Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-lcc!styx!ames!ucbcad!ucbvax!decvax!ima!johnl
From: johnl@ima.UUCP
Newsgroups: comp.lang.fortran,comp.compilers,comp.sources.wanted
Subject: fortran preprocessor for in-line code substitution
Message-ID: <560@ima.UUCP>
Date: Thu, 30-Apr-87 16:21:46 EDT
Article-I.D.: ima.560
Posted: Thu Apr 30 16:21:46 1987
Date-Received: Sun, 3-May-87 01:35:17 EDT
Sender: johnl@ima.UUCP
Reply-To: harvard!seismo!sierra!lamson ()
Organization: General Electric CRD, Schenectady, NY
Lines: 38
Approved: compilers@ima.UUCP
Xref: mnetor comp.lang.fortran:90 comp.compilers:10 comp.sources.wanted:1063
Posted-Date: 30 Apr 87 20:21:46 GMT

Original-sender: rochester!steinmetz!root@steinmetz..arpa


We are looking for software to read FORTRAN-77 source code and
do an in-line replacement of subroutine/function calls with the
source (also fortran) code from the called subroutine/function.

We're interested in such a tool for several reasons:

1) to allow (encourage) users to modularize code (good software
	engineering) without the penalty of subroutine call
	overhead for frequently called routines.

2) to develop portable applications on top of non-portable functions
	(bit manipulation, etc.).

3) to allow vectorizing/parallelizing compilers to "do their thing"
	across fortran module boundaries.

4) to restructure existing code in a more "parallel" sense (minimizing
	synchronization and communication).

Has anyone done something like this that is publically available?
Or are you looking for a good CS thesis project?  Failing that, do
enough people think this makes sense that we can get some support to
develop such a tool here, probably based on TOOLPACK.

p.s.  The next thing would be a tool to take a section of code
	out of a routine that had "evolved" beyond desirable size
	and "modularize" it automatically.
        Scott|  ARPA:      lamson@ge-crd.arpa
       Lamson|  UUCP:      lamson@sierra.steinmetz.ge.com
(518)387-5795|  GE DECnet: qtmvax::lamson
[There are a zillion Fortran source mungers out there, ratfor being the most
familiar to the Unix crowd.  I'd expect this has been done before, and invite
reports.  -John]
--
Send compilers articles to ima!compilers or, in a pinch, to Levine@YALE.ARPA
Plausible paths are { ihnp4 | decvax | cbosgd | harvard | yale | cca}!ima
Please send responses to the originator of the message -- I cannot forward
mail accidentally sent back to compilers.  Meta-mail to ima!compilers-request