Regression from 1.4.1 to 1.4.2: Segmentation fault with reified constraints

Thierry Martinez <[email protected]> Sat, 23 Feb 2013 22:04:50 +0100

Newsgroups	gmane.comp.gnu.prolog.bugs
Message-ID	<CA+MZgJzC4++YX-cJt9aBNVcv4_k-0iLv5FO9V79RjaP9bfNTEg@mail.gmail.com>

Hi!

The following program leads to a segmentation violation both under
Windows and Linux with GNU Prolog 1.4.2 (“here” is printed, the
segmentation violation occurs on the last line). This program
terminates correctly with GNU Prolog 1.4.1.

:- initialization(a).

a :-
   L = [A, B, C],
   fd_domain(L, 0, 1),
   B = A,
   C #==> A,
   \+ (
      #\ A #\/ #\ A,
      C = 1
   ),
   fd_domain(D, 0, 1),
   print(here),
   nl,
   D = 1.

Thanks you.
--
Thierry.

_______________________________________________
Bug-prolog mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/bug-prolog