,
Mareike Burg
,
Daniel Lohmann
Creative Commons Attribution 4.0 International license
Timing analysis for embedded real-time systems is crucial to guarantee the correct behavior and to calculate the Worst-Case Response Time (WCRT) of safety-critical applications. With the increasing requirements of such systems in automotive, industrial or avionic industries, consolidation of multiple real-time and general-purpose operating systems on a single high-performance Multiprocessor System-on-Chip (MPSoC) platform using Static Partitioning Hypervisors (SPHs) is becoming more prevalent. Although the strong separation is well suited to reduce interference between isolated domains in such mixed-criticality systems, cross-domain interactions must still be considered in real-time analysis. Previous work has focused on dynamic monitoring and enforcement of timing constraints in virtualized environments. In this paper, we present HyperSSE, the first approach for the static analysis of cross-domain interactions in hypervisor-based real-time systems. By hierarchically combining existing domain-local static analyses, and synchronizing the control flow at cross-domain interactions, we enable control-flow-sensitive whole-platform analysis including multiple real-time domains. Using abstract task models, this approach can integrate a coarser analysis of general-purpose operating systems, accelerators, and coprocessors with the precise timing analysis. We demonstrate the applicability of HyperSSE in an automotive case study with mixed-criticality software stacks running on Xen in a static partitioning configuration. The resulting Hypervisor State Transition Graph (HSTG) exposes deep knowledge about the platform interactions, enabling cross-domain timing analysis with reduction of pessimistic WCRT calculations. Additionally, HyperSSE can be used for placement optimizations for better predictability, verification of critical interaction paths, and is the foundation for further platform-level analyses, such as analysis of implicit interactions through transparent resource sharing.
@InProceedings{kassens_et_al:LIPIcs.ECRTS.2026.23,
author = {K\"{a}ssens, Andreas and Burg, Mareike and Lohmann, Daniel},
title = {{HyperSSE: Cross-Domain Static Analysis of Partitioned Real-Time Hypervisor Systems}},
booktitle = {38th European Conference on Real-Time Systems (ECRTS 2026)},
pages = {23:1--23:27},
series = {Leibniz International Proceedings in Informatics (LIPIcs)},
ISBN = {978-3-95977-429-1},
ISSN = {1868-8969},
year = {2026},
volume = {375},
editor = {Kritikakou, Angeliki},
publisher = {Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
address = {Dagstuhl, Germany},
URL = {https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.ECRTS.2026.23},
URN = {urn:nbn:de:0030-drops-266165},
doi = {10.4230/LIPIcs.ECRTS.2026.23},
annote = {Keywords: Static Analysis, Hypervisor, Real-Time Operating System, Cross-Domain}
}