INTRODUCTION

addressability and 64,000 bytes of IIO HOW THE 8088 PIPELINED

addressability.ARCHITECTURE INCREASES SYSTEM

PERFORMANCE

The iAPX 88 instruction set includes a full complement of arithmetic operations includ- ing addition, subtraction, multiplication, and division, on 8-bit or 16-bit quantities. This gives the 8088 the highest computational throughput of any 8-bit microprocessor for numerics intensive applications. The 8088 also has a complete set of string manipula- tion operations for performance and flexi- bility in applications where large amounts of data are involved.

To make efficient use of its megabyte of memory addressing, the 8088 provides the most powerful range of addressing modes available to the programmer; from simple immediate addressing (data contained in the instruction) to complex addressing built from four components (three registers plus imme- diate data). More details are provided on addressing modes later on in this chapter.

The 8088 has built-in hardware support for multi-processor systems to coordinate re- source sharing of memory or peripheral devices among multiple processors.

Finally, and possibly the most powerful advantage: the 8088 is 100% code compatible with the 16-bit 8086 CPU. All the power of the 8086 16-bit instruction set is available in the 8-bit 8088. So, iAPX 88 systems ani easily upgradable to iAPX 86 16-bit systems because of this complete instruction set compatibility.

Figure 1-3 shows how programs are executed over time in a standard microprocessor. First, the microprocessor must fetch the instruction to be performed, then it executes the instruction. Only after the execution is complete is the CPU ready to fetch in the next instruction, execute that instruction, etc. as the program proceeds from beginning to end.

The CPU hardware that executes instruc- tions must obviously wait until the instruction is fetched and decoded before execution begins. Therefore, in standard microprocessors, the execution hardware (primarily the control circuitry and the arithmetic and logic unit) spends a lot of time waiting for instructions to be fetched. The 8088 eliminates this wasted time by dividing the internal CPU into two independent func- tional units (Fig. 1-4).

EXECUTION ~STRUCTION BUS

UNITPIPELINE INTERFACE UNIT

V

SYS1EM BUS

PIPELINED ARCHITECTURE DELIVERS HIGHER PERFORMANCE WITH REDUCED BUS "DEAD TIME"

Figure 1-4. Pipelined Internal Architecture

FETCH EXECUTE FETCH EXECUTE FETCH··•

TIME -

Figure 1-3. Program Execution in Standard Microprocessor

1-2

Page 13
Image 13
Intel 210200-002 manual Execution ~STRUCTION BUS