Port-vax archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: VS2000 + NetBSD



Holm Tiffe wrote:

> Anders Magnusson wrote:
> 
> > Den 2018-04-11 kl. 19:11, skrev Holm Tiffe:
> > > ..that thing is going complicated.
> > >
> > > I have an VS3100 M76 with an installed NetBSD 1.5.3 on it that
> > > I used for Netbooting (Mopd) and I connected the fresh installed Disk to
> > > it and tried to boot..failed.
> > > Next I've netbooted the m76 and looked at the disk..all ok so far.
> > > I've erased the first 16 blocks from the disk, copied the first Blocks
> > > from the M76 disk and modified the disklable according the actuel
> > > filesystem sizes (VS2000 disk is smaller, only the last partition
> > > differs).
> > > With that disk the VS2000 with the modified Firmware could start the
> > > primary loader (sdboot) but the loader couldn't find boot or boot.vax
> > > the 2nd loader.
> > >
> > > I've found out that the primary loader on the M76 isn't the 1.5.3.
> > > sdboot since there is aoccurence of cd9660 in the code, I do hvave some
> > > other NetBSD Version on the disk that seems to be the source of that code,
> > > the xxboot from that Version written with disklabel from 1.5.3 to the
> > > disk makes an almost working primary bootstrap, but fails to boot the
> > > 2nd.
> > > I've tried to netboot that version, it results in an endless row of
> > > stray irq 18 ipl 31 on both machines. Don't know anymore what verion
> > > that is, think 1.6.0 or 1.6.2 or so..
> > >
> > > The begin of partition a on the disk is offset 16 blocks to the begin of
> > > the disk. How is the concrete layout of that 16 blocks, from where to
> > > where is the lable and where is the primary bootstrap?
> > All disks on all VAXen uses the same bootstrap code, which is located in
> > sectors 0-15 on the disk.  The disklabel is embedded in a bunch of code
> > in block 0, together with some magic numbers to make the firmware happy.
> > 
> > The disklabel itself is at LABELOFFSET, which is 64 bytes into block 0.
> > 
> > For the VS2k the boot block uses the provided ROM routines to load /boot,
> > so if something did not function as expected with the patched routines
> > the booting would fail.
> > 
> > -- Ragge
> 
> Thanks Ragge.
> 
> I'll look at that.
> My Problem is, that the disk that succesfully boots in the VS3100 M76
> won't boot in the VS2000.
> There must be something wrong in that old disklabel/bootstrap code
> besides of the fact that I can't find any reference that someone has
> successfully booted NetBSD on an modified VS2000 in the past.
> 
> The primary bootstrap xxboot has changed between 1.5.3 and 1.6, that are
> different files and I remember that I had all kind of trouble with
> disklabel and bootstraps on the different VAXstations that I tried to
> install in the past (to find somewhere on that list).
> 
> The oldest occurence of that problem that I know of was the old emulex
> uc07 controller from the "Applicon Workstation" that I had in th 90ies. Your
> remember for sure that I've sent the UC07 with some beer buddles
> ("Freiberger") to you in Sweden to find out why booting had not worked.
> (still don't know)
> Someone other sent me an UDA50 to this time (the Applicon was an MVII
> and had an QBUS UNIBUS Bridge) and I booted the machine from it, toggeled
> in the bootstrap code with ODT every time before booting..
> That Applicon worked as FTP server for some years afterwards with that
> UDA50.. Don't remember what Version of NetBSD that was, for sure before
> 1.5.3.
> The UC07 could boot VMS (4.something) but not NetBSD. The UDA50 could
> boot NetBSD, but the firmware from the CPU don't supported it.
> BTW: I think to remember that this MVAX CPU was only an double sized QBUS
> Board, here an KA630 is a quad sized one, which CPU type could that be?
> Have double sized CPUs existet or ist that my failing memory?
> 
> 
> Is there a newer then 1.5.3 NetBSD version from which is known that the
> GENERIC or Install Kernel works on an VS2000 on at least the original
> MFM disk or with netboot? (w/o that int18 ipl31 thing).
> 

Ok, I think I'm giving up now.
I've done several experiments now, begun to install 1.6.2 but the kernel
panics every time when one tries to copy some things over nfs on both
the netbooted VS2000 and the VS3100 M76.
Switched to 1.6.1, that's stable so far. I was able to create a
disklabel, make filesystems, copy base.tgz and etc.tgz to the disk.
Next I've installed the sdboot with installboot ...doesn't work at all.
Cleaned the first 16 sectors, reinstalled the label and used disklabel
-B this time.
The disk is now booting with the VS3100 M76, but not on the VS2000.

