X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
From: Cesar Rabak <csrabak AT yahoo DOT com DOT br>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: /beta/bnu217b.zip won't work in pure DOS (UPX bug??)
Date: Sat, 29 Mar 2008 15:18:08 -0300
Organization: Aioe.org NNTP Server
Lines: 34
Message-ID: <fsm127$jlm$1@aioe.org>
References: <399e419e-7325-4a50-91c1-09d15037dc7b AT a22g2000hsc DOT googlegroups DOT com> 	<47ecaff9$0$4851$9b4e6d93 AT newsspool4 DOT arcor-online DOT net> <4f07ca13-4bb0-4e50-9b59-e5017757b0a6 AT l42g2000hsc DOT googlegroups DOT com> 	<fsjusv$j9e$1 AT aioe DOT org> <ddff703f-1adc-4291-9036-7ea451ce7ebf AT d45g2000hsc DOT googlegroups DOT com>
NNTP-Posting-Host: sq0ONYDxsvAShPsNFcLVaA.user.aioe.org
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-Complaints-To: abuse AT aioe DOT org
In-Reply-To: 
User-Agent: Thunderbird 2.0.0.12 (Windows/20080213)
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Rugxulo escreveu:
> Hi,
> 
> On Mar 28, 6:28 pm, Cesar Rabak <csra DOT  DOT  DOT  AT yahoo DOT com DOT br> wrote:
>> Rugxulo escreveu:> Hi,
>>
>> [snipped]
>>
>>> Well, here's what pure MS-DOS 6.22 on real hardware says for me:
>>> DOSBox 0.72 reports almost the same thing (except ecx=00000005 and
>>> error=0002).
>> Just to grasp completely your report is the "almost the same thing" in
>> _the same phisical machine_?
> 
> No. The MS-DOS 6.22 computer is a 486 Sx/25 with 8 MB of RAM (heh), so
> it's real hardware. I ran DOSBox 0.72 (emulation) on this cpu right
> here, Pentium 4 "Northwood" 2.52 Ghz w/ 512 MB RAM (Win XP Home SP2).
> But DOSBox emulates a 486DX2, so the instructions supported are
> similar.

OK, due above my reasoning is moot by now, but I asked to see if it 
could be explained by malfunctioning memory modules (have been bitten by 
this, specially in older HW where sometimes people had more than 2MB of 
memory but used strictly DOS apps and never 'knew' higher memory to be 
marginally operational).

> 
> As mentioned in the other posts, it appears to be an errant CMOVNE
> (PPro/P2) instruction, which is not supported on 486s or original
> Pentiums.

Yes. I only saw the other post _after_ I've already posted my question.

regards.