Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!columbia!rutgers!sri-spam!sri-unix!hplabs!pyramid!prls!philabs!mcnc!ece-csc!uvacs!edison!steinmetz!jesup
From: jesup@steinmetz.UUCP (Jesup)
Newsgroups: net.micro.amiga
Subject: Re: Easy fix to fexecv compatibility problems.
Message-ID: <952@kbsvax.steinmetz.UUCP>
Date: Fri, 17-Oct-86 15:18:31 EDT
Article-I.D.: kbsvax.952
Posted: Fri Oct 17 15:18:31 1986
Date-Received: Wed, 22-Oct-86 23:26:34 EDT
References: <8610132307.AA10388@cory.Berkeley.EDU>
Reply-To: jesup@kbsvax.UUCP (Randell Jesup)
Organization: General Electric CRD, Schenectady, NY
Lines: 21

In article <8610132307.AA10388@cory.Berkeley.EDU> dillon@CORY.BERKELEY.EDU (Matt Dillon) writes:
>
>	There is one place in the code for Manx's fexecv where they try to
>get the BCPL enviroment (I think that is what broke).  I like my solution
>to the problem better:	Simply save the startup enviroment when you start
>up.  That is, the first instruction in the startup object module should be:
>
>	movem.l	D0-D7/A0-A6,_Startup_Regs
>
>	Or something like that.  Then, when you need specific BCPL stuff,
>just look at _Startup_Regs instead of meandering through DOS.

Ah, but can you be sure that the CLI will give you the same initial register
values that it would give a BCPL program?  I do believe that they are quite
different.
What broke in manx's fexec was the series of pointers & offsets leading to
'*bcpl', which holds the BPTR to the loaded seglist.  One of the offsets
changed by 4.  The CLI was using this value instead of cli_Module.

	Randell Jesup		jesup@ge-crd.arpa  jesup@kbsvax.uucp
				...!rochester!edison!steinmetz!jesup.uucp