[oe] [OE-core] [yocto] [RFT] GCC 6 Recipes

[Burton, Ross]

Took it for another ride on our AB last night:

http://errors.yoctoproject.org/Errors/Latest/?filter=ab893d4e1c3efc1ee73c6cfbbb9a64036c63d93c&type=commit

The systemd-sysusers stuff I failed to remove, and is broken.

There's a pango-native failure that appears to be caused by GIR.

And two SDK image test failures caused by transient network problems.

Looking good!

Ross

On 28 April 2016 at 16:29, Khem Raj <raj.khem at gmail.com> wrote:

>
> > On Apr 28, 2016, at 8:19 AM, Mark Hatle <mark.hatle at windriver.com>
> wrote:
> >
> > On 4/28/16 2:10 AM, Khem Raj wrote:
> >>
> >>> On Apr 27, 2016, at 11:06 AM, Mark Hatle <mark.hatle at windriver.com>
> wrote:
> >>>
> >>> On 4/27/16 12:53 PM, Khem Raj wrote:
> >>>> thanks I think most of them come to these categories. So any help in
> >>>> fixing them is welcome.
> >>>>
> >>>> 1. kernel older than 4.4, backport needed fixed into these branches or
> >>>> move to newer kernels.
> >>>> 2. uboot dont have gcc6 compiler header files, backport the needed
> fixes
> >>>> 2. meta-qt4 needs to use gnu++98
> >>>> 3. gcc6 providing its own stdlib.h in libstdc++
> >>>> 4. gcc6 swtiching to gnu++14 for C++ as default.
> >>>> 5. GPLv2-only versions of recipes are simply old will need fixing
> >>>
> >>> From a -very- brief look, I suspect each of these fixes will be fairly
> simple.
> >>> So as we move into the next oe-core phase, we should be able get
> through this
> >>> pretty quickly.
> >>
> >> I think yes, they are trivial, except the GPL-2.0 packages which might
> need patching
> >>
> >
> > Ha, actually that was what I was referring to.  The couple I looked at
> seemed to
> > be trivial.. :)
>
> yes they are just the fact that compiler has gone stricter compliance with
> standards compared
> to when code was written. In worst case we might downgrade -std for these
> recipe but that
> may not be enough in some cases
> >
> > If someone doesn't get to it first, I expect some of my guys will in
> about a month.
> >
>
> OK cool. I think we should make it merge worthy and get it early into
> master and fix the fallouts
>
>
> > --Mark
>
>