Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 7/17/84; site tropix.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!whuxlm!harpo!decvax!mcnc!philabs!cmcl2!seismo!rochester!ritcv!tropix!dls
From: dls@tropix.UUCP (David L. Snyder )
Newsgroups: net.lang.f77
Subject: Re: 4.2bsd f77 to m68000 unix
Message-ID: <244@tropix.UUCP>
Date: Mon, 17-Jun-85 20:00:15 EDT
Article-I.D.: tropix.244
Posted: Mon Jun 17 20:00:15 1985
Date-Received: Fri, 21-Jun-85 03:15:10 EDT
References: <453@ho95b.UUCP> <26610@lanl.ARPA> <2850003@acf4.UUCP> <875@sdcsla.UUCP> <361@moncol.UUCP> <2281@sunRe: 4.2 F77 to m68000 unix
Organization: Tropel Div., GCA, Rochester NY
Lines: 24

I am in the process of porting the Berkeley 4.2 f77 compiler to our
m68000 unix.  The idea is, of course, to generate 68k code, not VAX
code.  I have proceeded by:
	first, porting the code itself; i.e., getting rid of
	machine dependencies to the point that I can generate
	VAX code.

	second, gluing the "front" of f77 to the "back" of
	our C compiler, using its table, &tc.  The intermediate
	binary is the same for both.

	finally, trying to work back upstream, changing the
	f77 front half to accomodate our machine.

The 4.2 f77 is not a work of art.  I appears to have been hacked
away at by many semester's worth of Berkeley undergraduates.  I
understand that someone at Utah has been coordinating improvements,
bug fixes, &tc.  Can someone give me a pointer to this kind soul?

I would like to know if anyone out there has attempted this kind
of port before, and what advise you can give.  Any deep insights
into the workings of the 4.2 f77 would also be greatly appreciated.

Thank you.