Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utcs!mnetor!seismo!mcvax!ukc!kcl-cs!greenw
From: greenw@kcl-cs.UUCP
Newsgroups: net.math,net.micro,net.micro.pc,net.research,net.wanted
Subject: Graph Theory and Overlaid Program Structures
Message-ID: <713@neon.kcl-cs.UUCP>
Date: Sun, 13-Jul-86 14:21:09 EDT
Article-I.D.: neon.713
Posted: Sun Jul 13 14:21:09 1986
Date-Received: Mon, 14-Jul-86 23:04:26 EDT
Reply-To: jules@westf
Organization: Department of Computing, Kings College, University of London.
Lines: 27
Keywords: graph overlays
Xref: utcs net.math:3306 net.micro:14707 net.micro.pc:9016 net.research:509 net.wanted:9108

[Eat me]

I'm currently trying to write a program to generate automatically the
optimum overlay structure from the source of any given program. The host
machine is a Sun-2, and the target an IBM-PC, using Plink-86 Plus as the
linker (it handles overlays already).

The stuff I've done so far is based on normal graph theory plus Aho & Ullman's
stuff on dominators, and then my own knowledge of how Plink's overlay
structures work. Unbelieveably, it all seems to work (sort of!).

I would be most grateful if anyone could give me any pointers at all in any
area of this work - overlays, graph theory etc.

I'm not on the net (I'm borrowing a friend's login at the moment - ta Jules!),
but Jules will forward any replies by mail to me by surface mail. Obviously
I cannot enter into any dialogue on the net, but if you want to contact me
direct:

	Dave Gurr
	R4D Ltd.
	Europa House
	13-17 Ironmonger Row
	London EC1V 3QN
	ENGLAND

	Telephone: (+44) 1 608 2361