[PATCH V2 1/2] mfd: da9062: Remove IRQ requirement

[Lee Jones]

On Thu, 09 Feb 2023, Christoph Niedermaier wrote:

> This patch removes the requirement for an IRQ, because for the core
> functionality IRQ isn't needed. So this makes the DA9061/62 chip
> useable for designs which haven't connected the IRQ pin.
> 
> Signed-off-by: Christoph Niedermaier <cniedermaier at dh-electronics.com>
> ---
> Cc: Support Opensource <support.opensource at diasemi.com>
> Cc: Lee Jones <lee at kernel.org>
> Cc: Adam Thomson <Adam.Thomson.Opensource at diasemi.com>
> Cc: Liam Girdwood <lgirdwood at gmail.com>
> Cc: Mark Brown <broonie at kernel.org>
> Cc: Marek Vasut <marex at denx.de>
> Cc: kernel at dh-electronics.com
> Cc: linux-kernel at vger.kernel.org
> To: linux-arm-kernel at lists.infradead.org
> ---
> V2: - Rebase on current next 20230209
>     - Add Lee Jones to Cc list
> ---
>  drivers/mfd/da9062-core.c | 98 +++++++++++++++++++++++++++++++++++------------
>  1 file changed, 73 insertions(+), 25 deletions(-)
> 
> diff --git a/drivers/mfd/da9062-core.c b/drivers/mfd/da9062-core.c
> index 40cde51e5719..caa597400dd1 100644
> --- a/drivers/mfd/da9062-core.c
> +++ b/drivers/mfd/da9062-core.c
> @@ -212,6 +212,27 @@ static const struct mfd_cell da9061_devs[] = {
>  	},
>  };
>  
> +static const struct mfd_cell da9061_devs_without_irq[] = {

"_noirq"

> +	{
> +		.name		= "da9061-core",
> +	},
> +	{
> +		.name		= "da9062-regulators",
> +	},

Place the one line entries on one line please.

Even better, use MFD_CELL_NAME()

> +	{
> +		.name		= "da9061-watchdog",
> +		.of_compatible  = "dlg,da9061-watchdog",
> +	},

MFD_CELL_OF(<name>, NULL, NULL, NULL, 0, <compatible>);

> +	{
> +		.name		= "da9061-thermal",
> +		.of_compatible  = "dlg,da9061-thermal",
> +	},
> +	{
> +		.name		= "da9061-onkey",
> +		.of_compatible = "dlg,da9061-onkey",
> +	},
> +};
> +
>  static const struct resource da9062_core_resources[] = {
>  	DEFINE_RES_NAMED(DA9062_IRQ_VDD_WARN, 1, "VDD_WARN", IORESOURCE_IRQ),
>  };
> @@ -288,6 +309,35 @@ static const struct mfd_cell da9062_devs[] = {
>  	},
>  };
>  
> +static const struct mfd_cell da9062_devs_without_irq[] = {
> +	{
> +		.name		= "da9062-core",
> +	},
> +	{
> +		.name		= "da9062-regulators",
> +	},
> +	{
> +		.name		= "da9062-watchdog",
> +		.of_compatible  = "dlg,da9062-watchdog",
> +	},
> +	{
> +		.name		= "da9062-thermal",
> +		.of_compatible  = "dlg,da9062-thermal",
> +	},
> +	{
> +		.name		= "da9062-rtc",
> +		.of_compatible  = "dlg,da9062-rtc",
> +	},
> +	{
> +		.name		= "da9062-onkey",
> +		.of_compatible	= "dlg,da9062-onkey",
> +	},
> +	{
> +		.name		= "da9062-gpio",
> +		.of_compatible	= "dlg,da9062-gpio",
> +	},
> +};

As above.

