From owner-freebsd-current Tue Sep 29 07:43:56 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA02585 for freebsd-current-outgoing; Tue, 29 Sep 1998 07:43:56 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from whistle.com (s205m131.whistle.com [207.76.205.131]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA02563 for ; Tue, 29 Sep 1998 07:43:47 -0700 (PDT) (envelope-from archie@whistle.com) Received: (from smap@localhost) by whistle.com (8.7.5/8.6.12) id HAA22024; Tue, 29 Sep 1998 07:43:28 -0700 (PDT) Received: from bubba.whistle.com(207.76.205.7) by whistle.com via smap (V1.3) id sma022022; Tue Sep 29 07:43:07 1998 Received: (from archie@localhost) by bubba.whistle.com (8.8.7/8.6.12) id HAA21302; Tue, 29 Sep 1998 07:43:06 -0700 (PDT) From: Archie Cobbs Message-Id: <199809291443.HAA21302@bubba.whistle.com> Subject: Re: DEVFS? In-Reply-To: <199809252023.VAA03634@woof.lan.awfulhak.org> from Brian Somers at "Sep 25, 98 09:23:43 pm" To: brian@Awfulhak.org (Brian Somers) Date: Tue, 29 Sep 1998 07:43:06 -0700 (PDT) Cc: adhir@rocks.worldbank.org, current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Brian Somers writes: > I think Julian agrees - none unless the SLICE stuff is implemented. > I've no idea why it was taken out either - maybe Julian made the > diagnostics too verbose :-) > > Seriously, IMHO, the DEVFS/SLICE commit was as impressive as the > recent [EPC]-days (and D-day would have had a certain ring to it). > It was a far more radical idea though - both a strength and a > weakness! It would be a shame if the SLICE idea wasn't implemented in *some* form, either Julian's or Poul's model or somewhere in between. Maybe it's just not possible to reach 'rough concensus' but I doubt that. More likely, the relatively small number of people who understand all of the issues (not me) must simply find time to work out how to do it. This is not easy because folks are busy and there's a lot of complexity and everybody loves their own pet idea for how to do it. Also, there's a minimum amount of communication between the relevant parties that is required. However, it's the opportunities to re-architect major chunks of code that define whether an O/S evolves and improves or wallows in mediocrity. The non-modularity of FreeBSD's disklabel/driver/filesystem code will make future cool ideas (eg, striping) harder to implement. It would be a shame to miss an opportunity to fix this. As a success story, we have the unified buffer/VM cache. DEVFS is close, but needs to be finished so it can become 'standard'. Etc.. Seems to me like the SLICE idea just needs more thoughtful discussion (probably will have to wait until after '3-day' :-) -Archie ___________________________________________________________________________ Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message