Return to BSD News archive
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!simtel!news.kei.com!news.mathworks.com!gatech!howland.reston.ans.net!xlink.net!zib-berlin.de!news.tu-chemnitz.de!irz401!narcisa.sax.de!not-for-mail
From: j@narcisa.sax.de (J Wunsch)
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: boot problems with latest snap
Date: 2 Jun 1995 10:07:33 +0200
Organization: Private U**x site, Dresden.
Lines: 54
Distribution: usa
Message-ID: <3qmgs5$s5c@bonnie.tcd-dresden.de>
References: <PHILS.95May27225234@satori.tv.tek.com> <PHILS.95May28082804@satori.tv.tek.com> <3qhd1g$ht0@bonnie.tcd-dresden.de> <PHILS.95May31231311@satori.tv.tek.com>
Reply-To: joerg_wunsch@uriah.heep.sax.de
NNTP-Posting-Host: 192.109.108.139
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Phil Staub <phils@satori.tv.tek.com> wrote:
>But the SNAP floppy hangs IMMEDIATELY the first time the BIOS reads it
>to boot (i.e., after the BIOS's seek/recalibrate sequence on each drive).
>At this point, it takes a hard reset or power cycle to recover.
There have been changes to the bootblocks that were made to support an
automatic selection between keyboard and graphics console, and a
serial console. (The logic is that a missing keyboard falls back to
the serial console.) Anyway, there is no universal BIOS support to
detect if a keyboard is connected, so there has to be a keyboard probe
in the (very limited in size) primary bootstrap. This probe routine
apparently causes some grieve for several people, and hangs the
system. It has been decided to back this functionality out for 2.0.5,
in favour of the much larger user base who never want to use a serial
console, as opposed to the few people who would really like this
functionality.
I'm interested in knowing _all_ people who've got problems with the
keyboard probe in the boot blocks, since i'm intending to get it
finally working for 2.1.
>Question: I had to download a 1.44 Meg floppy version of the 2.0R boot
>floppy before I could get it to boot from my 1.44 Meg A: drive. But I
>didn't find a 1.44 Meg version of the SNAP floppy. Is there supposed
>to be one? Or was I correct in assuming that the same file can be
>installed on either kind of disk?
You're correct. For 2.0, the full disk has been somewhat more than
1.2 MB, so there have been two different images (with the 1.2 MB image
being slightly ``cleaned up''). We're now back at a single image
again, as it used to be all the time.
>My theory now is that BIOS (which normally detects media type from the
>boot sector) is being told by the boot sector that there are 15
>sectors per track (like a 1.2Meg disk), instead of the 18 that are on
>a 1.44 Meg disk. Apparently some BIOSes don't care, but unfortunately
>mine seems to not like it. Anybody want to buy a slightly used BIOS?
Nope. You're confusing this with DOS' BPB (BIOS parameter block, part
of the boot sector). The BIOS knows the drive type only out of the
CMOS, and handles it properly. Since the primary bootstrap is only 15
sectors in size, it's always possible to load it without knowing the
drive type (sector 1 is loaded by the BIOS boot loader, and will then
continue to load sectors 2 thru 15). If you look into the routine
devopen() in disk.c in the bootstrap loader (sys/i386/boot/disk.c),
you'll find how the bootstrap then finds about the sectors per track,
and uses it in order to finally load the kernel.
--
cheers, J"org private: joerg_wunsch@uriah.heep.sax.de
http://www.sax.de/~joerg/
Never trust an operating system you don't have sources for. ;-)