[PPL-devel] Compile error on last versions

[Roberto Bagnara]

Dario Meloni wrote:
> That was helpfull. I have installed gcc 4.2.2 and had no problem on  
> that. To test the my environment works correctly i am trying to  
> compile version 0.9 and then i'll try the beta version which i need.
> 
> I had some problems with the recognition of the gmp version. The  
> script wasn't able to determine the right version stating that mine  
> was prior of 4.1.3 and had the cxx interface disbled, i have version  
> 4.2.2.
> So i have deleted the test in the file ac_check_gmp.m4 and made it  
> pass in any case, this was working with gcc 4.0.1 but now i get this  
> error
> 
> checking for shared library run path origin... done
> checking how to link with libgmp... -L/opt/local/lib -lgmp
> checking how to link with libgmpxx... -lgmpxx -L/opt/local/lib -lgmp
> checking for the GMP library version 4.1.3 or above... yes  <-- hacked  
> test
> checking for mp_limb_t... yes
> checking size of mp_limb_t... configure: error: cannot compute sizeof  
> (mp_limb_t)
> 
> GMP is installed with a 32bit ABI and with --enable-cxx flag in /opt/ 
> local
> I am using this configure script (gmp and gcc are installed in the / 
> opt/local path)
> ./configure --prefix=/opt/julia --with-libgmpxx-prefix=/opt/local -- 
> with-cxx=/opt/local/bin/g++-mp-4.2

Dario,

I assume you have compiled GMP with the same compiler that you are
using to compile the PPL.  My suggestion is to

1) download http://www.cs.unipr.it/ppl/Download/ftp/snapshots/ppl-0.10pre14.tar.bz2
2) configure with your favorite options (BTW, is your C++ compiler executable
    really named `g++-mp-4.2'?);
3) if it fails locating GMP, please send me (compressed) the config.log
    file created by configure.

> I think i'll give up installing the library on mac os x, i can think  
> where the error is, if in gmp, the compilers or the library

It is a rather common experience for us that people have various problems
with Mac OS X.  Unfortunately, not having access to such machines, we
can only rely on people telling us exactly what happens and (possibly)
about the peculiarities of such systems.  However, I suggest we make
another attempt before giving up.
All the best,

     Roberto

-- 
Prof. Roberto Bagnara
Computer Science Group
Department of Mathematics, University of Parma, Italy
http://www.cs.unipr.it/~bagnara/
mailto:bagnara at cs.unipr.it