Re: [PATCH] iio: adc: meson-saradc: add calibration for meson gxbb

[Heiner Kallweit <hkallweit1@gmail.com>]

Am 04.03.2017 um 00:53 schrieb Martin Blumenstingl:
> Hi Heiner,
> 
> thanks for this patch!
> 
> On Fri, Mar 3, 2017 at 10:13 PM, Heiner Kallweit <hkallweit1@gmail.com> wrote:
>> The calibration on Meson GXBB is different from the one on GXM/GXL.
> I guess you are referring to the "12bit calibration" code from u-boot
> [0] which uses MESON_SAR_ADC_REG13_12BIT_CALIBRATION_MASK.

Indeed ..

> However, Amlogic's Linux driver uses the same "calibration logic" for
> anything pre-GXBB until GXL/GXM, see [1] and [2]
> 
Thanks for the links. Then most likely we can do the same.

>> This patch only supports GXBB as I don't have other spec and test hw.
>> Points 25% vref and 75% vref are used for calibration.
>>
>> Adding a calibration scale resulted in no improvement, therefore
>> only a simple calibration bias is supported.
> in early versions of my IIO driver I had "calibration" support
> (Amlogic Linux kernel driver style, which uses scaling), see [3]
> on my Khadas VIM (GXL) this resulted in values which were basically
> identical with the values reported by Amlogic's Linux driver (I can't
> remember if I compared this with the u-boot values also).
> do you still have a version of your scaling code at hand (if not then
> I'll try to hack something up this weekend) so I can give that a try
> on a GXL based device?
> 
I will send a v2 based on the scaling code.

> a few minor nits inline below
> 
>> Successfully tested on a Odroid C2.  Calibbias is 8 on my system.
>>
>> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
>> ---
>>  drivers/iio/adc/meson_saradc.c | 56 ++++++++++++++++++++++++++++++++++++++++--
>>  1 file changed, 54 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/iio/adc/meson_saradc.c b/drivers/iio/adc/meson_saradc.c
>> index 23710088..276375c7 100644
>> --- a/drivers/iio/adc/meson_saradc.c
>> +++ b/drivers/iio/adc/meson_saradc.c
>> @@ -173,7 +173,8 @@
>>         .channel = _chan,                                               \
>>         .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) |                  \
>>                                 BIT(IIO_CHAN_INFO_AVERAGE_RAW),         \
>> -       .info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE),           \
>> +       .info_mask_shared_by_type = BIT(IIO_CHAN_INFO_SCALE) |          \
>> +                               BIT(IIO_CHAN_INFO_CALIBBIAS),           \
>>         .datasheet_name = "SAR_ADC_CH"#_chan,                           \
>>  }
>>
>> @@ -233,6 +234,7 @@ struct meson_sar_adc_priv {
>>         struct clk                              *adc_div_clk;
>>         struct clk_divider                      clk_div;
>>         struct completion                       done;
>> +       int                                     calibbias;
>>  };
>>
>>  static const struct regmap_config meson_sar_adc_regmap_config = {
>> @@ -252,6 +254,14 @@ static unsigned int meson_sar_adc_get_fifo_count(struct iio_dev *indio_dev)
>>         return FIELD_GET(MESON_SAR_ADC_REG0_FIFO_COUNT_MASK, regval);
>>  }
>>
>> +static int meson_sar_adc_calib_val(struct iio_dev *indio_dev, int val)
>> +{
>> +       struct meson_sar_adc_priv *priv = iio_priv(indio_dev);
>> +
>> +       return clamp(val + priv->calibbias, 0,
>> +                    (1 << priv->data->resolution) - 1);
> any reason why you're not using GENMASK as your are doing in
> meson_sar_adc_read_raw_sample?
> 
I prefer to use GENMASK only if the value is supposed to be
used as an actual mask in a bit operation.

