From owner-freebsd-current Mon Dec 18 08:28:25 1995 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA04877 for current-outgoing; Mon, 18 Dec 1995 08:28:25 -0800 (PST) Received: from skiddaw.elsevier.co.uk (skiddaw.elsevier.co.uk [193.131.222.60]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA04869 for ; Mon, 18 Dec 1995 08:28:15 -0800 (PST) Received: from snowdon.elsevier.co.uk (snowdon.elsevier.co.uk [193.131.197.164]) by skiddaw.elsevier.co.uk (8.6.12/8.6.12) with ESMTP id QAA00761; Mon, 18 Dec 1995 16:26:17 GMT Received: from cadair.elsevier.co.uk (actually host cadair) by snowdon with SMTP (PP); Mon, 18 Dec 1995 16:25:59 +0000 Received: (from dpr@localhost) by cadair.elsevier.co.uk (8.6.12/8.6.12) id QAA05747; Mon, 18 Dec 1995 16:25:54 GMT From: Paul Richards Message-Id: <199512181625.QAA05747@cadair.elsevier.co.uk> Subject: Re: FreeBSD-current-stable ??? To: phk@critter.tfs.com (Poul-Henning Kamp) Date: Mon, 18 Dec 1995 16:25:54 +0000 (GMT) Cc: FreeBSD-current@FreeBSD.org (FreeBSD current mailing list) In-Reply-To: <22179.819302448@critter.tfs.com> from "Poul-Henning Kamp" at Dec 18, 95 05:00:48 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-current@FreeBSD.org Precedence: bulk In reply to Poul-Henning Kamp who said > > >I< don't know of any interface changes yet between 2.1 and 2.2 There > is no change in sysctl(3) there is no change to malloc. I would > argue that you missed your shot, unless you can show me a piece of > code that needs an ifdef for "post-2.1" That's really not the issue. I'm not expecting such clear incompatibilties that an #ifdef 2.1 would be required (although the vm header changes would in fact require that so there you are, I have an example). It's the runtime environment as much as anything that's an issue. So many bugs have been found because of the new malloc, which is a good thing but that wouldn't happen if all the folks who want to do user-land development stick to 2.1. The yp code in -current is evolving, the yp code in 2.1 might not even exist in -current. There are loads more examples I could think up if I didn't have better things to do. It should be clear to everyone that it's a dumb idea to go and tell people to work on user-land code under a different version of the OS. > > > 2) It's *IMPORTANT* to have people actually running -current even if they're > > not actively developing. If only a tiny number of active developers are > > hacking -current then we'll never have enough of an user base for that > > branch to find all the bugs that crop up from diverse use. > > Now this is true, but it doesn't mean that people should hack around > in their kernels, does it ? No, I wasn't disagreeing that there's lots of user-land things to do. I was arguing against telling them to work in a 2.1 environment. I think it might be instructive if you grabbed something from 2.1 and tried to compile it under -current. I think point 2 is much more important than point 1 anyway. -- Paul Richards. Originative Solutions Ltd. Internet: paul@netcraft.co.uk, http://www.netcraft.co.uk Phone: 0370 462071 (Mobile), +44 1225 447500 (work)