AMX for ARM/Thumb Timing Guide

It is the purpose of this AMX Timing Guide to provide actual AMX timing information and to introduce the factors which affect performance figures. Since instruction counts and cycle counts can no longer be used to measure execution times, it is necessary to provide specific measurements and to define, in detail, the conditions under which they were made. The remainder of this chapter will introduce the major factors which make timing measurement so difficult.

Chapter 2 introduces the timing tests used to measure typical task switching times in commonly occurring task to task synchronization scenarios. Chapter 3 defines the processing overhead necessary to support AMX timing services. Chapter 4 introduces the topic of interrupt latency and illustrates typical interrupt service timing. Timing data sheets are provided in Chapter 5. The data sheets provide specific timing figures for all of the measurements described in Chapters 2 to 4. The data sheets identify the exact characteristics of the hardware on which the measurements were made and describe the C compiler options used to compile AMX.

View Entire Paper | Previous Page | White Papers Search

If you found this page useful, bookmark and share it on: