Elnatan Reisner wrote: > One result of this, however, is that something like '*p > && 0' will become simply '0', and if p was null, this means that the > transformed program does not crash, whereas the original would have. I > would argue that this an acceptable transformation, but I'm curious to > hear if others disagree.
Well, CIL already does other transformations that remove bugs from programs (for instance, moving block-local variables to the top of the function can transform an undefined behavior into defined behavior). It is something that has to be taken into account when using CIL as a front-end for a correct static analyzer. This is what some of us do, but for this and other reasons, our front-end is a proper fork of CIL now. Pascal ------------------------------------------------------------------------------ Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev _______________________________________________ CIL-users mailing list CIL-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/cil-users
