Path: utzoo!attcan!uunet!ncrlnk!ncrcae!ece-csc!mcnc!ecsvax!urjlew
From: urjlew@ecsvax.uncecs.edu (Rostyk Lewyckyj)
Newsgroups: comp.arch
Subject: Re: A simple question on RISC
Summary: Minor clarification wanted
Message-ID: <5893@ecsvax.uncecs.edu>
Date: 23 Nov 88 17:48:02 GMT
References: <6544@xanth.cs.odu.edu> <75577@sun.uucp> <1618@imagine.PAWL.RPI.EDU> <12369@srcsip.UUCP>
Organization: UNC Educational Computing Service
Lines: 46

In article <12369@srcsip.UUCP>, shankar@src.honeywell.COM (Son of Knuth) writes:
% In article <11562@cup.portal.com> bcase@cup.portal.com (Brian bcase Case) writes:
% >>        mov     lower,i
% >>        bra     entry
% >>body    foo
% >>        add     step,i
% >>entry   cmp     i,upper
% >>        bge     body
% >>
% >>Statically, same number of instructions; dynamically, less except in the
% >>zero-trip case.  But look, the add, compare, and branch are consecutive.
% >
% >Ok, but notice the label in the middle of the add-compare-branch sequence?
% >I don't think it is possible to specify the ACB instruction in a version
% >that has a branch destination in the middle!  This is one of the points of
% >RISC:  Hey, *I'M* the compiler, let *me* decide on the semantics of
% >instruction sequences.
% 
% But this can be transformed into:
%           bra    entry
% body      foo
%           acb    step,i,upper,body  (or whatever the ACB operands are)
% entry     cmp    i,upper
%           bge body
% 
% This doesn't justify the ACB instruction - but it does point out its
% usefulness in the above example, assuming it doesn't take too many
% microcode cycles.


Is this acb instruction being discussed , a real instruction in
some machine? and Is it in any way different from the usual loop
closing instructions like IBM 360  BXLE or BXH?
In the dynamic evaluations, what costs/weights are being assigned
to instruction fetch/issue for the two cases? and if it's a close
comparison, what frequencies are expected for the zero trip case
versus non zero trip?
Also in these arguments and in real evaluations does anyone ever
try to assign costs to instruction set complexity? i.e. cost of
one instruction name space? register dependancies (I am thinking
of the funny even oodd rules in IBM 360)?
-----------------------------------------------
  Reply-To:  Rostyslaw Jarema Lewyckyj
             urjlew@ecsvax.UUCP ,  urjlew@tucc.bitnet
       or    urjlew@tucc.tucc.edu    (ARPA,SURA,NSF etc. internet)
       tel.  (919)-962-9107