> +    ID_ADS8688,
>>> +};
>>> +
>>> +enum ads868x_range {
>>> +    ADS868X_PLUSMINUS25VREF        =    0x00,
>>> +    ADS868X_PLUSMINUS125VREF    =    0x01,
>>> +    ADS868X_PLUSMINUS0625VREF    =    0x02,
>>> +    ADS868X_PLUS25VREF        =    0x05,
>>> +    ADS868X_PLUS125VREF        =    0x06,
>>> +};
>>> +
>>> +static const char ads868x_range_def[] = {
>>> +    ADS868X_PLUSMINUS25VREF,
>>> +    ADS868X_PLUSMINUS125VREF,
>>> +    ADS868X_PLUSMINUS0625VREF,
>>> +    ADS868X_PLUS25VREF,
>>> +    ADS868X_PLUS125VREF,
>>> +};
>>> +
>>> +static ssize_t ads868x_show_scales(struct device *dev,
>>> +                   struct device_attribute *attr, char *buf)
>>> +{
>>> +    int i, len = 0;
>>> +
>>> +    for (i = 0; i < ARRAY_SIZE(ads868x_range_def); i++)
>>> +        len += sprintf(buf + len, "%d ", ads868x_range_def[i]);
>>> +
>>> +    len += sprintf(buf + len, "\n");
>> These values don't look like scales in appropriate units?  They
>> look to be perhaps register values.
>>
>> Scale should be such that (raw_value + offset)*scale = real value
>> as per the ABI in Documentation/ABI/testing/sysfs-bus-iio
>>
>>> +
>>> +    return len;
>>> +}
>>> +
>>> +static IIO_DEVICE_ATTR(in_voltage_scale_available, S_IRUGO,
>>> +               ads868x_show_scales, NULL, 0);
>>> +
>>> +static struct attribute *ads868x_attributes[] = {
>>> +    &iio_dev_attr_in_voltage_scale_available.dev_attr.attr,
>>> +    NULL,
>>> +};
>>> +
>>> +static const struct attribute_group ads868x_attribute_group = {
>>> +    .attrs = ads868x_attributes,
>>> +};
>>> +
>>> +#define ADS868X_CHAN(index)                    \
>>> +{                                \
>>> +    .type = IIO_VOLTAGE,                    \
>>> +    .indexed = 1,                        \
>>> +    .channel = index,                    \
>>> +    .info_mask_separate = BIT(IIO_CHAN_INFO_RAW)        \
>>> +                  | BIT(IIO_CHAN_INFO_SCALE),    \
>> If address == channel then no point in setting it, use channel instead.
>>> +    .address = index,                    \
>> No buffer support as yet so no need for everything below here.
>>
>>> +    .scan_index = index,                    \
>>> +    .scan_type = {                        \
>>> +        .sign = 'u',                    \
>>> +        .realbits = 16,                    \
>>> +        .storagebits = 16,                \
>>> +        .endianness = IIO_BE,                \
>>> +    },                            \
>>> +}
>>> +
>>> +static const struct iio_chan_spec ads8684_channels[] = {
>>> +    ADS868X_CHAN(0),
>>> +    ADS868X_CHAN(1),
>>> +    ADS868X_CHAN(2),
>>> +    ADS868X_CHAN(3),
>>> +};
>>> +
>>> +static const struct iio_chan_spec ads8688_channels[] = {
>>> +    ADS868X_CHAN(0),
>>> +    ADS868X_CHAN(1),
>>> +    ADS868X_CHAN(2),
>>> +    ADS868X_CHAN(3),
>>> +    ADS868X_CHAN(4),
>>> +    ADS868X_CHAN(5),
>>> +    ADS868X_CHAN(6),
>>> +    ADS868X_CHAN(7),
>>> +};
>>> +
>>> +static int ads868x_prog_write(struct iio_dev *indio_dev, unsigned int addr,
>>> +                  unsigned int val)
>>> +{
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +    unsigned int tmp;
>>> +
>>> +    tmp = ADS868X_PROG_REG(addr) | ADS868X_PROG_WR_BIT | val;
>>> +    tmp <<= ADS868X_PROG_DONT_CARE_BITS;
>>> +    st->data[0].d32 = cpu_to_be32(tmp);
>>> +
>>> +    return spi_write(st->spi, &st->data[0].d8[1], 3);
>>> +}
>>> +
>>> +static int ads868x_reset(struct iio_dev *indio_dev)
>>> +{
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +    unsigned int tmp;
