From: Bob Supnik [bsupnik@nauticusnet.com]
Sent: Tuesday, April 16, 2002 10:02 AM
To: Info-VAX@Mvb.Saic.Com
Subject: Re: SIMH 2.9-5 - first official VAX release

There are clearly more bugs to find.  For example, installation of
layered products using the default (compressed) system libraries
fails.  Decompressing STARLET.OLB allows DEC C to install correctly.

The following bug was found post-release.  It does not appear to be
related to the compression problem but should be fixed anyway.  I want
to accumulate a few more days of feedback before updating the source
pool.

In vax_cpu.c, locate this code fragment:

case CASEL:
	r = (op0 - op1) & LMASK;		/* sel - base */
>>>	CC_CMP_B (r, op2);		/* r:limit, set cc's */
	if (((uint32) r) > ((uint32) op2))	/* r > limit? */
		JUMP (PC + ((op2 + 1) * 2));
	else {	temp = Read (PC + (r * 2), L_WORD, RA);
		BRANCHW (temp);  }
	break;

The flagged line should be CC_CMP_L (r, op2);

/Bob

On Mon, 15 Apr 2002 08:51:03 -0400, Bob Supnik
<bsupnik@nauticusnet.com> wrote:

>SIMH V2.9-5 was posted last night to http://simh.trailing-edge.com.
>It includes the first official release of the VAX.  The simulator can:
>
>1) restore the starter system from the hobbyist CD
>2) run a full system installation using the starter system and the
>hobbyist CD
>3) autogen
>4) reboot to the autogen'd system
>5) runs some simple commands, including shutdown
>
>Tim Stark and Kevin Handy found two show-stopper bugs:
>
>(Tim) Interrupts clear previous mode, rather than set previous mode to
>former current mode.
>(Kevin) PROBEx uses only bits<1:0> of its argument, rather than the
>entire argument.
>
>Thanks, Kevin and Tim, for tracing these bugs to their root cause.
>
>I am sure there are other bugs, but this seems like a good first step
>for wider use.
>
>/Bob Supnik