Hi Gustavo, We fixed the bug but a new release might take one or two more days.
We went through all changes in 3.0.0 again and checked whether they are covered by our tests (which they were not, of course. But that's fixed now). Retesting everything will take a day or two. So far, Gecode 3.0.* has been a total disaster and hence we decided to carefully go through everything again. The problem was really in our testing methodology: we did not check carefully enough that the new optimizations and in particular their bordercases are covered. All bugs discovered had in common that they were uncovered optimizations. Cheers and thanks Christian -- Christian Schulte, www.it.kth.se/~cschulte/ -----Original Message----- From: users-boun...@gecode.org [mailto:users-boun...@gecode.org] On Behalf Of Gustavo Gutierrez Sent: Tuesday, March 24, 2009 6:45 PM To: gecode list Subject: [gecode-users] minimodel question Hello, After migrating my code to gecode 3.0, I am having a problem when posting a constraint on a boolean relation. The file attached compiles with gecode 3.0 but when executing it I got a bus error (a segmentation fault, I think). The file contains mostly a simplification of my code but it is good enough to exhibit the behavior. The only constraint I am posting is: post(*this, tt(imp(~(x == 1), ~(y != 1)))); which seems to be valid according with the specification presented in modeling.pdf at page 54. Am I doing something wrong?, I am not familiar with the minimodel part of gecode at all, so maybe this is produced because I am doing something stupid. Thanks in advance, Gustavo _______________________________________________ Gecode users mailing list us...@gecode.org https://www.gecode.org/mailman/listinfo/gecode-users
