Mixing compilers

[Thomas Friedrichsmeier]

Hi,

On Sun, 23 Jun 2019 12:27:49 +0000
Gilles Caulier <caulier.gilles at gmail.com> wrote:
[...]
> This is my feedback :
> 
> - linking C API between both compiler work well.

Yes, we need only C linkage, and therefore compiling with MSVC works of
sorts. However, there are some messy points in the API. For instance
there is at least one point where we are expected to allocate a pointer
that will later be freed "on the other side" of the API. This crashes.

There appear to be other subtle issues, that I have not yet fully
debugged.

[...]

> About your problem of WebEngine which cannot be compiled with MinGW,
> the problem is know, and Qt bug has a report about this topic. The
> problem come from Google team which do not support this compiler with
> internal engine code. A possible solution again to cross compile
> QtWebEngine for Windows under Linux will be to use Clang instead GCC.
> 
> https://bugreports.qt.io/browse/QTBUG-42725

I know. And I was really angry at Qt for stepping into this, knowingly,
effectively killing off MinGW for Qt. But well, that's too late to
change, now...

Clang would seem like an interesting new ingredient to the mix.
Shouldn't it also be possible to compile on Windows, directly, but
telling to it compile for MinGW/GNU ABI?

Then would it actually seem feasible to mix MinGW and clang (or MSVC
and clang) inside a single craft installation?

> If you have a solution to mix C++ symbols with MinGW and MSVC, it
> will be interesting to have another feedbacks.

Nope, sorry. For all I have read on that, it's just impossible (unless
the use case is simple enough so you can realistically write a C-only
wrapper).

Regards
Thomas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://mail.kde.org/pipermail/kde-windows/attachments/20190623/e3b74e92/attachment.sig>