Re: [PATCH] iio: add support for Analog Devices ad7194 a/d converter

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Aside from the ABI stuff, I had a couple responses to your comments.

thanks,
Paul

>>>
>>> Signed-off-by: Paul Thomas <pthomas8589@xxxxxxxxx>
>>> ---
>>>  drivers/staging/iio/adc/Kconfig  |    7 +
>>>  drivers/staging/iio/adc/Makefile |    1 +
>>>  drivers/staging/iio/adc/ad7194.c |  462 ++++++++++++++++++++++++++++++++++++++
>>>  3 files changed, 470 insertions(+), 0 deletions(-)
>>>  create mode 100644 drivers/staging/iio/adc/ad7194.c
>>>
>>> diff --git a/drivers/staging/iio/adc/Kconfig b/drivers/staging/iio/adc/Kconfig
>>> index 8c751c4..871605b 100644
>>> --- a/drivers/staging/iio/adc/Kconfig
>>> +++ b/drivers/staging/iio/adc/Kconfig
>>> @@ -17,6 +17,13 @@ config AD7152
>>>        Say yes here to build support for Analog Devices capacitive sensors.
>>>        (ad7152, ad7153) Provides direct access via sysfs.
>>>
>>> +config AD7194
>>> +    tristate "Analog Devices AD7194 ADC driver"
>>> +    depends on SPI
>>> +    help
>>> +      Say yes here to build support for Analog Devices ad7194.
>>> +      Provides direct access via sysfs.
>>> +
>>>  config AD7291
>>>      tristate "Analog Devices AD7291 temperature sensor driver"
>>>      depends on I2C
>>> diff --git a/drivers/staging/iio/adc/Makefile b/drivers/staging/iio/adc/Makefile
>>> index 1d9b3f5..4da3c40 100644
>>> --- a/drivers/staging/iio/adc/Makefile
>>> +++ b/drivers/staging/iio/adc/Makefile
>>> @@ -31,6 +31,7 @@ obj-$(CONFIG_AD7298) += ad7298.o
>>>
>>>  obj-$(CONFIG_AD7150) += ad7150.o
>>>  obj-$(CONFIG_AD7152) += ad7152.o
>>> +obj-$(CONFIG_AD7194) += ad7194.o
>>>  obj-$(CONFIG_AD7291) += ad7291.o
>>>  obj-$(CONFIG_AD7314) += ad7314.o
>>>  obj-$(CONFIG_AD7745) += ad7745.o
>>> diff --git a/drivers/staging/iio/adc/ad7194.c b/drivers/staging/iio/adc/ad7194.c
>>> new file mode 100644
>>> index 0000000..a867662
>>> --- /dev/null
>>> +++ b/drivers/staging/iio/adc/ad7194.c
>>> @@ -0,0 +1,462 @@
>>> +/*
>>> + *  ad7194 - driver for Analog Devices AD7194 A/D converter
>>> + *
>>> + *  Copyright (c) 2011 Paul Thomas <pthomas8589@xxxxxxxxx>
>>> + *
>>> + *  This program is distributed in the hope that it will be useful,
>>> + *  but WITHOUT ANY WARRANTY; without even the implied warranty of
>>> + *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
>>> + *  GNU General Public License for more details.
>>> + *
>>> + *  This program is free software; you can redistribute it and/or modify
>>> + *  it under the terms of the GNU General Public License version 2 or
>>> + *  later as publishhed by the Free Software Foundation.
>>> + *
>>> + *  You need to have something like this in struct spi_board_info
>>> + *  {
>>> + *          .modalias       = "ad7194",
>>> + *          .max_speed_hz   = 2*1000*1000,
>>> + *          .chip_select    = 0,
>>> + *          .bus_num        = 1,
>>> + *  },
>>> + *
>>> + * Three sysfs files are used chX_value, chX_diff and chX_gain
>>> + * chX_value is read only and returns that channels value in volts.
>>> + * chX_gain is read/write, the chX_value is scaled by the gain so it retains
>>> + * it's proper units.
>>> + * chX_diff is read/write, if chX_diff is 0 then pseude mod is used,
>>> + * if chX_diff is 1 then differential mode is used.
