Re: [PATCH 1/9 v3] kernel-devsrc: restructure for out of tree (and on target) module builds

[Richard Purdie <richard.purdie@linuxfoundation.org>]

On Wed, 2018-03-07 at 08:35 -0500, Bruce Ashfield wrote:
> On Wed, Mar 7, 2018 at 3:40 AM, Burton, Ross <ross.burton@intel.com>
> wrote:
> > 
> > On 6 March 2018 at 18:11, Bruce Ashfield <bruce.ashfield@windriver.
> > com>
> > wrote:
> > > 
> > > 
> > > +RDEPENDS_${PN} = "bc python ${TCLIBC}-utils"
> > > +# 4.15+ needs these next two RDEPENDS
> > > +RDEPENDS_${PN} += "openssl-dev util-linux"
> > > +# and x86 needs a bit more for 4.15+
> > > +RDEPENDS_${PN} += "${@bb.utils.contains('ARCH', 'x86',
> > > 'elfutils', '',
> > > d)}"
> > 
> > Annoyingly this is breaking our multilib test case:
> > 
> > https://autobuilder.yocto.io/builders/nightly-multilib/builds/852/s
> > teps/BuildImages_4/logs/stdio
> > 
> > It's entirely possible that the test is broken in some way but it's
> > far too
> > early for me right now to comprehend the multilib matrix...
> And at this point, I'm basically out of cycles for hacking on devsrc
> :(
> 
> I'll see about it again at the start of next week, and failing that,
> bump it to the next release cycle.

Just as a quick brain dump from a glance at this, I think the issue is
the openssl-dev dependency. It sounds like the image tries to install
two sets of -dev packages and those files conflict. Quite why its doing
that and how/why they conflict I don't know but the -dev dependency is
where I'd start looking at this.

I will also mention that -dev packages are not so well looked at with
multilib, we probably don't test that area much :(.

Cheers,

Richard