[oe] chromium detected as stripped

[Raphael Kubo da Costa]

Trevor Woerner <twoerner at gmail.com> writes:

> On Wed, Sep 20, 2017 at 1:43 PM, Trevor Woerner <twoerner at gmail.com> wrote:
>> When I compile chromium61 with pyro, I get a strange error. Every time the
>> python wrapper (gcc_link_wrapper.py) is called for a native compile (g++), I
>> get:
> ...
>> This happens with both qemux86-64 and raspberrypi3. Unfortunately those log
>> files are 18M and 19M each, so I won't be attaching them. Why can't my build
>> find 'ld'? That seems quite strange. Probably some kind of ld vs gold issue?
>> I'm not using gold anywhere and have not enabled it, are you using gold?
>
>
> Turns out, I had to install binutils-gold on my host for this to work.
> Which suggests a hiccup in the build. Maybe the build needs to tweak
> the python wrapper? (As if using ninja and gn weren't enough, the
> build also wants to use a wrapper to call the compiler?!)

I think I've fixed both the gold and the Breakpad issues, as described
in my other reply.

> This still leaves the problem with my builds against master. I tried
> downgrading the compiler to 6%, but that causes a build issue in mesa,
> so the build can't even get close to chromium to see if that compile
> can succeed with gcc-6 instead of the default gcc-7.
>
> The regular procedure is to add things to master, then backport them
> to various branches if appropriate. If we can't get master to work,
> there won't be anything to backport. In any case meta-browser
> currently doesn't have a pyro branch.

Can you try my recipe with master again? So far I've only tested pyro to
avoid having the ground shift under my as I work on the recipe; I know
that at least Chromium master builds fine with GCC 7, so any build
errors caused by the compiler version are easily fixed with one or more
backports.