From owner-freebsd-hackers Thu Apr 23 23:24:21 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA01019 for freebsd-hackers-outgoing; Thu, 23 Apr 1998 23:24:21 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from smtp03.primenet.com (daemon@smtp03.primenet.com [206.165.6.133]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA00887 for ; Thu, 23 Apr 1998 23:23:14 -0700 (PDT) (envelope-from tlambert@usr09.primenet.com) Received: (from daemon@localhost) by smtp03.primenet.com (8.8.8/8.8.8) id XAA16013; Thu, 23 Apr 1998 23:23:12 -0700 (MST) Received: from usr09.primenet.com(206.165.6.209) via SMTP by smtp03.primenet.com, id smtpd015967; Thu Apr 23 23:23:04 1998 Received: (from tlambert@localhost) by usr09.primenet.com (8.8.5/8.8.5) id XAA07872; Thu, 23 Apr 1998 23:22:59 -0700 (MST) From: Terry Lambert Message-Id: <199804240622.XAA07872@usr09.primenet.com> Subject: Re: Protected mode instructions which reduce to noop. To: dec@phoenix.its.rpi.edu (David E. Cross) Date: Fri, 24 Apr 1998 06:22:59 +0000 (GMT) Cc: jlemon@americantv.com, perlsta@fang.cs.sunyit.edu, freebsd-hackers@FreeBSD.ORG In-Reply-To: from "David E. Cross" at Apr 23, 98 12:32:48 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > What VM type architecture were you referring to? > > I am referring to IBM's VM operating system. and here is the original > message from -hackers from nov-1997: Look into "Protected mode software architecture" from MindShare. The general problem is not that Intel does not allow this, but that implementing a VMM (Virtual Machine Manager) is not trivial. One big piece of windows 95 is the VMM. The other big problem is that the emulated machines lack simultanaeity; this is mostly because some of the instructions are very, very hard to emulate. If you look at the BOCHS code, you will see that it's perfectly possible to implement a VM on Intel. BOCHS doesn't use the native instructions that it can, and it doesn't use a kernel assist in the person of the "VM" bit in control register 4 on 486 and above chips. This makes BOCHS slower than it absolutely has to be, but it also makes it run on non-Intel machines. It would probably be worth implementing a full-on VMM (in the MindShare documented sense), if only to allow running of MS Os's in a window on FreeBSD. The IBM VM architecture is logically complete -- that is, nearly all of the instruction emulation implementation is in hardware, so a VMM is (relatively) *much* lighter weight. I had the opportunity to use VM/CMS and VM/UTS (I was writing an IBM3101 block mode terminal emulator for UNIX) back in the mid 80's, and I was impressed that their performance nearly equalled unemulated hardware. IBM is technically competent in most regards (brilliant in some, which is surprising; management tends to fear brilliant things). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message