From owner-freebsd-fs Wed Jun 17 15:16:50 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA20212 for freebsd-fs-outgoing; Wed, 17 Jun 1998 15:16:50 -0700 (PDT) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA20170; Wed, 17 Jun 1998 15:16:41 -0700 (PDT) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id WAA08666; Wed, 17 Jun 1998 22:15:23 GMT Date: Thu, 18 Jun 1998 07:15:23 +0900 (JST) From: Michael Hancock To: Terry Lambert cc: Matthew.Alton@anheuser-busch.com, FreeBSD-fs@FreeBSD.ORG, Scott.Smallie@anheuser-busch.com, Hackers@FreeBSD.ORG Subject: Re: Filesystem Development Toolkit In-Reply-To: <199806172020.NAA26615@usr01.primenet.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Ok, there are 3 problems: 1) Refcounting and locking layering violations remaining in 3 vops. 2) cn_path freeing layering violations. 3) vnode object coherence. 1 and 2 are in the same class of complexity, mostly grunt work but still requires someone reasonably knowledgeable of the bsd file systems. A solution to 3 that allows you to take advantage of caching at different layers is the tough one. An easier solution is to never cache, but always proxy to the right object. Heidemann has a nice paper on the tough solution, but it's hard to verify that a hands free approach to locking and distributed operations is correct. Which is why I'd like to look at Simon's DLM. Regards, Mike Hancock To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message