[PPL-devel] Re: minimized_constraints() problem in 0.4.2

[Roberto Bagnara]

Dear Ken,

it seems you have found a genuine bug in the PPL.
We are investigating and we hope to be back to you with a workaround soon.
All the best

      Roberto


kmixter at longshot.com wrote:
> Hi,
> 
> I'm doing a first pass at integrating PPL with the Action Language
> Verifier under Dr. Tevfik Bultan at UC Santa Barbara.  I've run into what
> appears to be a problem in the most recent 0.4.2 release. Please let me
> know as soon as possible if I'm doing something wrong or if it's a genuine
> problem.  I'd like to have an integration I can show next week.
> 
> The problem is that minimized_constraints is crashing when it gets down
> into strongly_minimize_constraints:
> 
> if (topologically_closed || !strict_inequals_saturate_all_rays) {
>     assert(cs_rows < cs.num_rows());
>     ...
> 
> The condition topologically_closed is true, but cs_rows == cs.num_rows().
> 
> Here is the testcase which reproduces the condition:
> -----------------------------------------
> #include "ppl.hh"
> using namespace Parma_Polyhedra_Library;
> 
> int main()
> {
>   Variable x2(0), x3(1), x5(2);
>   NNC_Polyhedron poly(3, Polyhedron::UNIVERSE);
> 
>   poly.add_constraint(x5 > x2);
>   poly.add_constraint(x3 >= x2);
>   poly.add_constraint(x2 + 1 > x3);
> 
>   poly.minimized_constraints();
>   return 0;
> }
> ------------------------------------------
> 
> Thanks,
> Ken Mixter

-- 
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