Efficient clock-cycle precise simulation at architecture level in C++

Typically the design of hardware and software for embedded systems is tightly coupled. In combination, hardware and software have to fulfil the specific requirements of a particular application. This represents a challenge and opportunity in order to look for a balanced solution between both parts. A fast simulator with adequate interfaces is able to support this desire based on experiments starting early during the design process. This paper presents SEARS, a high level simulation written in C++ which is fast enough to run real-world programs on simulated hardware and, nevertheless, generates sufficient time information to measure program execution time precisely. The simulator provides configuration layers for increased flexibility and an interface with familiar parts for both the hardware and the software engineer