The 8087 was the first math coprocessor for 16 bit processors designed by Intel (the I8231 was older but designed for the 8 bit Intel 8080); it was built to be paired with the Intel 8088 and 8086 microprocessors. The purpose of the 8087, the first of the x87 family, was to speed up computations on demanding applications involving floating point command mathematics. The performance enhancements went from 20% to 500% depending on the specific application. The 8087 could perform about 50 kFLOP/s, depending on the operation performed.
This coprocessor introduced about 60 new instructions available to the programmer, their mnemonics all beginning with "F" to differentiate them from the 8086 integer instructions. For example, in contrast to ADD/MUL, the 8087 provided FADD/FMUL. In binary format, all new instructions began with the bit pattern 11011, decimal 27, the same as the ASCII ESCAPE character, and so are sometimes referred to as "escape opcodes" (or the other way round).
The x87 family does not use a freely addressable register set such as the main registers of the x86 processors; instead the x87 registers form a 8-level deep stack structure ranging from st0 (top) to st7. The x87 instructions operate by pushing, calculating, and popping values. It is possible to exchange the contents of any of the eight registers with st0, so, by employing these instructions (exch st0..exch st7 or D9C8..D9CF), the x87 registers can in fact be used as seven freely addressable registers. These instructions were therefore optimized down to a zero clock penalty on the superscalar x86 processors (i.e. the Pentium and later models).
When Intel designed the 8087 it aimed to make a standard floating point format for future designs. In fact, one of the most successful things from a historical perspective of this coprocessor was the introduction of the first floating point standard for the x86 PCs: the IEEE 754 (although the Intel 8087 does not implement the eventual IEEE 754 standard in all its details, as the standard wasn't finished until 1985; the first Intel Coprocessor to implement it completely was the 80387). The 8087 provided two basic 32/64-bit floating point data types and an additional extended 80-bit internal support to improve accuracy over large and complex calculations. Apart from this, the 8087 offered an 80-bit/18-digit packed BCD (binary coded decimal) format and 16, 32 and 64-bit integer data types.
The 8087 differed from subsequent Intel coprocessors in that it was directly connected to the address and data buses. The 8088/86 looked for instructions that commenced with the '11011' sequence and relinquished control to the coprocessor. The coprocessor handed control back once the sequence of coprocessor instructions ended. There was a potential crash problem if the coprocessor instruction failed to decode to one that the coprocessor understood. Intel's later coprocessors did not connect to the buses in the same way, but were handed the instructions by the main processor. This yielded an execution time penalty, but the potential crash problem was avoided because the main processor would ignore the instruction if the coprocessor refused to accept it.
The 8087 was able to work out whether it was connected to a 8088 or a 8086 by monitoring the data bus during the reset cycle.
The 8087, announced in 1980, was superseded by the 80287, 80387DX/SX and the 80487. Intel 80486DX, Pentium and later processors include a built-in coprocessor on the CPU core.
The 8087 contains 45,000 transistors, 3 micrometre circuit technology (8086 has 29,000 transistors).
External links
- Intel 80x87 math coprocessors at cpu-collection.de
- Coprocessor.info : 8087 math coprocessor history information and pictures
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
This entry is from Wikipedia, the leading user-contributed encyclopedia. It may not have been reviewed by professional editors (see full disclaimer)
