Estimation tools are a key component of system-level methodologies, enabling a fast design space exploration. Estimation of software performance is essential in current software-dominated embedded systems. This work proposes an integrated methodology for system design and performance analysis. An analytic approach based on neural networks is used for high-level software performance estimation. At the functional level, this analytic tool enables a fast evaluation of the performance to be obtained with selected processors, which is an essential task for the definition of a "golden" architecture. From this architectural definition, a tool that refines hardware and software interfaces produces a bus-functional model. A virtual prototype is then generated from the bus-functional model, providing a global, cycle-accurate simulation model and offering several features for design validation and detailed performance analysis. Our work thus combines an analytic approach at functional level and a simulation-based approach at bus functional level. This provides an adequate trade-off between estimation time and precision. A multiprocessor platform implementing an MPEG4 encoder is used as case study, and the analytic estimation results in errors only up to 17% when compared to the virtual platform simulation. On the other hand, the analytic estimation takes only 17 seconds, against 10 minutes using the cycle-accurate simulation model.

Considering current time-to-market pressures, IP reuse is mandatory for the design of complex embedded systems-on-chip (SoC). The integration of IP components into a given design is the most complex task in the whole reuse process. This paper describes the IP integration approach implemented in the ROSES design environment, which presents a unique combination of features that enhance IP reuse: automatic assembly of interfaces between heterogeneous software and hardware IP components; easy adaptation to different on-chip communication structures and bus and core standards; generation of customized and minimal OSs for programmable components; and an architecture-independent high-level API embedded into SystemC that makes application software independent from system implementation. Application code is written by using communication functions available in this API. ROSES automatically assembles wrappers that implement these functions, such that the application code does not need to be modified in order to run in the final synthesized system.

This paper proposes the use of a high-level architecture exploration method for different MPEG4 video encoders using different customization parameters. The targeted architecture is a heterogeneous MP-SoC which may include up 2 coarse grain SIMD (task level SIMD) subsystems to perform the computations. The customization parameters are related to video resolution, frame rate, communication network, level of parallelism and CPU types. These parameters are determined during the high-level architecture exploration, by estimating the architecture performances at early stages of the design flow. Experiments shows that the error factor of these high-level performances estimations are less than 10% compared to those obtained with final manually implemented RTL architecture. This method was used successfully for exploration of different MPEG4 architecture configurations with different customization parameters. We consider these experiments a breakthrough because they show how a complex design can be mastered through a set of pragmatic choices.

One major challenge in System-on-Chip (SoC) design is the definition and design of interfaces between hardware and software. Traditional ASIC designer and software designer model HW/SW interface twice. Using two separate models introduces a discontinuity between hardware and software. This paper introduces a unified HW/SW component model to describe different parts of HW/SW interface at different abstraction levels. The benefits of using the proposed model are two fold: first, it provides a single model to present system design from abstract specification to mixed HW/SW implementation and second, it enables full system simulation at different abstraction level during refinement flow.

Most system-on-chip (SoC) design methodologies promote the reuse of pre-designed (hardware, software, and functional) components. However, as these components are heterogeneous, their integration requires complex interface sub-systems. These sub-systems can also be constructed by assembling pre-designed basic interface components. Hence, SoC design and validation involves component composition techniques to create hardware, software, and functional interface sub-systems by assembling basic interface components. We propose a unified methodology for automatic component integration that allows designers to reuse pre-designed components effectively. We also present ROSES, a design flow that uses this methodology to generate hardware, software, and functional interface sub-systems automatically starting from a system-level architectural model.

Abstract- This paper presents a full System-on-Chip (SoC) design flow from system specification to RT-level. A new approach to obtain a full path to implementation for SoC design is proposed. This approach combines architecture design space exploration using the VCC design environment and system synthesis using the ROSES design flow, allowing a true and complete system level design flow. The experiment with a VDSL application shows a significant reduction of design time. I.

This paper presents a high-level component-based methodology and design environment for application-specific multicore SoC architectures. Component-based design provides primitives to build complex architectures from basic components. This bottom-up approach allows design-architects to explore efficient custom solutions with best performances. This paper presents a high-level component-based methodology and design environment for application-specific multicore SoC architectures. The system specifications are represented as a virtual architecture described in a SystemC-like model and annotated with a set of configuration parameters. Our component-based design environment provides automatic wrapper-generation tools able to synthesize hardware interfaces, device drivers, and operating systems that implement a high-level interconnect API. This approach, experimented over a VDSL system, shows a drastic design time reduction without any significant efficiency loss in the final circuit.

This paper addresses performance estimation and architecture exploration issues within the context of hardware/software codesign. We introduce a new methodology to rapidly explore the large design space encountered in hardware/software systems. The proposed methodology is based on a fast and accurate estimation approach. This estimation approach takes advantage of both system and RT levels of abstraction, and combines both static and dynamic analysis techniques, in order to obtain the best trade-off between speed and accuracy. It has been implemented as an extension to a hardware/software codesign flow to enable the exploration of a large number of multiprocessor architecture solutions from the very start of the design process. The effectiveness of the proposed methodology is illustrated by a significant application example. Experimental results indicate strong advantages of the proposed methodology.

Although very popular and largely wanted, behavioral synthesis was never widely accepted by designers. This paper analyzes the reasons for this failure and introduces a new generation of behavioral synthesis tools with more practical synthesis schemes. The main breakthrough of this new generation is the redefinition of the behavioral synthesis flow to better profit from the power of modern RTL and FSM synthesis. The synthesis results for two large design examples: a 2-million transistors ATM shaper and a motion estimator for a video codec (H261 standard) are shown. They illustrate the effectiveness of this new approach when compared with RT-level design methodologies.