[Soot-list] Call Graph issues and advice needed

[Henddher Pedroza]

Thank you Bernhard,

Rohan Padhye noted this:

main() {
   e1();
   e2();
}

e1() {
   x = 10;
}

e2() {
   use x;
}

If you use a stub main then your analysis might conclude that x = 10 is constant at e2(), but that is not the case if the program starts at e2(). In IFDS terms, you really want results of thread starting at e2 to compute reachability from the zero node only, not all nodes that are true at the exit of the thread starting at e1.

You would probably be better off performing multiple IFDS solutions, one for each entry point, and then merging results for each program point across all such instances if you want to do some optimizations.

If I understood correctly, one would need to generate entry points per thread and compute IFDS on each set.

Thanks again.

On Apr 27, 2013, at 9:06 AM, Bernhard Berger <berber at tzi.de> wrote:

> Hi,
> 
>> As for the non-static entry point stuff, I'm going to ask Bernhard to 
>> elaborate on that, because he dealt with it a whole lot more than I did. 
>> IIRC it was because of the this pointer.
>> 
>> My guess is that you could patch Spark to deal with it by using CHA to 
>> resolve the this pointer whenever you have a non-static entry point. 
>> After all, you don't have any extra information at that point to know 
>> what you're pointing to…
> 
> a year or two ago I tried to use the entry point mechanism to get a call graph of android components. I used probe to dump and check the results of the cg construction and observed that calls where missing. AFAIR there were entry methods that called another method within the class (this.foobar()) and they were missing. Furthermore there were problems with calls on the parameters. I never debugged it but I inferred that the reason is that there is information missing on the concrete type of this and the parameters. At this point I switched to entry point generation (creating a dummy main, creating the objects and parameters and call the methods).
> 
> Regards,
> Bernhard
> 
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.mcgill.ca/pipermail/soot-list/attachments/20130427/d37b0acd/attachment.html