Fwd: [abc-users] Inter procedural control-flow analysis

From: Eric Bodden <bodden_at_st.informatik.tu-darmstadt.de>
Date: Mon, 14 Sep 2009 23:08:51 +0200

Sorry for the late response. I already sent this on the 1st but my
response got discarded by the list server.


---------- Forwarded message ----------
From: Eric Bodden <bodden_at_st.informatik.tu-darmstadt.de>
Date: 2009/9/1
Subject: Re: [abc-users] Inter procedural control-flow analysis
To: abc-users_at_comlab.ox.ac.uk

Hi Antoine.

> I need this information to affect the advice weaving but I read in the
> AOSD05 paper that the advice weaving phase precedes the analyses phase. Does
> that mean what I would like to do is impossible using ABC ?

It's true that many of the analyses that are implemented in abc happen
after weaving, as "ReweavingAnalysis". However, it is certainly
possible to set up abc in such a way that the analysis happens before

I am confused though, as to why you want to analyze the u-woven
program. In the woven program you can "see" the effect of applying an
advice right away, just by looking at the appropriate woven code. Are
you sure that you want the un-woven program?

> If not, what would be a good starting point to implement that feature ? I
> guess I need some intra and inter procedural control-flow analysis. Does
> some related work have already been implemented ?

You may want to look at Dehua"s work:


Eric Bodden
Software Technology Group
Technical University Darmstadt, Germany
Eric Bodden
Software Technology Group, Technische Universität Darmstadt, Germany
Tel: +49 6151 16-5478    Fax: +49 6151 16-5410
Mailing Address: S2|02 A209, Hochschulstraße 10, 64289 Darmstadt
Received on Mon Sep 14 2009 - 22:09:14 BST

This archive was generated by hypermail 2.2.0 : Sun Sep 20 2009 - 11:00:10 BST