https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123976
--- Comment #95 from Peter Dyballa <Peter_Dyballa at web dot de> --- (In reply to Iain Sandoe from comment #94) > (In reply to Peter Dyballa from comment #93) > > this would seem to suggest that one of the other patches you have applied > has somehow broken things ? It's not impossible, I'd write. I am waiting to see that makeinfo successfully does its job on the patched gcc/doc/ifn.texi file. Then I'd kill build, make distclean and extract the original three .MK files to see whether the changes applied there are the culprits. When they're not, then I would extract GCC 16.1 again into a clean directory and apply the patches inside Apple's GNU Emacs to have a stable trace of what the patch programnme is performing. Could be there is a hint of some file that is related to some Makefile. Then I would sort out this patch. I prefer to believe that someone forgot some rule to make make think of the strange case darwin on PowerPC. A 64 bit Mac with (Snow) Leopard would be nice to have as an alternative to my 32 bit PowerPC Mac. Besides this, is unwind-dw2_s.o also made on x86_64? It's this object file that needs through its source file at least one of the unmade symlinks.
