Messages in this thread | | | Date | Thu, 8 Nov 2012 13:58:53 +0100 | Subject | Re: [PATCH] gpiolib: iron out include ladder mistakes | From | Linus Walleij <> |
| |
On Tue, Nov 6, 2012 at 10:40 PM, Stephen Warren <swarren@wwwdotorg.org> wrote: > On 11/06/2012 09:21 AM, Linus Walleij wrote:
>> And we need to keep the static inlines in <linux/gpio.h> >> but here for the !CONFIG_GENERIC_GPIO case, and then we >> may as well throw in a few warnings like the other >> prototypes there, if someone would have the bad taste >> of compiling without GENERIC_GPIO even. > > Hmm. Is there way to avoid the duplication of the dummy implementations? > Having a prototype and a truly dummy implementation in one place, but a > WARNing/failing dummy implementation elsewhere, seems like it'll cause > issues.
Yeah :-/
This is not exactly elegant and is some side effect of the split between CONFIG_GENERIC_GPIO and CONFIG_GPIOLIB, the real fix is to get rid of all GENERIC_GPIO implementations in the kernel and switch everyone over to GPIOLIB.
Not that easy though :-( can't think of any nice fix.
> Does this patch mean the previous series causes "git bisect" failures?
Yeah once I have something that doesn't break x86 I might just squash collapse all of this into the gpioranges patch.
Yours, Linus Walleij
| |