linux-kernel - Re: [PATCH v3 4/9] irqchip/renesas-rzg2l: Drop IRQC_NUM

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <878qd6147f.ffs@tglx>
Date: Fri, 06 Feb 2026 12:50:12 +0100
From: Thomas Gleixner <tglx@...nel.org>
To: Biju <biju.das.au@...il.com>
Cc: Biju Das <biju.das.jz@...renesas.com>, linux-kernel@...r.kernel.org,
 Geert Uytterhoeven <geert+renesas@...der.be>, Prabhakar Mahadev Lad
 <prabhakar.mahadev-lad.rj@...renesas.com>, Biju Das
 <biju.das.au@...il.com>, linux-renesas-soc@...r.kernel.org
Subject: Re: [PATCH v3 4/9] irqchip/renesas-rzg2l: Drop IRQC_NUM_IRQ macro

On Fri, Feb 06 2026 at 11:16, Biju wrote:
> +/**
> + * struct rzg2l_hw_info - Interrupt Control Unit controller hardware info structure.
> + * @num_irq:		Total Number of interrupts
> + */
> +struct rzg2l_hw_info {
> +	u8	num_irq;

Odd data type. Whats wrong with a good old unsigned int?

> +};
> +
>  /**
>   * struct rzg2l_irqc_priv - IRQ controller private data structure
>   * @base:	Controller's base address
>   * @irqchip:	Pointer to struct irq_chip
>   * @fwspec:	IRQ firmware specific data
>   * @lock:	Lock to serialize access to hardware registers
> + * @info:	Pointer to struct rzg2l_hw_info

Why a pointer?

>   * @cache:	Registers cache for suspend/resume
>   */
>  static struct rzg2l_irqc_priv {
> @@ -81,6 +89,7 @@ static struct rzg2l_irqc_priv {
>  	const struct irq_chip		*irqchip;
>  	struct irq_fwspec		*fwspec;
>  	raw_spinlock_t			lock;
> +	const struct rzg2l_hw_info	*info;
>  	struct rzg2l_irqc_reg_cache	cache;
>  } *rzg2l_irqc_data;
>  
> @@ -136,7 +145,7 @@ static void rzg2l_irqc_eoi(struct irq_data *d)
>  	raw_spin_lock(&priv->lock);
>  	if (hw_irq >= IRQC_IRQ_START && hw_irq <= IRQC_IRQ_COUNT)
>  		rzg2l_clear_irq_int(priv, hw_irq);
> -	else if (hw_irq >= IRQC_TINT_START && hw_irq < IRQC_NUM_IRQ)
> +	else if (hw_irq >= IRQC_TINT_START && hw_irq < priv->info->num_irq)

Ah I see. To make this more expensive by accessing yet another cache
line. Simply embed a struct hwinfo into irqc_priv and copy the data into
it at probe time.

 
> -	if (hwirq > (IRQC_NUM_IRQ - 1))
> +	if (hwirq > (priv->info->num_irq - 1))

  hwirq >= priv->info.num_irq

This -1 logic is horrible and error prone.

Thanks,

        tglx