Path: sparky!uunet!usc!elroy.jpl.nasa.gov!ucla-cs!ucla-mic!scott
From: sc...@pita.cns.ucla.edu (Scott Burris)
Newsgroups: comp.unix.bsd
Subject: my experience with 386BSD
Message-ID: <1992Mar16.102239.5670@mic.ucla.edu>
Nntp-Posting-Host: pita.cns.ucla.edu
Reply-To: sc...@pita.cns.ucla.edu (Scott Burris)
Organization: UCLA Campus Network Services
Date: 16 Mar 92 10:22:38 PST
Lines: 45

I've been running 386BSD now over the weekend (the old 0.0 release, not the
new) and have a few observations, questions, etc.

I tried it on 2 cheapo clone 486/33 machines, both using the OPTI chipset.  
One machine, with a Paradise VGA card came up without any trouble.  The other, 
with an STB WIND/X board (using S3's 86C911 accelerator chip), sort of boots, 
hangs with the cursor in a weird place, then if you touch any key on the 
keyboard, the system panics and reboots so quickly, I'm not sure what it 
panics with.  So all further observations are based on the Paradise machine.

Writing a disk description in /etc/disktab and labeling the disk was
no problem, but every time I boot (and fsck) I get console messages about
an error in the bad block area.  I saw someone else post a message about
this, so I'd guess disklabel isn't initializing something correctly.

The kernel doesn't look like it was gen'ed with support for a second floppy
drive.  I also have the Western Digital ethernet card, but the kernel doesn't
seem to recognize it.  I was able to configure the loopback device, run
inetd, and telnet to myself.

The tty driver bugs are driving me batty.

I've learned real quickly that when the system starts acting weird, don't
try to keep going.  When I've done that, I've ended up with damaged
filesystems (lots of DUP blocks) and had to restore from diskettes.

I didn't initially realize that the system wasn't running a C-shell and
typed "!!" to repeat the previous command -- instant panic.

I compiled and ran the obligatory "Hello World" program without problems.
When I compiled a program with an infinite loop and ran it, either the
system hangs or there's some sort of scheduler bug, because everything seems
to go dead and there's no response to ^C.

All in all, it looks very promising.  Wish there was an Adaptec 1542B
driver though -- I'll have to write one if someone else isn't working
on it so I can use some bigger disks.

Is anyone keeping a master bug list that we can refer to?

--
----------
Scott Burris
UCLA Campus Network Services
cne...@oac.ucla.edu (213) 206-4860 - OR - sc...@pita.cns.ucla.edu

Path: sparky!uunet!usc!apple!netcomsv!resonex!michael
From: mic...@resonex.com (Michael Bryan)
Newsgroups: comp.unix.bsd
Subject: Re: my experience with 386BSD
Message-ID: <1992Mar16.154007.15180@resonex.com>
Date: 16 Mar 92 15:40:07 GMT
References: <1992Mar16.102239.5670@mic.ucla.edu>
Organization: Resonex Inc., Sunnyvale CA
Lines: 30

In article <1992Mar16....@mic.ucla.edu> 
sc...@pita.cns.ucla.edu (Scott Burris) writes:
>I tried it on 2 cheapo clone 486/33 machines, both using the OPTI chipset.  
>One machine, with a Paradise VGA card came up without any trouble.  The other, 
>with an STB WIND/X board (using S3's 86C911 accelerator chip), sort of boots, 
>hangs with the cursor in a weird place, then if you touch any key on the 
>keyboard, the system panics and reboots so quickly, I'm not sure what it 
>panics with.

My system also hangs/panics in this way.  It is a 486/25 clone, OPTI
chipset, 4 MB RAM, Orchid ProDesigner IIs SVGA, 125 MB Maxtor IDE.
Since my disk is used for DOS right now, and I want to keep it that
way until I get a second disk, I disabled the hard drive before attempting
to boot.  First I tried by just uncabling the disk, but leaving the BIOS
setup unchanged.  It boots/hangs/panics just as Scott describes.  The
panic message is "panic: trap", and other messages given that I cannot
read.  If I actually disable the drive in BIOS, it behaves differently.
It goes out and apparently reads in the kernel file, then gives the
message "Operator abort --- Unable to boot 386bsd, trying 386bsd.alt",
or something like that.  It then cycles quickly through several other
filenames, giving the error "File not found --- Unable to boot xxx,
trying yyy".  It eventually works its way back to 386bsd, and the
sequence cycles ad nauseum.

