SoC multiprocessor debugging and synchronisation using generic dynamic-connect debugger frontends

摘要

Current and future SoC designs will contain an increasing number of programmable units. To be able to tailor and debug these processors in their system context at the highest possible overall simulation speed, we propose a methodology and the necessary tooling for a multiprocessor debugging environment which allows a flexible runtime trade-off between observability and simulation speed. When selecting full observability for a processor core, the user gets state-of-the-art debugging and profiling capabilities on assembly as well as on C/C++ source code level. This tool-set has been applied on a complex networking SoC case study.