On Tue, 27 Mar 2018 15:08:31 +0200 Mason <slash.tmp@xxxxxxx> wrote: M> On 23/03/2018 18:47, Vadim Zeitlin wrote: M> M> > I'm running into a rather crazy problem that I can't minimize enough to M> > create a bug report for, so I'd like to ask here if anybody might have any M> > ideas about it by chance. The problem is that the behaviour of some piece M> > of C++ code changes, when compiling using MinGW g++ 7.2 (20180123 version M> > from Debian Buster) and "-O2", depending on whether both of -Wall and M> > -Woverloaded-virtual command line options are specified or not. M> M> That is an interesting problem ^_^ That's a polite way of putting it :-) M> I wanted to try reproducing, but it seems that Ubuntu only packages old M> versions of mingw-w64 (6.3 in 17.10; 5.3 in 16.04 LTS) M> M> Do you know if your issue occurs with older versions of mingw-w64? No, I don't think so. We did use 6.3 for this project before and didn't have any such problems with it and my test case works just fine with it too. M> In my opinion, if the problem does not occur on "vanilla" gcc, then M> the issue might lie with the mingw-w64 specific bits. It seems it's even worse than that as mingw-w64 from Fedora doesn't seem to have any problems with the test case neither. So it might be a Debian problem, and not a gcc one, after all. But looking at Debian-specific changes at https://anonscm.debian.org/cgit/collab-maint/gcc-mingw-w64.git/tree/debian/patches/ I really don't see anything even remotely related, so I still have no idea what's going on here. Regards, VZ
Attachment:
pgp1FY8nHFp_4.pgp
Description: PGP signature