>>> + * In this mode ch1_value selects ch1-ch2, and ch2_value selects ch2-ch1
>>> + */
>>> +
>>> +#define DEBUG 1
>>> +
>>> +/*From Table 15 in the datasheet*/
>>> +/*Register addresses*/
>>> +#define REG_STATUS  0
>>> +#define REG_MODE    1
>>> +#define REG_CONF    2
>>> +#define REG_DATA    3
>>> +#define REG_ID      4
>>> +#define REG_GPOCON  5
>>> +#define REG_OFFSET  6
>>> +#define REG_FS      7
>>> +
>>> +/*From Table 15 in the datasheet*/
>>> +#define COMM_ADDR_bp 3
>>> +#define COMM_READ_bm (1 << 6)
>>> +
>>> +#define CONF_CHOP_bm (1 << 23)
>>> +#define CONF_PSEUDO_bm (1 << 18)
>>> +#define CONF_BUF_bm (1 << 4)
>>> +#define CONF_CHAN_NEG_bp 8
>>> +#define CONF_CHAN_POS_bp 12
>>> +
>>> +
>>> +#define MODE_MD_SINGLE_gc (0x01 << 21)
>>> +#define MODE_MD_ZS_CAL_gc (0x04 << 21)
>>> +#define MODE_MD_FS_CAL_gc (0x05 << 21)
>>> +#define MODE_CLK_INTTRI_gc (0x02 << 18)
>>> +/*Table 8 in the datasheet provides options for the Filter Word*/
>>> +#define MODE_FILTER_WORD 1
>>> +#define SETTLE_MS 2
>>> +
>>> +#include <linux/device.h>
>>> +#include <linux/module.h>
>>> +#include <linux/init.h>
>>> +#include <linux/spi/spi.h>
>>> +#include <linux/err.h>
>>> +#include <linux/mutex.h>
>>> +#include <linux/delay.h>
>>> +
>>> +#include "../iio.h"
>>> +#include "../sysfs.h"
>>> +
>>> +#define DEVICE_NAME "ad7194"
>>> +#define NUM_CHANNELS 16
>>> +
> This wins the odd award.  The available gains seem unlikely to be tied to the
> number of channels.
Right! what was I thinking?

>>> +const uint8_t gains[NUM_CHANNELS] = {1, 0, 0, 8, 16, 32, 64, 128};
>>> +
>>> +struct ad7194_data {
>>> +    struct spi_device *spi_dev;
>>> +    struct iio_dev *indio_dev;
>>> +    uint8_t gain[NUM_CHANNELS];
>>> +    uint8_t diff[NUM_CHANNELS];
>>> +    struct mutex lock;
>>> +};
>>> +
>>> +static ssize_t show_gain(struct device *dev,
>>> +            struct device_attribute *attr, char *buf)
>>> +{
>>> +    struct iio_dev *dev_info = dev_get_drvdata(dev);
>>> +    struct ad7194_data *pdata = dev_info->dev_data;
>>> +    struct iio_dev_attr *this_attr = to_iio_dev_attr(attr);
>>> +
>>> +    return sprintf(buf, "%d\n", gains[pdata->gain[this_attr->address]]);
>>> +}
>>> +
>>> +static ssize_t set_gain(struct device *dev,
>>> +            struct device_attribute *attr, const char *buf, size_t count)
>>> +{
>>> +    uint8_t gain_real, i;
>>> +    struct iio_dev *dev_info = dev_get_drvdata(dev);
>>> +    struct ad7194_data *pdata = dev_info->dev_data;
>>> +    struct iio_dev_attr *this_attr = to_iio_dev_attr(attr);
>>> +
>>> +    gain_real = simple_strtol(buf, NULL, 10);
>>> +    if (gain_real == 0)
>>> +            return -EPERM;
>>> +    for (i = 0; i < NUM_CHANNELS; i++) {
>>> +            if (gains[i] == gain_real) {
>>> +                    pdata->gain[this_attr->address] =  i;
>>> +                    return count;
>>> +            }
>>> +    }
>>> +    return -EPERM;
>>> +}
>>> +
>>> +static ssize_t show_diff(struct device *dev,
>>> +            struct device_attribute *attr, char *buf)
>>> +{
>>> +    struct iio_dev *dev_info = dev_get_drvdata(dev);
>>> +    struct ad7194_data *pdata = dev_info->dev_data;
>>> +    struct iio_dev_attr *this_attr = to_iio_dev_attr(attr);
>>> +
>>> +    return sprintf(buf, "%d\n", pdata->diff[this_attr->address]);
>>> +}
>>> +
>>> +static ssize_t set_diff(struct device *dev,
>>> +            struct device_attribute *attr, const char *buf, size_t count)
>>> +{
>>> +    uint8_t diff;
>>> +    struct iio_dev *dev_info = dev_get_drvdata(dev);
>>> +    struct ad7194_data *pdata = dev_info->dev_data;
>>> +    struct iio_dev_attr *this_attr = to_iio_dev_attr(attr);
>>> +
>>> +    diff = simple_strtol(buf, NULL, 10);
> strtobool and set value to the resulting bool.
