wandboard / imx6 / exynos4 / cortex-a9 "wrong-endian bug" fixed

[Warner Losh]

On Feb 3, 2014, at 9:59 PM, Tim Kientzle wrote:

> 
> On Feb 2, 2014, at 6:59 PM, Warner Losh <imp at bsdimp.com> wrote:
> 
>> 
>> On Feb 2, 2014, at 7:37 PM, Ian Lepore wrote:
>> 
>>> On Sun, 2014-02-02 at 18:25 -0700, Tom Everett wrote:
>>>> I can confirm that a crochet image for Wandboard no longer crashes after
>>>> this fix.  It doesn't mount the FS yet, but that's a known issue, noted
>>>> here:
>>>> 
>>>> https://github.com/kientzle/crochet-freebsd/issues/29
>>> 
>>> That thread mentions needing to build a u-boot for wandboard.  Here's a
>>> patchset for doing that.  Apply this in your ports/sysutils dir, using:
>>> 
>>> patch -p0 <u-boot-wandboard.diff
>>> 
>>> This was copied from Tim's beaglebone work and changed a bit for
>>> wandboard.  
>>> 
>>> I've been meaning to ask about how we should handle this.  Do we want a
>>> u-boot port for every board, or is there some way to parameterize a
>>> single port so that it applies the right set of patches, or what?  I'm
>>> not very savvy about ports.
>> 
>> Usually there's a master port, and slave ports that parameterize. When we talked about this at BSDcan, I believe that was the model that was suggested for this problem..
> 
> Does that work when there’s a different source distro for
> many of the ports?  (Not to mention different patches for
> each one.)

Yes, but there may be complications...

> P.S.  I started a U-Boot port for BeagleBone but my
> time has evaporated.  If anyone wants to take that
> over and extend it for other boards, step right up…

ENOTIME...

Warner