[Soot-list] New variables added to chain overlap the original ones

[Eric Bodden]

Hi Christina.

I am sorry but what is this derivLocalsChain you are talking about?
Usually locals are added by calling body.getLocals().add(..)...

Also, what do you mean by overlap. I would love to help but I am not
quite sure I understand the problem.

Eric

On 22 May 2012 15:32, Cristina Ilie <crysgirlro at yahoo.com> wrote:
> Hello,
>
> My tool which uses soot introduces sometimes new definitions of new
> variables in the derivLocalsChain. The problem
> is that some new variables overlap with the existing ones. This occurs when
> I output directly bytecode...
>
> For example here [1] the IntLinkedStack object overlap on the fifth position
> with another DerivType object which was in
> the original code (actually the DerivType was a double but it existed in the
> original code! I only change the type to
> double and float - doesn't matter this idea...I just exapained).
>
> The new object introduced by me is an IntLinkedStack object and I add it
> through "derivLocalsChain.add(intStackLocal);"
> where intStackLocal is a JimpleLocal.
>
> I tried also to build the initalization stmt by adding the <init> method (I
> loaded the IntLinkedStack class) and
> after this the SpecialInvoke stmt. But the result is the same... How to
> avoid this overlapping of variables?
>
> Cristina
>
> [1]9  new IntLinkedStack [46]
>      12  dup
>      13  invokespecial IntLinkedStack() [42]
>      16  astore 5
>      18  new DerivType [26]
>      21  dup
>      22  invokespecial DerivType() [3]
>      25  astore 5
>
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>



-- 
Eric Bodden, Ph.D., http://bodden.de/
Head of Secure Software Engineering Group at EC SPRIDE
Principal Investigator in Secure Services at CASED
Tel: +49 6151 16-75422    Fax: +49 6151 16-72051
Room 3.2.14, Mornewegstr. 30, 64293 Darmstadt