Hi,
Thanks for the patch.
On 09/11/2016 12:52 PM, Gwenhael Goavec-Merou wrote:
> +static struct ad9834_platform_data *ad9834_parse_dt(struct spi_device *spi)
> +{
> + struct ad9834_platform_data *pdata;
> + struct device_node *np = spi->dev.of_node;
> +
> + pdata = devm_kzalloc(&spi->dev, sizeof(*pdata), GFP_KERNEL);
> + if (!pdata)
> + return ERR_PTR(-ENOMEM);
> +
> + if (of_property_read_u32(np, "mclk", &pdata->mclk))
> + return ERR_PTR(-ENODEV);
The input clock should be using the standard clock bindings.
> + if (of_property_read_u32(np, "freq0", &pdata->freq0))
> + return ERR_PTR(-ENODEV);
> + if (of_property_read_u32(np, "freq1", &pdata->freq1))
> + return ERR_PTR(-ENODEV);
> + if (of_property_read_u16(np, "phase0", &pdata->phase0))
> + return ERR_PTR(-ENODEV);
> + if (of_property_read_u16(np, "phase1", &pdata->phase1))
> + return ERR_PTR(-ENODEV);
> + pdata->en_div2 = of_property_read_bool(np, "en_div2");
> + pdata->en_signbit_msb_out = of_property_read_bool(np,
> + "en_signbit_msb_out");
The other attributes seem to be more runtime configuration data, rather than
hardware description. Maybe just choose a fixed default.
- Lars
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
