[Soot-list] Chicken-and-egg problem

[Rohan Padhye]

Hello,

Isn't it possible to manually run a call graph phase from inside a 
"wjpp" phase transformer? That way, you can have method bodies of all 
reachable methods (or their superset) loaded, can run any 
inter-procedural transformations, and then the subsequent actual "cg" 
phase will be run as usual constructing a call graph on the transformed 
program, which might be better than the first version.

Will this mess up anything? (I don't really know all of Soot's internals 
and how bodies and call graphs are dependent)

On Sunday 05 May 2013 01:46 PM, Eric Bodden wrote:
> Hello.
>
> That's really hard to do without major modifications to the way Soot works. I guess you would need to discover call edges on the fly, which means that you cannot use Heros as it is right now.
>
> Eric
>
>
> On 02.05.2013, at 01:13, Marc-André Laverdière-Papineau <marc-andre.laverdiere-papineau at polymtl.ca> wrote:
>
>> Hello,
>>
>> I am trying to make transformations in the wjpp phase in order to get a
>> better call graph. That will be used for IFDS analyses later.
>>
>> But here is the problem: I am not guaranteed to have all the information
>> within a single method. So I need an interprocedural analysis.
>>
>> But I don't have a call graph, nor any points-to information.
>>
>> So how do I break that cycle?
>>
>> Thanks and regards,
>>
>> -- 
>> Marc-André Laverdière-Papineau
>> Doctorant - PhD Candidate
>> _______________________________________________
>> Soot-list mailing list
>> Soot-list at sable.mcgill.ca
>> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
> --
> Eric Bodden, Ph.D., http://sse.ec-spride.de/ http://bodden.de/
> Head of Secure Software Engineering Group at EC SPRIDE
> Tel: +49 6151 16-75422    Fax: +49 6151 16-72051
> Room 3.2.14, Mornewegstr. 30, 64293 Darmstadt
>
>
>
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list

-- 
Regards,
Rohan Padhye

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.mcgill.ca/pipermail/soot-list/attachments/20130505/0cc161d7/attachment.html