>>> +
>>> +    tmp = ADS868X_CMD_REG(ADS868X_CMD_REG_RST);
>>> +    tmp <<= ADS868X_CMD_DONT_CARE_BITS;
>>> +    st->data[0].d32 = cpu_to_be32(tmp);
>>> +
>>> +    return spi_write(st->spi, &st->data[0].d8[0], 4);
>>> +}
>>> +
>>> +static int ads868x_read(struct iio_dev *indio_dev, unsigned int chan)
>>> +{
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +    int ret;
>>> +    unsigned int tmp;
>>> +    struct spi_transfer t[] = {
>>> +        {
>>> +            .tx_buf = &st->data[0].d8[0],
>>> +            .len = 4,
>>> +            .cs_change = 1,
>>> +        }, {
>>> +            .tx_buf = &st->data[1].d8[0],
>>> +            .rx_buf = &st->data[1].d8[0],
>>> +            .len = 4,
>>> +        },
>>> +    };
>>> +
>>> +    tmp = ADS868X_CMD_REG(ADS868X_CMD_REG_MAN_CH(chan));
>>> +    tmp <<= ADS868X_CMD_DONT_CARE_BITS;
>>> +    st->data[0].d32 = cpu_to_be32(tmp);
>>> +
>>> +    tmp = ADS868X_CMD_REG(ADS868X_CMD_REG_NOOP);
>>> +    tmp <<= ADS868X_CMD_DONT_CARE_BITS;
>>> +    st->data[1].d32 = cpu_to_be32(tmp);
>>> +
>>> +    ret = spi_sync_transfer(st->spi, t, ARRAY_SIZE(t));
>> Fiddly!  Ah well.
>>> +    if (ret < 0)
>>> +        return ret;
>>> +
>>> +    return be32_to_cpu(st->data[1].d32) & 0xffff;
>>> +}
>>> +
>>> +static int ads868x_read_raw(struct iio_dev *indio_dev,
>>> +                struct iio_chan_spec const *chan,
>>> +                int *val, int *val2, long m)
>>> +{
>>> +    int ret;
>>> +    unsigned long scale_mv;
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +
>>> +    switch (m) {
>>> +    case IIO_CHAN_INFO_RAW:
>>> +        mutex_lock(&indio_dev->mlock);
>>> +        ret = ads868x_read(indio_dev, chan->channel);
>>> +        mutex_unlock(&indio_dev->mlock);
>>> +        if (ret < 0)
>>> +            return ret;
>>> +        *val = ret;
>>> +        return IIO_VAL_INT;
>>> +    case IIO_CHAN_INFO_SCALE:
>>> +        scale_mv = st->vref_mv;
>>> +        switch (st->range[chan->channel]) {
>>> +        case ADS868X_PLUSMINUS25VREF:
>>> +            scale_mv *= 5;
>>> +            break;
>>> +        case ADS868X_PLUSMINUS125VREF:
>>> +        case ADS868X_PLUS25VREF:
>>> +            scale_mv = scale_mv * 25 / 10;
>>> +            break;
>>> +        case ADS868X_PLUSMINUS0625VREF:
>>> +        case ADS868X_PLUS125VREF:
>>> +            scale_mv = scale_mv * 125 / 100;
>>> +            break;
>>> +        default:
>>> +            return -EINVAL;
>>> +        }
>>> +        *val = scale_mv;
>>> +        *val2 = chan->scan_type.realbits;
>>> +        return IIO_VAL_FRACTIONAL_LOG2;
>>> +    }
>>> +
>>> +    return -EINVAL;
>>> +}
>>> +
>>> +static int ads868x_write_raw(struct iio_dev *indio_dev,
>>> +            struct iio_chan_spec const *chan,
>>> +            int val, int val2, long mask)
>>> +{
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +    unsigned int tmp;
>>> +    int ret, i;
>>> +
>>> +    switch (mask) {
>>> +    case IIO_CHAN_INFO_SCALE:
>>> +        ret = -EINVAL;
>>> +        for (i = 0; i < ARRAY_SIZE(ads868x_range_def); i++)
>>> +            if (val != ads868x_range_def[i])
>>> +                continue;
>> This logic is a bit flipped around to the more conventional loop
>> and break with the error set by checking if we reached the end
>> of the search range or not.  I think I'd slightly prefer the more
>> conventional option.  Personally I prefer to see error cases
>> as the 'exception' rather than set on the more natural flow through
>> the code.
