On Wed, 5 Feb 2020 19:15:07 +0200
Alexandru Tachici <alexandru.tachici@xxxxxxxxxx> wrote:
> This patch makes the ad7192_probe fail in case
> regulator_get_voltage will return an error.
>
> Signed-off-by: Alexandru Tachici <alexandru.tachici@xxxxxxxxxx>
There is a subtlety in here we should handle.
> ---
> drivers/staging/iio/adc/ad7192.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/staging/iio/adc/ad7192.c b/drivers/staging/iio/adc/ad7192.c
> index bf3e2a9cc07f..4780ddf99b13 100644
> --- a/drivers/staging/iio/adc/ad7192.c
> +++ b/drivers/staging/iio/adc/ad7192.c
> @@ -899,10 +899,13 @@ static int ad7192_probe(struct spi_device *spi)
>
> voltage_uv = regulator_get_voltage(st->avdd);
>
I had to dig a bit to check this as it's not documented.
regulator_get_voltage returns negative for error, not 0.
0 typically means the voltage has not been set or has been
explicitly set to 0.
So if it is set to 0 I suppose we could try setting it as it
might be a variable high precision reference?
Perhaps for now just spitting out an error is the best plan.
> - if (voltage_uv)
> + if (voltage_uv) {
> st->int_vref_mv = voltage_uv / 1000;
> - else
> + } else {
> + ret = voltage_uv;
> dev_err(&spi->dev, "Device tree error, reference voltage undefined\n");
> + goto error_disable_avdd;
> + }
>
> spi_set_drvdata(spi, indio_dev);
> st->devid = spi_get_device_id(spi)->driver_data;
> @@ -957,6 +960,7 @@ static int ad7192_probe(struct spi_device *spi)
> ret = iio_device_register(indio_dev);
> if (ret < 0)
> goto error_disable_clk;
> +
We shouldn't have stray white space changes in a patch making a real change.
Thanks,
Jonathan
> return 0;
>
> error_disable_clk:
