From owner-freebsd-emulation Sat May 4 00:00:28 1996 Return-Path: owner-emulation Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id AAA16335 for emulation-outgoing; Sat, 4 May 1996 00:00:28 -0700 (PDT) Received: from zygorthian-space-raiders.MIT.EDU (ZYGORTHIAN-SPACE-RAIDERS.MIT.EDU [18.70.0.61]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id AAA16326 for ; Sat, 4 May 1996 00:00:25 -0700 (PDT) Received: (from mycroft@localhost) by zygorthian-space-raiders.MIT.EDU (8.7.4/8.6.11) id DAA16643; Sat, 4 May 1996 03:00:08 -0400 (EDT) To: Pace Willisson Cc: "Jordan K. Hubbard" , Sujal Patel , emulation@freebsd.org Subject: Re: New DOS emulator snapshot (fwd) References: <199605022057.QAA17724@mytus.blitz.com> From: mycroft@mit.edu (Charles M. Hannum) Date: 04 May 1996 02:59:53 -0400 In-Reply-To: Pace Willisson's message of Thu, 02 May 1996 16:57:57 -0400 Message-ID: Lines: 27 X-Mailer: September Gnus v0.80/Emacs 19.30 Sender: owner-emulation@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Pace Willisson writes: > > I'm sorry for the confusion. I didn't notice the "netbsd" part in the recent > announcement of a new snapshot - I guess they are not as concerned with > avoiding significant changes. I'll wait until I see an announcement > of a freebsd update, and I'll give it another shot. Rather than guessing about my intent, why don't you just *ask*? Or is it more fun to randomly guess and see how much you can annoy people? As I just wrote to Paul Vixie: I made an offer (to Keith) some time ago to work with someone at BSDI to make it run again under BSD/OS. I tried it myself with the last snapshot, but it didn't work for some reason, and I gave up in disgust. The VM86 interface in BSD/OS is *truly* baroque. Since then, I've completely eliminated most of the interrupt handling kluges and quit using sigreturn(2) directly except at startup time, so the code should be much more portable. Making it work under BSD/OS is mainly a matter of tweaking the single call to sigreturn(2) at startup (to deal with the fact that BSD/OS's sigreturn(2) doesn't work as documented) and saving/restoring the eflags bits that the kernel munches.