AMD 250 manual Types of Instructions, Instructions, Macro-ops and Micro-ops, Mrom

25112 Rev. 3.06 September 2005 Software Optimization Guide for AMD64 Processors

Table 1 summarizes the differences between AMD64 instructions, macro-ops, and micro-ops.

Table 1.	Instructions, Macro-ops and Micro-ops
Comparing		AMD64 instructions	Macro-ops	Micro-ops
Complexity		Complex	Average	Simple
		A single instruction may	A single macro-op may	A single micro-op
		specify one or more of	specify—at most—one	specifies only one of the
		each of the following	integer or floating-point	following primitive
		operations:	operation and one of the	operations:
		• Integer or floating-point	following operations:	• Integer or floating-point
		operation	• Load	• Load
		• Load	• Store	• Store
		• Store	• Load and store to the
			same address
Encoded length		Variable (instructions are	Fixed (all macro-ops are	Fixed (all micro-ops are
		different lengths)	the same length)	the same length)
Regularized		No (field locations and	Yes (field locations and	Yes (field locations and
instruction fields		definitions vary among	definitions are the same	definitions are the same
		instructions)	for all macro-ops)	for all micro-ops)

Types of Instructions

Instructions are classified according to how they are decoded by the processor. There are three types of instructions:

Instruction Type	Description
DirectPath Single	A relatively common instruction that the processor decodes directly into one macro-op
	in hardware.
DirectPath Double	A relatively common instruction that the processor decodes directly into two macro-
	ops in hardware.
VectorPath	A sophisticated or less common instruction that the processor decodes into one or
	more (usually three or more) macro-ops using the on-chip microcode-engine ROM
	(MROM).

Chapter 1 Introduction 5