Re: [PATCH 2/2] irqchip/aspeed-scu-ic: Add support for AST2700 SCU interrupt controllers

[Thomas Gleixner <tglx@linutronix.de>]

On Mon, Aug 04 2025 at 13:34, Ryan Chen wrote:
> +#define ASPEED_AST2700_SCU_IC0_EN_REG	0x1d0
> +#define ASPEED_AST2700_SCU_IC0_STS_REG	0x1d4
> +#define ASPEED_AST2700_SCU_IC0_SHIFT	0
> +#define ASPEED_AST2700_SCU_IC0_ENABLE	\
> +	GENMASK(3, ASPEED_AST2700_SCU_IC0_SHIFT)

Let it stick out, you have 100 characters

>  struct aspeed_scu_ic {
>  	unsigned long irq_enable;
>  	unsigned long irq_shift;
>  	unsigned int num_irqs;
>  	unsigned int reg;
> +	unsigned int en_reg;
> +	unsigned int sts_reg;
> +	bool split_ier_isr;

Please reformat this struct declaration according to:

https://www.kernel.org/doc/html/latest/process/maintainer-tip.html#struct-declarations-and-initializers

>  	struct regmap *scu;
>  	struct irq_domain *irq_domain;
>  };
> @@ -52,33 +83,51 @@ static void aspeed_scu_ic_irq_handler(struct irq_desc *desc)
>  	unsigned long status;
>  	struct aspeed_scu_ic *scu_ic = irq_desc_get_handler_data(desc);
>  	struct irq_chip *chip = irq_desc_get_chip(desc);
> -	unsigned int mask = scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT;
> +	unsigned int mask;

https://www.kernel.org/doc/html/latest/process/maintainer-tip.html#variable-declarations

>  	chained_irq_enter(chip, desc);
>  
> -	/*
> -	 * The SCU IC has just one register to control its operation and read
> -	 * status. The interrupt enable bits occupy the lower 16 bits of the
> -	 * register, while the interrupt status bits occupy the upper 16 bits.
> -	 * The status bit for a given interrupt is always 16 bits shifted from
> -	 * the enable bit for the same interrupt.
> -	 * Therefore, perform the IRQ operations in the enable bit space by
> -	 * shifting the status down to get the mapping and then back up to
> -	 * clear the bit.
> -	 */
> -	regmap_read(scu_ic->scu, scu_ic->reg, &sts);
> -	enabled = sts & scu_ic->irq_enable;
> -	status = (sts >> ASPEED_SCU_IC_STATUS_SHIFT) & enabled;
> -
> -	bit = scu_ic->irq_shift;
> -	max = scu_ic->num_irqs + bit;
> -
> -	for_each_set_bit_from(bit, &status, max) {
> -		generic_handle_domain_irq(scu_ic->irq_domain,
> -					  bit - scu_ic->irq_shift);
> -
> -		regmap_write_bits(scu_ic->scu, scu_ic->reg, mask,
> -				  BIT(bit + ASPEED_SCU_IC_STATUS_SHIFT));
> +	if (scu_ic->split_ier_isr) {
> +		mask = scu_ic->irq_enable;
> +		regmap_read(scu_ic->scu, scu_ic->en_reg, &sts);
> +		enabled = sts & scu_ic->irq_enable;
> +		regmap_read(scu_ic->scu, scu_ic->sts_reg, &sts);
> +		status = sts & enabled;
> +
> +		bit = scu_ic->irq_shift;
> +		max = scu_ic->num_irqs + bit;
> +
> +		for_each_set_bit_from(bit, &status, max) {
> +			generic_handle_domain_irq(scu_ic->irq_domain, bit - scu_ic->irq_shift);
> +
> +			regmap_write_bits(scu_ic->scu, scu_ic->sts_reg, mask, BIT(bit));
> +		}
> +	} else {
> +		mask = scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT;
> +		/*
> +		 * The SCU IC has just one register to control its operation and read
> +		 * status. The interrupt enable bits occupy the lower 16 bits of the
> +		 * register, while the interrupt status bits occupy the upper 16 bits.
> +		 * The status bit for a given interrupt is always 16 bits shifted from
> +		 * the enable bit for the same interrupt.
> +		 * Therefore, perform the IRQ operations in the enable bit space by
> +		 * shifting the status down to get the mapping and then back up to
> +		 * clear the bit.
> +		 */
> +		regmap_read(scu_ic->scu, scu_ic->reg, &sts);
> +		enabled = sts & scu_ic->irq_enable;
> +		status = (sts >> ASPEED_SCU_IC_STATUS_SHIFT) & enabled;
> +
> +		bit = scu_ic->irq_shift;
> +		max = scu_ic->num_irqs + bit;
> +
> +		for_each_set_bit_from(bit, &status, max) {
> +			generic_handle_domain_irq(scu_ic->irq_domain,
> +						  bit - scu_ic->irq_shift);
> +
> +			regmap_write_bits(scu_ic->scu, scu_ic->reg, mask,
> +					  BIT(bit + ASPEED_SCU_IC_STATUS_SHIFT));
> +		}

