Re: [PATCH] i2c: reorganize muxes to a standard pattern

[Jean Delvare <khali-PUYAD+kWke1g9hUCZPvPmw@public.gmane.org>]

On Mon, 28 Nov 2011 10:58:30 +0100, Peter Korsgaard wrote:
> >>>>> "Wolfram" == Wolfram Sang <w.sang-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org> writes:
> 
>  >> > http://thread.gmane.org/gmane.linux.drivers.i2c/7171/focus=7244
>  >> 
>  >> Ah, yes, that's the discussion I was looking for, thanks for digging it
>  >> out. My point wasn't totally wrong back then, but Wolfram's is simply
>  >> better, I admit.
> 
>  Wolfram> I agree :) Okay, so I'll update the documentation as well. What about
>  Wolfram> include/linux/gpio-i2cmux.h and its users? I'd like consistency, but
>  Wolfram> renaming header files is not too nice...
> 
> Indeed. If we were to rename it we should also rename struct
> gpio_i2cmux_platform_data.
> 
> I don't feel strongly about it - It will break for existing users, but
> there's probably not too many of those. Your call.

I see exactly 1 user of <linux/gpio-i2cmux.h> in the upstream kernel
tree, and that is gpio-i2cmux itself. So I'd say no big deal renaming
it, and actually if we intend to rename header files and/or structures,
the sooner the better.

-- 
Jean Delvare