On Tue, May 30, 2017 at 11:03 AM, Geert Uytterhoeven <[email protected]> wrote: > On Mon, May 29, 2017 at 6:45 PM, Linus Walleij <[email protected]> > wrote: >> On Mon, May 29, 2017 at 2:57 PM, Arnd Bergmann <[email protected]> wrote: >> >>> The driver cannot access the of_node field when CONFIG_OF is disabled: >>> >>> drivers/gpio/gpio-ingenic.c: In function 'ingenic_gpio_probe': >>> drivers/gpio/gpio-ingenic.c:324:10: error: 'struct gpio_chip' has no member >>> named 'of_node' >>> >>> Fixes: b0653ce39a0d ("gpio: Add gpio-ingenic driver") >>> Signed-off-by: Arnd Bergmann <[email protected]> >> >> I already sent a combined patch fixing these two deps earlier today, >> good that we are several at it :) > > Hmm... issues like these are actually the reasons why the of_node member > of struct device does not depend on CONFIG_OF?
It's a bit of kludginess-vs-footprint thing I guess, you gain a pointer, you loose a bit of Kconfig and #ifdefs. Yours, Linus Walleij