This is horrible, really. Either rework the code so that both chips can
share it with minimal conditionals or provide seperate handlers. It's
not rocket science.
  
>  	chained_irq_exit(chip, desc);
> @@ -87,30 +136,42 @@ static void aspeed_scu_ic_irq_handler(struct irq_desc *desc)

>  static void aspeed_scu_ic_irq_mask(struct irq_data *data)
>  static void aspeed_scu_ic_irq_unmask(struct irq_data *data)

>  {
>  	struct aspeed_scu_ic *scu_ic = irq_data_get_irq_chip_data(data);
>  	unsigned int bit = BIT(data->hwirq + scu_ic->irq_shift);
> -	unsigned int mask = bit |
> -		(scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT);
> -
> -	/*
> -	 * Status bits are cleared by writing 1. In order to prevent the unmask
> -	 * operation from clearing the status bits, they should be under the
> -	 * mask and written with 0.
> -	 */
> -	regmap_update_bits(scu_ic->scu, scu_ic->reg, mask, bit);
> +	unsigned int mask;
> +
> +	if (scu_ic->split_ier_isr) {
> +		mask = bit;
> +		regmap_update_bits(scu_ic->scu, scu_ic->en_reg, mask, bit);
> +	} else {
> +		mask = bit | (scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT);
> +
> +		/*
> +		 * Status bits are cleared by writing 1. In order to prevent the unmask
> +		 * operation from clearing the status bits, they should be under the
> +		 * mask and written with 0.
> +		 */
> +		regmap_update_bits(scu_ic->scu, scu_ic->reg, mask, bit);
> +	}

This also wants to be consolidated or seperated.

