Axel Simon wrote:
Roberto Bagnara wrote: there is now a snapshot of the PPL 0.6 development branch in
http://www.cs.unipr.it/ppl/Download/ftp/snapshots/ppl-0.6pre2.tar.bz2
In the C interface you will find the function ppl_Polyhedron_maximize() that does what you need (and more). Beware: this new code is almost untested. Please let us know how it goes.
Ok, thanks, I'll give it a go.
The snapshot mentioned above had bugs and an interface for the new mazimization and minimization methods that I did not like. A new snapshot is in
http://www.cs.unipr.it/ppl/Download/ftp/snapshots/ppl-0.6pre3.tar.bz2
P.S. Any progress in debugging the ConSys iteration problem?
No, not yet. There reason might be that Haskell uses a pre-installed gmp package which is version 4.0.1. I compiled PPL against 4.2.1. Thus PPL used the headers and gmpxx.a of the newer version while the gmp.a is from Haskell. I know that it's asking for trouble, but the binary representation of mpz's haven't changed and I don't get any link errors. I'll try to build PPL against version 4.0.1 now. Is that possible?
Unfortunately not: we require GMP 4.1.2 (the latest release, as of today).