From owner-freebsd-commit Mon Sep 11 17:28:52 1995 Return-Path: commit-owner Received: (from majordom@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id RAA13206 for freebsd-commit-outgoing; Mon, 11 Sep 1995 17:28:52 -0700 Received: (from majordom@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id RAA13193 for cvs-all-outgoing; Mon, 11 Sep 1995 17:28:49 -0700 Received: (from majordom@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id RAA13180 for cvs-sys-outgoing; Mon, 11 Sep 1995 17:28:47 -0700 Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id RAA13174 ; Mon, 11 Sep 1995 17:28:35 -0700 Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id RAA02342; Mon, 11 Sep 1995 17:28:09 -0700 From: "Rodney W. Grimes" Message-Id: <199509120028.RAA02342@GndRsh.aac.dev.com> Subject: Re: cvs commit: src/sys/kern vnode_if.sh To: nate@rocky.sri.MT.net (Nate Williams) Date: Mon, 11 Sep 1995 17:28:08 -0700 (PDT) Cc: jkh@time.cdrom.com, pst@shockwave.com, bde@freefall.freebsd.org, CVS-commiters@freefall.freebsd.org, cvs-sys@freefall.freebsd.org In-Reply-To: <199509120008.SAA04023@rocky.sri.MT.net> from "Nate Williams" at Sep 11, 95 06:08:57 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 2537 Sender: commit-owner@FreeBSD.org Precedence: bulk > > Rodney W. Grimes writes: > Paul Traina writes: > > > Maybe it's time to ask the general question: > > > > Why do we care about non-ansi compilers? > > Jordan responds: > > > we should go to full c++ style prototypes and obey only those > > sylististic conventions necessary to making things like `ctags' work. > > That's about as far as I think it's necessary to go with "backwards > > compatability" in source code. > > Rod's replies with: > > Then you have not thought about the facts that the BSD source code > > base is used in other OS's that do _not_ have a fully ansi compliant > > compiler and are not going to be getting one any day soon. > > BSD source code is used, but making FreeBSD source code portable to > legacy systems shouldn't be a goal IMHO. > > > I love the fact that I can take large hunks of BSD user land code and > > haul them over to my discrepent old Domain/IX SR 9.1 system and compile > > them up to make the system somewhat more palatable, and hell if I am > > going to go port gcc to this thing :-(. > > My response to this is 'who cares'? We are not in the business of > supporting Domain/IX SR 9.1, and a significant percentage of our time is > spent maintaining this ability with no gain to FreeBSD. If a Domain/IX > person wants to use BSD code, the 4.4lite and 4.4lite2 tapes are just as > available to them as they are to us. If we've made fixes to the code, > then it's my opinion that they can do the same thing we've done with a > lot of code and back-port the changes. > > There is plenty enough work to go around without making more work by > supporting pre-ANSI/C compilers. I will support this stance under one condition, that finally a _FREEBSD_ policy guide is written. Until then I consider these things noise, and it is one of the reasons I dropped away, as every one seems to want to hack, but no one wants to set policy and direction properly in documents. There is not a single written policy any place on anything that I can find except ones either I have drafted and placed here and there which are now either in the Attic or in the 1.1.5.1 repository, and ones that Jordan has sent to the mailling lists. Yea, I know, if I want this go do it, well, I am not interested in cramming stuff down the throughts of folks here, so I am going to go off and hack like every one else and forget about this noise. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD