[Soot-list] Return Stmt

[Eric Bodden]

Yes, anything that calls itself "Java" must by definition generate
correct bytecode.

Eric

On 11/05/07, Mario Mendez <mario at cs.unm.edu> wrote:
> Yep, when talking about 'block' I was referring to the
> soot.toolkits.graph.Block class.
>
> One question, then: what do you mean by 'incorrect' bytecode? How can
> that be generated? Can we assume that standard Java compilers output
> correct bytecode?
>
> Thank you, Mario
>
> Richard L. Halpert wrote:
> > Soot has a whole slew of basic-block graph classes in addition to the
> > unit graphs that we are all familiar with.  Look in
> > soot/toolkits/graph.  Every return statement *should* be the last
> > statement in some block, though after looking at the BlockGraph code, I
> > suspect that a block graph constructed from incorrect bytecode could
> > break this rule.  If your bytecode is correct, then it should hold.
> >
> > Note that most flow analyses in Soot expect to be given unit graphs.
> >
> > -Rich
> >
> > On 5/11/07, *Chris Pickett* < chris.pickett at mail.mcgill.ca
> > <mailto:chris.pickett at mail.mcgill.ca>> wrote:
> >
> >     Eric Bodden wrote:
> >      > Oh, sorry I got that wrong. No a basic block actually ends by
> >      > definition after each jump, return or before each statement that is
> >      > jumped to. So I guess that should always holds.
> >
> >     Oh yeah... oops :)
> >
> >     Chris
> >
> >     _______________________________________________
> >     Soot-list mailing list
> >     Soot-list at sable.mcgill.ca <mailto:Soot-list at sable.mcgill.ca>
> >     http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
> >     <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
> _______________________________________________
> Soot-list mailing list
> Soot-list at sable.mcgill.ca
> http://mailman.cs.mcgill.ca/mailman/listinfo/soot-list
>


-- 
Eric Bodden
Sable Research Group
McGill University, Montréal, Canada