Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!decvax!decwrl!ucbvax!ULKYVX.BITNET!RDROYA01
From: RDROYA01@ULKYVX.BITNET.UUCP
Newsgroups: mod.computers.68k
Subject: Is this a way to handle the exception problem?
Message-ID: <8605161256.AA15293@ucbjade.Berkeley.Edu>
Date: Fri, 16-May-86 09:39:00 EDT
Article-I.D.: ucbjade.8605161256.AA15293
Posted: Fri May 16 09:39:00 1986
Date-Received: Sat, 17-May-86 04:38:35 EDT
Sender: mwm@ucbvax.BERKELEY.EDU
Organization: University of Louisville
Lines: 16
Approved: info-68k@ucbvax.berkeley.edu



What I have decided to do about the exception processing problem is create a
handler that intercepts all bdos calls while a specific program is running
in high memory.  That program will actually controll the whole system.  I
intend to do this by resetting the trap #2 vector and testing for D0=0 and
D0=PRGM_LD.  I will have the program save the old vectors and two flags.  If
the RESET flag is TRUE, then it will put the original vectors back and cause
a warm boot.  If the PRGM_LD flag is set (never will be at first), it will
put the program in supervisor state and call the CCP to load the program.
The privilidge error is the only problem with the CCP call now.  If the
resident program intercepts the RESET command, then when the CCP returns, it
will return to the calling program rather than resetting.  It should also be
able to grab exit() status this way.  What are the potential problems here?

                RDROYA01@ULKYVX.BITNET