Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!pt.cs.cmu.edu!a.gp.cs.cmu.edu!koopman
From: koopman@a.gp.cs.cmu.edu (Philip Koopman)
Newsgroups: comp.lang.forth
Subject: Re: Advanced Beginners
Message-ID: <8130@pt.cs.cmu.edu>
Date: 23 Feb 90 12:24:35 GMT
References: <514.UUL1.3#5129@willett.UUCP>
Organization: Carnegie-Mellon University, CS/RI
Lines: 32

In article <514.UUL1.3#5129@willett.UUCP>, ForthNet@willett.UUCP (ForthNet articles from GEnie) writes:
> F.SERGEANT [Frank]           at 20:39 CST
>  To: David Albert  Re: inner interpreter (next) vs CALL/RET on 8088/8086
> in #1 & #2 because of a number of other variabls.  My  Intel book for the 8088
> says 

I have recently seen some work that suggests that the Intel numbers for 80x86
series processors are extremely optimistic on real programs.  The
book claims that you should add about 5% for prefetch delays, etc.
Apparently some folks took measurements that suggest the real number
is 30%-40% on an 80C286.

If you are thinking of tuning your Forth for clock cycles, get an
80286 reference manual and use it as well since there are an
awful lot of ATs out there.  You will find out
that the clock cycle counts are considerably different for
almost all instructions.  Harris makes an 80C286 and will probably
ship you an "80C286 Hardware Reference Manual" if you call the
800 sales line.

PUSH reg = 3 clocks
POP reg = 5 clocks
LODS = 9 clocks
CALL = about 9 clocks (depends on prefetch fill)
RET = about 13 clocks
JMP reg = about 9 clocks

  Phil Koopman                koopman@greyhound.ece.cmu.edu   Arpanet
  2525A Wexford Run Rd.
  Wexford, PA  15090
Senior Scientist at Harris Semiconductor, adjunct professor at CMU.
I don't speak for them, and they don't speak for me.