FPGAs tackle microcontroller tasks: Part 2 - 'Flexible' CPUs

By Robert Blake, Altera Corporation
(01/23/08, 12:10:00 AM EST) -- Automotive DesignLine

FPGAs offer flexibility to handle the explosion in electronic content in today's automotive systems.

Part 1 of this feature dealt with the burgeoning growth in automotive electronics applications, which strains current electronics architectures, and how FPGAs offer control, processing, and cost solutions.

Like an FPGA, a CPU should be sufficiently flexible enough to provide greater design latitude, depth, and breadth for automotive electronics applications. For example, the Nios II embedded processor (below) does not reside on a rigid, unchangeable, pre-specified silicon device like conventional microcontrollers (ìCs). Instead, it is automatically generated, based on the specifications defined by the system designer with the aid of available tools.


View a full-size image

The embedded processor is then loaded into the FPGA with the remainder of the logic required for the entire circuit. Thus, the processor core can be parameterized based on the requirements for the specific auto electronics application.

The embedded processor is built on a standard reduced instruction set computing (RISC) architecture with separate address and data buses, each 32-bits wide. Both operate via separate caches and can be continued separately into the bus system. The designer determines whether or not to use separate memory for code and data or keep both in shared memory.

Many of the functional units that every processor contains are present in this embedded processor. But the settings determine their character. For example, the hardware multiplier, barrel shifter, and hardware divider can be selected as options. The same is true for instruction and data cache, which can be varied in size or completely excluded.

To read the full article, click here