Sorry for top posting. I only have OWA access here...

Yeah, likely a missing include, which root cause can be that icc is not fully 
MSVC compliant. 
Since the files are generated by MIDL tool it seems to assume that a compiler 
from the same toolchain willl be used to compile the generated files, thus it 
might generate code with non-standard constructs. 

If ICC can link with MSVC object files, a possibly (ugly) workaround is to 
compile the iaccessible2 files with MSVC. You would have to create an 
EXTRA_COMPILER (IIRC) where you use cl.exe for the iaccessible2 files (passing 
the same compiler flags of course).

I'm not an expert on this myself (even though I added these), but I've lately 
realized that some of these files can possibly be removed. Some of the other 
generated files cause headaches in other situations, too (e.g. static builds).

Jan Arve

________________________________________
Fra: interest-bounces+jan-arve.saether=theqtcompany....@qt-project.org 
<interest-bounces+jan-arve.saether=theqtcompany....@qt-project.org> på vegne av 
Thiago Macieira <thiago.macie...@intel.com>
Sendt: 22. januar 2015 16:50
Til: qt-interest
Emne: Re: [Interest] Building Qt5 with Intel C++ under Windows

On Thursday 22 January 2015 09:46:27 Carsten Schneemann wrote:
> Hi Emre,
>
> Well, there are not so many enlightening things to say.
>
> Intel C++ on Windows is not among the "supported platforms" for Qt 5:
> http://doc.qt.io/QtSupportedPlatforms/index.html

Correct. It's supported by me, on a best-effort basis (tier 2 style). That
means the actual Qt release may fail until I get a chance to try it and fix the
issues.

In fact, I'm more reactive: if you have a problem and report it, I'll try fix
it. So if you report it during the beta or release candidate period, I'll
probably have enough time before the release.

> Which doesn't mean it can't be made to work. As far as I understand, Intel
> C++ and Visual C++ should in general be compatible enough s.th. code which
> compiles with MSVC (and this is a supported platform) should also compile
> with Intel C++. Except for little details which e.g. cause the current
> headache with iAccessible2.

Right.

It's a missing #include, it has to be.
--
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center

_______________________________________________
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest
_______________________________________________
Interest mailing list
Interest@qt-project.org
http://lists.qt-project.org/mailman/listinfo/interest

Reply via email to