From owner-freebsd-ports Thu Dec 4 01:55:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id BAA07898 for ports-outgoing; Thu, 4 Dec 1997 01:55:08 -0800 (PST) (envelope-from owner-freebsd-ports) Received: from home.gtcs.com (home.gtcs.com [206.54.69.238]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id BAA07893 for ; Thu, 4 Dec 1997 01:55:04 -0800 (PST) (envelope-from bruce@gtcs.com) Received: from home.gtcs.com (localhost.gtcs.com [127.0.0.1]) by home.gtcs.com (8.8.5/8.8.5) with SMTP id CAA15177 for ; Thu, 4 Dec 1997 02:52:54 -0700 (MST) Date: Thu, 4 Dec 1997 02:52:53 -0700 (MST) From: Bruce Gingery To: freebsd-ports@freebsd.org Subject: Q: When one dist makes several useful ports? Message-ID: Disposition-Notification-To: bgingery@gtcs.com MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-ports@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've done a Quick-and-Dirty port of bochs v971118 - a PC Emulator that reportedly can run: DOS 5.0 and up (perhaps back to 3.2 or 3.3?) Windows 3.1 Windows95 Minix and in the future, most likely Linux BeOS - PC and FreeBSD each under it. It uses its own embedded user-level vn device to emulate DOS Diskette and FAT supporting HDD -in-a-file- similar to other products like SoftPC. It was an easy port to do, with GNU-configure already in place, but the varied options *do* raise a question. It can be compiled to support 1. an 8088/8086 2. an i286 OR 3. an i386 (with 8086 mode too) Now, because the "installed memory" needs to both be compiled and user-level configured, it would appear at this point that at least TWO distinct ports would be advisable from this package. 1. Minimal 8086 (for old DOS apps, including games) 2. 16M full-AT class machine able to run Windows95 or Minix and better support Windows3.1 Yet as one might guess, MOST of the files are shared between the two optional compiles - including a VGA BIOS image, a PC BIOS image, and all of the documentation. Would it therefore be best to make *three* portballs out of this? 1. Common components 2. top-end support 3. minimal-impact .. with #2 and #3 both dependent on #1? I don't have time for this, so if anyone would like to take what I've started and run with it -- WONDERFUL. I'm just hitting some constraints where "wine" just won't cut it, for my minimal need for DOS/Doze compat. At the moment, I have (what seems to be) a fully functional port of the full-blown version. portlint returns ONE fatal error, as I did the same as the Wine port in specifying the DATE in the DIST block - no warnings. Bruce Gingery