I've tried disabling all shadowed RAM/ROM, and booting in non-Turbo
mode, all to no avail.  Figured I'd add my $0.02 here so that the
appropriate people might be made aware of such problems.

-- 
Michael Bryan	        mic...@resonex.com
This offer law where prohibited by void.

Newsgroups: comp.unix.bsd
Path: sparky!uunet!usc!sdd.hp.com!hp-col!bdale
From: bd...@col.hp.com (Bdale Garbee)
Subject: Re: my experience with 386BSD
Sender: no...@col.hp.com (notes)
Message-ID: <1992Mar17.054513.15447@col.hp.com>
Date: Tue, 17 Mar 1992 05:45:13 GMT
References: <1992Mar16.154007.15180@resonex.com>
Organization: HP Colorado Springs Division
Lines: 16

mic...@resonex.com (Michael Bryan) writes:
> >sort of boots, 
> >hangs with the cursor in a weird place, then if you touch any key on the 
> >keyboard, the system panics and reboots so quickly, I'm not sure what it 
> >panics with.
> 
> My system also hangs/panics in this way.

All 4 systems I tried hang in this same way.  Toshiba T2200SX notebook machine
(actually leaves the panic messages on the screen, don't remember what they
were, after hitting a key), two machines with 16Mhz 386DX processors, Symphony
chipset, and AMI bios, and an HP Vectra QS/20.

Bleepingly annoying.

Bdale

Path: sparky!uunet!mcsun!uknet!warwick!mrccrc!icdoc!frigate!lmjm
From: lm...@doc.ic.ac.uk (Lee M J McLoughlin)
Newsgroups: comp.unix.bsd
Subject: Re: my experience with 386BSD
Message-ID: <LMJM.92Mar23234929@raquel.doc.ic.ac.uk>
Date: 23 Mar 92 23:49:29 GMT
References: <1992Mar16.102239.5670@mic.ucla.edu>
Sender: use...@doc.ic.ac.uk
Organization: Dept. of Computing, Imperial College, London. U.K.
Lines: 29
In-Reply-To: scott@pita.cns.ucla.edu's message of 16 Mar 92 10:22:38 PST
Nntp-Posting-Host: raquel.doc.ic.ac.uk

I've been trying to get the latest release of 0.0 up and running on an
system whose motherboard is uses a 386DX at 25/40 Mhz, Cyrix
coprocessor, symphony chipset and AMI bios, 4M main memory, 40 Mhz
slow disk and an ST506 controller.

The bios allows me to reset about 40 different options!

I found that in turbo mode I just couldn't see the hard disk, the
pc<color> bug.  So I'm running the machine in slow 25Mhz mode, which
is a real shame.  Initially I was getting very odd problems talking to
the hard disk, a colleage spent some time looking at this and found
that accessing both the hard disk and floppy was very sensitive to the
bus clocking rates.  (I'm told its unusually for the bios to allow the
bus clocking rates to be changed.)  By slowing the bus down a bit I
was able to get at the hard disk and the floppy ok.

The machine is now in a state where all seems to work ok except that
on accessing the ethernet things seem to hang.  A "route add ..."
command hangs the machine immediately.  "telnet machine" connects me
to the machine I get the login: banner but then the machine hangs.

Still these teething problems not withstanding it is real fun.

	Lee
--
--
Lee McLoughlin.                          Phone: 071 589 5111 X 5085
Dept of Computing, Imperial College,     Fax: 071 581 8024
180 Queens Gate, London, SW7 2BZ, UK.    Email: L.McLo...@doc.ic.ac.uk