[U-Boot] Please pull u-boot-marvell/master

[Chris Packham]

On Thu, May 17, 2018 at 8:48 PM Stefan Roese <sr at denx.de> wrote:

> Hi Chris,

> On 17.05.2018 10:45, Chris Packham wrote:
> > On Thu, May 17, 2018 at 8:36 PM Chris Packham <judge.packham at gmail.com>
> > wrote:
> >
> >
> >> On Thu, May 17, 2018 at 3:37 AM Tom Rini <trini at konsulko.com> wrote:
> >>>
> >>> On Wed, May 16, 2018 at 02:05:03PM +0200, Stefan Roese wrote:
> >>>
> >>>> Hi Tom,
> >>>>
> >>>> please pull the Kirkwood DT patches from Chris.
> >>>>
> >>>> Thanks,
> >>>> Stefan
> >>>>
> >>>> The following changes since commit
> > 0315d6959fdd9d2a4d89016c311e9c8c8d239a10:
> >>>>
> >>>>    ARM: mvebu: a38x: Add missing SPDX license identfier (2018-05-15
> > 09:08:00 -0400)
> >>>>
> >>>> are available in the Git repository at:
> >>>>
> >>>>    git://www.denx.de/git/u-boot-marvell.git
> >>>>
> >>>> for you to fetch changes up to
> > a7925926dc7f81fb5c137d2b4b7c6457694b1ac0:
> >>>>
> >>>>    ARM: kirkwood: Add device-tree for sheevaplug (2018-05-16 10:29:53
> > +0200)
> >>>>
> >>>
> >>> NAK.  With gcc-7.3
> >>> (
> >
https://mirrors.edge.kernel.org/pub/tools/crosstool/files/bin/x86_64/7.3.0/)
> >>> I get:
> >>>         arm:  +   openrd_base openrd_ultimate openrd_client
> >>> +(openrd_base,openrd_ultimate,openrd_client) u-boot-nodtb.bin exceeds
> > file size limit:
> >>> +(openrd_base,openrd_ultimate,openrd_client)   limit:  393216 bytes
> >>> +(openrd_base)   actual: 395632 bytes
> >>> +(openrd_base)   excess: 2416 bytes
> >>> +(openrd_base,openrd_ultimate,openrd_client) make[1]: ***
> > [Makefile:1014: u-boot-nodtb.bin] Error 1
> >>> +(openrd_ultimate)   actual: 395664 bytes
> >>> +(openrd_ultimate)   excess: 2448 bytes
> >>> +(openrd_client)   actual: 395680 bytes
> >>> +(openrd_client)   excess: 2464 bytes
> >>>
> >>> And I will be pushing those changes to finally switch travis over to
> >>> gcc-7.3 soon (I had thought I had posted them a long while back so
that
> >>> I could have pulled them in right after v2018.05, but I forgot, so
here
> >>> we are now..).
> >
> >> Odd I did get those failures initially but then I rebased against
master
> > and didn't see them again. I've been using gcc 7.3 throughout. At the
time
> > I assumed there were some other changes that had shrunk openrd enough
for
> > the dtb to fit. I see Stefan's patch to move the env location. Is that
> > sufficient or do you want me to either drop openrd or find something
that
> > can be left out?
> >
> > (gmail seems to have forgotten how to wrap quotes)
> >
> > Sure enough my local branch with this series based on commit
cc95535867f4
> > ("bootm.c: Correct the flush_len used in bootm_load_os()") builds
> > openrd_ultimate is fine. But if I rebase onto commit f2d0f5e7ab3b ("ARM:
> > re-enable MVGBE for edminiv2") I see the build size errors. I'll do some
> > digging. Chances are my other MVGBE patch has enabled something
unwanted.

> Thanks Chris. But perhaps its also good to increase the image size
> for those openrd boards, since we most likely will increase the
> size in the near future even further (addition of DM support etc).
> Please see:

> http://patchwork.ozlabs.org/patch/915135/

I'm fine with that. We'll probably end up needing it.

I've just sent a patch to not automatically select PHYLIB anyway since my
MVGBE changes weren't the no-op they should have been.