>>> +    if (diff)
>>> +            pdata->diff[this_attr->address] = 1;
>>> +    else
>>> +            pdata->diff[this_attr->address] = 0;
>>> +    return count;
>>> +}
>>> +
>>> +static inline int ad7194_read_reg8(struct spi_device *spi,
>>> +            int reg, uint8_t *val)
>>> +{
>>> +    int ret;
>>> +    uint8_t tx[1], rx[1];
>>> +
>>> +    tx[0] = (COMM_READ_bm | (reg << COMM_ADDR_bp));
>>> +
>>> +    ret = spi_write_then_read(spi, tx, 1, rx, 1);
>>> +    *val = rx[0];
>>> +    return ret;
>>> +}
>>> +
>>> +static inline int ad7194_read_reg24(struct spi_device *spi,
>>> +            int reg, uint32_t *val)
>>> +{
>>> +    int ret;
>>> +    uint8_t tx[1], rx[3];
>>> +
>>> +    tx[0] = (COMM_READ_bm | (reg << COMM_ADDR_bp));
> Don't allocate single element arrays.  Just makes things
> harder to read.
I was trying to keep consistency between the _reg8 & _reg24 versions
as well as between the rx & tx buffers. It makes the
spi_write_then_read(spi, tx, 1, rx, 3) read nicely, but if that's a
no-no then I can change it.

>
>>> +
>>> +    ret = spi_write_then_read(spi, tx, 1, rx, 3);
>>> +    *val = (rx[0] << 16) + (rx[1] << 8) + rx[2];
>>> +    return ret;
>>> +}
>>> +
>>> +static inline int ad7194_write_reg24(struct spi_device *spi,
>>> +            int reg, uint32_t *val)
>>> +{
>>> +    uint8_t tx[4];
>>> +
>>> +    tx[0] = (reg << COMM_ADDR_bp);
>>> +    tx[1] = (*val >> 16) & 0xff;
>>> +    tx[2] = (*val >> 8) & 0xff;
>>> +    tx[3] = (*val >> 0) & 0xff;
>>> +
>>> +    return spi_write(spi, tx, 4);
> IIRC spi_write requires dma safe buffers (cache line aligned
> buffers on the heap).
Could you expand here?

>>> +}
>>> +
>>> +static ssize_t show_voltage(struct device *dev,
>>> +            struct device_attribute *attr, char *buf)
>>> +{
>>> +    uint32_t conf, mode, val, sign, pos_chan, neg_chan = 0;
>>> +    uint8_t status, gain, diff;
>>> +    int32_t whole, fract;
>>> +    int ret;
>>> +
>>> +    struct iio_dev *dev_info = dev_get_drvdata(dev);
>>> +    struct ad7194_data *pdata = dev_info->dev_data;
> Pdata is a bad name choice.  THat implies platform data (by conventional
> use).  Here it is instance specific data I think? state is a typical
> variable name, or chip.
>
>>> +    struct spi_device *spi = pdata->spi_dev;
>>> +    struct iio_dev_attr *this_attr = to_iio_dev_attr(attr);
>>> +
>>> +    pos_chan = this_attr->address;
>>> +    gain = pdata->gain[this_attr->address];
>>> +    diff = pdata->diff[this_attr->address];
> Just general point. Don't bother with local variables for stuff that
> is only used once.  Just makes reading harder.
>>> +
>>> +    if (diff == 1) {
>>> +            /*if in0_diff is true, reading in0_input still returns
>>> +            * in0, but it is in0-in1, if you read in1_input
>>> +            * then you get in1-in0 */
> In a nutshell that explains why we don't use the interface you've gone
> with but have an explicit one for differential channels. (see the max1363
> driver for examples).
