Re: [PATCH 3/3] bcm63xx: early registering of gpiochip

[Ralf Baechle <ralf@linux-mips.org>]

On Mon, Jun 01, 2009 at 08:18:20PM +0200, Florian Fainelli wrote:

> This patch makes the gpiochip be registered
> earlier in the prom code. This allows GPIO-based
> board detection to be made earlier.
> 
> Tested-by: Maxime Bizon <mbizon@freebox.fr>
> Signed-off-by: Florian Fainelli <florian@openwrt.org>
> ---
> diff --git a/arch/mips/bcm63xx/gpio.c b/arch/mips/bcm63xx/gpio.c
> index b78d3fd..997fcaa 100644
> --- a/arch/mips/bcm63xx/gpio.c
> +++ b/arch/mips/bcm63xx/gpio.c
> @@ -120,9 +120,8 @@ static struct gpio_chip bcm63xx_gpio_chip = {
>  	.ngpio			= BCM63XX_GPIO_COUNT,
>  };
>  
> -static int __init bcm63xx_gpio_init(void)
> +int __init bcm63xx_gpio_init(void)
>  {
>  	printk(KERN_INFO "registering %d GPIOs\n", BCM63XX_GPIO_COUNT);
>  	return gpiochip_add(&bcm63xx_gpio_chip);
>  }
> -arch_initcall(bcm63xx_gpio_init);
> diff --git a/arch/mips/bcm63xx/prom.c b/arch/mips/bcm63xx/prom.c
> index d97ceed..1e5bb15 100644
> --- a/arch/mips/bcm63xx/prom.c
> +++ b/arch/mips/bcm63xx/prom.c
> @@ -13,6 +13,7 @@
>  #include <bcm63xx_cpu.h>
>  #include <bcm63xx_io.h>
>  #include <bcm63xx_regs.h>
> +#include <bcm63xx_gpio.h>
>  
>  void __init prom_init(void)
>  {
> @@ -40,6 +41,9 @@ void __init prom_init(void)
>  
>  	/* do low level board init */
>  	board_prom_init();
> +
> +	/* early register gpio */

That word order sounds odd ...

> +	bcm63xx_gpio_init();
>  }
>  
>  void __init prom_free_prom_memory(void)
> diff --git a/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h b/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h
> index 72cee75..8bc812f 100644
> --- a/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h
> +++ b/arch/mips/include/asm/mach-bcm63xx/bcm63xx_gpio.h
> @@ -1,6 +1,8 @@
>  #ifndef BCM63XX_GPIO_H
>  #define BCM63XX_GPIO_H
>  
> +int __init bcm63xx_gpio_init(void);

It is not necessary to declare a function as __init.  But if you do you
better include <linux/init.h> first.  With this patch you'd rely on init.h
having previously included which eventually when the context changes will
blow up.

And not last, the patch seems to reject ...

  Ralf