Second, today's embedded debug solutions are built for single processors, not taking into account requirements of debugging parallelism in an application distributed to 10, 20 or 50 processors.
Lastly, today's compilers are targeted to single processors only. Problems of expressing parallelism in application programming, then mapping and distributing software on a MPSoC platform, remain largely unaddressed.
Who is up for the challenge? Will it be driven by the software side or the hardware side? Who can afford the investment? Who is willing to pay the price?
While hardware designers are used to paying a price for fast simulation and debug, software designers are used to free or almost free tools in the age of GCC and GDB. This trend is exacerbated by the expectation of software developers to get a software development environment presented with the silicon they are programming. Vertically integrated semiconductor houses such as Texas Instruments and Philips understand this and have hired more software developers recently than hardware developers to support and differentiate their silicon.
The bottom line for requirements and trends suggest that the new system design automation paradigm can only be addressed in close cooperation with hardware and software designers. It will have to be funded by the hardware world in which users — software developers — expect to receive as part of the silicon offering a state-of-the-art software development environment supporting the platform. Otherwise, they will switch platforms or remap the application to a new, highly programmable MPSoC specifically taking into account their application requirements.
Providing this new methodology for system design automation is the next great challenge, particularly with the need to cross traditional industry boundaries and attitudes between EDA and embedded software. When a design team can apply functionality across a range of processors, successfully debug and simultaneously validate software and hardware, this new design paradigm will yield significant benefits for all. It will also enable accelerated growth in consumer electronics based on cheaper, faster, more power efficient and reliable chips.
If — and only if — system design automation becomes reality, then the next generation of convergence in consumer devices will be possible. We will be able to think back at and smile while confirming that our PDA has computed directions correctly using the integrated GPS, while phoning our children, all using one device. Quo vadis, indeed.
Simon Davidmann is president and CEO of Imperas Inc. He has been an executive with six U.S. EDA startups and one of the original developers of the HILO logic simulation system.
