From owner-freebsd-questions  Mon Jul  8 17:30:41 1996
Return-Path: owner-questions
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id RAA04421
          for questions-outgoing; Mon, 8 Jul 1996 17:30:41 -0700 (PDT)
Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id RAA04386;
          Mon, 8 Jul 1996 17:30:35 -0700 (PDT)
Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id SAA15606; Mon, 8 Jul 1996 18:27:07 -0600 (MDT)
Date: Mon, 8 Jul 1996 18:27:07 -0600 (MDT)
Message-Id: <199607090027.SAA15606@rocky.mt.sri.com>
From: Nate Williams <nate@mt.sri.com>
To: Terry Lambert <terry@lambert.org>
Cc: gpalmer@freebsd.org (Gary Palmer), ALHACK@am.pnu.com,
        questions@freebsd.org
Subject: Re: FreeBSD vs. Caldera Linux
In-Reply-To: <199607082249.PAA22765@phaeton.artisoft.com>
References: <199607061154.MAA21585@palmer.demon.co.uk>
	<199607082249.PAA22765@phaeton.artisoft.com>
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

[ ELF support in -current ]

Terry writes:
> Well, then I'll call it a mistake to have not considered it before
> it was this close to release... 8-(.
> 
> [ ... ]
[ Gary responds with a list of files necessary for the upgrade ]

> See the number of non-linux related files touched? Since then there
> have been countless bug fixes for various aspects of the `emulator',
> and with the different VM (etc) structures in -current and -stable, I
> wouldn't have recommended anyone try back-porting this.

> 
> How many non-linux related files were touched for other features that
> changed between 2.1R and 2.1.5?

Lots.  Maybe 'touched' was a poor word.  Many files were 'fixed' in the
2.1 -> 2.1.5 upgrade, but very few new features were added, and a couple
of them shouldn't have been (/dev/random stuff).  The ELF stuff is *new*
code, and as such doesn't fit the bill for the 'target' of the stable
release.

> I don't think a "weight of printout" argument is really applicable in
> this case.

It certainly is.  The 'weight of printout' implies that the code is both
new *and* fairly untested on a large scale.  And, simply because it
exists in -current doesn't mean it has been tested given the instablity
of current until recently plus the fact that very few folks actually
*use* the ELF stuff in -current.

The whole raison-d-etre of 'stable' is to be a bug-fix release of 2.1
with only known, tested, 'evolutionary' code in it, vs. new, faster,
'revolutionary' code.


Nate