Loop Level Analysis of Security and Network Applications

-It has been known that loops constitute the most executed segments of programs and therefore are the best candidates for hardware implementation. We present a set of profiling tools that are specifically dedicated to loop profiling and do support combined function and loop profiling. One tool relies on an instruction set simulator and can therefore be augmented with architecture and micro-architecture features simulation while the other is based on compile-time instrumentation of gcc and therefore has no slow down compared to the original program Instruction profiling tools can be broadly classified into two categories – instrumentation based instruction profilers and simulation based instruction profilers. An instrumentation based profiler instruments the compiler to add counters to various basic blocks of the program. During execution the counter values are written to a separate file. On the other hand, a simulation based instruction profiler uses an instruction set simulator to accomplish instruction profiling. Simulation based profilers can be further classified into static profilers and dynamic profilers. In dynamic profiling the instruction profile is obtained during the execution of the code on an instruction-set simulator while in static profiling the execution is written to a trace and the trace is processed to get instruction counts. For very large applications, the trace generated by a static profiler can grow to unmanageable proportions. Even though a dynamic profiling method is slow compared to the compilerbased instrumentation approach, a variety of architectural parameters can be tuned and studied while a program gets profiled on the simulator.