[U-Boot] [PATCH v2 00/34] env: Move environment code to use location drivers

[Tom Rini]

On Tue, Jul 25, 2017 at 03:40:42PM -0600, Simon Glass wrote:
> Hi Tom,
> 
> On 25 July 2017 at 11:49, Tom Rini <trini at konsulko.com> wrote:
> >
> > On Sun, Jul 23, 2017 at 09:19:35PM -0600, Simon Glass wrote:
> >
> > > The environment code is ripe for improvement in various ways. It has lots
> > > of duplication and inconsistencies between how things work with different
> > > environment locations. It does not properly use Kconfig. Error checking
> > > and handling is patchy.
> > >
> > > This series makes a start at improving things:
> > >
> > > - moves the rest of the CONFIG_ENV_IS_IN_... options to Kconfig
> >
> > I would like to see just this part first, by itself, so we can be sure
> > it's right.  I took a quick stab at trying this, and here's a few
> > things:
> > - The freescale SECURE_BOOT boards aren't right.
> >   board/freescale/common/Kconfig must have it select ENV_IS_NOWHERE and
> >   the other options must depend on !CHAIN_OF_TRUST.
> > - Patch 3 seem to, due to moving things slightly, cause tiny size
> >   changes.  On balance, this is fine.  But in terms of "did Kconfig
> >   migration happen correctly" it is not, since it makes it easier for
> >   something to slip past.
> >
> > So, I'm going to grab and update the migration portions of this series
> > and have them as a zero size (aside from boards with CMD_CONFIG on)
> > change, hopefully shortly.  And then we can still take in the rest of
> > this series, after comments, this release.  Sound good?  Thanks!
> >
> 
> I was unable to make this series avoid effective CONFIG changes as in
> some cases the environment is in one place in SPL (e.g.
> ENV_IS_NOWHERE) and another in U-Boot proper. I assumed that this was
> just going to have to be resolved by maintainers if it is a problem.
> 
> Anyway please do whatever you wish and I can rebase the other patches
> afterwards. We have time for people to chime in and adjust things
> later.

OK, thanks!  My first pass at size changes is looking good, I think the
*_SECURE_* things might have been what tripped up on the SPL-vs-not as
the code was messy to lead to ENV_IS_NOWHERE being in both.  This is
actually going quicker than I thought it might, and I should push the
Kconfig part in today or tomorrow.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20170725/588b9205/attachment.sig>