On Fri, Dec 19, 2008 at 11:32:22AM -0600, Peter O'Gorman wrote: > Hi, > > When building qt-3.3.8 and wxGTk on Tru64 UNIX 5.1 > (alphaev67-dec-osf5.1) with gcc-4.2.4, we got linker failures about > duplicate non-virtual thunks, e.g. from qt: > /usr/ccs/bin/ld: > .obj/release-shared-mt/qmotifdnd_x11.o: non-virtual thunk to > QDragMoveEvent::~QDragMoveEvent(): multiply defined > .obj/release-shared-mt/qmotifdnd_x11.o: non-virtual thunk to > QDragMoveEvent::~QDragMoveEvent(): multiply defined > .obj/release-shared-mt/qmotifdnd_x11.o: non-virtual thunk to > QDragEnterEvent::~QDragEnterEvent(): multiply defined > .obj/release-shared-mt/qmotifdnd_x11.o: non-virtual thunk to > QDragEnterEvent::~QDragEnterEvent(): multiply defined > > This appears to be because the non-virtual thunks are not output weak on this > platform, so, those that appear in multiple objects will cause linker errors > like this (in the above case, there are _ZThn16_N14QDragMoveEventD0Ev and > _ZThn16_N14QDragMoveEventD1Ev in both qdnd_x11.o and qmotifdnd_x11.o). >
We did a regression hunt, since we knew it worked in 3.4.3 and failed in 4.2.4. The problems start with r109149 - this patch: http://gcc.gnu.org/ml/gcc-patches/2005-12/msg01905.html With this patch reverted we can build qt on Tru64 with gcc-4.2.4. I have not yet built trunk on this platform, when I tired bootstrap failed, will try again now, but I don't believe this has changed much from 4.2.4. Is this patch itself to blame, or does it simply expose a latent bug? Will reverting this patch in the gcc that we ship cause any real problems? Thanks, Peter -- Peter O'Gorman po...@thewrittenword.com
