From owner-freebsd-hackers  Sat Nov 18 00:24:35 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id AAA02804
          for hackers-outgoing; Sat, 18 Nov 1995 00:24:35 -0800
Received: from cls.net (freeside.cls.de [192.129.50.1])
          by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id AAA02799
          for <hackers@freebsd.org>; Sat, 18 Nov 1995 00:24:31 -0800
Received: by mail.cls.net (Smail3.1.29.1)
	  from allegro.lemis.de (192.109.197.134) with smtp
	  id <m0tGiXO-0011jBC@cls.net>; Sat, 18 Nov 95 08:22 GMT
From: grog@lemis.de (Greg Lehey)
Organisation: LEMIS, Schellnhausen 2, 36325 Feldatal, Germany
Phone: +49-6637-919123
Fax:   +49-6637-919122
Reply-To: grog@lemis.de (Greg Lehey)
Received: (grog@localhost) by allegro.lemis.de (8.6.9/8.6.9) 
	id JAA16694; Sat, 18 Nov 1995 09:01:44 +0100
Message-Id: <199511180801.JAA16694@allegro.lemis.de>
Subject: Re: elm problem - "solved"
To: terry@lambert.org (Terry Lambert)
Date: Sat, 18 Nov 1995 09:01:44 +0100 (MET)
Cc: hackers@freebsd.org (FreeBSD Hackers)
In-Reply-To: <199511180023.RAA06399@phaeton.artisoft.com> from "Terry Lambert" at Nov 17, 95 05:23:24 pm
X-Mailer: ELM [version 2.4 PL23]
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 1007      
Sender: owner-hackers@freebsd.org
Precedence: bulk

Terry Lambert writes:
> 
> > >I really have great difficulty understanding why this change was made.
> > >It means that old FreeBSD and BSD/386 binaries won't work correctly
> > >under FreeBSD or BSD/OS Versions 2.  I can't see any advantage at all
> > >in this change.  Grrrr.
> > 
> > They wouldn't have worked anyway, because off_t was smaller.  Apparently
> > 4.4 took this breakage as an opportunity to improve the layout of
> > `struct flock'.  For some reason, there are no compatibility syscalls for
> > fcntl() or flock() like there are for lseek(), mmap(), and other syscalls
> > affected by the off_t change.  It is too late to fix this now, because
> > splitting the syscalls would break all 4.4 binaries that use them.
> 
> Why not use an aternate call table ala execution class for ABI support
> for IBCS2, Linux, etc.?

How are you going to distinguish between the objects?  In the case of
ibcs2 and Linux, the object files have different magic.  Not so in the
case of old BSD objects.

Greg