I wrote: > FWIW, mono compiles quite nicely with gcc 2.96, except for a few C > twists. However, make check freezes when it tries to verify > ByteFX.Data.dll. I'll seek your help again as soon as I can figure out > what's happening.
Astonishingly, that seems to be a bug in glib's memory allocation. I didn't file a bug on bugzilla.gnome.org (yet); however, I reported it on the mailing list, see: http://www.nabble.com/Astonishing-allocation-bug-in-glib-2.16.4-compiled-with-gcc-2.96-to18428394.html Why didn't that come out before? Or am I missing something? I configured mono as ./configure \ --with-gnu-ld\ --enable-minimal=profiler,reflection_emit,large_code,com,ssa\ --with-xen_opt=no\ --with-moonlight=no\ --with-x=no\ --with-sigaltstack=no Is excluding SSA uncommon enough to justify the possibility that a bug in glib has remained unnoticed until now? I get the bug on "Compiling 120 ByteFX.Data.MySqlClient.MySqlCommandBuilder" in make check. Everytime. That is on the mono-107638 snapshot that I retrieved a few days ago. In case someone is willing to have a try, I've saved a copy in http://www.tana.it/sw/mono/mono-107638.tar.bz2 while glib-2.16.4 is still current on http://ftp.gnome.org/pub/GNOME/sources/glib/2.16/ Any hint is welcome TIA Ale -- please cc to me your response _______________________________________________ Mono-list maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-list