>  static int da9062_clear_fault_log(struct da9062 *chip)
>  {
>  	int ret;
> @@ -625,7 +675,7 @@ static int da9062_i2c_probe(struct i2c_client *i2c)
>  {
>  	const struct i2c_device_id *id = i2c_client_get_device_id(i2c);
>  	struct da9062 *chip;
> -	unsigned int irq_base;
> +	unsigned int irq_base = 0;
>  	const struct mfd_cell *cell;
>  	const struct regmap_irq_chip *irq_chip;
>  	const struct regmap_config *config;
> @@ -645,21 +695,16 @@ static int da9062_i2c_probe(struct i2c_client *i2c)
>  	i2c_set_clientdata(i2c, chip);
>  	chip->dev = &i2c->dev;
>  
> -	if (!i2c->irq) {
> -		dev_err(chip->dev, "No IRQ configured\n");
> -		return -EINVAL;
> -	}
> -
>  	switch (chip->chip_type) {
>  	case COMPAT_TYPE_DA9061:
> -		cell = da9061_devs;
> -		cell_num = ARRAY_SIZE(da9061_devs);
> +		cell = i2c->irq ? da9061_devs : da9061_devs_without_irq;
> +		cell_num = i2c->irq ? ARRAY_SIZE(da9061_devs) : ARRAY_SIZE(da9061_devs_without_irq);

This is hideous.

Why not just NULLify the resources below instead?

>  		irq_chip = &da9061_irq_chip;
>  		config = &da9061_regmap_config;
>  		break;
>  	case COMPAT_TYPE_DA9062:
> -		cell = da9062_devs;
> -		cell_num = ARRAY_SIZE(da9062_devs);
> +		cell = i2c->irq ? da9062_devs : da9062_devs_without_irq;
> +		cell_num = i2c->irq ? ARRAY_SIZE(da9062_devs) : ARRAY_SIZE(da9062_devs_without_irq);
>  		irq_chip = &da9062_irq_chip;

Still setting this despite no IRQs?

>  		config = &da9062_regmap_config;
>  		break;
  __
 _||_
 \  /
  \/

[...]

if (i2c->irq <= 0)
  cell->resources = NULL;
  cell->num_resources = 0;

> @@ -695,29 +740,32 @@ static int da9062_i2c_probe(struct i2c_client *i2c)
>  	if (ret)
>  		return ret;
>  
> -	ret = da9062_configure_irq_type(chip, i2c->irq, &trigger_type);
> -	if (ret < 0) {
> -		dev_err(chip->dev, "Failed to configure IRQ type\n");
> -		return ret;
> -	}
> +	if (i2c->irq) {
> +		ret = da9062_configure_irq_type(chip, i2c->irq, &trigger_type);
> +		if (ret < 0) {
> +			dev_err(chip->dev, "Failed to configure IRQ type\n");
> +			return ret;
> +		}
>  
> -	ret = regmap_add_irq_chip(chip->regmap, i2c->irq,
> -			trigger_type | IRQF_SHARED | IRQF_ONESHOT,
> -			-1, irq_chip, &chip->regmap_irq);
> -	if (ret) {
> -		dev_err(chip->dev, "Failed to request IRQ %d: %d\n",
> -			i2c->irq, ret);
> -		return ret;
> -	}
> +		ret = regmap_add_irq_chip(chip->regmap, i2c->irq,
> +				trigger_type | IRQF_SHARED | IRQF_ONESHOT,
> +				-1, irq_chip, &chip->regmap_irq);
> +		if (ret) {
> +			dev_err(chip->dev, "Failed to request IRQ %d: %d\n",
> +				i2c->irq, ret);
> +			return ret;
> +		}
>  
> -	irq_base = regmap_irq_chip_get_base(chip->regmap_irq);
> +		irq_base = regmap_irq_chip_get_base(chip->regmap_irq);
> +	}
>  
>  	ret = mfd_add_devices(chip->dev, PLATFORM_DEVID_NONE, cell,
>  			      cell_num, NULL, irq_base,
>  			      NULL);
>  	if (ret) {
>  		dev_err(chip->dev, "Cannot register child devices\n");
> -		regmap_del_irq_chip(i2c->irq, chip->regmap_irq);
> +		if (i2c->irq)
> +			regmap_del_irq_chip(i2c->irq, chip->regmap_irq);
>  		return ret;
>  	}
>  
> -- 
> 2.11.0
> 

-- 
Lee Jones [李琼斯]