Date: Wed, 3 Jan 1996 18:59:46 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: msmith@atrad.adelaide.edu.au (Michael Smith) Cc: terry@lambert.org, jdl@jdl.com, jkh@time.cdrom.com, obrien@cs.ucdavis.edu, freebsd-hackers@freebsd.org Subject: Re: X for install Message-ID: <199601040159.SAA16479@phaeton.artisoft.com> In-Reply-To: <199601040143.MAA09628@genesis.atrad.adelaide.edu.au> from "Michael Smith" at Jan 4, 96 12:13:19 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > I maintain that if Microsoft can do it, *anyone* can do it. > > Ok Terry, you win. VM86() it is; now can you or one of the other VM > wizards dribble enough of it out so that we can set up a context in > which it is safe to make BIOS calls? Well, this didn't *necessarily* need a VM86(). But that *is* one soloution... I was thinking on the lines of a weenie DOS program to blow the 32 bit offsets in the partition table before install. > I certainly don't have the low-level iNTeL smarts required to do this, > and all of the discussions about it so far have always escalated to the > point where it looks like a monster project, and then petered out. Speaking of Peter, Peter Wemm was getting his mind around locore.s when I left on my extended vacation back in December, so he might be the ideal candidate. 8-) 8-). I have someone else who I'd nominate, since he has been all over the Win95 VM system very recently, but he'd probably kill me for it... > I seem to recally you saying that the NetBSD folks had it working; if this > is the case, what are the critical differences between their code and > ours that would need to be changed? They had a VM86() mode working well enough for DOSEMU. This is not quite enough for a "call this BIOS call on my behalf in a Virtual 8086 machine that prevents kernel reentrancy, please". Which is really what needs to be done, unless you dick with the DOS stack for the low INT calls (there's a nice TSR book that describes doing exactly this with Borland Turbo C). > In the current scenario, what we want is an equivalent of IBM's OS/2 int13 > driver, and most likely similar for int15 (EISA) and while we're at it > int10 would be nice (BIOS video). Actually, you *don't* want video BIOS for anything more than mode setting. Most VGA cards that emulate EGA and CGA registers (Paradise, etc.) disable interrrupts instead of waiting for the vertical blank interrupt to handle screen draws. A no-no if you have active serial ports or a network card, floppy tape, etc. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199601040159.SAA16479>