Am Mittwoch, 19. Juli 2006 19:44 schrieb Bennett Helm: > With most recent 1.5svn, I get a crash on saving an existing > document. (It doesn't occur when saving a new document or saving-as a > new document, but it does occur when saving-as and overwriting an > existing document.) > > See the backtrace, below. > > Bennett > > -------- > > Program received signal SIGABRT, Aborted. > 0x9004802c in kill () > (gdb) bt > #0 0x9004802c in kill () > #1 0x9012dfb4 in abort () > #2 0x00176424 in __cxxabiv1::__terminate () at /usr/include/gcc/ > darwin/3.3/c++/bits/stl_alloc.h:656 > #3 0x00176468 in std::terminate () at /usr/include/gcc/darwin/3.3/c+ > +/bits/stl_alloc.h:656 > #4 0x00144c34 in __cxa_throw () at /usr/include/gcc/darwin/3.3/c++/ > bits/stl_alloc.h:656 > #5 0x001b4ad4 in boost::filesystem::copy_file ([EMAIL PROTECTED], > [EMAIL PROTECTED], noclobber=3221181360) at ../../boost/boost/detail/ > sp_counted_base_gcc_ppc.hpp:167
This part of the backtrace looks strange. Why is noclobber not true or false? Why is boost::filesystem::copy_file mentioned, but the error is in boost/boost/detail/sp_counted_base_gcc_ppc.hpp? Does it work if you compile Lars' boost 1.34 snapshot branch? Or, if you don't want to do that, if you replace boost/boost/detail/sp_counted_base_gcc_ppc.hpp by http://boost.cvs.sourceforge.net/*checkout*/boost/boost/boost/detail/sp_counted_base_gcc_ppc.hpp?revision=1.5 ? Georg
