[Soot-list] a problem in the points-to analysis using SPARK

[Marc-Andre Laverdiere-Papineau]

Hello,

I don't have a perfect answer, but I would like to ask you two things.

1) Would you please build from the 'develop' branch and see what you get
2) Paste the settings you are giving for Spark (all the cg.* phase)

Marc-André Laverdière-Papineau
Doctorant - PhD Candidate

On 13-03-14 10:47 PM, Zhang Yufeng wrote:
> I have tried soot2.5 and soot2.3.
> The results are the same.
> The jimple code for the method Use.go() is :
> =======================
> public void go()
>      {
>          Use r0;
>          Container $r1, r2, r4, $r7;
>          Item r3, r5, $r6, $r8;
>          r0 := @this: Use;
>          $r1 = new Container;
>          specialinvoke $r1.<Container: void <init>()>();
>          r2 = $r1;
>          $r6 = new Item;
>          specialinvoke $r6.<Item: void <init>()>();
>          r3 = $r6;
>          virtualinvoke r2.<Container: void setItem(Item)>(r3);
>          $r7 = new Container;
>          specialinvoke $r7.<Container: void <init>()>();
>          r4 = $r7;
>          $r8 = new Item;
>          specialinvoke $r8.<Item: void <init>()>();
>          r5 = $r8;
>          virtualinvoke r4.<Container: void setItem(Item)>(r5);
>          return;
>      }
> ====================
> The points-to set obtained by statement reachingObjects(local) for r2
> (Container c1) is :
> new set: empty
> old set: bits:{15, 20, 25, 45, 473, 796, 6863, 6865}
> AllocNode 15 DEFAULT_CLASS_LOADER in method null,
> AllocNode 20 PRIVILEGED_ACTION_EXCEPTION in method null,
> AllocNode 25 Pair AbstractObject,Any_subtype_of_java.lang.Thread in method null,
> AllocNode 45 Pair AbstractObject,Any_subtype_of_java.security.PrivilegedActionException in method null,
> AllocNode 473 Pair AbstractObject,Any_subtype_of_java.lang.ClassLoader in method null,
> AllocNode 796 Pair AbstractObject,Any_subtype_of_java.lang.Object in method null,
> AllocNode 6863 new Container in method <Use: void go()>,
> AllocNode 6865 new Container in method <Use: void go()>,
> points-to set for r4 (Container c2) is :
> new set: empty
> old set: bits:{15, 20, 25, 45, 473, 796, 6863, 6865}
> AllocNode 15 DEFAULT_CLASS_LOADER in method null,
> AllocNode 20 PRIVILEGED_ACTION_EXCEPTION in method null,
> AllocNode 25 Pair AbstractObject,Any_subtype_of_java.lang.Thread in method null,
> AllocNode 45 Pair AbstractObject,Any_subtype_of_java.security.PrivilegedActionException in method null,
> AllocNode 473 Pair AbstractObject,Any_subtype_of_java.lang.ClassLoader in method null,
> AllocNode 796 Pair AbstractObject,Any_subtype_of_java.lang.Object in method null,
> AllocNode 6863 new Container in method <Use: void go()>,
> AllocNode 6865 new Container in method <Use: void go()>,
> These are the same.
> I guess that this is because some options are not configured properly.
> Best.
> Yufeng
> ------------------------------------------------------------------------
> *发件人：* Quentin Sabah
> *发送时间：* 2013-03-15  03:13:39
> *收件人：* soot-list
> *抄送：*
> *主题：* Re: [Soot-list] a problem in the points-to analysis using SPARK
>  > I am debugging on this example.
>  > I find that the statement
>  > Scene.v().getPointsToAnalysis().reachingObjects(local loc);
>  > gets the same reaching points-to sets for variable 'Container c1' and 'Container c2'.
>  > This is strange.
> It is strange. What is the content of these points-to sets using toString(), and their class type using getClass().toString() ?
> What version of soot are you using? Have you tried different versions?
> --
> Quentin Sabah, CIFRE Ph.D. student
> Grenoble University
> INRIA-SARDES                   | STMicroelectronics/AST
> Montbonnot, France             | Grenoble, France
> mailto:quentin.sabah at inria.fr  | mailto:quentin.sabah at st.com
> phone: +33 476 61 52 42        | phone: +33 476 58 44 14
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>
>
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>