Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!uw-beaver!rice!titan!preston
From: preston@titan.rice.edu (Preston Briggs)
Newsgroups: comp.arch
Subject: Re: VLIW Architecture
Keywords: VLIW
Message-ID: <1921@brazos.Rice.edu>
Date: 4 Oct 89 19:33:43 GMT
References: <1914@brazos.Rice.edu>
Sender: root@rice.edu
Reply-To: preston@titan.rice.edu (Preston Briggs)
Organization: Rice University, Houston
Lines: 49

In article <1914@brazos.Rice.edu> 
preston@titan.rice.edu (Preston Briggs) writes:

>	DO i = 1, 2*n
>	    x(i) = f(y(i))
>	    z(i) = g(x(i-1))
>	ENDDO

becomes

>	x1 = x(0)
>	DO i = 1, 2*n
>	    x0 = f(y(i))
>	    x(i) = x0
>	    z(i) = g(x1)
>	    x1 = x0
>	ENDDO

becomes

>	x1 = x(0)
>	DO i = 1, n
>	    x0 = f(y(i))
>	    x(i) = x0
>	    z(i) = g(x1)
>	    x1 = f(y(i+1))
>	    x(i+1) = x1
>	    z(i) = g(x0)
>	ENDDO


but this last is wrong.
It should be

	x1 = x(0)
	DO i = 1, 2*n, 2
	    x0 = f(y(i))
	    x(i) = x0
	    z(i) = g(x1)
	    x1 = f(y(i+1))
	    x(i+1) = x1
	    z(i+1) = g(x0)
	ENDDO

I'm sorry for any confusion.
Think of this as an example of why
I'd prefer an optimizer to transform my program.

	Preston