🚀 go-pugleaf

In article <2306@sphinx.uchicago.edu> <4478@amd.AMD.COM> ayac071@ut-ngp.UUCP (William T. Douglass)
                                writes:
>In article <4478@amd.AMD.COM> ching@amd.UUCP (Mike Ching) writes:
>>In article <2306@sphinx.uchicago.edu> cjdb@sphinx.uchicago.edu (Charles Blair) writes:
>>         The NOPs are never executed and don't slow down the
>>processor.
>
>Actually, I thought that NOPs did an OR of the AX register on itself (or some
>similar operation that causes no change to data.)  As such, they DO affect the
>execution time involved.
>
>Could someone knowledgeable confirm or refute this?
>
>Bill Douglass
>ayac071@ngp.UUCP

In the "The 80386 Programmer's Reference Manual" NOP is listed as

Opcode: 90h
Clocks: 3
Description: No operation
             NOP is an alias mnemonis for the XCHG (E)AX,(E)AX

So NOP does take some time.  The only chip that I have heard of that takes no
time on NOP's is the novix forth chip which merges a nop in with the next
instruction (so I guess you can actually say that it takes a fraction of a
cycle).  All in all it has to influence performance in some way since it has
to be fetched in the minimal case.
Of Course in the case of jump optimization who cares??  You never get to the
NOP anyways!! After every JMP the prefetch que is fetched also so the NOP
doesn't effect execution in that way.

Michael Hunter         UUCP  : ....{seismo, rutgers, ames}!cit-vax!oxy!bagpiper
Box 241                ARPA  : oxy!bagpiper@csvax.caltech.edu
Occidental College     BITNET: oxy!bagpiper@hamlet.bitnet
Los Angeles, CA 90041  CSNET : oxy!bagpiper%csvax.caltech.edu@relay.cs.net