Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site decwrl.UUCP
Path: utzoo!linus!decvax!decwrl!dec-rhea!dec-mrvax!ddb
From: ddb@mrvax.DEC (DAVID DYER-BENNET MRO1-2/L14 DTN 231-4076)
Newsgroups: net.lang.c
Subject: setjmp/longmp
Message-ID: <4104@decwrl.UUCP>
Date: Fri, 2-Nov-84 16:16:58 EST
Article-I.D.: decwrl.4104
Posted: Fri Nov  2 16:16:58 1984
Date-Received: Sat, 3-Nov-84 21:39:17 EST
Sender: daemon@decwrl.UUCP
Organization: DEC Engineering Network
Lines: 11

Saving all registers at the time of the setjmp, and restoring them at
the time of the longjmp, is not sufficient to ensure that all accessible
variables have their setjmp-time values restored after the longjmp.
Stack variables will have whatever their last set value was.  Furthermore,
since there is no way to tell which variables are living where AT THE TIME
OF THE SETJMP (or at any other time), it is thus impossible to determine
which variables have which values after a setjmp reached by a longjmp.
This behavior makes setjmp/longjmp completely useless.

			-- David Dyer-Bennet
			-- ...decwrl!dec-rhea!dec-mrvax!ddb