KA410-W V2.3           

F...E...D...C...B...A...9...8...7...6...5?..4...3_..2_..1...


 ?  C  0080  0000.4001
??  5  0001  0000.0002

>>> b dka2


-DKA2

>> NetBSD/vax boot [1.11 Wed Apr  9 05:16:34 UTC 2003] <<
>> Press any key to abort autoboot 0
open netbsd.vax: Unknown error: code 3149680
> boot netbsd
open netbsd: Unknown error: code 3149680
netbsd: boot failed: Unknown error: code 3149680
> boot netbsd.gz
open netbsd.gz: Unknown error: code 3149680
netbsd.gz: boot failed: Unknown error: code 3149680
> boot netbsd.old
open netbsd.old: Unknown error: code 3149680
netbsd.old: boot failed: Unknown error: code 3149680
> boot gennetbsd
open gennetbsd: Unknown error: code 3149680
gennetbsd: boot failed: Unknown error: code 3149680
>
It seems that the 2nd bootstrap is not seeing any files from the disk.
That means, first bootstrap was loaded, 2nd to but the 2nd can't handle
the VS2000 SCSI disks here.

Netbooting works fine so far:

>>> b esa0


-ESA0

>> NetBSD/vax boot [Jan  6 2002 22:13:30] <<
>> Press any key to abort autoboot 0
Trying BOOTP
Using IP address: 192.168.50.116
myip:  (192.168.50.116)
root addr=192.168.50.49 path=/data/exports/vs3138/
1569636+211352 [74+156512+112929]=0x1f4b84
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002
    The NetBSD Foundation, Inc.  All rights reserved.
Copyright (c) 1982, 1986, 1989, 1991, 1993
    The Regents of the University of California.  All rights reserved.

NetBSD 1.6.1 (GENERIC) #0: Wed Apr  9 05:42:03 UTC 2003
    autobuild%tgm.daemon.org@localhost:/autobuild/netbsd-1-6/vax/OBJ/autobuild/netbsd-1-6/src/sys/arch/vax/compile/GENERIC

VAXstation 2000
cpu: KA410
total memory = 6124 KB
avail memory = 2356 KB
using 102 buffers containing 408 KB of memory
mainbus0 (root)
vsbus0 at mainbus0
vsbus0: interrupt mask 0
le0 at vsbus0 csr 0x200e0000 vec 120 ipl 14 maskbit 5 buf
0x3bc000-0x3cbfff
le0: address 08:00:2b:06:69:f3
le0: 32 receive buffers, 8 transmit buffers
dz0 at vsbus0 csr 0x200a0000 vec 304 ipl 14 maskbit 6
dz0: 4 lines
lkkbd0 at dz0
wskbd0 at lkkbd0 (mux ignored)
lkms0 at dz0
wsmouse0 at lkms0 (mux ignored)
hdc0 at vsbus0 csr 0x200c0000 vec 774 ipl 14 maskbit 0
si0 at vsbus0 csr 0x200c0080 vec 770 ipl 14 maskbit 1
si0: NCR5380, SCSI ID 7
scsibus0 at si0: 8 targets, 8 luns per target
smg0 at vsbus0 csr 0x200f0000 didn't interrupt
scsibus0: waiting 2 seconds for devices to settle...
sd0 at scsibus0 target 2 lun 0: <SEAGATE, ST31200N, 8590> SCSI2 0/direct
fixed
sd0: 1006 MB, 2700 cyl, 9 head, 84 sec, 512 bytes/sect x 2061108 sectors
sd0: async, 8-bit transfers
Kernelized RAIDframe activated
boot device: le0
root on le0
nfs_boot: trying DHCP/BOOTP
nfs_boot: DHCP next-server: 192.168.50.49
nfs_boot: my_addr=192.168.50.116
nfs_boot: my_mask=255.255.255.0
nfs_boot: gateway=192.168.50.254
root on unic:/data/exports/vs3138/
root file system type: nfs
/etc/rc.conf is not configured.  Multiuser boot aborted.
Enter pathname of shell or RETURN for sh: 
Terminal type? [unknown] vt102
Terminal type is vt102.                                                 
We recommend creating a non-root account and using su(1) for root
access.
# 

