Skip to content

Latest commit

 

History

History
62 lines (46 loc) · 1.85 KB

README.md

File metadata and controls

62 lines (46 loc) · 1.85 KB

About TACLe Benchmarks

This is the starting point for the repository of the TACLe benchmark collection. See also: http://www.tacle.eu/index.php/activities/taclebench

Documentation

The TACLe benchmark collection has been described in a paper at the WCET 2016 workshop. The paper is included as PDF in the doc folder. If you use the TACLe benchmarks please cite following paper:

@INPROCEEDINGS{TACLeBench,
  author = {Heiko Falk and Sebastian Altmeyer and Peter Hellinckx and Bj{\"o}rn
	Lisper and Wolfgang Puffitsch and Christine Rochange and Martin Schoeberl
	and Rasmus Bo S{\o}rensen and Peter W{\"a}gemann and Simon Wegener},
  title = {{TACLeBench}: A Benchmark Collection to Support Worst-Case Execution
	Time Research},
  booktitle = {16th International Workshop on Worst-Case Execution Time Analysis
	(WCET 2016)},
  year = {2016},
  editor = {Martin Schoeberl},
  volume = {55},
  series = {OpenAccess Series in Informatics (OASIcs)},
  pages = {2:1--2:10},
  address = {Dagstuhl, Germany},
  publisher = {Schloss Dagstuhl--Leibniz-Zentrum f\"ur Informatik}
}

Getting Started

All benchmarks are self-contained and can be compiled just with

cc/gcc/clang *.c

Current Status

the TACLeBench group is currently in the process of cleaning up and unifying the benchmarks. A version 2.0 will be available in the relative near future.

The WCET paper is based on the version V1.9. To switch to this version use:

git checkout V1.9

Regression Tests

https://www4.cs.fau.de/Research/TACLeBench

Getting Involved

You are welcome to contribute and help. To get involved contact:

Heiko for the mailing list (http://www.tuhh.de/es/esd/people/hfalk) and Martin for GitHub access (http://www2.imm.dtu.dk/~masca/).