Functional Verification of Programmable Embedded Architectures

It is widely acknowledged that the cost of validation and testing comprises a s- nificant percentage of the overall development costs for electronic systems today, and is expected to escalate sharply in the future. Many studies have shown that up to 70% of the design development time and resources are spent on functional verification. Functional errors manifest themselves very early in the design flow, and unless they are detected up front, they can result in severe consequence- both financially and from a safety viewpoint. Indeed, several recent instances of high-profile functional errors (e. g. , the Pentium FDIV bug) have resulted in - creased attention paid to verifying the functional correctness of designs. Recent efforts have proposed augmenting the traditional RTL simulation-based validation methodology with formal techniques in an attempt to uncover hard-to-find c- ner cases, with the goal of trying to reach RTL functional verification closure. However, what is often not highlighted is the fact that in spite of the tremendous time and effort put into such efforts at the RTL and lower levels of abstraction, the complexity of contemporary embedded systems makes it difficult to guarantee functional correctness at the system level under all possible operational scenarios. The problem is exacerbated in current System-on-Chip (SOC) design meth- ologies that employ Intellectual Property (IP) blocks composed of processor cores, coprocessors, and memory subsystems. Functional verification becomes one of the major bottlenecks in the design of such systems.

Validation of programmable architectures, consisting of processor cores, coprocessors, and memory subsystems, is one of the major bottlenecks in current System-on-Chip design methodology. A critical challenge in validation of such systems is the lack of a golden reference model. As a result, many existing validation techniques employ a bottom-up approach to design verification, where the functionality of an existing architecture is, in essence, reverse-engineered from its implementation. Traditional validation techniques employ different reference models depending on the abstraction level and verification task, resulting in potential inconsistencies between multiple reference models.

This book presents a top-down validation methodology that complements the existing bottom-up approaches. It leverages the system architect’s knowledge about the behavior of the design through architecture specification using an Architecture Description Language (ADL). The authors also address two fundamental challenges in functional verification: lack of a golden reference model, and lack of a comprehensive functional coverage metric.

Functional Verification of Programmable Embedded Architectures: A Top-Down Approach is designed for students, researchers, CAD tool developers, designers, and managers interested in the development of tools, techniques and methodologies for system-level design, microprocessor validation, design space exploration and functional verification of embedded systems.