Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 14 Mar 1998 15:17:41 +1100 (EST)
From:      John Birrell <jb@cimlogic.com.au>
To:        thyerm@camtech.net.au (Matthew Thyer)
Cc:        Studded@dal.net, current@FreeBSD.ORG
Subject:   Re: trouble booting
Message-ID:  <199803140417.PAA16394@cimlogic.com.au>
In-Reply-To: <350A0175.F8D28907@camtech.net.au> from Matthew Thyer at "Mar 14, 98 02:33:01 pm"

next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Thyer wrote:
> I'm curious too.
> 
> I hope someone else might comment on the possible benefits of
> making a new kernel first !
> 
> I think I read somewhere once that occasionally a make world
> may require this...  i.e. the new code uses new features in
> the kernel during compilation or something like that.
> 
> Is this real anyone ?  Or am I just wasting 8 minutes or so
> doing this.

Last year, NetBSD designed a new stat syscall. The old one continued
to exist (same number), but any new compiles would use __stat13 when
you called stat(). In this case, you had to build and boot a new kernel
before building libc. The old programs continued to work with the new
kernel, but new programs failed with the old kernel because of the
missing syscall.

So it depends on what the changes are. If you don't follow -current
closely (and continuously like I didn't last year 8-), then this sort
of thing bites.

-- 
John Birrell - jb@cimlogic.com.au; jb@freebsd.org
CIMlogic Pty Ltd, GPO Box 117A, Melbourne Vic 3001, Australia +61 418 353 137

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199803140417.PAA16394>