From: Ruben V. B. <van...@gm...> - 2012-08-16 18:34:59
|
2012/8/11 Martin Mitáš <mi...@mo...> > > Dne 11.8.2012 19:06, Kai Tietz napsal(a): > > No, > > > > please ignore this advice from Earnie. It is just partial true and > > also misleading. Indeed the decoration of symbols with an '@' is > > either caused by the stdcall-convention, or by the fastcall. > > Nevertheless there is a way to have in export-table no '@' decoration. > > Just use for 32-bit the --kill-at option of ld (usable by gcc's > > frontend via -Wl,--kill-at option). > > > > For 64-bit this option isn't necessary, as for 64-bit there is just on > > calling-convention in fact, and it has no special decoration. > > > > Hope this answers your question. > > I am aware of this and I already do so in my DLL project. > > In short this thread is just a report to Ruben that the clang thing has > a bug. > When making a DLL using __declspec(dllexport) and --kill-at when linking > then > (1) compiling with gcc + linking with gcc -Wl,--kill-at does the right > thing, > (2) compiling with clang + linking with gcc -Wl,--kill-at fails, > > i.e. that compiling with clang is not equivalent replacement for gcc as > Ruben has suggested, at least in this particular use case. > I never meant to say it was as solid and full-featured as GCC on Windows. What I meant was that the program you call ("clang"/"clang++", i.e. the "compiler driver") can be used as a drop-in replacement for "gcc"/"g++", unlike for example Visual Studio's "cl". Now that I have fixed my 4.7.1-1-release debacle, and actually have useful Linux64 builds, I can check out this problem. Ruben > > Regards, > Morous > > > > ------------------------------------------------------------------------------ > Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. Discussions > will include endpoint security, mobile security and the latest in malware > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Mingw-w64-public mailing list > Min...@li... > https://lists.sourceforge.net/lists/listinfo/mingw-w64-public > |