From owner-freebsd-current Mon Mar 13 8:18:13 2000 Delivered-To: freebsd-current@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 9D0FB37BB85; Mon, 13 Mar 2000 08:18:08 -0800 (PST) (envelope-from mjacob@feral.com) Received: from semuta.feral.com (semuta [192.67.166.70]) by feral.com (8.9.3/8.9.3) with ESMTP id IAA03217; Mon, 13 Mar 2000 08:18:05 -0800 Date: Mon, 13 Mar 2000 08:18:24 -0800 (PST) From: Matthew Jacob Reply-To: mjacob@feral.com To: Poul-Henning Kamp Cc: current@FreeBSD.ORG Subject: Re: B_WRITE cleanup patch, please test! In-Reply-To: <20502.952948995@critter.freebsd.dk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > This patch is the first step towards the stackable BIO system as > sketched out on http://www.freebsd.org/~phk/Geom/ > > Please test & review. There's no code to test- it's just a sketch. It's fine as a start, but it's important that you clarify the role of node device drivers in informing the geometry device about what's what (to get information about physical limits) and how errors are to be flagged (if an out of range request comes floating thru). Presumably the latter is just marking a transaction with B_ERROR and setting b_errno to something specific that would say that the block is out of range (insert argument over 'correct' errno here), and that the range involved is the physical device limit (and what about commands that overlap the end of the device?). Presumably the former, if to give the geometry stack something worthwhile to chew on, can handle the case of devices that resize and devices where, unlike most devices currently, geometry really does matter. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message