[CLAM] fftw library (double precision) not found

David Garcia Garzon david.garcia at iua.upf.es
Fri May 16 09:44:32 PDT 2003

> > I'm installing on a Linux version 2.4.18-27.7.x, Red Hat Linux 7.3
> > 2.96-112, gcc version 2.96 20000731.
> It's possible that this gcc/g++ version gives some problem to you,
> since is not an official GNU's release. The fully supported gcc/g++
> version is 2.95, and partially 3.2...
> Little to do about this, since I'm compiling on one of the
> university's computers.

Not at all. RH7.3 comes also with g++ 3.0. You should do some trick and 
download the externals libraries that we have already compiled for RH7.3 and 

The trick is to create a 'bin' directory in your home with links to the 3.0 
versions of gcc binaries (g++, gcc and any other it demands you). That 
binaries have a '3.0' sufix, make the links without that sufix, for example:
	ln -s /usr/bin/gcc-3.0 ~/bin/gcc
	ln -s /usr/bin/g++-3.0 ~/bin/g++
	ln -s /usr/bin/cpp-3.0 ~/bin/cpp
  ... I think that there was another binary but if it complaints just do 
another link for it.
Then put that bin directory at the beggining of your PATH.
	export PATH=~/bin:$PATH

Then you can checkout from our cvs, the following modules containing the 
external libraries compiled for the gcc3.0 for RH7,3
I have not found any version for the fftw library but, anyway, your problems 
as Albert said, are about configuring CLAM for 'floats' and fftw for 
'doubles' or viceversa. Check the 'configure' options for CLAM.

Ask for any other problem you have. I hope you will enjoy CLAM. :-)
Happy coding.

David García Garzón <david.garcia at removespam.iua.upf.es>
Phone: 034 93 542 21 99
Music Technology Group, Institut Universitari de l'Audiovisual
Universitat Pompeu Fabra

More information about the clam-users mailing list