Skip to content
Technical University of Munich

Program Optimization (IN2053)

Lecturer (assistant)
Florian Zuleger [L]
Florian Sextl
Number 0240795803
Type lecture with integrated exercises
Duration 6 SWS
Term Winter semester 2026/27
Language of instruction English
Position within curricula See TUMonline
Dates See TUMonline

Admission information

See TUMonline
Note: see https://www.in.tum.de/i02/lehre/wintersemester-1920/vorlesungen/program-optimization/

Description

The lecture starts with basic dataflow analyses like availability of
expressions or liveness of variables together with the optimizing
transformations enabled by these analyses. Motivated by the examples, we
develop the theoretical background on complete lattices and monotone functions
necessary for program analysis methods. Subsequently, we discuss more
sophisticated analyses such as constant propagateion. By discussing interval
analysis, we present general methods for the case where ordinary fixpoint
computation will not terminate. Further topics are:
- interprocedural analysis;
- pointer analysis;
- fixpoint algorithms.
We also consider hardware dependent optimizations such as:
- register allocation;
- instruction selection;
- instruction scheduling
and also optimizations to improve the cache behavior.

Prerequisites

see https://www.in.tum.de/i02/lehre/wintersemester-1920/vorlesungen/program-optimization/

Teaching and learning methods

see https://www.in.tum.de/i02/lehre/wintersemester-1920/vorlesungen/program-optimization/

Links

To top