marble hard-coding -O3 optimizations

[Albert Astals Cid]

El Dissabte, 4 de febrer de 2012, a les 16:15:40, Torsten Rahn va escriure:
> Not sure whether that solves the issue but thanks for pointing out :)

I'm with Rex here, debug should be debuggeable to the maximum and having a -O3 
there problably means that the compiler does some stuff to make it faster, 
which is cool, but probably means it's not maximally debuggeable anymore.

Ok course this does not help if our packagers compile with Debug instead of 
RelWithDebInfo (no idea if this is the case or not).

Albert

> 
> On Samstag, 4. Februar 2012 15:13:28 Rex Dieter wrote:
> > Torsten Rahn wrote:
> > > Yes this has been intentional. I had done this right from the start
> > > since compilation without this flag would lead to unacceptable
> > > performance in Debug mode at least.
> > 
> > Did you know that cmake provides a "RelWithDebInfo" CMAKE_BUILD_TYPE
> > too?
> > 
> > :)
> > 
> > -- rex
> > 
> > > On Samstag, 4. Februar 2012 07:23:26 Rex Dieter wrote:
> > >> While working on some marble packaging for fedora today, noticed
> > >> that it hard-codes -O3 optimizations in several CMakeLists.txt. 
> > >> Is there some good reason to do this override of the usual
> > >> (Release mode or otherwise) optimization flags?  I checked git
> > >> history and couldn't find anything mentioning this.
> > 
> > _______________________________________________
> > kde-edu mailing list
> > kde-edu at mail.kde.org
> > https://mail.kde.org/mailman/listinfo/kde-edu
> 
> _______________________________________________
> kde-edu mailing list
> kde-edu at mail.kde.org
> https://mail.kde.org/mailman/listinfo/kde-edu