Hi, See: https://sourceforge.net/p/podofo/mailman/message/35915862/
On Mon, May 3, 2021 at 2:05 PM Christopher Creutzig <ccreu...@mathworks.com> wrote: > Hi list, > > > > PdfEncodingFactory.cpp uses a broken form of double-checked locking > <https://preshing.com/20130930/double-checked-locking-is-fixed-in-cpp11/> > to initialize its encoding instances. Not a big deal, as these objects > don’t do much; technically, that is a data race and can lead to undefined > behaviour, but realistically, I would be surprised to see anything worse > than a small memory leak, if even that. > It is undefined behaviour in C++. Actually it should cause problems only on weakly ordered processors like ARM. So you should never see it on x64. > > Does PoDoFo require C++11 or newer (where there are simple fixes > available)? Will it ever? > > If podofo cannot depend on C++11 then I would suggest to use "single"-checked locking (just remove the first check): //if(!s_pWinAnsiEncoding) // First check //{ Util::PdfMutexWrapper wrapper( PdfEncodingFactory::s_mutex ); if(!s_pWinAnsiEncoding) // Double check s_pWinAnsiEncoding = new PdfWinAnsiEncoding(); //} return s_pWinAnsiEncoding; Another solution would be to use some additional library that provides atomic primitives for older C++ but I do not think that it would be worth it in this case. > > > > Cheers, > > Christopher > > > > The MathWorks GmbH | Friedlandstr.18 | 52064 Aachen | District Court > Aachen | HRB 8082 | Managing Directors: Bertrand Dissler, Steven D. Barbo, > Jeanne O’Keefe > > > _______________________________________________ > Podofo-users mailing list > Podofo-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/podofo-users >
_______________________________________________ Podofo-users mailing list Podofo-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/podofo-users
