Re: [PATCH-next] irqchip/renesas-rzv2h: Fix potentially mismatched datatype

[Jiri Slaby <jirislaby@kernel.org>]

Hi,

On 31. 10. 24, 20:36, Advait Dhamorikar wrote:
> This patch updates the type of hw_irq to unsigned long to
> match irq_hw_number_t.
> 
> The variable hw_irq is defined as unsigned int at places,
> However when it is initialized using irqd_to_hwirq(), it returns
> an irq_hw_number_t, which inturn is a typedef for unsigned long.

"in turn"

But what's the purpose of this? First, why wouldn't you use 
irq_hw_number_t then?

Nevertheless, the HW does not support hw irqs > uint (it supports 
1+16+32, actually). So why all this in the first place?

> @@ -265,9 +265,9 @@ static int rzv2h_tint_set_type(struct irq_data *d, unsigned int type)
>   	u32 titsr, titsr_k, titsel_n, tien;
>   	struct rzv2h_icu_priv *priv;
>   	u32 tssr, tssr_k, tssel_n;
> -	unsigned int hwirq;
> +	unsigned long hwirq;
>   	u32 tint, sense;
> -	int tint_nr;
> +	unsigned long tint_nr;

Switching tint_nr to unsigned might still be a good thing to avoid weird 
signed overflows in the future. But I don't think it warrants for a patch...

thanks,
-- 
js
suse labs