[Soot-list] Query regarding SPARK points-to analysis

[Aniya Aggarwal]

Hi Eric,
Thanks a lot for your response. Yes, I am getting the same points-to sets
as you mentioned. I was probably misinterpreting the output of Points-to
analysis.

Regards
Aniya


Thanks & Regards,
Aniya Aggarwal
MT-12034
M.Tech CSE (Data Engineering)

On Mon, Nov 3, 2014 at 2:30 PM, Bodden, Eric <eric.bodden at sit.fraunhofer.de>
wrote:

> Hello Aniya.
>
> I am a bit confused by your interpretation. Points-to sets normally
> contain allocation sites, not variables. Hence, all reaching* methods
> should normally return a set containing alloc sites, in your example in
> particular:
>
> obj: {new PointPoolObject()}
> arr: {new PointPoolObject[POINTS_COUNT];}
> arr.<elements>: {new PointPoolObject()}
>
> Did you try to use Spark's default options. They should give you a result
> like the one above...
>
> Cheers,
> Eric
>
>
> On 02.11.2014, at 17:38, Aniya Aggarwal <aniya1234 at iiitd.ac.in> wrote:
>
> > Hi All,
> > I am trying to do points-to analysis using SPARK framework. For the code
> mentioned below, the analysis yields the following points-to sets for 'obj'
> and 'arr'..
> >
> > arr: {obj}
> > obj: empty points-to set
> >
> > The analysis can detect 'obj' in the points-to set of 'arr' using
> reachingObjectsOfArrayElement() as the elements of the array 'arr'
> points-to 'obj' in the first loop. But in the second loop, even though
> 'obj' points to an element of the array 'arr', the latter is not detected
> in the points-to set of 'obj' while using reachingObjects(Local l).
> >
> > I am using SPARK with the following options.
> >
> > propagator=worklist, simple-edges-bidirectional=false, on-fly-cg=true,
> set-impl=double, double-set-old=hybrid, double-set-new=hybrid
> >
> > Please let me know if this is something that SPARK cannot handle or some
> option is set incorrectly. Kindly suggest an alternative to detect 'arr' in
> the points-to set of 'obj'.
> >
> > //Test code
> > private void testPool( final int POINTS_COUNT){
> > PointPoolObject obj;
> > PointPoolObject[] arr = new PointPoolObject[POINTS_COUNT];
> >
> > for (int i = 0; i < POINTS_COUNT; i++) {
> > obj = new PointPoolObject();
> > arr[i] = obj;
> > }
> > for (int i = 0; i < POINTS_COUNT; i++) {
> > obj = arr[i];
> > }
> > }
> >
> > Thanks & Regards,
> > Aniya Aggarwal
> > MT-12034
> > M.Tech CSE (Data Engineering)
> > _______________________________________________
> > Soot-list mailing list
> > Soot-list at CS.McGill.CA
> > https://mailman.CS.McGill.CA/mailman/listinfo/soot-list
>
> --
> Prof. Eric Bodden, Ph.D., http://sse.ec-spride.de/ http://bodden.de/
> Head of Secure Software Engineering at Fraunhofer SIT, TU Darmstadt and EC
> SPRIDE
> Tel: +49 6151 16-75422    Fax: +49 6151 869-127
> Room B5.11, Fraunhofer SIT, Rheinstraße 75, 64295 Darmstadt
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.CS.McGill.CA/pipermail/soot-list/attachments/20141103/adab427f/attachment-0003.html