From owner-cvs-release  Tue Mar 24 01:10:35 1998
Return-Path: <owner-cvs-release>
Received: (from daemon@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id BAA25127
          for cvs-release-outgoing; Tue, 24 Mar 1998 01:10:35 -0800 (PST)
          (envelope-from owner-cvs-release)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA25107;
          Tue, 24 Mar 1998 01:10:16 -0800 (PST)
          (envelope-from bde@godzilla.zeta.org.au)
Received: (from bde@localhost)
	by godzilla.zeta.org.au (8.8.7/8.8.7) id UAA08120;
	Tue, 24 Mar 1998 20:06:25 +1100
Date: Tue, 24 Mar 1998 20:06:25 +1100
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199803240906.UAA08120@godzilla.zeta.org.au>
To: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-release@FreeBSD.ORG,
        jkh@FreeBSD.ORG
Subject: Re: cvs commit: src/release/sysinstall misc.c
Sender: owner-cvs-release@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>  Modified files:        (Branch: RELENG_2_2)
>    release/sysinstall   misc.c 
>  Log:
>  Remove MNT_NOATIME option; it makes the atime predate the ctime which
>  is a worisome side-effect.  Sorry, Bruce.

But it's normal for the atime to predate the ctime for files that have
never been read.  This should be the normal case for freshly installed
files, since installs should not waste time reading what they have
just written.

The case where timestamps are copied from a source file system doesn't
seem to apply here, but is interesting. cp -p apparently copies the
initial atime of the source.  If the source hasn't been read for a
while,  this gives a target atime way behind the target ctime, not
to mention quite different from the source atime if the source fs is
mounted without -noatime.

Bruce