>> +}
>> +
>>  static int meson_sar_adc_wait_busy_clear(struct iio_dev *indio_dev)
>>  {
>>         struct meson_sar_adc_priv *priv = iio_priv(indio_dev);
>> @@ -303,7 +313,7 @@ static int meson_sar_adc_read_raw_sample(struct iio_dev *indio_dev,
>>         fifo_val = FIELD_GET(MESON_SAR_ADC_FIFO_RD_SAMPLE_VALUE_MASK,
>>                              regval);
>>         fifo_val &= GENMASK(priv->data->resolution - 1, 0);
>> -       *val = fifo_val;
>> +       *val = meson_sar_adc_calib_val(indio_dev, fifo_val);
>>
>>         return 0;
>>  }
>> @@ -529,6 +539,10 @@ static int meson_sar_adc_iio_info_read_raw(struct iio_dev *indio_dev,
>>                 *val2 = priv->data->resolution;
>>                 return IIO_VAL_FRACTIONAL_LOG2;
>>
>> +       case IIO_CHAN_INFO_CALIBBIAS:
>> +               *val = priv->calibbias;
>> +               return IIO_VAL_INT;
>> +
>>         default:
>>                 return -EINVAL;
>>         }
>> @@ -754,6 +768,38 @@ static irqreturn_t meson_sar_adc_irq(int irq, void *data)
>>         return IRQ_HANDLED;
>>  }
>>
>> +static int meson_sar_adc_calib_gxbb(struct iio_dev *indio_dev)
>> +{
>> +       struct meson_sar_adc_priv *priv = iio_priv(indio_dev);
>> +       int ret, act0, act1, set0, set1;
> I don't have better names at the moment, but the "act" and "set"
> variables sound a bit strange
> 
Finding good variable names is the hardest part in coding ..
Maybe we can do it like in the vendor driver and use nominal/value
instead.

>> +
>> +       /* use points 25% and 75% for calibration */
>> +       set0 = (1 << priv->data->resolution) / 4;
>> +       set1 = (1 << priv->data->resolution) * 3 / 4;
> just like I mentioned above: any reason why you're not using GENMASK
> here as well?
> 
>> +
>> +       meson_sar_adc_set_chan7_mux(indio_dev, CHAN7_MUX_VDD_DIV4);
>> +       ret = meson_sar_adc_get_sample(indio_dev,
>> +                                      &meson_sar_adc_iio_channels[7],
>> +                                      MEAN_AVERAGING, EIGHT_SAMPLES, &act0);
>> +       if (ret < 0)
>> +               goto out;
>> +
>> +       meson_sar_adc_set_chan7_mux(indio_dev, CHAN7_MUX_VDD_MUL3_DIV4);
>> +       ret = meson_sar_adc_get_sample(indio_dev,
>> +                                      &meson_sar_adc_iio_channels[7],
>> +                                      MEAN_AVERAGING, EIGHT_SAMPLES, &act1);
>> +       if (ret < 0)
>> +               goto out;
>> +
>> +       priv->calibbias = (set0 + set1 - act0 - act1) / 2;
>> +
>> +       ret = 0;
>> +out:
>> +       meson_sar_adc_set_chan7_mux(indio_dev, CHAN7_MUX_CH7_INPUT);
>> +
>> +       return ret;
>> +}
>> +
>>  static const struct iio_info meson_sar_adc_iio_info = {
>>         .read_raw = meson_sar_adc_iio_info_read_raw,
>>         .driver_module = THIS_MODULE,
>> @@ -901,6 +952,12 @@ static int meson_sar_adc_probe(struct platform_device *pdev)
>>         if (ret)
>>                 goto err;
>>
>> +       if (priv->data == &meson_sar_adc_gxbb_data) {
>> +               ret = meson_sar_adc_calib_gxbb(indio_dev);
>> +               if (ret)
>> +                       goto err_hw;
>> +       }
>> +
>>         platform_set_drvdata(pdev, indio_dev);
>>
>>         ret = iio_device_register(indio_dev);
>> --
>> 2.12.0
>>
> 
> 
> Regards,
> Martin
> 
> 
> [0] https://github.com/khadas/u-boot/blob/d69d067db45d3002df0ad92558285e4671adb203/drivers/adc/saradc.c#L163
> [1] https://github.com/hardkernel/linux/blob/odroidc-3.10.y/drivers/amlogic/input/saradc/saradc.c#L92
> [2] https://github.com/khadas/linux/blob/6cb8e8669d2fb051f681f403c89ddf907a229586/drivers/amlogic/input/saradc/saradc.c#L108
> [3] https://github.com/xdarklight/linux/blob/9b3887d228b93ac0657ba454a924a45afedc2064/drivers/iio/adc/meson_saradc.c
>