Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!gem.mps.ohio-state.edu!brutus.cs.uiuc.edu!apple!well!farren
From: farren@well.UUCP (Mike Farren)
Newsgroups: comp.sys.amiga
Subject: Re: Lattice 5.04 problems
Keywords: lmk just exiting
Message-ID: <14759@well.UUCP>
Date: 29 Nov 89 16:20:32 GMT
References: <3760@scolex.sco.COM>
Reply-To: farren@well.UUCP (Mike Farren)
Organization: Whole Earth 'Lectronic Link, Sausalito, CA
Lines: 35

In article <3760@scolex.sco.COM> brianm@sco.COM (Brian Moffet) writes:
>
>I am having a problem with Lattice's LMK program just exiting
>after it prints out the Copyright information on the screen.
>I have installed the update to get me to Lattice 5.04 and since
>I had not used LMK before this point, I cannot tell if the problem
>existed before (I can't find my original disks.)

No, this didn't exist before, and yes, it IS a problem with the new
version of LMK.  The discussion on the Lattice BBS seems to indicate
that it's a problem with some signal bit or other in the process structure,
and one of the folks posted the following small program, which seems to
allow LMK to work properly:

#include <libraries/dosextens.h>

_main()
{
	struct Task *t;
	
	t = FindTask(0L);
	t->tc_SigAlloc &= 0x7fffffff;
	t->tc_SigRecvd &= 0x7fffffff;
}

As you can see, all it does is clear the high-order bit of the CLI signal
words.  I've found that if you run this before you run LMK, LMK works
almost always.  If it does not, closing the CLI and opening a new CLI
does the trick.

At any event, Lattice is aware of this bug, and (I hope) is going to fix
it next update.

-- 
Mike Farren 				     farren@well.sf.ca.usa