Re: [PATCH v2 4/4] irqchip/aspeed-scu-ic: Add support AST2700 SCU interrupt controllers

[Thomas Gleixner <tglx@linutronix.de>]

On Sun, Aug 31 2025 at 10:14, Ryan Chen wrote:

> The AST2700 continues the multi-instance SCU interrupt controller model
> introduced in the AST2600, with four independent interrupt domains
> (scu-ic0 to 3).
>
> Unlike earlier generations that combine interrupt enable and status bits
> into a single register, the AST2700 separates these into distinct IER and
> ISR registers. Support for this layout is implemented by using register
> offsets and separate chained IRQ handlers.
>
> The variant table is extended to cover AST2700 IC instances, enabling
> shared initialization logic while preserving support for previous SoCs.
>
> Signed-off-by: Ryan Chen <ryan_chen@aspeedtech.com>
> ---
>  drivers/irqchip/irq-aspeed-scu-ic.c | 123 +++++++++++++++++++++-------
>  1 file changed, 95 insertions(+), 28 deletions(-)
>
> diff --git a/drivers/irqchip/irq-aspeed-scu-ic.c b/drivers/irqchip/irq-aspeed-scu-ic.c
> index cbfc35919281..ffdd9b4e44c1 100644
> --- a/drivers/irqchip/irq-aspeed-scu-ic.c
> +++ b/drivers/irqchip/irq-aspeed-scu-ic.c
> @@ -17,12 +17,16 @@
>  
>  #define ASPEED_SCU_IC_STATUS		GENMASK(28, 16)
>  #define ASPEED_SCU_IC_STATUS_SHIFT	16
> +#define AST2700_SCU_IC_STATUS		GENMASK(15, 0)
>  
>  struct aspeed_scu_ic_variant {
>  	const char		*compatible;
>  	unsigned long	irq_enable;
>  	unsigned long	irq_shift;
>  	unsigned int	num_irqs;
> +	bool			split_ier_isr;

How does that end up aligned?

> +	unsigned long	ier;
> +	unsigned long	isr;
>  };
>  
>  #define SCU_VARIANT(_compat, _shift, _enable, _num) { \
> @@ -30,13 +34,20 @@ struct aspeed_scu_ic_variant {
>  	.irq_shift		=	_shift,		\
>  	.irq_enable		=	_enable,	\
>  	.num_irqs		=	_num,		\
> +	.split_ier_isr	=	_split,		\

Ditto.

> +	.ier			=	_ier,		\
> +	.isr			=	_isr,		\

But what's worse is that '_split, _ier and _isr' come out of thin air as
SCU_VARIANT does not have corresponding arguments. So how is that
supposed to work?

>  }
>  
>  struct aspeed_scu_ic {
> @@ -45,9 +56,12 @@ struct aspeed_scu_ic {
>  	unsigned int		num_irqs;
>  	void __iomem		*base;
>  	struct irq_domain	*irq_domain;
> +	bool				split_ier_isr;

Sigh...

> +	unsigned long		ier;
> +	unsigned long		isr;
>  };
>  
> -static void aspeed_scu_ic_irq_handler(struct irq_desc *desc)
> +static void aspeed_scu_ic_irq_handler_combined(struct irq_desc *desc)
>  {
>  	struct aspeed_scu_ic *scu_ic = irq_desc_get_handler_data(desc);
>  	struct irq_chip *chip = irq_desc_get_chip(desc);
> @@ -84,33 +98,69 @@ static void aspeed_scu_ic_irq_handler(struct irq_desc *desc)
>  	chained_irq_exit(chip, desc);
>  }
>  
> +static void aspeed_scu_ic_irq_handler_split(struct irq_desc *desc)
> +{

...

>  static void aspeed_scu_ic_irq_mask(struct irq_data *data)
>  {
>  	struct aspeed_scu_ic *scu_ic = irq_data_get_irq_chip_data(data);
> -	unsigned int mask = BIT(data->hwirq + scu_ic->irq_shift) |
> -		(scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT);
>  
> -	/*
> -	 * Status bits are cleared by writing 1. In order to prevent the mask
> -	 * operation from clearing the status bits, they should be under the
> -	 * mask and written with 0.
> -	 */
> -	writel(readl(scu_ic->base) & ~mask, scu_ic->base);
> +	if (scu_ic->split_ier_isr) {
> +		writel(readl(scu_ic->base) & ~BIT(data->hwirq + scu_ic->irq_shift),
> +		       scu_ic->base + scu_ic->ier);
> +	} else {
> +		unsigned int mask = BIT(data->hwirq + scu_ic->irq_shift) |
> +			(scu_ic->irq_enable << ASPEED_SCU_IC_STATUS_SHIFT);
> +
> +		/*
> +		 * Status bits are cleared by writing 1. In order to prevent the mask
> +		 * operation from clearing the status bits, they should be under the
> +		 * mask and written with 0.
> +		 */
> +		writel(readl(scu_ic->base) & ~mask, scu_ic->base);
> +	}

So you have two different handlers. Why can't you provide two different
mask/unmask/ functions along with a seperate irq chip instead of
cluttering the code with conditionals. Thes two variants share no code
at all.

> -	irq_set_chained_handler_and_data(irq, aspeed_scu_ic_irq_handler,
> -					 scu_ic);
> +	if (scu_ic->split_ier_isr)
> +		irq_set_chained_handler_and_data(irq, aspeed_scu_ic_irq_handler_split,
> +						 scu_ic);
> +	else
> +		irq_set_chained_handler_and_data(irq, aspeed_scu_ic_irq_handler_combined,
> +						 scu_ic);
>

Please get rid of the line break. You have 100 characters....

Thanks,

        tglx