On Thu, 6 Dec 2018 11:10:52 +0200
Mircea Caprioru <mircea.caprioru@xxxxxxxxxx> wrote:
> When using the internal clock the device has an option in which the clock
> output is available on MCLK2 pin. This patch adds a dt binding for enabling
> this property.
>
> Signed-off-by: Mircea Caprioru <mircea.caprioru@xxxxxxxxxx>
I'd rather we looked at getting this out of staging, with proper binding
docs before we add too many new features, but I suppose this is small
enough that it isn't too much of a problem other that we are defining
a binding that might not go down well long term.
If nothing else we should be supporting this as a clock source
to allow it for example to be looped into another identical device
as the input.
Here we are putting it out on a pin with no 'consumers'.
Jonathan
> ---
> drivers/staging/iio/adc/ad7192.c | 9 ++++++++-
> 1 file changed, 8 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/staging/iio/adc/ad7192.c b/drivers/staging/iio/adc/ad7192.c
> index 8a4e6ede42b3..a11c8a82bb7c 100644
> --- a/drivers/staging/iio/adc/ad7192.c
> +++ b/drivers/staging/iio/adc/ad7192.c
> @@ -611,6 +611,10 @@ static const struct iio_chan_spec ad7193_channels[] = {
> static int ad7192_clock_select(struct spi_device *spi, struct ad7192_state *st)
> {
> int ret;
> + bool clock_out_en;
> +
> + clock_out_en = of_property_read_bool(spi->dev.of_node,
> + "adi,int-clock-output-enable");
>
> st->clock_sel = AD7192_CLK_EXT_MCLK2;
> st->mclk = devm_clk_get(&spi->dev, "clk");
> @@ -626,7 +630,10 @@ static int ad7192_clock_select(struct spi_device *spi, struct ad7192_state *st)
> return PTR_ERR(st->mclk);
>
> /* use internal clock */
> - st->clock_sel = AD7192_CLK_INT;
> + if (!clock_out_en)
> + st->clock_sel = AD7192_CLK_INT;
> + else
> + st->clock_sel = AD7192_CLK_INT_CO;
> st->fclk = AD7192_INT_FREQ_MHZ;
> }
> }
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