disklabel is fine: (and the VS3100 can boot this!)

# disklabel sd0c
# /dev/rsd0c:
type: SCSI
disk: ST31200N
label: fictitious
flags:
bytes/sector: 512
sectors/track: 84
tracks/cylinder: 9
sectors/cylinder: 756
cylinders: 2700
total sectors: 2061108
rpm: 3600
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0           # microseconds
track-to-track seek: 0  # microseconds
drivedata: 0

8 partitions:
#        size    offset     fstype  [fsize bsize cpg/sgs]
 a:    262144        16     4.2BSD   1024  8192   112   # (Cyl.    0*- 346*)
 b:    262144    262160       swap                      # (Cyl.  346*- 693*)
 c:   2061108         0     unused      0     0         # (Cyl.    0 - 2726*)
 d:    131072    524304     4.2BSD   1024  8192   116   # (Cyl.  693*- 866*)
 e:   1024000    655376     4.2BSD   1024  8192   112   # (Cyl.  866*- 2221*)
 f:    381732   1679376     4.2BSD   1024  8192   112   # (Cyl. 2221*- 2726*)
#

and so is the root fs:

# mount /dev/sd0a /mnt
# ls -l /mnt
total 2084
-rw-r--r--   2 root  wheel      939 Apr  9  2003 .cshrc
-rw-r--r--   2 root  wheel      493 Apr  9  2003 .profile
drwxr-xr-x   2 root  wheel      512 Apr 12 08:48 altroot
drwxr-xr-x   2 root  wheel      512 Apr 12 09:40 bin
-r-xr-xr-x   1 root  wheel    65224 Apr 12 11:11 boot
-r-xr-xr-x   1 root  wheel    65224 Apr  9  2003 boot.old
-r-xr-xr-x   1 root  wheel    65224 Apr 12 11:11 boot.vax
drwxr-xr-x   4 root  wheel     1536 Apr 12 09:57 dev
drwxr-xr-x  17 root  wheel     1536 Apr 12 09:56 etc
drwxr-xr-x   2 root  wheel      512 Apr 12 08:45 home
drwxr-xr-x   2 root  wheel      512 Apr  9  2003 mnt
-rw-r--r--   1 root  wheel  1894085 Apr 12 11:12 netbsd.vax
-r-xr-xr-x   1 root  wheel    11264 Apr  9  2003 pcs750.bin
drwxr-xr-x   2 root  wheel      512 Apr 12 09:56 root
drwxr-xr-x   2 root  wheel     2048 Apr  9  2003 sbin
drwxr-xr-x   2 root  wheel      512 Apr  9  2003 stand
drwxrwxrwt   2 root  wheel      512 Apr  9  2003 tmp
drwxr-xr-x   2 root  wheel      512 Apr 12 08:45 usr
drwxr-xr-x   2 root  wheel      512 Apr 12 08:45 var
# 

I don't know much of what I could do here other than giving up,
since I don't know what exactly is going on with the patched
firmware or what prevents the 2nd bootstrap to see the SCSI disk.

Conclusion: No, a VS2000 booting NetBSD from a SCSI Disk isn't
possible. Trying newer NetBSD Releases makes no sense, since they
are simply to big. VMS may work.

Someone that want's to correct me?

Regards,

Holm
-- 
      Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, 
     Freiberger Straße 42, 09600 Oberschöna, USt-Id: DE253710583
info%tsht.de@localhost Fax +49 3731 74200 Tel +49 3731 74222 Mobil: 0172 8790 741



Home | Main Index | Thread Index | Old Index