This project investigates the detection of functional errors caused by timing faults in hardware-software systems. A hardware-software system can be defined as one in which hardware and software must be designed together, and must interact to properly implement system functionality. The widespread use of these systems in cost-critical and life-critical applications motivates the need for a systematic approach to verify functionality. The complexity of the verification problem for large, heterogeneous hardware-software systems necessitates the development of simulation-based covalidation approaches, which uniformly validate hardware, software, and their interaction.

Hardware-software systems are typically composed of several behaviors, or processes, which may be mapped to different hardware and software components. The correctness of process interactions depends on the synchronization between the executions of communicating processes. Correct synchronization depends on the adherence of the system to numerous timing relationships between inter-process communication events. Verifying these timing relationships is central to the hardware-software covalidation problem, and is the focus of this project.

The goal of this research effort is to provide hardware-software system designers with a set of CAD tools that partially automate the timing covalidation process. The CAD tools are being developed to support a generic codesign/covalidation flow. At different intermediate steps of the codesign process, cosimulation is performed using a functional test sequence. Designers need the ability to evaluate the completeness of an existing functional test sequence, and the ability to automatically generate a functional test sequence. To support this codesign flow, we explore the following specific research objectives:

1. Develop a design fault model to describe design errors in hardware-software systems, specifically targeting errors in the timing of inter-process communication events.

2. Develop a fault simulation approach for the design fault model which evaluates the degree to which a given test sequence detects all design errors.

3. Develop techniques for automatic test pattern generation (ATPG) for a hardware-software description. The ATPG tool developed must target the proposed design fault model in order to ensure the detection of design errors.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Communication Foundations (CCF)
Application #
0418725
Program Officer
Sankar Basu
Project Start
Project End
Budget Start
2003-08-31
Budget End
2007-08-31
Support Year
Fiscal Year
2004
Total Cost
$255,879
Indirect Cost
Name
University of California Irvine
Department
Type
DUNS #
City
Irvine
State
CA
Country
United States
Zip Code
92697