--- layout: fc_discuss_archives title: Message 1 from Frama-C-discuss on October 2009 ---
jun shen a ?crit : > I am interested in the details of slicing, because I might do some > academic work above Frama-C and I have to make sure that the > foundation is what I expected. Interesting. Maybe you can tell us a bit more about the subject you are working on ? > By the way, can you send me the document now? It doesn't matter that > it is in French. In fact, after a short discussion with the team, it appears that this document is not distributed with Frama-C because it doesn't reflect the status of the implementation... It is rather a preliminary document that was made to decide what to do. Sorry that there is no detailed documentation about the implementation, but I would be happy to answer to your questions if I can. > My question is whether you have implemented Susan Horwitz's PDG > algorithm described in her paper. And what are the differences between > your pdg and hers? If I remember well, the PDG is similar to Horwitz one, except for the inter-procedural part, because to handle the function calls, the PDG uses the result of another dependence analysis. You can see the results of this other analysis by using the -deps option of frama-c. Moreover, I don't know if you have seen that you can see the PDG of a function ? If not, you can have a look at an answer I have ealier made on the list : http://lists.gforge.inria.fr/pipermail/frama-c-discuss/2009-September/001403.html Hope this will help, -- Anne.