From owner-freebsd-hackers Tue Jan 9 00:51:13 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id AAA22720 for hackers-outgoing; Tue, 9 Jan 1996 00:51:13 -0800 (PST) Received: from synthcom.com (beacon.synthcom.com [198.145.98.253]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id AAA22715 for ; Tue, 9 Jan 1996 00:51:08 -0800 (PST) Received: from synthcom.com (synthcom.com [198.145.98.1]) by synthcom.com (8.6.9/8.6.9) with SMTP id AAA04080; Tue, 9 Jan 1996 00:47:09 -0800 Date: Tue, 9 Jan 1996 00:47:09 -0800 (PST) From: Neil Bradley To: Robert Nordier cc: Michael Smith , hackers@freebsd.org Subject: Re: DOS File system fixes In-Reply-To: <199601090822.KAA01144@eac.iafrica.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org Precedence: bulk On Tue, 9 Jan 1996, Robert Nordier wrote: > Wasn't it Michael Smith who said: > > Robert Nordier stands accused of saying: > > > What is the current status of this part of the project, is more help > > > needed, and what steps should one take to get involved? > > You're involved already; welcome aboard 8) > > We want to set up an environment wherein it is possible to make real-mode > > BIOS (and possibly other) calls in order to work with devices for which > > we have no protected-mode support. > Of course, under something like Windows, through DPMI, even a "straight" > BIOS call (like int 0x13) tends to actually pull in a whole lot more: > IO.SYS, DBLSPACE, the Windows WDCTRL device, SMARTDRV, etc.... Whereas, > from FreeBSD, it would be just BIOS. It would be heinous to support something like DBLSPACE. ;-) > Maintaining the BIOS environment is a way that is acceptable to both FBSD > and the BIOS might also get tricky. There are parts of the BIOS you > probably just don't want to get called: like 0x15/0x87 which is only a > helper routine but attempts a return to protected mode with interrupts > disabled. The other option is to write BIOS equivalent routines to do the dirty work. This would work OK with ST506 style interfaces, but SCSI might be a bit trickier. Does FreeBSD nuke the C0000-DFFFF region? If so, the SCSI BIOS would be gone, and any INT 13's to the BIOS would be non-existent. Something like INT 14h could be simulated very easily with IOCTL calls to the sio's. It would be pretty difficult to include XMS, as XMS tends to like linear memory segments. It may even have to be a configuration option for bootup. -->Neil ------------------------------------------------------------------------------- Synthcom System's homepage: http://www.synthcom.com/ Europa Upgrade, Synth patches (D-50, Xpander/Matrix 12), used gear pricelist