S
Stefan Hepp
Researcher at Vienna University of Technology
Publications - 11
Citations - 364
Stefan Hepp is an academic researcher from Vienna University of Technology. The author has contributed to research in topics: Worst-case execution time & Cache. The author has an hindex of 9, co-authored 11 publications receiving 317 citations. Previous affiliations of Stefan Hepp include University of Vienna.
Papers
More filters
Journal ArticleDOI
T-crest
Martin Schoeberl,Sahar Abbaspour,Benny Akesson,Neil Audsley,Raffaele Capasso,Jamie Garside,Kees Goossens,Sven Goossens,Scott Hansen,Reinhold Heckmann,Stefan Hepp,Benedikt Huber,Alexander Jordan,Evangelia Kasapaki,Jens Knoop,Yonghui Li,Daniel Prokesch,Wolfgang Puffitsch,Peter Puschner,Andre Rocha,Claudio Silva,Jens Sparsø,Alessandro Tocchi +22 more
TL;DR: Within the T-CREST project the authors propose novel solutions for time-predictable multi-core architectures that are optimized for the WCET instead of the average-case execution time.
Journal ArticleDOI
Patmos: a time-predictable microprocessor
TL;DR: This paper designs and optimize a processor, called Patmos, for low WCET bounds rather than for high average-case performance, a way out of this dilemma: a processor designed for real-time systems.
Proceedings ArticleDOI
The T-CREST approach of compiler and WCET-analysis integration
TL;DR: It is shown that a predictable architecture and the tight and seamless integration of compilation and WCET analysis is beneficial to achieve the initial two goals of good worst-case performance and the availability of high-quality bounds on the WCET of computation tasks.
Proceedings ArticleDOI
A Method Cache for Patmos
TL;DR: A method cache is presented as a time-predictable solution for instruction caching and the method cache caches whole methods (or functions) and simplifies worst-case execution time analysis in Patmos.
Proceedings ArticleDOI
Static profiling of the worst-case in real-time programs
TL;DR: The criticality metric is formally defined and some of its properties with respect to dominance in control-flow graphs are investigated, and an algorithm is proposed that reduces the overhead of computing the metric to cover complete real-time programs.