From owner-freebsd-current  Mon Oct 20 15:59:41 1997
Return-Path: <owner-freebsd-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id PAA26509
          for current-outgoing; Mon, 20 Oct 1997 15:59:41 -0700 (PDT)
          (envelope-from owner-freebsd-current)
Received: from rvc1.informatik.ba-stuttgart.de (rvc1.informatik.ba-stuttgart.de [141.31.112.22])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA26491
          for <freebsd-current@freefall.FreeBSD.org>; Mon, 20 Oct 1997 15:59:27 -0700 (PDT)
          (envelope-from helbig@Informatik.BA-Stuttgart.DE)
Received: (from helbig@localhost)
	by rvc1.informatik.ba-stuttgart.de (8.8.7/8.8.5) id AAA27784;
	Tue, 21 Oct 1997 00:28:40 +0200 (MET DST)
From: Wolfgang Helbig <helbig@Informatik.BA-Stuttgart.DE>
Message-Id: <199710202228.AAA27784@rvc1.informatik.ba-stuttgart.de>
Subject: Re: bad system call - world build
In-Reply-To: <199710201433.AAA01025@word.smith.net.au> from Mike Smith at "Oct 21, 97 00:03:06 am"
To: mike@smith.net.au (Mike Smith)
Date: Tue, 21 Oct 1997 00:28:40 +0200 (MET DST)
Cc: kuku@gilberto.physik.RWTH-Aachen.DE, grog@lemis.com,
        freebsd-current@freefall.FreeBSD.org
X-Mailer: ELM [version 2.4ME+ PL30 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> > > > ===> share/termcap
> > > > ex - /a/src/share/termcap/termcap.src < /a/src/share/termcap/reorder > /dev/null
> > > > Bad system call - core dumped
> ...
> > > This is almost a FAQ.  Build a new kernel first, boot with it, then
> > > you should be able to continue.
> > 
> > Not quite, Greg. Normally I know what to do when proc.h has been changed
> > (ps/w etc. weirdnesses) but in this case I'm not sure whether
> > building a new kernel (3.0-current) and booting it on top of a
> > set of (up to then) 2.2.2-binaries would be a could remedy here.
> 
> Funnily enough, Greg is quite right.  I almost feel that I'm insulting 
> him by backing him up, but Chris, this *is* a FAQ, and if you were 
> reading the -current list like you're supposed to, you would know all 
> about this.
> 
> You do need a newer kernel to get past this.  The issue *was*announced* 
> when the change was made.  There was some attempt made to put in place 
> code to deal with the new system call, but it appears to have been 
> unsuccessful and at any rate is only needed for the bootstrapping case. 
> Going from 2.x to 3.x is going to get harder, not easier.
> 
> > I will more likely follow the advice of your forespeaker, namely
> > excluding the termcap subdir from the worldbuild.
> 
> This is stupid, and fails to address the problem.

Hmm. As far as I recall, we had some interface change of
mount(8) in -current about half year ago which made it
impossible to successfully mount with a new kernel and old userland
environmemt.

So even if this is a FAQ the answer depends...  Most of the time
building userland first gives you less trouble than building
the kernel first.  Just think about outdated config(8) which is
still part of userland.

I don't think my suggestion is that stupid.

Wolfgang