>
>>> +            if ((pos_chan % 2) == 0)
>>> +                    neg_chan = pos_chan+1;
>>> +            else
>>> +                    neg_chan = pos_chan-1;
>>> +
>>> +            conf = CONF_CHOP_bm | CONF_BUF_bm |
>>> +                    (pos_chan << CONF_CHAN_POS_bp) |
>>> +                    (neg_chan << CONF_CHAN_NEG_bp) | gain;
>>> +    } else {
>>> +            conf = CONF_CHOP_bm | CONF_PSEUDO_bm | CONF_BUF_bm |
>>> +                    (pos_chan << CONF_CHAN_POS_bp) | gain;
>>> +    }
>>> +
>>> +    ret = mutex_lock_interruptible(&pdata->lock);
>>> +    if (ret != 0)
>>> +            return ret;
>>> +
>>> +    ret = ad7194_write_reg24(spi, REG_CONF, &conf);
>>> +    if (ret != 0)
>>> +            goto out;
>>> +
>>> +    mode = MODE_MD_SINGLE_gc | MODE_CLK_INTTRI_gc | MODE_FILTER_WORD;
>>> +    ret = ad7194_write_reg24(spi, REG_MODE, &mode);
>>> +    if (ret != 0)
>>> +            goto out;
>>> +
>>> +    msleep_interruptible(SETTLE_MS);
>>> +
>>> +    ret = ad7194_read_reg8(spi, REG_STATUS, &status);
>>> +    if (ret != 0)
>>> +            goto out;
>>> +    status = (status >> 6) & 0x01;
>>> +
>>> +    ret = ad7194_read_reg24(spi, REG_DATA, &val);
>>> +    if (ret != 0)
>>> +            goto out;
>>> +
>>> +    sign = (val & 0x800000) >> 23;
>>> +    if (sign)
>>> +            fract = (val & 0x7fffff);
>>> +    else
>>> +            fract = 0x7fffff - (val & 0x7fffff);
>>> +    fract = ((int64_t)fract*4095000) >> 23;
>>> +    fract = fract / gains[gain];
>>> +    whole = fract / 1000000;
>>> +    fract = fract % 1000000;
> As a general rule, we'd push this into userspace, but the interface
> allows for either so we can keep this as is if you really want to.
> I'd like to see a little comment explaining what the calcuation is
> though!
>>> +
>>> +    if (status == 0) {
>>> +            mutex_unlock(&pdata->lock);
>>> +            if (sign)
>>> +                    return sprintf(buf, "%d.%.6d\n", whole, fract);
>>> +            else
>>> +                    return sprintf(buf, "-%d.%.6d\n", whole, fract);
>>> +    } else {
>>> +            ret = -EAGAIN;
>>> +            goto out;
>>> +    }
>>> +
>>> +out:
>>> +    mutex_unlock(&pdata->lock);
>>> +    return ret;
>>> +}
>>> +
>
>
> These don't even vaguely conform to the abi. See
> drivers/staging/iio/Documentation/sysfs-bus-iio.  Mostly this will
> get cleaned up in converting the driver over to chan_spec based
> registration.  Hehe, this reminds me why we introduced the chan spec
> stuff in the first place!
