From owner-freebsd-current Sun Aug 24 11:36:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA14407 for current-outgoing; Sun, 24 Aug 1997 11:36:12 -0700 (PDT) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id LAA14400 for ; Sun, 24 Aug 1997 11:36:10 -0700 (PDT) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA01251; Sun, 24 Aug 1997 11:35:40 -0700 From: Terry Lambert Message-Id: <199708241835.LAA01251@phaeton.artisoft.com> Subject: Re: another mount arg incompatibility To: bde@zeta.org.au (Bruce Evans) Date: Sun, 24 Aug 1997 11:35:40 -0700 (MST) Cc: bde@zeta.org.au, current@FreeBSD.ORG In-Reply-To: <199708241628.CAA16264@godzilla.zeta.org.au> from "Bruce Evans" at Aug 25, 97 02:28:50 am X-Mailer: ELM [version 2.4 PL24] 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 > `mount -u -t msdos ...' fails early because mount_msdos doesn't support > the MNT_UPDATE flag although msdosfs sort of supports it. Support for > MNT_FORCE is similarly broken. Support for MNT_SYNCHRONOUS is slightly > more broken - it is a general vfs flag (it mainly affects vn_write(), so > it belongs in MNT_STD more that some of the other options in MNT_STD > (it doesn't apply to inherently readonly file systems, but neither does > MNT_NOATIME, and support for quotas is not standard...). Perhaps the > support shouldn't be limited in individual mount utilities. However, > many kernel mount routines just ignore unsupported flags. This code should be centralized. The flags processing is relatively independent of the per FS mount processing. Same thing goes for root vs. non-root mounts, and who does the work covering the vnode, IMO. The per VFS mount code is very bogus. This crap is why some FS's can't be your root, and others export poorly via NFS. 8-(. The CD9660 is particularly gross, using options and then backing up enforcement with autodetect, when it could just autodetect and keep it's mouth shut. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.