On Tue, 2017-03-07 at 12:15 +0200, Andy Shevchenko wrote:
> In order to make GPIO ACPI library stricter prepare users of
> gpiod_get_index() to correctly behave when there no mapping is
> provided by firmware.
>
> Here we add explicit mapping between _CRS GpioIo() resources and
> their names used in the driver.
While the first patch is correct, this one doesn't.
It missed index conversion when GPIO ACPI mapping table is present.
Please ignore both, I think I will make it similar to what I did for
st21nfca.
>
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
> ---
> - assign mapping table only in ACPI case
> drivers/nfc/pn544/i2c.c | 15 +++++++++++++++
> 1 file changed, 15 insertions(+)
>
> diff --git a/drivers/nfc/pn544/i2c.c b/drivers/nfc/pn544/i2c.c
> index 2df3176205c4..bfe099f3d347 100644
> --- a/drivers/nfc/pn544/i2c.c
> +++ b/drivers/nfc/pn544/i2c.c
> @@ -874,11 +874,25 @@ static void pn544_hci_i2c_fw_work(struct
> work_struct *work)
> }
> }
>
> +static const struct acpi_gpio_params enable_gpios = { 1, 0, false };
> +static const struct acpi_gpio_params firmware_gpios = { 2, 0, false
> };
> +
> +static const struct acpi_gpio_mapping acpi_pn544_gpios[] = {
> + { "enable-gpios", &enable_gpios, 1 },
> + { "firmware-gpios", &firmware_gpios, 1 },
> + { },
> +};
> +
> static int pn544_hci_i2c_acpi_request_resources(struct i2c_client
> *client)
> {
> struct pn544_i2c_phy *phy = i2c_get_clientdata(client);
> struct gpio_desc *gpiod_en, *gpiod_fw;
> struct device *dev = &client->dev;
> + int ret;
> +
> + ret = acpi_dev_add_driver_gpios(ACPI_COMPANION(dev),
> acpi_pn544_gpios);
> + if (ret)
> + return ret;
>
> /* Get EN GPIO from ACPI */
> gpiod_en = devm_gpiod_get_index(dev, "enable", 1,
> GPIOD_OUT_LOW);
> @@ -1092,6 +1106,7 @@ static int pn544_hci_i2c_remove(struct
> i2c_client *client)
> pdata->free_resources();
> }
>
> + acpi_dev_remove_driver_gpios(ACPI_COMPANION(&client->dev));
> return 0;
> }
>
--
Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
Intel Finland Oy