>>> +static IIO_DEVICE_ATTR(ch1_value, S_IRUGO, show_voltage, NULL, 0);
>>> +static IIO_DEVICE_ATTR(ch2_value, S_IRUGO, show_voltage, NULL, 1);
>>> +static IIO_DEVICE_ATTR(ch3_value, S_IRUGO, show_voltage, NULL, 2);
>>> +static IIO_DEVICE_ATTR(ch4_value, S_IRUGO, show_voltage, NULL, 3);
>>> +static IIO_DEVICE_ATTR(ch5_value, S_IRUGO, show_voltage, NULL, 4);
>>> +static IIO_DEVICE_ATTR(ch6_value, S_IRUGO, show_voltage, NULL, 5);
>>> +static IIO_DEVICE_ATTR(ch7_value, S_IRUGO, show_voltage, NULL, 6);
>>> +static IIO_DEVICE_ATTR(ch8_value, S_IRUGO, show_voltage, NULL, 7);
>>> +static IIO_DEVICE_ATTR(ch9_value, S_IRUGO, show_voltage, NULL, 8);
>>> +static IIO_DEVICE_ATTR(ch10_value, S_IRUGO, show_voltage, NULL, 9);
>>> +static IIO_DEVICE_ATTR(ch11_value, S_IRUGO, show_voltage, NULL, 10);
>>> +static IIO_DEVICE_ATTR(ch12_value, S_IRUGO, show_voltage, NULL, 11);
>>> +static IIO_DEVICE_ATTR(ch13_value, S_IRUGO, show_voltage, NULL, 12);
>>> +static IIO_DEVICE_ATTR(ch14_value, S_IRUGO, show_voltage, NULL, 13);
>>> +static IIO_DEVICE_ATTR(ch15_value, S_IRUGO, show_voltage, NULL, 14);
>>> +static IIO_DEVICE_ATTR(ch16_value, S_IRUGO, show_voltage, NULL, 15);
>>> +
>>> +static IIO_DEVICE_ATTR(ch1_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 0);
>>> +static IIO_DEVICE_ATTR(ch2_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 1);
>>> +static IIO_DEVICE_ATTR(ch3_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 2);
>>> +static IIO_DEVICE_ATTR(ch4_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 3);
>>> +static IIO_DEVICE_ATTR(ch5_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 4);
>>> +static IIO_DEVICE_ATTR(ch6_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 5);
>>> +static IIO_DEVICE_ATTR(ch7_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 6);
>>> +static IIO_DEVICE_ATTR(ch8_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 7);
>>> +static IIO_DEVICE_ATTR(ch9_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 8);
>>> +static IIO_DEVICE_ATTR(ch10_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 9);
>>> +static IIO_DEVICE_ATTR(ch11_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 10);
>>> +static IIO_DEVICE_ATTR(ch12_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 11);
>>> +static IIO_DEVICE_ATTR(ch13_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 12);
>>> +static IIO_DEVICE_ATTR(ch14_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 13);
>>> +static IIO_DEVICE_ATTR(ch15_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 14);
>>> +static IIO_DEVICE_ATTR(ch16_gain, S_IWUSR | S_IRUGO, show_gain, set_gain, 15);
>>> +
>
> Umm.. What are these?  I think based on quick look at the datasheet that you
> are using them to switch the _value attributes from unipolar to differential?
> Please register them as separate channels (see max1363 for lots and lots of
> examples).  It can be a bit fiddly to maintain the internal state but it does
> give us a coherent general interface.
>>> +static IIO_DEVICE_ATTR(ch1_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 0);
>>> +static IIO_DEVICE_ATTR(ch2_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 1);
>>> +static IIO_DEVICE_ATTR(ch3_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 2);
>>> +static IIO_DEVICE_ATTR(ch4_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 3);
>>> +static IIO_DEVICE_ATTR(ch5_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 4);
>>> +static IIO_DEVICE_ATTR(ch6_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 5);
>>> +static IIO_DEVICE_ATTR(ch7_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 6);
>>> +static IIO_DEVICE_ATTR(ch8_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 7);
>>> +static IIO_DEVICE_ATTR(ch9_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 8);
>>> +static IIO_DEVICE_ATTR(ch10_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 9);
>>> +static IIO_DEVICE_ATTR(ch11_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 10);
>>> +static IIO_DEVICE_ATTR(ch12_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 11);
>>> +static IIO_DEVICE_ATTR(ch13_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 12);
>>> +static IIO_DEVICE_ATTR(ch14_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 13);
>>> +static IIO_DEVICE_ATTR(ch15_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 14);
>>> +static IIO_DEVICE_ATTR(ch16_diff, S_IWUSR | S_IRUGO, show_diff, set_diff, 15);
>>> +
>>> +static struct attribute *ad7194_attributes[] = {
>>> +    &iio_dev_attr_ch1_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch2_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch3_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch4_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch5_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch6_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch7_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch8_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch9_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch10_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch11_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch12_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch13_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch14_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch15_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch16_value.dev_attr.attr,
