Yes, here locally we have some additional changes/fixes in some places as well. And if talk about mpc85xx gpio support it's better to have driver in drivers/gpio vs static definitions in header file and I have it. But currently I need this minimal patch to be accepted to send our board support patch based on this one.
Oleks
On 03/27/2015 11:46 AM, Joakim Tjernlund wrote:
On Fri, 2015-03-27 at 10:51 -0400, Oleksandr G Zhadan wrote:
Extra gpio bank argument at low level may be discussable, but IMPO we should not brake standard int gpio_set_value(unsigned gpio, int value) definition.
And the reason of this patch is to fix:
- Incompatibility in functions gpio_free() and gpio_set_value()
definitions between <asm/gpio.h> and <asm-generic/gpio.h> (void vs int)
- Compilation error when build common/cmd_gpio.c for mpc85xx (p1020)
Sure, when I saw "mpc85xx gpio .." I remembered seeing this flaw so I figured I point it out before I forget again.
This also makes me remember, drivers/spi/fsl_espi.c is broken for anything but reading an EEPROM.
Jocke