[PATCH v2] mtd/maps: gpio-addr-flash: new driver for GPIO assisted flash addressing

[Mike Frysinger]

On Sun, Sep 20, 2009 at 18:04, David Woodhouse wrote:
> On Wed, 2009-06-03 at 08:51 -0400, Mike Frysinger wrote:
>> This driver lets people use GPIO's for additional address lines in case
>> their processor does not have enough address lines already.
>>
>> Signed-off-by: Mike Frysinger <vapier at gentoo.org>
>> Signed-off-by: Bryan Wu <cooloney at kernel.org>
>
> In file included from /home/dwmw2/git/mtd-2.6/arch/x86/include/asm/gpio.h:20,
>                 from drivers/mtd/maps/gpio-addr-flash.c:27:
> include/asm-generic/gpio.h: In function ‘gpio_get_value_cansleep’:
> include/asm-generic/gpio.h:168: error: implicit declaration of function ‘gpio_get_value’
> include/asm-generic/gpio.h: In function ‘gpio_set_value_cansleep’:
> include/asm-generic/gpio.h:174: error: implicit declaration of function ‘gpio_set_value’
> drivers/mtd/maps/gpio-addr-flash.c: In function ‘gpio_flash_probe’:
> drivers/mtd/maps/gpio-addr-flash.c:232: error: implicit declaration of function ‘gpio_request’
> drivers/mtd/maps/gpio-addr-flash.c:236: error: implicit declaration of function ‘gpio_free’
> drivers/mtd/maps/gpio-addr-flash.c:240: error: implicit declaration of function ‘gpio_direction_output’
> make[2]: *** [drivers/mtd/maps/gpio-addr-flash.o] Error 1
> make[1]: *** [drivers/mtd/maps] Error 2
> make: *** [_module_drivers/mtd] Error 2
>
> Setting CONFIG_GPIOLIB seems to fix it -- but we can't make it depend on
> that, can we? Maybe it should be GPIOLIB || !GENERIC_GPIO?

it works fine on Blackfin systems w/out gpiolib, so yes, forcing a
depend on that would be awful.

i think you mean:
depend GPIOLIB || GENERIC_GPIO
-mike