Abstract Interpretation and Program Transformation
Information about this course will be distributed via the WWW, so this page
will be updated as the course develops. Remember to press Reload to see the
latest version. At the moment it is very much under construction!
This course is the second half of a 10-point course on implementing functional
languages. The first half of the course covered
compiled graph reduction, and was taught by Thomas Johnsson. This half of
the course covers program analysis and transformation. The draft course plan
on which the entire course is based is available here.
Teaching Form
The course will be taught via a mixture of lectures, recommended reading,
student presentations, and practical projects.
The lecture times and presentation times are available.
Examination
To pass the course, you will be
expected to:
- Implement a program analysis phase and accompanying optimisation in the
compiler you have developed in the first half of this course (see this).
- Implement a program transformation phase in your compiler, or
construct a program generator by partial evaluation.
- Give a presentation of a relevant article to the other participants.
Participants
Here are the email addresses of the course participants:
md2oven@mdstud.chalmers.se,henrikp,pareto,nazari,oloft,tato,ahlberg,bengtj,bond,heldal,masa,agat,pollack,andrew,gustavss,gustun,fede,frito,johank@ce.chalmers.se,nafun-6@mdstud.chalmers.se,johnsson,ilya
(Addresses which caused errors: csguest2)
Course Materials
- Abstract
Interpretation: a Semantics-Based Tool for Program Analysis, N D
Jones and F Nielson, in Handbook of Logic in Computer Science,
Oxford University Press, 1994, 527-629.
Summary: This is a broad overview of Abstract Interpretation,
to be a large chapter (around 100 pages) in the above-mentioned
handbook. It consists of three main parts: an Introduction with
motivation and Descriptions of the main methods used in the field; a
mathematical development of the logical relations approach with several
applications; and short descriptions of a broad spectrum of
Semantics-Based Program Analyses.
- Partial Evaluation 1996,
the draft proceedings of the 1996 Dagstuhl workshop on PE. A collection
of 23 papers. Torben
Mogensen's paper looks to give a very nice introduction to the area.
John Hughes
Last modified: Mon Mar 4 10:23:30 MET 1996