On Mon, Feb 11, 2013 at 01:30:53PM +0400, Alexander Shiyan wrote:
> Hello.
>
> ...
> > > diff --git a/arch/arm/boards/clep7212/Makefile b/arch/arm/boards/clep7212/Makefile
> > > index a63aeae..676e867 100644
> > > --- a/arch/arm/boards/clep7212/Makefile
> > > +++ b/arch/arm/boards/clep7212/Makefile
> > > @@ -1,2 +1 @@
> > > -obj-y += lowlevel.o clep7212.o
> > > -pbl-y += lowlevel.o
> > > +obj-y += clep7212.o
> > > diff --git a/arch/arm/boards/clep7212/lowlevel.c b/arch/arm/boards/clep7212/lowlevel.c
> >
> > I think it's better to keep the reset vector in board specific code.
> > Then you can have a board specific reset vector, which simply calls
> > a SoC specific one. This also solves the Kconfig problem with the
> > PLL setup. So you would have this in your board:
> >
> > void __naked __bare_init barebox_arm_reset_vector(void)
> > {
> > arm_cpu_lowlevel_init();
> >
> > clps711x_barebox_entry();
> > }
> >
> > You could add different more specific clps711x_* functions which will
> > result in different CPU speeds, or you could pass a PLL value to the
> > entry function, if that better fits your needs.
> >
> > The rest of this series looks fine.
>
> Can we leave this basic clps711x_barebox_entry in mach-clps711x/lowlevel.c
> and just additionally mark it as "__weak" for ability to override?
I'm not a big fan of __weak functions. Also, if you mark it as weak this
would mean that you would have to duplicate the other setup the original
weak function does.
Sascha
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
