From owner-cvs-all Mon Nov 5 23:51:43 2001 Delivered-To: cvs-all@freebsd.org Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by hub.freebsd.org (Postfix) with ESMTP id 6E96E37B416; Mon, 5 Nov 2001 23:51:37 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.9.3/8.9.3) with UUCP id IAA09119; Tue, 6 Nov 2001 08:51:36 +0100 (CET) Received: (from j@localhost) by uriah.heep.sax.de (8.11.6/8.11.6) id fA67c2T64997; Tue, 6 Nov 2001 08:38:02 +0100 (MET) (envelope-from j) Date: Tue, 6 Nov 2001 08:38:02 +0100 From: Joerg Wunsch To: John Baldwin , "Kenneth D. Merry" Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/boot/i386/cdboot Makefile cdboot.s Message-ID: <20011106083802.H43204@uriah.heep.sax.de> Reply-To: Joerg Wunsch References: <20011105232259.B80377@uriah.heep.sax.de> <20011105154810.A21812@panzer.kdm.org> <20011105154810.A21812@panzer.kdm.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from jhb@FreeBSD.org on Mon, Nov 05, 2001 at 02:53:18PM -0800 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG As Kenneth D. Merry wrote: > The code Joerg wrote was done in mid-1997, and IIRC, it would only > work properly with version 1.23 or maybe 1.25 of the Adaptec 2940UW > SCSI BIOS. The other BIOS versions didn't do El Torito right at > that point. None of them did it right at that point. ;-) Adaptec 1.23 was the only one that did at least that part right that was supposed to be implemented. That is, BIOS int 0x13 extensions (which are needed to read the CD-ROM using BIOS functions), plus the floppy-emulation booting. Non-emulation booting wasn't even supposed to be implemented, according to your Adaptec contacts... Rev 1.26 then broke again the BIOS int 0x13 extensions. Any ATAPI BIOS i could find around that time was too stupid to work at all, they all broke the int 0x13 extensions. At some point, i think i've also tried to boot off a Tekram using a Symbios Logic chip. No idea about non-emulation booting, but at least the part that was done in Adaptec 1.23 also worked there. On-board SDMS BIOSes couldn't boot CD-ROMs at all then. I'm really wondering, WinNT must have been around at that time. Ah, hmm, probably only 3.5x, so the logo program you mention, John, might have appeared afterwards only. Anyway, after discovering that none of the existing BIOSes even bothered to implement non-emulation booting, i discarded the idea to use it. Anyway, i'm glad all this is possible now. Of all that El Torito cr*p, non-emulation booting at least looked like the most logical thing to me. Btw. John, did you find a way using only BIOS functions and the El Torito descriptor blocks to find out where the session you've been booted from actually is rooted at? I didn't. So when booting from something that was not the first session, the starting block number of the session to boot needed to be manually entered at the boot prompt. :( -- cheers, J"org .-.-. --... ...-- -.. . DL8DTL http://www.sax.de/~joerg/ NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message