Re: 4.3 BSD on a uVAX (KA630) ?
- From: Holm Tiffe <holm@xxxxxxxxxxxxxx>
- Date: Sat, 11 Feb 2012 10:22:33 +0100
On 10.02.2012 21:28, Holm Tiffe wrote:
On 10.02.2012 20:44, Johnny Billquist wrote:On 2012-02-10 18.11, Holm Tiffe wrote:On 10.02.2012 12:20, Johnny Billquist wrote:On 2012-02-10 11:13, Holm Tiffe wrote:
I have an KA630 CPU, 16 Mbytes of RAM, an Emulex UC08 double SCSI to
MSCP/TMSCP Controller and a DELQA (and DEUNA) here.
I want to install an old BSD on this beast, just for fun.
I think to remember that there was an issue with MSCP disks, what was
this exactly? Which release supports MSCP Disks and TMSCP boot?
I think I can boot up BSD the first time using simh and writing my own
boot tape (done this with PDP11 stuff already).
So are there any hints on this?
I'm not aware of any problems, apart from the fact that I'm not sure
4.3BSD supports the uVAX II. I thought it was only the early Unibus
machines. And the follow up problem there was/is then that booting from
tape is not supported at all by VMB.
VMB is the boot program. On those early VAXen, VMB was actually loaded
from the boot device. And VMB in turn then booted VMS, or Ultrix, or
whatever. BSD back in those days did not actually use VMB to boot, but
instead had their own boot program, which you had to somehow get on to
the boot device, and then make the system use that image to boot BSD.
The FE (Front End) usually runs a command file on startup, which loads
VMB (or whatever), sets various registers right, and then starts the
execution. The exception was the VAX-11/750, which did not have any kind
of FE, and instead booted from the VAX directly. So, the VAX-11/750
instead have different boot roms, in order to be able to boot from
various devices, very much the same way as PDP-11s did. And for that
machine, you then needed to have a boot block on the disk, which was
read in, and started executing, which did the rest of the booting. VMB
can understand some file system things, and other tricks. Later VAXen
have VMB in the boot roms instead, removing the need for a FE and FE
Even harder to boot a VAX-11/750 from tape in other words.
Oh, by the way, if you have a DEUNA, it is not at all usable, since
a Unibus controller, and the KA630 is a Qbus machine.
Ok Johnny, I've mixed up DEUNA with DEQNA..., I do have the qbus
DEUNA - Digital Ethernet Unibus Network Adapter
DEQNA - Digital Ethernet Qbus Network Adapter
And then you have the DELUA and DELQA. Where the L stands for "Lance".
...andd now we are at that point:
>>> b mua0
?06 HLT INST
PC = 20040314
Performing normal system tests.
On the tape is the boot.fs from NetBSD 1.6.2 with 512 byte blocks
(written on FreeBSD
$ mt status
Mode Density Blocksize bpi Compression
Current: 0x22:QIC-2GB(C) 512 bytes 40640 0x3
0: 0x22:QIC-2GB(C) 512 bytes 40640 0x3
1: 0x22:QIC-2GB(C) 512 bytes 40640 0x3
2: 0x22:QIC-2GB(C) 512 bytes 40640 0x3
3: 0x22:QIC-2GB(C) 512 bytes 40640 0x3
Current Driver State: at rest.
File Number: 0 Record Number: 0 Residual Count 0
$ dd if=boot.fs of=/dev/sa0
2400+0 records in
2400+0 records out
1228800 bytes transferred in 51.376840 secs (23917 bytes/sec)
This is NetBSD 1.6.2
$ dd if=/home/holm/boot.fs of=/dev/sa0
4800+0 records in
4800+0 records out
2457600 bytes transferred in 44.420745 secs (55326 bytes/sec)
This is NetBSD 4.0.1
Tried 4.0.1 boot.fs too, it behaves identical.
Im pretty sure that the controller the disk and the tape drive are ok.
I've installed both of them (NVRAM from the Emulex) on this system
with the EMULEX's Firmware, verified the disk and run some tape tests.
I know, that the drive, an Tandberg SLR5, works pretty well with RT11,
XXDP, RSX11 and 2.11 BSD on a Dilog SC703 wich is an Q-BUS TMSCP(-only)
to SCSI Controller.
I'll verify what happens with the actual Emulex UC08 with and PDP11/52
CPU tomorrow, but I'm sure this works. I'll try to install RT11 on the
disk with the current drives.
So here is the result with 11/53's CPU Test, same card cage,
same Controller, nothing other is installed besides the two grant continuity cards (removed the additional mem Boards for thest with the KA630 too, it has on mbyte on board, should be sufficient to try to boot)
9 8 7 6 5 4 3 2 1
Commands are Help, Boot, List, Map, Test and Wrap.
Type a command then press the RETURN key: B MU0
?KMON-F-File not found DK:STRTSB.COM
SWAP .SYS 28 RT11SB.SYS 97
DU .SYS 10 MU .SYS 14
LP .SYS 2 PIP .SAV 30
DUP .SAV 52 DIR .SAV 20
8 Files, 253 Blocks
65214 Free blocks
This is the installed RT-11 V5.07 Minimal system on DU0, needed to install the rest of the tape.
It seems that (as expected) there is nothing wrong with the Emulex, Disk and Tape. I've made the install tape myself using simh and my own programm to write simh's tape files with correct block sizes out to the
Tandberg drive in my PC (using FreeBSD). I've checket an DLT drive too, is working also flawlessly.
The Problem must be something with the NetBSDs boot tape I've made.
There are no bootblocks to be installed before boot.fs according the docs at:
>Creating boot tapes
>Fetch the bootable bootfs image from
>and write it on the beginning of the tape. Under NetBSD this is done >(for MSCP tape, with SCSI tape the name is nrst0) via:
> # mt -f /dev/nrmt0 rewind
> # dd if=boot.fs of=/dev/nrmt0
> # mt -f /dev/nrmt0 rewoffl
>Under Ultrix the tape name is different:
Since I've successfully formated and installed the disk and the tape drive with the KA630 CPU (using the ODT or how it is called here) I think the KA630 is ok too.
I do have an external TK50 drive too (in its box with an additional
Z80 Based Board to make a SCSI Drive out of it). Besides of the unreliable TK50 Tape cardridges that I have I couldn't find any difference in the behavior of this drive regarding the usage with the PDP11 so the drive is currently in my external storage.
So here the call again for ideas....
- Re: 4.3 BSD on a uVAX (KA630) ?
- From: Holm Tiffe
- Re: 4.3 BSD on a uVAX (KA630) ?
- Prev by Date: Re: 4.3 BSD on a uVAX (KA630) ?
- Next by Date: Re: 4.3 BSD on a uVAX (KA630) ?
- Previous by thread: Re: 4.3 BSD on a uVAX (KA630) ?
- Next by thread: Re: 4.3 BSD on a uVAX (KA630) ?