On Thu, 15 Nov 2012, Sascha Hauer wrote:
> On Thu, Nov 15, 2012 at 11:39:58AM -0500, Robert P. J. Day wrote:
> >
> > yup, i'm testing that shortly, but my question was more along the
> > lines of, rather than having to disable all NAND-related
> > functionality *manually*, how easy/worthwhile would it be to create an
> > additional make target that reflects a beagle *without* NAND flash?
> >
> > one significant difference between a classic beagle and a current xM
> > is the total lack of NAND flash, and that's not a minor difference.
> > it means you can remove all barebox functionality related to that --
> > NAND routines and support, MTD stuff, JFFS2(?) support, and so on.
> > this could be automated by creating a couple new configs, say:
> >
> > make dm3730_beaglexm_xload_defconfig
> > make dm3730_beaglexm_defconfig
> >
> > so that the result really matches the underlying board.
> >
> > is this worth doing? it would seem to be fairly easy, i can whip
> > something up and test it.
>
> I don't think this is worth it. We should rather keep the number of
> defconfigs low. Splitting them means we have to maintain them and
> keep them in sync. Also it means to increase the time needed to
> compile all defconfigs (which I do routinely, currently on every
> commit).
ok, that makes sense. i'll still have a couple questions later on
the proper configuration of barebox for flash-less systems like the
xM. thanks.
rday
--
========================================================================
Robert P. J. Day Ottawa, Ontario, CANADA
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
