Date: Sun, 24 Dec 2000 16:13:27 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Richard Dawe <richdawe AT bigfoot DOT com>
cc: DJGPP workers <djgpp-workers AT delorie DOT com>
Subject: Re: dup(), dup2() and FSEXTs
In-Reply-To: <3A45FBE1.7BEA9458@bigfoot.com>
Message-ID: <Pine.SUN.3.91.1001224161016.10353B-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
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


On Sun, 24 Dec 2000, Richard Dawe wrote:

> I'm just writing an FSEXT for /dev/zero. I've been trying to add support
> for __FSEXT_dup and __FSEXT_dup2, but I notice that they are not
> supported. There was a discussion about this a while ago. Mark E submitted
> some patches, but these don't seem to have made it into CVS. Have anything
> happened about this issue since the last discussion?

These changes are not in CVS because (IIRC) the discussion raised some 
issues which were not resolved.

dup and dup2 are special because FSEXT machinery itself uses dup, and it 
needs it to be the _real_ dup.  In addition, again IIRC, there was 
uncertainty as to what would a user want to accomplish by hooking a dup 
call.

I believe you should be able to see all the deliberations in the mail 
archives on DJ's server.  We can discuss this again, if something is 
unclear.