Sender: itm AT dra DOT hmg DOT gb
Message-ID: <332E8249.167EB0E7@dra.hmg.gb>
Date: Tue, 18 Mar 1997 11:53:45 +0000
From: Ian Miller <itmiller AT dra DOT hmg DOT gb>
Organization: DRA
MIME-Version: 1.0
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
CC: DJGPP workers <djgpp-workers AT delorie DOT com>
Subject: Re: The "Can't backup file..." patch Win95+LFN bug
References: <Pine DOT SUN DOT 3 DOT 91 DOT 970318142027 DOT 17465O-100000 AT is>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Eli Zaretskii wrote:
> 
> On Mon, 17 Mar 1997, Ian Miller wrote:
> 
> > So, IMHO, either this is because I built my libc with LFN undefined
> > or the distributed patch.exe was built with an old libc. (This is
> > patch 2.1 as distributed with DJGPP v2.01 BTW.)
> 
> I don't see how building libc with LFN undefined can influence
> anything at run time.

Neither can I, but I'm no expert so I thought I'd better mention it :)

> And patch.exe from pat21b.zip that I have (size
> 117248 bytes, date Oct 19, 1996) is built with a library that was
> itself built on Oct 19, so this is the latest libc.  You can see what
> version of the library is linked in using `strings':
> 
>                 strings -a patch.exe | less
> 
> Page down to the end of the output, then back up and look for a string
> like so:
> 
>    $Id: DJGPP libc built Oct 19 1996 11:26:11 by gcc 2.7.2.1 $
> 
> If you have RCS installed, it's even easier: type "ident patch.exe"
> and you will see the above line.

Yeah, I'll check this when I get home. But the fact remains that the
binary I build from the patch21s sources does not have this bug while
the binary I extract from patch21b does (and we know it's not just me).
Maybe my patch21b.zip is out of date, but I don't think so. Like I
said, I'll check this later on and report back tomorrow.

Regards,
-- 
Ian Miller, Dorset, UK
DJGPP 2.01, Win95 DOS box (LFN undefined, FNCASE=y)