Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!sri-unix!rutgers!labrea!aurora!ames!sdcsvax!jww
From: jww@sdcsvax.UCSD.EDU (Joel West)
Newsgroups: comp.sys.mac
Subject: Re: LSC and MultiFinder/Juggler
Message-ID: <3687@sdcsvax.UCSD.EDU>
Date: Thu, 20-Aug-87 09:38:40 EDT
Article-I.D.: sdcsvax.3687
Posted: Thu Aug 20 09:38:40 1987
Date-Received: Sat, 22-Aug-87 10:08:11 EDT
References: <wV-DDAy00V4IzGo0fR@andrew.cmu.edu> <2708@husc6.UUCP>
Organization: Palomar Software, Inc., Vista, CA
Lines: 13

In article <2708@husc6.UUCP>, singer@endor.harvard.edu (Andrew Singer) writes:
> Once it has restored the dispatch table, the context-switch code should
> assume that the foreground application can get in on any trap.  Ideally
> no traps should be issued until the context switch is complete.  

Although, as regular readers know, I'm very opposed to the 'blame Apple
for everything' school of developers, I believe THINK is right in calling
this a bug.

MultiFinder is supposed to maintain separate trap patches for each
application, and swap them as part of each context swap.  Clearly
you don't want to swap the patches in before you swap the corresponding
value of CurrentA5.