Re: [PATCH v7 5/5] OMAP: McBSP: Split and move read/write functions to mach-omap1/2

[Jarkko Nikula <jhnikula@gmail.com>]

On Wed, 9 Dec 2009 21:34:30 +0100
Janusz Krzysztofik <jkrzyszt@tis.icnet.pl> wrote:

> > > > Almost ready with it, one more question: what do you think about
> > > > splitting and moving omap_mcbsp_read()/_write() there as well? If you
> > > > agree, should I submit 2 patches, one with this cleanup, the other one
> > > > actually introducing cache support, or is one combined OK?
> > >
> > > Sounds good to me!
> >
...
> diff -upr git.orig/arch/arm/mach-omap1/mcbsp.c git/arch/arm/mach-omap1/mcbsp.c
> --- git.orig/arch/arm/mach-omap1/mcbsp.c	2009-12-09 15:49:52.000000000 +0100
> +++ git/arch/arm/mach-omap1/mcbsp.c	2009-12-09 16:20:43.000000000 +0100
>  
> +void omap_mcbsp_write(struct omap_mcbsp *mcbsp, u16 reg, u32 val)
...
> diff -upr git.orig/arch/arm/mach-omap2/mcbsp.c git/arch/arm/mach-omap2/mcbsp.c
> --- git.orig/arch/arm/mach-omap2/mcbsp.c	2009-12-09 15:49:52.000000000 +0100
> +++ git/arch/arm/mach-omap2/mcbsp.c	2009-12-09 16:20:43.000000000 +0100
>  
> +void omap_mcbsp_write(struct omap_mcbsp *mcbsp, u16 reg, u32 val)

These functions must be unique, otherwise multi-build is not possible
(no idea can we do it for OMAP1?).

IMO, the _write and _read functions in ./plat-omap/mcbsp.c are clean
after the patch 3/5 anyway so probably we don't need this splitting?

-- 
Jarkko