On Mon, Nov 22, 2010 at 5:51 PM, Scott Hannahs <shann...@users.sourceforge.net> wrote: > > On Nov 22, 2010, at 16:48, Alexander Hansen wrote: > >> On Mon, Nov 22, 2010 at 4:25 PM, Scott Hannahs >>> >>> Which implies that version 4.3.2 of the libgmp package is installed. >>> But this is installed in /sw/lib as libgmpxx.dylib and the >>> libgmp.dylib file is version 3.5.2 >> >> Packages and their libraries needn't have matching versions. > > Ok. So my naive idea is a bit off.... > >>> Did some version update leave the older version around? Should the >>> dependencies for ppl be on libgmpxx4? >> >> We're going to need more info to see what's really going on. Check >> the config.log file in the ppl build directory, and it should show the >> commands that are being run to check for gmp, as well as the low-level >> output from them. > > Ok. That is interesting..... Illegal instruction??? Something seems wrong > with configure and it is trying to execute a command that doesn't exist? > Somehow the variable is not set? > > It was created by the Parma Polyhedra Library configure 0.10.2, which was > generated by GNU Autoconf 2.63. Invocation command line was > > $ ../configure --prefix=/sw --mandir=/sw/share/man --infodir=/sw/share/info > --disable-static --enable-pch > --with-java=/System/Library/Frameworks/JavaVM.framework/Home CFLAGS=-g -O2 > -pipe -fpch-prep > rocess -fPIC CXXFLAGS=-g -O2 -pipe -fpch-preprocess -fPIC > CPPFLAGS=-I/System/Library/Frameworks/JavaVM.framework/Home/include > --enable-check=thorough -C > > ........ > > configure:15513: checking how to link with libgmp > configure:15902: result: -lgmp > configure:15942: checking how to link with libgmpxx > configure:16331: result: -lgmpxx -lgmp > configure:16375: checking for the GMP library version 4.1.3 or above > configure:16495: g++ -o conftest -Winvalid-pch -g -O2 -frounding-math -g > -O2 -pipe -fpch-preprocess -fPIC > -I/System/Library/Frameworks/JavaVM.framework/Home/include -I/sw/include > -L/sw/lib conftest. > cpp -lgmpxx -lgmp >&5 > configure:16499: $? = 0 > configure:16505: ./conftest > ../configure: line 16507: 9561 Illegal instruction ./conftest$ac_exeext > configure:16509: $? = 132 > configure: program exited with status 132 > > That is all I got. Somehow the C++ routine that is supposed to call the > library to test the version doesn't create an executable but g++ returns > with no error? I have the stock gcc installed through developer tools > % g++ --version > powerpc-apple-darwin9-g++-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5493) > > -Scott > > >
I didn't reproduce the issue on my 10.5/PowerPC setup. One possibility occurs to me, though. If the machine is a G4, and your gmp was built on a G5 (e.g. if you're using an unofficial binary distribution), errors such as this show up pretty often, since gmp does CPU optimizations. -- Alexander K. Hansen akh AT finkproject DOT org Fink User Liaison and Documenter ------------------------------------------------------------------------------ Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! Tap into the largest installed PC base & get more eyes on your game by optimizing for Intel(R) Graphics Technology. Get started today with the Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. http://p.sf.net/sfu/intelisp-dev2dev _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net http://news.gmane.org/gmane.os.apple.fink.devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel
