This paper presents an extensive survey of trends in embedded processor use with an emphasis on emerging applications in wireless communication, multimedia, and general telecommunications. We demonstrate the importance of application-specific instructionset processors (ASIP's) in high-volume, low cost applications. We also examine some of the underlying trends of the applications in which embedded processors are used. This is followed by a description of embedded software development tool requirements. High-performance software compilation emerges as a key requirement. Finally, specific industrial case studies of products in MPEG, videophone, and low-cost digital signal processor (DSP) applications are used to illustrate the architecture design tradeoffs, and highlight specific tool requirements. A companion paper in this issue [1] presents a comprehensive survey of embedded software development tools, focusing mostly on retargetable software compilation.

No abstract

This paper describes a new and eflective approach t o register and interconnect optimisation, which is applicable in a dual context : to reduce chip area in highlevel synthesis, and to reduce resource load (and thus execution time) in retargetable code generation. The key idea is to carefully optimise the way in which data is transferred between functional units. The impact on high-level synthesis will be demonstrated with a practical design from the area of telecommunications.

No abstract

Software productivity for embedded systems is greatly limited by the fragmentation of platforms and associated development tools. Platform virtualization environments, like Java and Microsoft .NET, help alleviate the problem, but they are limited to host functionalities running on the system microcontroller. Due to the ever increasing demand for processing power, it is desirable to extend their benefits to the rest of the system. We present an experimental framework based on GCC that validates the choice of CLI as a suitable processor-independent deployment format. In particular, we illustrate our GCC port to CLI and we evaluate the generated bytecode in terms of code size and performance. We inject it back into GCC through a CLI front-end that we also illustrate, and we complete the compilation down to native code. We show that using CLI does not degrade performance. Compared to other CLI solutions, we offer a full development flow for the C language, generating a subset of pure CLI that does not require any virtual machine support other than a JIT compiler. It is therefore well suited for deeply embedded media processors running high performance media applications.