Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!ames!oliveb!pyramid!batcomputer!braner
From: braner@batcomputer.tn.cornell.edu (braner)
Newsgroups: comp.sys.atari.st
Subject: orphaned virtual GEM workstations
Message-ID: <1060@batcomputer.tn.cornell.edu>
Date: Tue, 19-May-87 02:12:36 EDT
Article-I.D.: batcompu.1060
Posted: Tue May 19 02:12:36 1987
Date-Received: Wed, 20-May-87 03:54:56 EDT
Reply-To: braner@tcgould.tn.cornell.edu.UUCP (braner)
Distribution: world
Organization: Cornell Theory Center, Cornell University, Ithaca NY
Lines: 18
Summary: can you close them w/o rebooting?

[]

When you write a program that opens a GEM virtual workstation, you
need to close it (with v_clsvwk()) before you exit the program.  If you
don't, GEM keeps that workspace open, and opens a _new_ one next time
you run the program.  Eventually it runs out of "handles" and wierd
things happen.  You then have to reboot.

The problem is: if your program crashes unintentionally, without doing
the v_clsvwk(), you're in trouble!  Question: can you close it later,
without knowing the handle number?  Is there a way to find out whether
virtual workstations are open and what their handles are?  Is it safe
to call v_clsvwk() on a fictitious handle?  What handle numbers would
be good guesses to try and close?  (If all else fails, you might want
to have your program print out the handle numbers (to a file?) while
in debugging phase.)

- Moshe Braner