Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles; site ea.UUCP
Path: utzoo!watmath!clyde!burl!mgnetp!ihnp4!inuxc!pur-ee!uiucdcs!ea!kel
From: kel@ea.UUCP
Newsgroups: net.bugs.4bsd
Subject: A loop in 4.2 f77 do_fio - (nf)
Message-ID: <1700003@ea.UUCP>
Date: Tue, 29-May-84 15:45:00 EDT
Article-I.D.: ea.1700003
Posted: Tue May 29 15:45:00 1984
Date-Received: Sat, 2-Jun-84 08:43:47 EDT
Lines: 28
Nf-ID: #N:ea:1700003:000:831
Nf-From: ea!kel    May 29 14:45:00 1984

#N:ea:1700003:000:831
ea!kel    May 29 14:45:00 1984

As I have gotten some inquiries after my admittedly inadequate description
of a forever looping condition in the 4.2bsd do_fio, here's more:

The bug I was bit by is evoked by formats of the form 'a0', 'i0', etc.
for example:

      character * 80 thing
      data thing/'This is a test thing.  This has been a test thing.'/
      write(6, 1001) thing
1001  format(a0)
      stop
      end

	will write output to stdout forever.

As will be apparent by now, this problem is an artifact of a very
obscure dialect (bastardization?) of fortran.  By the way, when I
cooked up this example, I discovered that the formatter detects
an invalid format if you put the 'a0' directly into the write
statement.

If you get a fix for this, it might be nice for me to have, though
I've been warned about it, since I don't have source.

				Ken