building with clang - or, where does -fdelayed-template-parsing come from?

[Boudewijn Rempt]

Hm... well, I actually don't know either :-(. I've tried to figure it out this afternoon, but it's a mystery to me as well.

On Monday 24 March 2014 Mar 09:52:06 René J.V. Bertin wrote:
> Hello,
> 
> Since OS X requires the use of clang for anything involving C++ these days, I thought it'd be useful to test building calligra with that compiler on Linux too. There are other advantages: clang is (usually) faster, and recent versions can even generate more efficient code than gcc does. And of course it is capable as acting as a cross-compiler, allowing to use it with distcc in a heterogeneous environment :)
> 
> Except that it doesn't like the -fdelayed-template-parsing option, at least when building Sheets. Removing that option from the failing build commands make them succeed, but I cannot certify that this results in code with the exact same functionality.
> 
> I've tried to trace down where the option comes from, to no avail. The pattern 'fdelayed-template-parsing' cannot be found in either Calligra's CMake* and *.cmake files, nor in my cmake installation (/usr/share/cmake*}. It is also not a result of using distcc. When I use gcc (4.8) as my compiler, the option is also given, but in a much more limited number of cases, judging from the number of flags.make files matching the pattern.
> 
> So how might I hunt down the origin of this option, and the reason it is generated? Interestingly I didn't stumble across it when I built Calligra with clang on Mac OS X...
> 
> Thanks,
> R.
> _______________________________________________
> calligra-devel mailing list
> calligra-devel at kde.org
> https://mail.kde.org/mailman/listinfo/calligra-devel

-- 
Boudewijn Rempt
http://www.valdyas.org, http://www.krita.org, http://www.boudewijnrempt.nl