>  }
>  
> +static int __init aspeed_ast2700_scu_ic0_of_init(struct device_node *node,
> +						 struct device_node *parent)
> +{
> +	struct aspeed_scu_ic *scu_ic = kzalloc(sizeof(*scu_ic), GFP_KERNEL);
> +
> +	if (!scu_ic)
> +		return -ENOMEM;
> +
> +	scu_ic->irq_enable = ASPEED_AST2700_SCU_IC0_ENABLE;
> +	scu_ic->irq_shift = ASPEED_AST2700_SCU_IC0_SHIFT;
> +	scu_ic->num_irqs = ASPEED_AST2700_SCU_IC0_NUM_IRQS;
> +	scu_ic->split_ier_isr = true;
> +	scu_ic->en_reg = ASPEED_AST2700_SCU_IC0_EN_REG;
> +	scu_ic->sts_reg = ASPEED_AST2700_SCU_IC0_STS_REG;
> +
> +	return aspeed_scu_ic_of_init_common(scu_ic, node);
> +}
> +
> +static int __init aspeed_ast2700_scu_ic1_of_init(struct device_node *node,
> +						 struct device_node *parent)
> +{
> +	struct aspeed_scu_ic *scu_ic = kzalloc(sizeof(*scu_ic), GFP_KERNEL);
> +
> +	if (!scu_ic)
> +		return -ENOMEM;
> +
> +	scu_ic->irq_enable = ASPEED_AST2700_SCU_IC1_ENABLE;
> +	scu_ic->irq_shift = ASPEED_AST2700_SCU_IC1_SHIFT;
> +	scu_ic->num_irqs = ASPEED_AST2700_SCU_IC1_NUM_IRQS;
> +	scu_ic->split_ier_isr = true;
> +	scu_ic->en_reg = ASPEED_AST2700_SCU_IC1_EN_REG;
> +	scu_ic->sts_reg = ASPEED_AST2700_SCU_IC1_STS_REG;
> +
> +	return aspeed_scu_ic_of_init_common(scu_ic, node);
> +}
> +
> +static int __init aspeed_ast2700_scu_ic2_of_init(struct device_node *node,
> +						 struct device_node *parent)
> +{
> +	struct aspeed_scu_ic *scu_ic = kzalloc(sizeof(*scu_ic), GFP_KERNEL);
> +
> +	if (!scu_ic)
> +		return -ENOMEM;
> +
> +	scu_ic->irq_enable = ASPEED_AST2700_SCU_IC2_ENABLE;
> +	scu_ic->irq_shift = ASPEED_AST2700_SCU_IC2_SHIFT;
> +	scu_ic->num_irqs = ASPEED_AST2700_SCU_IC2_NUM_IRQS;
> +	scu_ic->split_ier_isr = true;
> +	scu_ic->en_reg = ASPEED_AST2700_SCU_IC2_EN_REG;
> +	scu_ic->sts_reg = ASPEED_AST2700_SCU_IC2_STS_REG;
> +
> +	return aspeed_scu_ic_of_init_common(scu_ic, node);
> +}
> +
> +static int __init aspeed_ast2700_scu_ic3_of_init(struct device_node *node,
> +						 struct device_node *parent)
> +{
> +	struct aspeed_scu_ic *scu_ic = kzalloc(sizeof(*scu_ic), GFP_KERNEL);
> +
> +	if (!scu_ic)
> +		return -ENOMEM;
> +
> +	scu_ic->irq_enable = ASPEED_AST2700_SCU_IC3_ENABLE;
> +	scu_ic->irq_shift = ASPEED_AST2700_SCU_IC3_SHIFT;
> +	scu_ic->num_irqs = ASPEED_AST2700_SCU_IC3_NUM_IRQS;
> +	scu_ic->split_ier_isr = true;
> +	scu_ic->en_reg = ASPEED_AST2700_SCU_IC3_EN_REG;
> +	scu_ic->sts_reg = ASPEED_AST2700_SCU_IC3_STS_REG;
> +
> +	return aspeed_scu_ic_of_init_common(scu_ic, node);
> +}

You seriously have no better idea than this copy & pasta orgy?

static struct scu_variant variants = {
	SCU("aspeed,ast2400-scu-ic", ......, whatever...),
        ...
	SCU("aspeed,ast2700-scu-ic0", 0x1D0, 0x1D4, 0, GENMASK(3, 0),
            4, whatever...),
        ...
	SCU("aspeed,ast2700-scu-ic3", 0x10C, 0x108, 0, GENMASK(1, 0),
            2, ......),
} __initdata;

static struct scu_variant *find_variant(struct device_node *node)
{
        for (int i = 0; i < ARRAY_SIZE(variant); i++) {
        	if (!strcmp(variant[i].name, node->name))
                	return &variant[i];
	}                               
        return NULL;
}

static int __init ast_scu_of_init(struct device_node *node, struct device_node *parent)
{
        struct variant *v = find_variant(node);

        if (!v)
          	...

        scu_ic = kzalloc(...);
        *scu_ic = v->scu;
        ...

You get the idea.

This rework needs to come first and then you can add your new 2700 muck
on top.

Thanks,

        tglx