>>
>>> +
>>> +            ret = 0;
>>> +            st->range[chan->channel] = val;
>>> +            mutex_lock(&indio_dev->mlock);
>>> +            tmp = ADS868X_PROG_REG_RANGE_CH(chan->channel);
>>> +            ads868x_prog_write(indio_dev, tmp, val);
>> check for write errors.  Also the cached value should be set after
>> the write has succeeded not before.
>>
>>> +            mutex_unlock(&indio_dev->mlock);
>> mlock is really intended as a lock on the device operating mode, but
>> I suppose using it for other things in a driver that doesn't
>> handle device state changes (no buffer support) is fine if not
>> as obvious as a local lock might be.
>>
>>> +        break;
>>> +    default:
>>> +        ret = -EINVAL;
>>> +    }
>>> +
>>> +    return ret;
>>> +}
>>> +
>>> +static const struct iio_info ads868x_info = {
>>> +    .read_raw = &ads868x_read_raw,
>>> +    .write_raw = &ads868x_write_raw,
>>> +    .attrs = &ads868x_attribute_group,
>>> +    .driver_module = THIS_MODULE,
>>> +};
>>> +
>>> +static const struct ads868x_chip_info ads868x_chip_info_tbl[] = {
>>> +    [ID_ADS8684] = {
>>> +        .channels = ads8684_channels,
>>> +        .num_channels = ARRAY_SIZE(ads8684_channels),
>>> +        .iio_info = &ads868x_info,
>>> +    },
>>> +    [ID_ADS8688] = {
>>> +        .channels = ads8688_channels,
>>> +        .num_channels = ARRAY_SIZE(ads8688_channels),
>>> +        .iio_info = &ads868x_info,
>>> +    },
>>> +};
>>> +
>>> +static int ads868x_probe(struct spi_device *spi)
>>> +{
>>> +    struct ads868x_state *st;
>>> +    struct iio_dev *indio_dev;
>>> +    bool ext_ref;
>>> +    enum ads868x_id drv_id;
>> I'm unconvinced this local drv_id adds anything in terms of readability.
>> If it's just here to avoid overly long lines, note the checkpatch warning
>> for that should be taken as a recomendation only!
>>
>>> +    int ret;
>>> +
>>> +    indio_dev = devm_iio_device_alloc(&spi->dev, sizeof(*st));
>>> +    if (indio_dev == NULL)
>>> +        return -ENOMEM;
>>> +
>>> +    st = iio_priv(indio_dev);
>>> +
>>> +    if (spi->dev.of_node)
>>> +        ext_ref = of_property_read_bool(spi->dev.of_node,
>>> +                        "vref-supply");
>>> +    else
>>> +        ext_ref = false;
>>> +
>>> +    if (ext_ref) {
>>> +        st->reg = devm_regulator_get(&spi->dev, "vref");
>>> +        if (!IS_ERR_OR_NULL(st->reg)) {
>> As far as I can see from a quick look at get_regulator docs
>> (driver/regulator/core.c) which is called from the devm call
>> the null condition cannot occur.  Hence IS_ERR is probably
>> more appropriate (and more commonly used).
