Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!topaz!ll-xn!cit-vax!nike!ucbcad!ucbvax!LLL-ICDC.ARPA!oberman
From: oberman@LLL-ICDC.ARPA ("Oberman, Kevin")
Newsgroups: mod.computers.vax
Subject: RE: DCL CALL trivia
Message-ID: <8608061105.AA20038@ucbvax.Berkeley.EDU>
Date: Tue, 5-Aug-86 10:49:00 EDT
Article-I.D.: ucbvax.8608061105.AA20038
Posted: Tue Aug  5 10:49:00 1986
Date-Received: Thu, 7-Aug-86 06:34:39 EDT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: "Oberman, Kevin" <oberman@lll-icdc>
Organization: The ARPA Internet
Lines: 27
Approved: info-vax@sri-kl.arpa

>One our operators was playing with DCL CALL in the following .com
>$ CALL TEST
>        .
>        .
>        .
>$ SUBROUTINE TEST:
>        .
>        .
>$ ENDSUBROUTINE
>
>  The label, of course, should be the first thing on that line,
>but you would thing that DCL would handle it a little more robustly.
>It gives you a dump and kicks you out on your ear!
>
>  A one-liner, $CALL TEST, will do the same thing.  We have SPRed it.

Any CALL to a non-existant subroutine will cause the process to crash. We
detected the problem at the Dallas DECUS Symposium. Since it was in the
VMS Campground with several VMS developer present, they were well aware
of it. Not to mention a bit embarressed that we had found such an obvious
bug in about 10 minutes.

					Kevin Oberman
					Lawrence Livermore Nat'l Lab
					ARPA: oberman@lll-icdc
					(415) 422-6955
------