OldComp.cz

Hezké řešení - přemýšlím jak to nejlépe otestovat. Mám tu takový jednoduchý prográmek, který jsem kdysi spáchal v TC. Jaká by byla rychlost v porovnání se standarníma knihovníma funkacema TC?

The original IBM-PC, and the XT, accept the original floating
point chip, the 8087. Later processors accept corresponding
chips: the 287 for the 286, the 387 for the 386, etc. From a
programming standpoint, the 8087 and 287 are nearly identical:
the 287 adds the instructions FSETPM and FSTSW AX, and ignores
the instructions FENI and FDISI. There is, however, a rather
nasty design flaw in the 8087, that was corrected in the 287.

To understand the flaw, you must understand how the 86 and 87
work as coprocessors. Whenever the 86 sees a floating point
instruction, it communicates the instruction, and any associated
memory operands, to the 87. Then the 86 goes on to its next
instruction, operating in parallel with the 87. That's OK, so
long as the following instructions don't do one of the following:

1. Execute another floating point instruction; or

2. Try to read the results of the still-executing floating
point instruction.

If they do, then you must provide an instruction called WAIT (or
synonymously FWAIT), which halts the 86 until the 87 is finished.
For almost all floating point instructions, it should not be
necessary to provide an explicit FWAIT; the 86 ought to know that
it should wait. For the 8087, it IS necessary to give an
explicit FWAIT before each floating point instruction: that is
the flaw.

Because of the flaw, all assemblers supporting the 8087 will
silently insert an FWAIT code (hex 9B) before all 87
instructions, except those few (the FN instructions other than
FNOP) not requiring the FWAIT. A86 will insert the opcode as
well, when it is assembling for the original 8087.

The are three ways to tell A86 whether it is assembling for an
8087 or a 287-or-later processor. First, A86 will use a default
for the processor on which it is currently assembling: no .287
for an 8086, 8088, 186, or NEC; .287 for a 286 or later. Second,
this can be overridden by the switch +F (the F must be
capitalized), to signal that the 287 is the target processor, or
-F to specify the 8087. Third, an 8087 setting can be further
overridden in the source code, with the directive ".287",
compatible with Microsoft's assembler.

When A86 is assembling for the 287 or later, it ceases outputting
FWAIT directives that are unnecessary for the 287, ignores the
instructions FENI, FDISI, FNENI, and FNDISI, and honors the
instructions FSETPM and FSTSW AX.

WARNING: The most common mistake 87 programmers make is to try to
read the results of an 87 operation in 86 memory, before the
results are ready. At least on my computer, the system often
crashes when you do this! If your program runs correctly when
single stepped, but crashes when set loose, then chances are you
need an extra explicit FWAIT somewhere.

Díky za hezké téma a zajímavé příklady.
Každopádně by to chtělo hardware - aby to nebyla jen teorie. Já mám hezkou 386, ale bez kopru Snad se časem něco najde...

387 uz ma FSINCOS. .)

387 uz ma FSINCOS. .)

287 ma asi uz jinak resene to pripojeni koprocesoru, fakt vaham, kde ma byt FWAIT a kde ne. V tom vzorovem prikladu je jen jedno, a az na konci..

8086 assembler je strasna hruza oproti Z80. Koukam ze napisi XOR AL,AL...2 byty! Coze? MOV AL,0 je pakto same. Kratke instrukce pro rotace akumulatoru..kde? Invertovani akumulatoru? Nikde? INC AL jsou zase 2 byty!

Online kompilace: