|
||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT
Blocking for resource accesses can have a crucial impact on a real-time system's ability to meet its deadlines. In priority-based systems, uncontrolled blocking can lead to potentially unbounded priority inversion. To counteract this, basic priority inheritance and priority ceiling protocols have been defined. However, in large systems, it is still difficult to determine the maximum amount of blocking that a thread can suffer when using these protocols. The Real-Time Specification for Java provides a framework for execution and online feasibility analysis of priority-based systems. In this paper, we investigate possible extensions to this framework to support the measurement and monitoring of blocking times. We implement and evaluate the proposed mechanism within the context of the open source jRate implementation of the RTSJ running on top of the MaRTE OS. REFERENCES
Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.
INDEX TERMS
Primary Classification:
Additional Classification:
Keywords:
|
||||||||||||||||||||||||||||||||||||||||||||||