On Thu, Aug 16, 2012 at 07:57:09PM +0200, laurent pagani wrote: > Jack, > > 1) I am not sure to understand what "a clean bootstrap of fink" is and how to > do it. Going to fink….org, it says there is no binary to install fink for > 10.7 and it looks painful to restart from older versions and do some upgrade. > So there is something I don't understand here. I am no guru, sorry.
How did you ever get fink installed on Lion then? There is no supported upgrade path from fink 10.6 to 10.7. Fink on 10.7 was a clean break with its own tree for 10.7 forward that allowed us to transition to x86_64-only and cleanly adapt to building everything under the clang compilers. If you had x86_64 fink installed under 10.6 and upgraded that machine to 10.7, you need to move /sw aside and start over. http://www.finkproject.org/download/10.7-upgrade.php?phpLang=en Roughly the steps are... 1) Download http://downloads.sourceforge.net/fink/fink-0.34.2.tar.gz 2) Move aside sw with the command sudo mv /sw /sw.old 3) Uncompress and bootstrap fink with the commands tar -zxvf fink-0.34.2.tar.gz cd fink-0.34.2 ./bootstrap (you can just hit return at all of the prompts to take the defaults for the bootstrap process). 4) I would then setup fink for cvs access and build gcc47 as follows /sw/bin/init.csh or /sw/bin/init.sh depending on the shell used fink selfupdate-cvs fink update-all fink install gcc47 > > 2) why /sw/bin/gfortran does not work when calling the actual > /sw/lib/gcc4.7/bin/gfortran-fsf-4.7 does while the former is only a symlink > to the latter ? > > Laurent > > > Le 16 août 2012 à 19:07, Jack Howarth a écrit : > > > On Thu, Aug 16, 2012 at 06:30:41PM +0200, laurent pagani wrote: > >> Hum, > >> > >> Looks like there has been a double confusion. You meant fink reinstall > >> gcc46 and not 47 to indeed return to the 46 version (which answers part of > >> my question). > >> So I am now trying to reinstall gcc46 to see if it works anew. > > > > > > Laurent, > > Assuming your current deb packages are built correctly and haven't been > > purged from the system, > > 'fink reinstall gcc46' will simply reinstall the existing gcc46 deb package > > and not rebuild it. > > Note that building fink while a populated /usr/local exists is totally > > unsupported. You might > > consider moving /usr/local aside as well as /sw and doing a clean bootstrap > > of fink. If you can't > > build a usable gcc47 package from that then your problem lies elsewhere. > > Jack > > > >> > >> Laurent > >> > >> Le 16 août 2012 à 17:57, Jack Howarth a écrit : > >> > >>> On Thu, Aug 16, 2012 at 05:36:21PM +0200, laurent pagani wrote: > >>>>> What do you get for... > >>>>> > >>>>> howarth% /sw/bin/gfortran -v > >>>>> Using built-in specs. > >>>>> COLLECT_GCC=/sw/bin/gfortran > >>>>> COLLECT_LTO_WRAPPER=/sw/lib/gcc4.7/libexec/gcc/x86_64-apple-darwin12.0.0/4.7.2/lto-wrapper > >>>>> Target: x86_64-apple-darwin12.0.0 > >>>>> Configured with: ../gcc-4.7-20120710/configure --prefix=/sw > >>>>> --prefix=/sw/lib/gcc4.7 --mandir=/sw/share/man > >>>>> --infodir=/sw/lib/gcc4.7/info > >>>>> --enable-languages=c,c++,fortran,lto,objc,obj-c++,java --with-gmp=/sw > >>>>> --with-libiconv-prefix=/sw --with-ppl=/sw --with-cloog=/sw > >>>>> --with-mpc=/sw --with-system-zlib --x-includes=/usr/X11R6/include > >>>>> --x-libraries=/usr/X11R6/lib --program-suffix=-fsf-4.7 > >>>>> --enable-cloog-backend=isl > >>>>> Thread model: posix > >>>>> gcc version 4.7.2 20120710 (prerelease) (GCC) > >>>>> > >>>> > >>>> Monte-Carlo>/sw/bin/gfortran -v > >>>> Utilisation des specs internes. > >>>> COLLECT_GCC=/sw/bin/gfortran > >>>> Target: x86_64-apple-darwin11.4.0 > >>>> Configur'e avec: ../gcc-4.7.1/configure --prefix=/sw > >>>> --prefix=/sw/lib/gcc4.7 --mandir=/sw/share/man > >>>> --infodir=/sw/lib/gcc4.7/info > >>>> --enable-languages=c,c++,fortran,lto,objc,obj-c++,java --with-gmp=/sw > >>>> --with-libiconv-prefix=/sw --with-ppl=/sw --with-cloog=/sw > >>>> --with-mpc=/sw --with-system-zlib --x-includes=/usr/X11R6/include > >>>> --x-libraries=/usr/X11R6/lib --program-suffix=-fsf-4.7 > >>>> --enable-cloog-backend=isl > >>>> Mod`ele de thread: posix > >>>> gcc version 4.7.1 (GCC) > >>>> Using built-in specs. > >>>> > >>>>> Also, have you tried pruning your .cshrc or .bashrc so that it only > >>>>> contains > >>>> > >>>> running tcsh, so > >>>> > >>>>> test -r /sw/bin/init.csh && source /sw/bin/init.csh > >>>> > >>>> I removed my .cshrc and included only the line above. The problem > >>>> remains. > >>>> > >>>>> > >>>>> I have never heard of this problem before. > >>>>> Jack > >>>>> ps Have you tried... > >>>>> > >>>>> /sw/lib/gcc4.7/bin/gfortran-fsf-4.7 -o prov prov.f > >>>> > >>>> This one works!!! > >>> > >>> Please keep in mind that the gcc4x split-off packages all conflict and > >>> you can't go in and just randomly change the symlinks that they contain. > >>> Unless you purged your built debs, you should have been able to simply > >>> done... > >>> > >>> fink reinstall gcc47 > >>> > >>> Note that the gcc4x-compiler packages provide access to the compilers > >>> through their suffixed names (gcc-fsf-4.7, g++-fsf-4.7, gfortran-fsf-4.7) > >>> but only the conflicting gcc4x split-off's provide the overlapping common > >>> names (gcc-4, g++-4, gfortran, etc). If you had gcc46 installed and then > >>> only installed gcc47-compiler/gcc47-shlibs and then tried to manually > >>> redirect the symlinks in gcc46 to those in gcc47, you likely didn't > >>> catch all of the instances. It seems like a huge amount of effort in > >>> order to avoid reinstalling the matching gcc47 split-off package. > >>> > >>>> > >>>>> Also is there anything unusual about your configuration (such as are > >>>>> you using case-sensitive HFS volumes or splitting the /sw directory > >>>>> across volumes due to lack of disk space)? > >>>> > >>>> neither of these. > >>>> > >>>> Laurent > >>>> > >>>> LERMA, UMR8112 du CNRS > >>>> Observatoire de Paris > >>>> 75014 Paris > >>>> France > >>>> Tel: (33) 1 40 51 20 13 > >>>> Fax: (33) 1 40 51 20 02 > >>>> e-mail: laurent.pag...@obspm.fr > >>>> http://aramis.obspm.fr/~pagani > >>>> > >>>> "Falling leaves are a good sign. It's when you see them jumping back > >>>> onto the trees that you're in trouble" Lucy van Pelt (C. Schulz) > >> > >> Laurent Pagani > >> LERMA, UMR8112 du CNRS > >> Observatoire de Paris > >> 75014 Paris > >> France > >> Tel: (33) 1 40 51 20 13 > >> Fax: (33) 1 40 51 20 02 > >> e-mail: laurent.pag...@obspm.fr > >> http://aramis.obspm.fr/~pagani > >> > >> "Falling leaves are a good sign. It's when you see them jumping back onto > >> the trees that you're in trouble" Lucy van Pelt (C. Schulz) > > Laurent Pagani > LERMA, UMR8112 du CNRS > Observatoire de Paris > 75014 Paris > France > Tel: (33) 1 40 51 20 13 > Fax: (33) 1 40 51 20 02 > e-mail: laurent.pag...@obspm.fr > http://aramis.obspm.fr/~pagani > > "Falling leaves are a good sign. It's when you see them jumping back onto the > trees that you're in trouble" Lucy van Pelt (C. Schulz) ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net List archive: http://news.gmane.org/gmane.os.apple.fink.devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel
