On Mon, Dec 08, 2008 at 02:05:15PM -0200, Werner Almesberger wrote:
> At least on 2410 and 2442, GPIOs don't end at G10 but at H10 or
> even J12.
>
> Port J is treated as a special case throughout much of the S3C
> code, so I'm not sure what the plan is for it now. However, it
> seems safe to include port H.
The re-work of the s3c24xx gpio (mainly the configuration code to
match the s3c generic code) has been put back by trying to find the
problems resuming the s3c64xx from suspend-to-ram.
> Signed-off-by: Werner Almesberger <[EMAIL PROTECTED]>
>
> ---
>
> Index: ktrack/arch/arm/mach-s3c2410/include/mach/gpio-core.h
> ===================================================================
> --- ktrack.orig/arch/arm/mach-s3c2410/include/mach/gpio-core.h
> 2008-12-08 13:27:19.000000000 -0200
> +++ ktrack/arch/arm/mach-s3c2410/include/mach/gpio-core.h 2008-12-08
> 13:29:10.000000000 -0200
> @@ -24,7 +24,7 @@
> {
> struct s3c_gpio_chip *chip;
>
> - if (pin > S3C2410_GPG10)
> + if (pin > S3C2410_GPH10)
> return NULL;
If this code was going to live in the kernel for more than a revision,
I'd ask for it to be dependant on the cpu version, however I will try
and sort out scrubbing the old s3c24xx gpio code out and moving completley
to gpiolib as soon as possible.
> chip = &s3c24xx_gpios[pin/32];
> Index: ktrack/arch/arm/plat-s3c24xx/gpiolib.c
> ===================================================================
> --- ktrack.orig/arch/arm/plat-s3c24xx/gpiolib.c 2008-12-08
> 13:29:13.000000000 -0200
> +++ ktrack/arch/arm/plat-s3c24xx/gpiolib.c 2008-12-08 13:29:37.000000000
> -0200
> @@ -133,6 +133,16 @@
> .ngpio = 16,
> },
> },
> + [7] = {
> + .base = S3C24XX_GPIO_BASE(S3C2410_GPH0),
> + .pm = __gpio_pm(&s3c_gpio_pm_2bit),
> + .chip = {
> + .base = S3C2410_GPH0,
> + .owner = THIS_MODULE,
> + .label = "GPIOH",
> + .ngpio = 11,
> + },
> + },
> };
>
> static __init int s3c24xx_gpiolib_init(void)
--
Ben
Q: What's a light-year?
A: One-third less calories than a regular year.
