--- layout: fc_discuss_archives title: Message 7 from Frama-C-discuss on May 2013 ---
Once again hi :) I'm having more and more problems with the WP plugin in this new version of frama-c After applying the previous patches suddenly i cant use the 'unsafe casts' option in one of my functions I get the following error: Current source was: ungetc.c:40 The full backtrace is: Raised at file "hashtbl.ml", line 353, characters 18-27 Called from file "hashtbl.ml", line 361, characters 22-38 Unexpected error (Not_found). code to reproduce the error: http://pastebin.com/AjDHf2jJ Thanks 2013/5/7 Cristiano Sousa <cristiano.sousa126 at gmail.com> > Hi, > > it's interesting that you mention you can prove it with z3. In my > experience with the WP plugin, I cant prove anything with Z3 (4.3.2) save > one or other random PO. > > I usually rely on CVC3 (2.4.1) > > > 2013/5/6 Lo?c Correnson <loic.correnson at cea.fr> > >> Thanks for the report. >> >> It's seems to be related to a problem in alt-ergo with the record theory >> : the addition of the helper assertions trigger more fact on record >> comparisons. >> Without these helpers, alt-ergo give up very quickly (less than 1s), >> which can be a sign of incompleteness. >> Goals are proven with Z3, though. >> >> Regarding cvc3, it is indeed a why3 issue, if any. >> L. >> >> >> Le 4 mai 2013 ? 19:07, Cristiano Sousa a ?crit : >> >> Hi, >> >> I'm having a odd problem with proving a simple contract with the WP >> plugin. I'm forced to insert the assertions HELPER and HELPER2 in order to >> discharge all POs, when in the Oxygen release I didn't need this kind of >> assertions. Without HELPER, BOUND cannot be proved and without HELPER2 the >> post condition cannot be proved. Note: the rest of the assertions were >> inserted to help me pinpoint the problem. >> >> http://pastebin.com/p5AdPN2n >> >> Also, in the previous release of frama-c there was a problem with cvc3 >> processes not being killed. This problem now happens more ofter but i'm >> guessing this is a problem related to why and not frama-c. >> >> Thanks! >> _______________________________________________ >> Frama-c-discuss mailing list >> Frama-c-discuss at lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/frama-c-discuss >> >> >> >> _______________________________________________ >> Frama-c-discuss mailing list >> Frama-c-discuss at lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/frama-c-discuss >> > > > > -- > Cumprimentos, > Cristiano Sousa > -- Cumprimentos, Cristiano Sousa -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gforge.inria.fr/pipermail/frama-c-discuss/attachments/20130509/747af05e/attachment.html>