Message-ID: <CAHp75VdNymzseF7Dt9kL8GBPLM0MGBQg-YQabKUKxEryM8nxOQ@mail.gmail.com>
Date: Mon, 28 Apr 2025 10:04:22 +0300
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Jonathan Santos <Jonathan.Santos@...log.com>
Cc: linux-iio@...r.kernel.org, devicetree@...r.kernel.org, 
	linux-kernel@...r.kernel.org, linux-gpio@...r.kernel.org, andy@...nel.org, 
	nuno.sa@...log.com, Michael.Hennerich@...log.com, marcelo.schmitt@...log.com, 
	jic23@...nel.org, robh@...nel.org, krzk+dt@...nel.org, conor+dt@...nel.org, 
	marcelo.schmitt1@...il.com, linus.walleij@...aro.org, brgl@...ev.pl, 
	lgirdwood@...il.com, broonie@...nel.org, jonath4nns@...il.com, 
	dlechner@...libre.com
Subject: Re: [PATCH v6 08/11] iio: adc: ad7768-1: add support for
 Synchronization over SPI

On Mon, Apr 28, 2025 at 3:14 AM Jonathan Santos
<Jonathan.Santos@...log.com> wrote:
>
> The synchronization method using GPIO requires the generated pulse to be
> truly synchronous with the base MCLK signal. When it is not possible to
> do that in hardware, the datasheet recommends using synchronization over
> SPI, where the generated pulse is already synchronous with MCLK. This
> requires the SYNC_OUT pin to be connected to SYNC_IN pin.

to the SYNC_IN

> Use trigger-sources property to enable device synchronization over SPI
> and multi-device synchronization, as an alternative to adi,sync-in-gpios
> property.

...

> +static int ad7768_send_sync_pulse(struct ad7768_state *st)
> +{
> +       if (st->en_spi_sync)
> +               return regmap_write(st->regmap, AD7768_REG_SYNC_RESET, 0x00);

> +       if (st->gpio_sync_in) {

Dup check, the following have already it.

> +               gpiod_set_value_cansleep(st->gpio_sync_in, 1);

Yes, I see the original code, but still the Q is why no delay. Perhaps
a comment explaining that the GPIO op is slow enough (?) to add.

> +               gpiod_set_value_cansleep(st->gpio_sync_in, 0);
> +       }
> +
> +       return 0;
> +}

...

> +static struct gpio_desc *ad7768_trigger_source_get_gpio(struct device *dev,
> +                                                       struct fwnode_handle *fwnode)
> +{
> +       const char *value;
> +       int ret;
> +
> +       ret = fwnode_property_read_string(fwnode, "compatible", &value);
> +       if (ret)
> +               return ERR_PTR(ret);
> +
> +       if (strcmp("gpio-trigger", value))
> +               return ERR_PTR(-EINVAL);

Reinvention of fwnode_device_is_compatible().

> +       return devm_fwnode_gpiod_get_index(dev, fwnode, NULL, 0,
> +                                          GPIOD_OUT_LOW, "sync-in");
> +}

...

> +static int ad7768_trigger_sources_get_sync(struct device *dev,
> +                                          struct ad7768_state *st)
> +{
> +       struct fwnode_reference_args args;
> +       struct fwnode_handle *fwnode = NULL;

Redundant assignment AFAICS.

> +       int ret;
> +
> +       /*
> +        * The AD7768-1 allows two primary methods for driving the SYNC_IN pin
> +        * to synchronize one or more devices:
> +        * 1. Using an external GPIO.
> +        * 2. Using a SPI command, where the SYNC_OUT pin generates a
> +        *    synchronization pulse that drives the SYNC_IN pin.
> +        */
> +       if (!device_property_present(dev, "trigger-sources")) {
> +               /*
> +                * In the absence of trigger-sources property, enable self
> +                * synchronization over SPI (SYNC_OUT).
> +                */
> +               st->en_spi_sync = true;
> +               return 0;
> +       }
> +
> +       ret = fwnode_property_get_reference_args(dev_fwnode(dev),

In this case the above is better to be also fwnode_property_present().
You save a double call to dev_fwnode().

> +                                                "trigger-sources",
> +                                                "#trigger-source-cells",
> +                                                0,
> +                                                AD7768_TRIGGER_SOURCE_SYNC_IDX,
> +                                                &args);
> +       if (ret)
> +               return ret;
> +
> +       fwnode = args.fwnode;
> +       /*
> +        * First, try getting the GPIO trigger source and fallback to
> +        * synchronization over SPI in case of failure.
> +        */
> +       st->gpio_sync_in = ad7768_trigger_source_get_gpio(dev, fwnode);
> +       if (IS_ERR(st->gpio_sync_in)) {
> +               /*
> +                * For this case, it requires one argument, which indicates the
> +                * output pin referenced.
> +                */
> +               if (args.nargs < 1)
> +                       goto err_not_supp;
> +
> +               if (args.args[0] != AD7768_TRIGGER_SOURCE_SYNC_OUT)
> +                       goto err_not_supp;
> +
> +               /*
> +                * Only self trigger is supported for now, i.e.,
> +                * external SYNC_OUT is not allowed.
> +                */
> +               if (fwnode->dev == dev) {

?!?! What is this?!

For the reference:
https://elixir.bootlin.com/linux/v6.15-rc3/source/include/linux/fwnode.h#L51

> +                       st->en_spi_sync = true;
> +                       goto out_put_node;
> +               }
> +
> +               goto err_not_supp;
> +       }
> +
> +       goto out_put_node;
> +
> +err_not_supp:
> +       ret = dev_err_probe(dev, -EOPNOTSUPP,
> +                           "Invalid synchronization trigger source");

Missing \n, and can be one line anyway (we don't complain about long
strings ending with string literals for ages, way before the 100
character limit).

> +out_put_node:
> +       fwnode_handle_put(args.fwnode);
> +       return ret;
> +}

-- 
With Best Regards,
Andy Shevchenko

Powered by blists - more mailing lists