From: sandmann AT clio DOT rice DOT edu (Charles Sandmann)
Message-Id: <10308111336.AA16307@clio.rice.edu>
Subject: Re: inode problem in `rm'
To: djgpp-workers AT delorie DOT com
Date: Mon, 11 Aug 2003 08:36:32 -0500 (CDT)
Cc: eliz AT elta DOT co DOT il ('Eli Zaretskii')
In-Reply-To: <000a01c35fee$6c9fe820$0101a8c0@acp42g> from "Andrew Cottrell" at Aug 11, 2003 07:53:26 PM
X-Mailer: ELM [version 2.5 PL2]
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> > I'm not sure what incarnation of the ported Fileutils I'm 
> > using, but I think it's from alpha 1.  "ident rm.exe" reports 
> > the following:
> > 
> >      $Id: stub.asm built 2003-04-25 16:01:12 by djasm $
> >      $Id: DJGPP libc built Apr 25 2003 16:04:32 by gcc 3.2.2 $
> > 
> > If there's a better version, please tell me where to get it.
> This version should have the fix in it that I was thinking of, but I think
> this is a new bug.
> 
> When trying to reproduce this one make sure that the directory structure is
> a bit complex and it will show up. 
> 
> I copied my contrib & manifest directories to another drive and tried the
> following:-
> 	E:\dj204>rm -rf contrib manifest
> 
> With the following results:-
> DJ204 E:\dj204>rm -rf contrib manifest
> rm: ERROR: the directory `contrib/alfont/freetype' initially had
> device/inode
> numbers 4/268435467, but now (after a chdir into it), the numbers for `.'
> are 4/268435468.  That means that while rm was running, the directory
> was replaced with either another directory or a link to another directory.
> 
> The following are the results from Richard's coreutils 5.0 (FSF & Richards
> patch):-
> DJ204 E:\dj204>rm -rf contrib manifest
> rm: `contrib/alfont/freetype/builds/amiga' changed dev/ino

Since our inodes are not very reliable, and the security issues under
a DJGPP port are not as big an issue, I believe we should remove this
code from the DJGPP ports.

At the very least put filutils 4.0 back in the current directory, since
it's rm works.