Xref: utzoo comp.lang.c:8683 comp.lang.misc:1368
Path: utzoo!utgpu!water!watmath!clyde!att-cb!att-ih!ihnp4!inuxc!iuvax!pur-ee!hankd
From: hankd@pur-ee.UUCP (Hank Dietz)
Newsgroups: comp.lang.c,comp.lang.misc
Subject: Re: Languages vs. machines (was Re: The need for D-scussion)
Summary: Pointer traversal can be vectorized...
Message-ID: <7792@pur-ee.UUCP>
Date: 30 Mar 88 05:24:18 GMT
References: <12176@brl-adm.ARPA> <1988Mar11.215238.976@utzoo.uucp> <1226@PT.CS.CMU.EDU>
Organization: Purdue University Engineering Computer Network
Lines: 18

In article <1226@PT.CS.CMU.EDU>, edw@IUS1.CS.CMU.EDU (Eddie Wyatt) writes:
> This really doesn't belong in comp.lang.c but.....
> 
> > Compilers can
> > hide lots of annoying differences, but they can't change a nifty linked
> > VAX-optimal data structure into a simple vectorizable Cray-optimal array.
> 
>    I ask you to prove the above statement.  

Better still, a disproof exists in a paper which I'll be presenting at this
year's ICPP...  it describes a loop transformation which often can perform
this kind of conversion.  As to its optimality, well, who can really say?
I'm willing to send interested folk advance copies of the paper.

					Prof. Hank Dietz
					School of EE
					Purdue University
					West Lafayette, IN  47907