Pentium™ Processor

Pentium Processor Optimization Tools covers advanced program optimization techniques for the Intel 80x86 family of chips, including the Pentium. The book starts by providing a review and history of the optimization tool. The text then discusses the 80x86 programming language; Pentium and its tools; and the superscalar Pentium programming. The operation of the floating-point unit; techniques for including assembly language routines in C or C++ programs; and the protected-mode programming are also considered. The book further tackles optimizations and code alignment; as well as the background and technical capabilities of the PowerPC vs. the Pentium and their future technical directions. Computer programmers and students taking related courses will find the book invaluable.

Introduction

Who Is This Book For

What Is on the Disk

Why Learn Assembly Language for the Pentium

How to Proceed

Acknowledgments

I Review and Historical Context

1 Number Systems

Hex

Signed Numbers

Numeric Overflow

Data Sizes

Little Endian vs. Big Endian

2 What is Assembly

Introduction

Tools and Terminology

What Are Compilers, Interpreters and Assemblers

3 The 8086 Family History and Architecture

Compatibility Lessons

Math Co-processors

The 80286

32-bit 80386

RISCy 80486

The 80586

The Competition

The P6

II 80x86 Family Background

4 8086 Architecture and Instruction Set

8088 Architecture

The 8088 Instruction Set

Program Control and Branching

Flag Manipulations

BCD Instructions

String Instructions

Interrupts

Miscellaneous Instructions

Flag Summary

5 Writing Beginning Programs

What Do All Those Statements Mean

Labels and Identifiers

Procedures

@Data

Defining Data Items

Using DOS System Functions

The END Directive

Memory Models

6 Assembly Tools

Editing

Assembling

Linking

Debugging

DEBUG32

7 The Instruction Set Evolves: The 186 to the 386

The Lost Brother, the 80186

The 80286

The 80386

New 386 Addressing Modes

New 386 Instructions

Protected Mode

III Introduction to Pentium and Tools

8 The 80486 and Pentium

Pentium

Bigger Cache

New Pentium Instructions

Summary

9 Superscalar Programming

Dual Integer Pipelines

Branch Prediction Logic

Optimized Cycle Times

10 Integer and Floating-Point Pipeline Operation

The Memory Cache

Pipelines

Address Generation Interlock AGI)

Paired Pipelines

486 Pipeline Delays

Pentium Pipeline Delays

Pentium Floating-Point Pipeline

FPU Pipeline Delays

Concurrent Integer and FPU Processing

11 Using the Pentium Optimizer Program

How It Works

Address Generation Interlocks

ICE

12 Timing with a Software Timer

Built-in Pentium Timer

Software Timer

Timer Software Function Reference

Percent Speed Changes

IV Superscalar Pentium Programming

13 Optimization Warm-Ups

String Instruction Optimizations

14 String Search and Translate

String Search

String Translations

Atomic Programming

Coding Challenge

Reality Check

Case-Independent String Searching

Case-Independent String Scan

Case-Independent String Compare

Conclusions

15 Checksums and Extended Precision Addition

Step 1

Step 2

Step 3

Step 4

Step 5

Step 6

Coming Completely Undone

Summary

False Steps

Extended Precision Addition

V Advanced Topics

16 Floating-Point Math

FPU Matrix Optimizations

Which Array Declaration Is Best

Optimizing with Assembly

17 Interfacing to C

Inline Assembly

Inline Assembly Example

Linking Separate Modules

Calling Conventions

Full C-to-Assembly Templates

Examples of Calling Assembly Routines from C

Using the Extended PROC directive

Fastcall

Fastcall Registers

Timing C Code

18 Protected-Mode Programming

Introduction to Protected Mode

DPMI, DOS Protected-Mode Interface

Protected-Mode Segments

Converting Code to Protected Mode

Mixed 16-Bit and 32-Bit Protected-Mode Programming

Full Segment Definitions

Protected Mode Timing

32-bit Protected-Mode Code Template

Large Data Segments

Timing 32-Bit Code

Cloaking Developers Toolkit

19 Final Notes and Optimizations

Speed vs. Code Size

LEA, the Multi-purpose Instruction

Code and Data Alignment

Local Stack Variables

Measuring and Correcting the Data Misalignment Penalty

Code Alignment

Further Reading

Where We've Been

VI PowerPC vs. Pentium 20

20. Power PC VS. PENTIUM

What Is RISC

What Is CISC

What Is RISC, Really

Which Is Better, RISC or CISC

Is the Pentium RISC or CISC

Superscalar Processors

Superscalar Techniques and Terminology

What Is in the PowerPC

Is the PowerPC Less Expensive

Future Processor Designs

Appendices

A Instruction Set Reference

Part I

Part II

B Optimization Cross-Reference by Instruction

C Optimization Guidelines by CPU

8088

286

386

486

Pentium

D Simple Instructions for Pentium Pairing

E Instruction Pairing Rules for Pentium

F Single-Byte Instructions

Pairable One-byte Instructions

Non-pairable One-byte Instructions

Prefix Bytes

G Quick Reference for Important Instruction Timings

Important Combinations for the Pentium

H Undocumented Pentium Registers

I Debug32 Command Summary

New Instructions

J Improving Performance

Other Pentium-class Machines

K Glossary of Terms

L Products Mentioned

Index