From owner-freebsd-current Fri Oct 24 10:27:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA05720 for current-outgoing; Fri, 24 Oct 1997 10:27:43 -0700 (PDT) (envelope-from owner-freebsd-current) Received: from smtp03.primenet.com (smtp03.primenet.com [206.165.5.84]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA05708; Fri, 24 Oct 1997 10:27:38 -0700 (PDT) (envelope-from tlambert@usr08.primenet.com) Received: (from daemon@localhost) by smtp03.primenet.com (8.8.7/8.8.7) id KAA15493; Fri, 24 Oct 1997 10:27:31 -0700 (MST) Received: from usr08.primenet.com(206.165.6.208) via SMTP by smtp03.primenet.com, id smtpd015472; Fri Oct 24 10:27:27 1997 Received: (from tlambert@localhost) by usr08.primenet.com (8.8.5/8.8.5) id KAA13692; Fri, 24 Oct 1997 10:27:22 -0700 (MST) From: Terry Lambert Message-Id: <199710241727.KAA13692@usr08.primenet.com> Subject: Re: Recursive mount [ was Re: -STABLE reboots ] To: dwmalone@maths.tcd.ie (David Malone) Date: Fri, 24 Oct 1997 17:27:22 +0000 (GMT) Cc: tlambert@primenet.com, kato@migmatite.eps.nagoya-u.ac.jp, current@freebsd.org, freebsd-stable@freebsd.org In-Reply-To: <9710241231.aa02348@salmon.maths.tcd.ie> from "David Malone" at Oct 24, 97 12:31:28 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-freebsd-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Is this a production environment patch? > > I've a dual processor machine with about 1000 undergrads with > accounts on it, so its really a matter of time before someone > discovers how to knock it over that way. Then I don't object. ;-). > > There's not really any > > conceptual difference between root and non-root mounts, once the > > greation of a mount struct instance is abstracted from it's mapping > > into the FS hierarchy by moving the latter into common code. > > I take it that this would factor the recursive lock problem out > of the VFS code into the generic mount code, and make it a bit > more straight forward to fix? Plus enable several nomadic computing features, plus make it possible to boot from any FS type. Etc. 8-). Actually, it should drop recursion out of the picture entirely. The only check you'd need to make is at the time you map the file into the directory hierarchy. And that's trivial because you know the device for the parentfs, the child fs, and the parent of the parent... up to the root. It's a straight linear traversal (not a bad thing, since mount is relatively rare, and the number of fs's is limited). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.