CAN IP-core for high-end microprocessors Authors

Reinhard Arlt, Andreas Block, Tobias Höger · 2012

here are still no CAN controllers integrated into most high-end microprocessor chips and a common way to add CAN using these processors is simply to connect one or more stand-alone CAN controllers (e.g. the SJA 1000 by NXP) to such systems. In most of such application the CAN interface build around the stand-alone CAN controller is one of the bottlenecks. There are numerous 8-bit accesses to the chip necessary to handle one CAN frame and most of these accesses are done in the interrupt routine with each access taking quite a while. The reason for this is that most available standalone CAN controllers are designed to work with relative low-performance micro-controllers. The consequences are rather slow 8-bit or 16-bit interfaces. Interfacing such a device to

Read the paper · More papers on PaperTik