[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20241219123444.19bd1074@jic23-huawei>
Date: Thu, 19 Dec 2024 12:34:44 +0000
From: Jonathan Cameron <jic23@...nel.org>
To: Matti Vaittinen <mazziesaccount@...il.com>
Cc: Matti Vaittinen <matti.vaittinen@...rohmeurope.com>, Lars-Peter Clausen
<lars@...afoo.de>, Rob Herring <robh@...nel.org>, Krzysztof Kozlowski
<krzk+dt@...nel.org>, Conor Dooley <conor+dt@...nel.org>,
linux-iio@...r.kernel.org, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/3] iio: dac: Support ROHM BD79703 DAC
On Thu, 19 Dec 2024 13:39:37 +0200
Matti Vaittinen <mazziesaccount@...il.com> wrote:
> The ROHM BD79703 is a 6 channel digital to analog converter.
>
> Based on the data-sheet examples the hardware would support setting the
> DAC word without changing the actual output. The data-sheet is not too
> specific on how the enabling the output of new voltage set by DAC
> should be done - hence this is not implemented by the driver.
I took a quick look and that does seem rather mysterious!
Normally there is either a 'sync' register or pin for setups like this
but no sign of either. Maybe it's legacy from some other use of the
same silicon.
>
> The BD79703 would also support two specific "PULL_DOWN" modes. These
> aren't currently supported by the driver either.
>
> Add a very basic support for controlling the channel outputs one-by-one.
>
> Signed-off-by: Matti Vaittinen <mazziesaccount@...il.com>
Short and sweet. One totally trivial formatting thing I can fix whilst
applying and a passing comment below. will leave this a little while
to see what other feedback it gets and for DT folk to look at the binding.
Jonathan
> diff --git a/drivers/iio/dac/rohm-bd79703.c b/drivers/iio/dac/rohm-bd79703.c
> new file mode 100644
> index 000000000000..a3df31d81b7c
> --- /dev/null
> +++ b/drivers/iio/dac/rohm-bd79703.c
> @@ -0,0 +1,162 @@
> +
> +static int bd79703_probe(struct spi_device *spi)
> +{
> + struct device *dev = &spi->dev;
> + struct bd79703_data *data;
> + struct iio_dev *idev;
> + int ret;
> +
> + idev = devm_iio_device_alloc(dev, sizeof(*data));
> + if (!idev)
> + return -ENOMEM;
> +
> + data = iio_priv(idev);
> +
> + data->regmap = devm_regmap_init_spi(spi, &bd79703_regmap_config);
> + if (IS_ERR(data->regmap))
> + return dev_err_probe(dev, PTR_ERR(data->regmap),
> + "Failed to initialize Regmap\n");
> +
> + ret = devm_regulator_get_enable(dev, "vcc");
> + if (ret)
> + return dev_err_probe(dev, ret, "Failed to enable VCC\n");
> +
> + ret = devm_regulator_get_enable_read_voltage(dev, "vfs");
> + if (ret < 0)
> + return dev_err_probe(dev, ret, "Failed to get Vfs\n");
> +
> + data->vfs = ret;
> + idev->channels = bd79703_channels;
> + idev->num_channels = ARRAY_SIZE(bd79703_channels);
> + idev->modes = INDIO_DIRECT_MODE;
> + idev->info = &bd79703_info;
> + idev->name = "bd79703";
> +
> + /* Initialize all to output zero */
I can't remember what we tend to do for defaults but I guess this
is safe enough. Alternative would be start the device with HI impedance
(power down state)
> + ret = regmap_write(data->regmap, BD79703_REG_OUT_ALL, 0);
> + if (ret)
> + return ret;
> +
> + return devm_iio_device_register(dev, idev);
> +}
> +
> +static const struct spi_device_id bd79703_id[] = {
> + { "bd79703", },
> + {}
{ }
Keep it consistent. This is the style I prefer though not sure I'll do
a full tidy up of this any time soon.
> +};
> +MODULE_DEVICE_TABLE(spi, bd79703_id);
> +
> +static const struct of_device_id bd79703_of_match[] = {
> + { .compatible = "rohm,bd79703", },
> + { }
> +};
> +MODULE_DEVICE_TABLE(of, bd79703_of_match);
Powered by blists - more mailing lists