>>
>>> +            ret = regulator_enable(st->reg);
>>> +            if (ret)
>>> +                return ret;
>>> +
>>> +            ret = regulator_get_voltage(st->reg);
>>> +            if (ret < 0)
>>> +                goto error_out;
>>> +        st->vref_mv = ret / 1000;
>>> +        }
>>> +    } else {
>>> +        /* Use internal reference */
>>> +        st->vref_mv = ADS868X_VREF_MV;
>>> +    }
>>> +
>>> +    drv_id = spi_get_device_id(spi)->driver_data;
>>> +    st->chip_info =    &ads868x_chip_info_tbl[drv_id];
>>> +
>>> +    spi->mode = SPI_MODE_1;
>>> +
>>> +    spi_set_drvdata(spi, indio_dev);
>>> +
>>> +    st->spi = spi;
>>> +
>>> +    indio_dev->name = spi_get_device_id(spi)->name;
>>> +    indio_dev->dev.parent = &spi->dev;
>>> +    indio_dev->modes = INDIO_DIRECT_MODE;
>>> +    indio_dev->channels = st->chip_info->channels;
>>> +    indio_dev->num_channels = st->chip_info->num_channels;
>>> +    indio_dev->info = st->chip_info->iio_info;
>>> +
>>> +    /* Reset ADS868x */
>>> +    mutex_lock(&indio_dev->mlock);
>>> +    ads868x_reset(indio_dev);
>>> +    mutex_unlock(&indio_dev->mlock);
>>> +
>>> +    ret = iio_device_register(indio_dev);
>>> +    if (ret)
>>> +        goto error_out;
>>> +
>>> +    return 0;
>>> +
>>> +error_out:
>>> +    if (!IS_ERR_OR_NULL(st->reg))
>>> +        regulator_disable(st->reg);
>>> +
>>> +    return ret;
>>> +}
>>> +
>>> +static int ads868x_remove(struct spi_device *spi)
>>> +{
>>> +    struct iio_dev *indio_dev = spi_get_drvdata(spi);
>>> +    struct ads868x_state *st = iio_priv(indio_dev);
>>> +
>>> +    iio_device_unregister(indio_dev);
>>> +
>>> +    if (!IS_ERR_OR_NULL(st->reg))
>>> +        regulator_disable(st->reg);
>>> +
>>> +    return 0;
>>> +}
>>> +
>>> +static const struct spi_device_id ads868x_id[] = {
>>> +    {"ads8684", ID_ADS8684},
>>> +    {"ads8688", ID_ADS8688},
>>> +    {}
>>> +};
>>> +MODULE_DEVICE_TABLE(spi, ads868x_id);
>>> +
>>> +static const struct of_device_id ads868x_of_match[] = {
>>> +    { .compatible = "ti,ads8684" },
>>> +    { .compatible = "ti,ads8688" },
>>> +    { }
>>> +};
>>> +MODULE_DEVICE_TABLE(of, ads868x_of_match);
>>> +
>>> +static struct spi_driver ads868x_driver = {
>>> +    .driver = {
>>> +        .name    = "ads868x",
>>> +        .owner    = THIS_MODULE,
>>> +    },
>>> +    .probe        = ads868x_probe,
>>> +    .remove        = ads868x_remove,
>>> +    .id_table    = ads868x_id,
>>> +};
>>> +module_spi_driver(ads868x_driver);
>>> +
>>> +MODULE_AUTHOR("Sean Nyekjaer <sean.nyekjaer@prevas.dk>");
>>> +MODULE_DESCRIPTION("Texas Instruments ADS868x driver");
>>> +MODULE_LICENSE("GPL v2");
>>>
> 
> -- 
> To unsubscribe from this list: send the line "unsubscribe linux-iio" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html