X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-bounces using -f
Message-ID: <3C216A3B.A51B96@yahoo.com>
From: CBFalconer <cbfalconer AT yahoo DOT com>
Organization: Ched Research
X-Mailer: Mozilla 4.75 [en] (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
Newsgroups: comp.os.msdos.djgpp
Subject: Re: ANNOUNCE: DJGPP V2.03 Refresh Beta
References: <200112181536 DOT fBIFajg13078 AT delorie DOT com> <3C1F8F7D DOT F2F2476C AT yahoo DOT com> <2427-Wed19Dec2001180855+0200-eliz AT is DOT elta DOT co DOT il> <3C210979 DOT 6D23A0A1 AT yahoo DOT com> <9628-Thu20Dec2001013941+0200-eliz AT is DOT elta DOT co DOT il>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 48
Date: Thu, 20 Dec 2001 05:54:34 GMT
NNTP-Posting-Host: 12.90.167.230
X-Complaints-To: abuse AT worldnet DOT att DOT net
X-Trace: bgtnsc04-news.ops.worldnet.att.net 1008827674 12.90.167.230 (Thu, 20 Dec 2001 05:54:34 GMT)
NNTP-Posting-Date: Thu, 20 Dec 2001 05:54:34 GMT
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Eli Zaretskii wrote:
> 
> > From: CBFalconer <cbfalconer AT yahoo DOT com>
> > Newsgroups: comp.os.msdos.djgpp
> > Date: Wed, 19 Dec 2001 22:14:51 GMT
> > >
> > > ???  How exactly do you ``check things by binary identity''?  If you
> > > use something like RCS's `ident' program, you should be able to see
> > > the difference.  If you use any other methods, please tell what they
> > > are.
> >
> > Before futzing with somebody elses code, I like to check that I
> > can duplicate their results.  The quickest way is to do a
> > compile/assemble/whatnot and a binary compare on the final
> > product.  I may have to allow for different embedded dates, etc,
> > and may have to play games with directory structures to get the
> > match, but when I do it is quite positive.
> 
> What would that comparison show?  It isn't useful, since differences
> in versions of compilers/assemblers/linkers will produce a totally
> differen binary.

It is *only* feasible if using the same tools.  But it has saved
my bacon many times when there was some ghastly mess of versions
of the end product components lying about.  Before trying to fix
it, I needed to physically grasp the thing I was trying to fix.  I
had the binary that was in the field, and I had eighteen different
'backups' of the source.  The original instigator was long gone
with invective on both sides, and I was hired to fix it.

I remember one case where I had to try two different versions of
the compiler, and three versions of the libraries, and then attack
the 30 or so confused sources.  Luckily there were some object
files left corresponding to the final binary, and I could use
these to probe and eliminate cases.

Thus my fixation on knowing that I have duplicated the toolset. 
Then you advance to telling the elephant from the rock (poke it
and see what it does).  Hard to do if you can't tell what you are
poking.

-- 
Chuck F (cbfalconer AT yahoo DOT com) (cbfalconer AT XXXXworldnet DOT att DOT net)
   Available for consulting/temporary embedded and systems.
   (Remove "XXXX" from reply address. yahoo works unmodified)
   mailto:uce AT ftc DOT gov  (for spambots to harvest)