Mir's building. And Mir's not building.

[Kevin DuBois]

As for the phone CI problem, it looks like there was a package transition
from "libprotobuf-lite9" in vivid to "libprotobuf-lite9v5" in wily. The job
is building on vivid, so the package depends on "libprotobuf-lite9". When
deployed on the wily phone image, the package resolver can't figure out
where to get "libprotobuf-lite9" (as its been renamed), and fails. The
solution is to build and deploy on the same system.

On Tue, Aug 18, 2015 at 12:47 PM, Alan Griffiths <
alan.griffiths at canonical.com> wrote:

> On 18/08/15 14:35, Alan Griffiths wrote:
> > On 18/08/15 13:09, Alan Griffiths wrote:
> >> On 18/08/15 10:15, Daniel van Vugt wrote:
> >>> The wily archive transition is now complete and you probably noticed
> >>> you can once again build lp:mir on a fully updated wily system.
> ...
> >>
> >>> * Clang builds fail, including in CI.
> >>>   No one owns this yet.
> > Reproduced, investigating.
>
> as best we can figure it clang++ is not sufficiently compatible with g++5
> and in a way that we can't fix or workaround.
> https://llvm.org/bugs/show_bug.cgi?id=23529 summarises the reason: for
> some interfaces (std::string) g++ now uses an "abi_tag" that clang doesn't
> grok.
>
> While that state remains we won't be able to build Mir with clang on
> Wily.
>
> We're planning to try switching the clang builds to run on Vivid.
>
>
>
> --
> Mir-devel mailing list
> Mir-devel at lists.ubuntu.com
> Modify settings or unsubscribe at:
> https://lists.ubuntu.com/mailman/listinfo/mir-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.ubuntu.com/archives/mir-devel/attachments/20150818/e5919525/attachment.html>