Frama-C-discuss mailing list archives

This page gathers the archives of the old Frama-C-discuss archives, that was hosted by Inria's gforge before its demise at the end of 2020. To search for mails newer than September 2020, please visit the page of the new mailing list on Renater.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Frama-c-discuss] Getting a node from its id for a particular pdg

  • Subject: [Frama-c-discuss] Getting a node from its id for a particular pdg
  • From: anne.pacalet at sophia.inria.fr (Anne Pacalet)
  • Date: Wed, 18 Mar 2009 16:44:52 +0100
  • In-reply-to: <1237387890.3508.114.camel@supelec>
  • References: <1237387890.3508.114.camel@supelec>
Jonathan-Christofer Demay a ?crit :
> Let's say I need to get a node from its id for a particular pdg, for
> now, in order to get this, I use Db.Pdg.iter_nodes looking for the
> correct id (I'm not sure it's worth creating a hashtable for this), but
> I was wondering if there were an easier way to do it (I did search but
> found nothing) or if maybe frama-c's dev would consider adding such
> feature.

I can add something like that easily, because it already exists 
internally, but using the nodes key given by [Db.Pdg.node_key] (not the 
id). But I am not sure to understand why you need that :
where do you get the node id from ?

> Also, if I'm talking about this on the ML, it's because I've also
> considered the fact that maybe if I need such feature, it's because I
> went wrong somewhere.

Let's see ;-)

> So just in case, I will tell why I need this. Let's consider this simple
> piece of C code:
> 
> 1.	b += z;
> 	
> 2.	c = toto(b,a);
> 
> 3.	c++;
> 
> 4.	e = toto(c,a);
> 
> For the statement 4, I ask for the corresponding nodes (using
> Db.Pdg.find_simple_stmt_nodes), then I ask for all its dependencies
> (using Db.Pdg.all_dpds). I get statements 2 and 3.

Can you provide more information about the 'toto' function  ?
Because I suspect that if [Db.Pdg.all_dpds] doesn't return the node 1,
it is because [c] doesn't depend on [b] in the call.

> But since statement 2 is a call, from which the result might depend (in
> worst case scenario) on all it's dependencies, I want to add those to
> the dependencies of the statement 4.

The dependencies in the call are given by the value analysis,
so normally, it is a sur-approximation at all that can possibly happen.

Do you know that for simple cases like that, you should consider to use
[Db.Pdg.extract] to generate a .dot file so you can have a look at the 
graph. I think it helps a lot at the beginning...

Hope this help, and tell me it you still want :
   Db.Pdg.find_node : node_key -> node.

-- 
Anne Pacalet  -
INRIA - 2004, route des Lucioles BP.93 F-06902 Sophia Antipolis Cedex.
Tel : +33 (0) 4 9715 5345