Re: [PATCH] ARM at91sam9260ek: remove unused reset entry

[Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>]

On 11:47 Wed 10 Oct     , Sascha Hauer wrote:
> On Wed, Oct 10, 2012 at 11:39:21AM +0200, Jean-Christophe PLAGNIOL-VILLARD wrote:
> > On 10:46 Wed 10 Oct     , Sascha Hauer wrote:
> > > On Mon, Oct 08, 2012 at 04:24:18PM +0400, Antony Pavlov wrote:
> > > > On 8 October 2012 16:15, Jean-Christophe PLAGNIOL-VILLARD
> > > > <plagnioj@jcrosoft.com> wrote:
> > > > > On 13:02 Mon 08 Oct     , Sascha Hauer wrote:
> > > > >> The at91sam9260ek has a custom reset function which does the same thing
> > > > >> as the default function. Also it does not define MACH_HAS_LOWLEVEL_INIT
> > > > >> as it should do. Remove the function.
> > > > >>
> > > > >> Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
> > > > >> ---
> > > > > Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
> > > > >
> > > > > btw we should rename the command reset reboot because it start to be confusing
> > > > 
> > > > IMHO it is a very good idea!
> > > 
> > > I would rather use a different name for the reset functions which are in
> > > the startup code. In the end the reset command just resets the board.
> > > 'reboot' sounds like something is being shutdown before restarting a
> > > board.
> > > 
> > > Changing a command is quite painful since all environment scripts will
> > > be broken. I don't think reset is used in the environment, but the
> > > API change to the mount command always nags me when using a new barebox
> > > which has an older environment in flash. I think we really have to be
> > > more careful when changing commands.
> > reset means reset the screen in linux
> > 
> > and I check nearly no one use reset in the env
> > execpt sam9g45 and defaultenv2
> > 
> > when I ask to do not update the var to pass the arch number on arm you give
> > the invert comment env evolve and we do not want to loose tis freedom
> 
> You are right, but still we have a problem in this area and I wanted to
> mention it. One way out could for example be that we introduce some
> version counter which we increase when incompatible changes occur. Then
> we could fall back to the default env. I haven't thought that to an end
> though.
agreed, I've add such info in the env for it's own format

maybe with a question to the end use the first time with a timeout
if he really want to keep the current one

Best Regards,
J.

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox