Daniel Große

Exact sat-based toffoli network synthesis

Compact realizations of reversible logic functions are of interest in the design of quantum computers. Such reversible functions are realized as a cascade of Toffoli gates. In this paper, we present the first exact synthesis algorithm for reversible functions using generalized Toffoligates. Our iterative algorithm formulates the synthesis problem with d Toffoli gates as a sequence of Boolean Satisfiability (SAT) instances. Such an instance is satisfiable if there exists a network representation with d gates. Thus, we can guarantee minimality. In addition to fully specified reversible functions, the algorithm can be applied to incompletely specified functions. For a set of benchmarks experimental results are given.

WoLFram- A Word Level Framework for Formal Verification

Due to high computational costs of formal verification on pure Boolean level, proof techniques on the word level, like Satisfiability Modulo Theories (SMT), were proposed. Verification methods originally based on Boolean satisfiability (SAT) can directly benefit from this progress. In this work we present the word level framework WoLFram that enables the development of applications for formal verification of systems independent of the underlying proof technique. The framework is partitioned into an application layer, a core engine and a back-end layer. A wide range of applications is implemented, e.g.~equivalence and property checking including algorithms for coverage/property analysis, debugging and robustness checking. The back-end supports Boolean as well as word level techniques, like SMT and Constraint Solving (CSP). This makes WoLFram a stable backbone for the development and quick evaluation of emerging verification techniques.

The system verification methodology for advanced TLM verification

The IEEE-1800 SystemVerilog [20] system description and verification language integrates dedicated verification features, like constraint random stimulus generation and functional coverage, which are the building blocks of the Universal Verification Methodology (UVM)[3], the emerging standard for electronic systems verification. In this article, we introduce our System Verification Methodology (SVM) as a SystemC library for advanced Transaction Level Modeling (TLM) testbench implementation. As such, we first present SystemC libraries for the support of verification features like functional coverage and constrained random stimulus generation. Thereafter, we introduce the SVM with advanced TLM support based on SystemC and compare it to UVM and related approaches. Finally, we demonstrate the application of our SVM by means of a testbench for a two wheel self-balancing electric vehicle.

Completeness-driven development

Due to the steadily increasing complexity, the design of embedded systems faces serious challenges. To meet these challenges additional abstraction levels have been added to the conventional design flow resulting in Electronic System Level (ESL) design. Besides abstraction, the focus in ESL during the development of a system moves from design to verification, i.e. checking whether or not the system works as intended becomes more and more important. However, at each abstraction level only the validity of certain properties is checked. Completeness, i.e. checking whether or not the entire behavior of the design has been verified, is usually not continuously checked. As a result, bugs may be found very late causing expensive iterations across several abstraction levels. This delays the finalization of the embedded system significantly. In this work, we present the concept of Completeness-Driven Development (CDD). Based on suitable completeness measures, CDD ensures that the next step in the design process can only be entered if completeness at the current abstraction level has been achieved. This leads to an early detection of bugs and accelerates the whole design process. The application of CDD is illustrated by means of an example.

Improvements for constraint solving in the systemc verification library

For verification of complex system-on-chip designs often constraint-based randomization is used. This allows to simulate scenarios that may be difficult to generate manually. For the system description language SystemC the SystemC Verification (SCV) Library has been introduced. Besides advanced verification features like data introspection and transaction recording the SCV library enables constraint-based randomization forSystemC models. However, the SCV library has two disadvantages that restrict their practical use: There is no support of bit operators in SCV constraintsand the SCV constraint solver cannot guarantee a uniform distribution of the constraint solutions. In this paper we provide a detailed analysis of these problems and present solutions that have been integrated in the library.

Towards analyzing functional coverage in SystemC TLM property checking

For Electronic System Level (ESL) design SystemC has become the standard language due to its excellent support of Transaction Level Modeling (TLM). But even if the complexity of the systems can be handled using the abstraction levels offered by TLM - the most abstract one is untimed and focuses on functionality - still verification is the major bottleneck. In particular, as untimed TLM models are the reference for the following refinement steps their correctness has to be ensured. Thus, formal verification approaches have been developed to prove properties for these models. However, even if several properties have been checked this does not guarantee that the complete functionality of the TLM model has been verified. Thus, in this paper we consider the problem of functional coverage analysis in formal TLM property checking. We present a coverage approach which can analyze whether the property set unambiguously describes all transactions in a SystemC TLM model. The developed coverage analysis method identifies uncovered scenarios and hence allows to close all coverage gaps. As an example we consider an automated teller machine and we show the benefits of the proposed approach.

Designing a RISC CPU in Reversible Logic

Driven by its promising applications, reversible logic received significant attention. As a result, an impressive progress has been made in the development of synthesis approaches, implementation of sequential elements, and hardware description languages. In this paper, these recent achievements are employed in order to design a RISC CPU in reversible logic that can execute software programs written in an assembler language. The respective combinational and sequential components are designed using state-of-the-art design techniques.

Behavior Driven Development for circuit design and verification

The design of hardware systems is a challenging and erroneous task where about 70% of the effort in designing these systems is spent on verification. In general, testing and verification are usually tasks that are being applied as a post-process to the implementation. In this paper, we propose a new design flow based on Behavior Driven Development (BDD), an agile technique for the development of software in which acceptance tests written in natural language play a central role and are the starting point in the design flow. We advance the flow such that the specifics that arise when modeling hardware are taken into account. Furthermore, we present a technique that allows for the automatic generalization of test cases to properties that are suitable for formal verification. This allows the designer to apply formal verification techniques based on test cases without specifying properties. We implemented our approach and evaluated the flow for an illustrative example that successfully demonstrates the advantages of the proposed flow.

Precise error determination of approximated components in sequential circuits with model checking

Error metrics are used to evaluate the quality of an approximated circuit or to trade-off several approximated candidates in design exploration. Precisely determining the error of an approximated circuit is a hard problem since the errors accumulate over time depending on the composition and nature of individual components. In this paper, we present methods based on model checking to precisely determine error behavior in sequential circuits that contain approximated combinational components. Our experiments show that such an analysis is very significant and crucial to properly deduce the effects of approximations.

Simulation-based equivalence checking between SystemC models at different levels of abstraction

Today for System-on-Chips (SoCs) companies Electronic System Level(ESL) design is the established approach. Abstraction and standardized communication interfaces based on SystemC Transaction Level Modeling (TLM) have become the core component for ESL design. The abstract models in ESL flows are stepwise refined down to hardware. In this context verification is the major bottleneck: After each refinement step the resulting model is simulated again with the same testbench. The simulation results have to be compared to the previous results to check the functional equivalence of both models. For models at lower levels of abstraction strong approaches exist to formally prove equivalence. However, this is not possible here due to the TLM abstraction. Hence, in practice equivalence checking in ESL flows is based on simulation. Since implementing the necessary verification environment requires a huge effort, we propose an equivalence checking framework in this paper. Our framework allows to easily compare variable accesses in different SystemC models. Therefore, the two models are co-simulated using a client-server architecture. In combination with multi-threading our approach is very efficient as shown by the experiments. In addition, the time required for debugging is reduced by the framework since the respective source code references where the variable accesses did not match are presented to the user.