Path: utzoo!attcan!uunet!husc6!bloom-beacon!tut.cis.ohio-state.edu!mailrus!ames!amdahl!apple!baum
From: baum@Apple.COM (Allen J. Baum)
Newsgroups: comp.arch
Subject: Re: Superoptimiser.
Message-ID: <13157@apple.Apple.COM>
Date: 1 Jul 88 18:39:58 GMT
References: <28200172@urbsdc>
Reply-To: baum@apple.UUCP (Allen Baum)
Organization: Apple Computer, Inc.
Lines: 25

[]
>In article <28200172@urbsdc> aglew@urbsdc.Urbana.Gould.COM writes:
>
>>>Branches tend to be deadly to fast machines. Delay slots or no, it can still
>>>give the cache/instruction stack indigestion.
>>
>>Gosh!  You guys aren't thinking big enough.  How about multiple
>>parallel pipelines to compute all the various instruction threads
>>in parallel and just keep the results of the one that is actually
>>taken?  Takes Big Bucks?  Sure.  But when you're using gallium
>>arsenide, who cares?
>
>The IBM 360 model 97 (I always get the model number wrong;
>anyway, one of the early 360s) did this.

I don't believe that any model of IBM 360 or 370 (including /91, /92, or /195)
ever executed multiple threads and threw away the unused one. The did have
a high degree of overlap, and out-of-order execution, but that is all (and
was enough). The only rumors I've heard about multiple threading machines
from IBM was a machine that was never fully built, killed by the promise of
the 360/91, which (I believe) John Cocke worked on, called the ACS. I've heard
rumors of all kinds of features that this machine was supposed to have, some
of which were pretty hairy. If only it had ben 360 compatible, it may have
seen the light of day.
--
{decwrl,hplabs,ihnp4}!nsc!apple!baum		(408)973-3385