Sergio Rojas wrote: > I though that "--with-mpfr=$MPFRDIR and --with-gmp=$GMPDIR" > were taken care of that. Well, now the error is as follows Those options tell the compiler and link editor where to find the files during compilation and linking. They don't tell the operating system's dynamic loader where to find the libraries at runtime. > (after setting LD_LIBRARY_PATH to the respectives libraries dir > and also LDFLAGS="-L$MPFRDIR -R$MPFRDIR -L$GMPDIR -R$GMPDIR"): You really shouldn't have to set LDFLAGS. Just the --with options alone should take care of it. > & if [ -f amd64/libgcc_s.so.1 ]; then mv -f amd64/libgcc_s.so.1 amd64/libgcc_s.so.1.backup; else true; fi && mv amd64/libgcc_s.so.1.tmp amd64/libgcc_s.so.1 && ln -s libgcc_s.so.1 amd64/libgcc_s.so > ld: fatal: unwind table: file /home/srojas/MySoftware/my_install/./gcc/amd64/crtend.o: section .eh_frame: bad cie version 0: offset 0xfffffd7fffdfa200 This is a bug in Sun's linker, see <http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33100> which includes a workaround (which will be in 4.3.3.) > By the way, is there a way to track at which stage of the compilation the script is stopping? If you watch the build messages you'll regularly see e.g. "Configuring stage 1 in i386-pc-solaris2.10/libgcc" scroll by, so just scroll up from the error to the first such line. (It helps to log the output and then you can grep instead.) You can also 'cat stage_current' which should tell you the current state of the tree at any given time. Brian