>>> +    &iio_dev_attr_ch1_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch2_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch3_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch4_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch5_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch6_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch7_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch8_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch9_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch10_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch11_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch12_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch13_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch14_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch15_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch16_gain.dev_attr.attr,
>>> +    &iio_dev_attr_ch1_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch2_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch3_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch4_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch5_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch6_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch7_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch8_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch9_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch10_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch11_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch12_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch13_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch14_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch15_diff.dev_attr.attr,
>>> +    &iio_dev_attr_ch16_diff.dev_attr.attr,
>>> +    NULL,
>>> +};
>>> +
>>> +static const struct attribute_group ad7194_attribute_group = {
>>> +    .attrs = ad7194_attributes,
>>> +};
>>> +
>>> +static const struct iio_info ad7194_info = {
>>> +    .attrs = &ad7194_attribute_group,
>>> +    .driver_module = THIS_MODULE,
>>> +};
>>> +
>>> +static int __devinit ad7194_probe(struct spi_device *spi)
>>> +{
>>> +    int ret, err;
>>> +    struct ad7194_data *pdata;
>>> +
>>> +    /* Configure the SPI bus */
>>> +    spi->mode = (SPI_MODE_0);
>>> +    spi->bits_per_word = 8;
>>> +    spi_setup(spi);
>>> +
>>> +    pdata = kzalloc(sizeof(struct ad7194_data), GFP_KERNEL);
>>> +    if (!pdata) {
>>> +            err = -ENOMEM;
>>> +            goto exit;
>>> +    }
>>> +
>>> +    dev_set_drvdata(&spi->dev, pdata);
>>> +
>>> +    pdata->spi_dev = spi;
>>> +
>>> +    pdata->indio_dev = iio_allocate_device(0);
>>> +    if (pdata->indio_dev == NULL) {
>>> +            ret = -ENOMEM;
>>> +            goto error_free;
>>> +    }
>>> +
>>> +    mutex_init(&pdata->lock);
>>> +    pdata->indio_dev->name = "ad7194";
>>> +    pdata->indio_dev->dev.parent = &spi->dev;
>>> +    pdata->indio_dev->info = &ad7194_info;
>>> +    pdata->indio_dev->dev_data = (void *)pdata;
> Devdata is going away I'm afraid (only still there in
> 1 driver and that's because Jon is having issues getting
> the kernel up and running on his board to check I haven't
> messed up the conversion!)  Please do this with
> iio_allocate_device(sizeof(*pdata)), then use iio_priv
> to get to the resulting private data.
>
>>> +
>>> +    ret = iio_device_register(pdata->indio_dev);
>>> +    if (ret)
>>> +            goto error_free_dev;
>>> +
>>> +    return 0;
>>> +
>>> +error_free_dev:
>>> +    iio_free_device(pdata->indio_dev);
>>> +error_free:
>>> +    kfree(pdata);
>>> +exit:
>>> +    return err;
>>> +}
>>> +
>>> +static int __devexit ad7194_remove(struct spi_device *spi)
>>> +{
>>> +    struct ad7194_data *pdata = dev_get_drvdata(&spi->dev);
>>> +
>>> +    dev_set_drvdata(&spi->dev, NULL);
>>> +    iio_device_unregister(pdata->indio_dev);
>>> +    iio_free_device(pdata->indio_dev);
>>> +    kfree(pdata);
>>> +    return 0;
>>> +}
>>> +
>>> +static struct spi_driver ad7194_driver = {
>>> +    .driver = {
> I've never understood why people like having a define
> for the part name.  Much better to just put it here
> where everyone expects to find it!
>
>>> +            .name = DEVICE_NAME,
>>> +            .bus = &spi_bus_type,
>>> +            .owner = THIS_MODULE,
>>> +    },
>>> +
>>> +    .probe = ad7194_probe,
>>> +    .remove = __devexit_p(ad7194_remove),
>>> +};
>>> +
>>> +static int __init ad7194_init(void)
>>> +{
>>> +    return spi_register_driver(&ad7194_driver);
>>> +}
>>> +
>>> +static void __exit ad7194_exit(void)
>>> +{
>>> +    spi_unregister_driver(&ad7194_driver);
>>> +}
>>> +
>>> +module_init(ad7194_init);
>>> +module_exit(ad7194_exit);
>>> +
>>> +MODULE_AUTHOR("Paul Thomas <pthomas8589@xxxxxxxxx>");
>>> +MODULE_DESCRIPTION("TI AD7194 A/D driver");
>>> +MODULE_LICENSE("GPL");
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-iio" in
>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-iio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[Index of Archives]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Input]     [Linux Kernel]     [Linux SCSI]     [X.org]

  Powered by Linux