[pups] 2.11BSD disklabel-programm

[pups] 2.11BSD disklabel-programm

[lothar felten]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2676 bytes --]

hi there!

well the pdp was once a pdp 11/23+ so
enclosure and backplane should be
the same as a 11/23 (ba 23). backplane should
be Q22/CD configuration, but
i´ll open the box and look for the 501-
number to be sure. on the back there
is a sticker saying this is a 11/73. the
cpu-board is a 11/83.(i´ll pass
the number too). do all 11/83 use PMI ? but
the memory seems to work, or has
the cpu board some memory on it?
when i picked up the box they booted it, i
suppose this configuration was the
way they used it there and should have
worked.
the RD54 controller has a 50pin ribboncable
wich goes to a small board (wich
was hanging on the backside) and a small
frontpanel (from a 11/83 or 73) was
hanging on it. it looks like

*************
*     *'''''*    O is a round hole (to hold a
batch?)
*  O  *'''''*    '' is a big hole (power
switch i suppose)
*     *'''''*
*************
*  X     B  *    X = run on/off ?(green led)
B = reset ?
*  X     X  *    X = write protect(red led) X
= online(green led)    for disk 0 ?
*  X     X  *    X = write protect(red led) X
= online(green led)    for disk 1 ?
*************
X is a switch with led B is a button with led
i never saw a pdp with this frontpanel, and
since there is nothing written on it,
i tried to compare with a picture found on
the web, but i´m not sure if i´m right.
can someone tell me if i´m right ?
i asked those guys from where i picked up the
box, but they told me that the last guy
using the pdp left some years ago, and in
1999 they just powered it off. this explains
also the small paper sticking on the pdp that
showed how to login and shutdown the box.
root password is written on it *g*.

tomorrow i´ll open the box again.

thanks for your fast response.

-- lothar


FWIW, I don't know about the tape error, but
that layout looks OK apart
>from the fact that if it's an 11/83, the
memory shold be in the first slot
and the CPU in the second.  The essential
difference between an 11/73 and
an 11/83 is that the 11/83 uses PMI memory.
Assuming your backplane is the
right one, in a BA23 or BA123 box, and that
your memory is a single 4MB
board, you should swap them round, otherwise
what you actually have is an
11/73.

I assume your RD54 controller is a genuine
DEC RQDX3, so it's in the right
place.  It's possible you have an old version
of the firmware on it, but it
should still work even if you do.

all the planes in the backplane are genuine
dec parts.

--
Pete

What enclosure? BA23? If yes you have empty
slots between the cards as
this box has 3 Q/CD slots on top and 5 Q/Q
slots below. Have a look at
the QBus HOWTO at
http://vaxarchive.sevensages.org/
--

tschuess,
          Jochen

[pups] 2.11BSD disklabel-programm

[jkunz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2585 bytes --]

On 19 Jan, lothar felten wrote:

> well the pdp was once a pdp 11/23+ so
> enclosure and backplane should be
> the same as a 11/23 (ba 23). backplane should
> be Q22/CD configuration, 
Only the first 3 slots are Q/CD. The other 5 slots are Q/Q in serpentine
bus grant wiring. Tony said that the memory should be in the first
slot. So I would recommend you should order your cards like this:

A/B Slot                          | C/D Slot
*memory (quad)--------------------|---------------------------------->
*cpu (quad)-----------------------|---------------------------------->
*controller for TK50 (double)---->|*empty---------------------------->
*network controller (double)----->|*controller for RD54 disks (double)
*controller for RL02 disks (quad)-|---------------------------------->

> i´ll open the box and look for the 501-
> number to be sure. 
501-? This is no Sun. ;-)

> on the back there is a sticker saying this is a 11/73. 
AFAIK some 11/73 labeld boxen where sold with a 11/83 CPU in the first
slot and the memory in the second. In this way the CPU can not use PMI
memory and accesses the memory via the QBus. That is much slower then
PMI. You can convert this "fake" 11/73 to a 11/83 by swaping CPU and
memory cards. 

> cpu-board is a 11/83.(i´ll pass
> the number too). 
Look for the Mxxxx number. It should be M8190 = KDJ11-B according to the
field-guide. 

> do all 11/83 use PMI ?
AFAIK a 11/83 CPU can use QBus and (or?) PMI memory. If it is
configured with QBus memory it is calld a "11/73". But keep in mind
that there is a "real" 11/73 CPU (M8192 = KDJ11-A). See above...

> the memory seems to work, or has
> the cpu board some memory on it?
According to the field-guide not.

> when i picked up the box they booted it, i
> suppose this configuration was the
> way they used it there and should have
> worked.
Hmm. Are there Q/CD only BA23 backplanes? 

> *************
> *     *'''''*    O is a round hole (to hold a batch?)
> *  O  *'''''*    '' is a big hole (power switch i suppose)
> *     *'''''*
Yes. Yes.

> *************
> *  X     B  *    X = run on/off ?(green led)
> B = reset ?
> *  X     X  *    X = write protect(red led) X
> = online(green led)    for disk 0 ?
> *  X     X  *    X = write protect(red led) X
> = online(green led)    for disk 1 ?
> *************
> X is a switch with led B is a button with led
I never saw a front panel like that, but your assumption sounds right.
All my front panels have only one disk write protect / online switch.
-- 



tschuess,
          Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz

[pups] 2.11BSD disklabel-programm

[Pete Turnbull]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 4780 bytes --]

On Jan 19, 20:47, lothar felten wrote:
> hi there!
>
> well the pdp was once a pdp 11/23+ so
> enclosure and backplane should be
> the same as a 11/23 (ba 23).

If I were to be really picky, I'd say you meant microPDP-11/23.  11/23-plus
would mean a BA11-S enclosure with a different type of backplane :-)

> backplane should
> be Q22/CD configuration, but
> i´ll open the box and look for the 501-
> number to be sure.

Not the 54- number, that's only the PCB part.  The backplane itself has a
model number; for a BA23 it should be H9728-A.

In an H9728-A the top three slots are Q22/CD-interconnect, and the rest are
Q22/Q22 serpentine (the 11/23-plus backplanes are different).  When I
replied to your earlier email I assumed you just gave the order of the
boards, not the layout.  The layout should be (reverse the positions of CPU
and memory if you wish):

    ---------KDJ11-B-CPU---------
    ---------MSV11-memory--------
    ------------RLV12------------
    ----RQDX3----   ----TQK50----
        empty       ----DEQNA----

I'm guessing you have a single memory board, probably an MSV11-Q (M7551),
and an RLV12 (M8061, one quad board) rather than an RLV11 (two quad boards)
-- if not, that makes a difference to the layout.  I'm also guessing at a
DEQNA (M7504) rather than any other Ethernet controller, but it makes no
difference to the placement, so long as it's a dual-height board.  If you
added another dual-height board it would go under the RQDX3 (M7555), the
next would go under that, and the next under the DELQA, etc.  The
arrangement of the slots after the first three is called "serpentine" or
occasionally "zig-zag".

> on the back there
> is a sticker saying this is a 11/73. the
> cpu-board is a 11/83.(i´ll pass
> the number too).

All the KDJ11-B processors, whether 11/73 or 11/83, use the same printed
circuit board and module number.  There were some differences about whether
an FPU could be fitted (due to an error on the original boards); those that
would not take an FPU were only sold as KDJ11-BC and all had 15MHz clocks.
 Others with 15MHz clocks were sold as KDJ11-BB (upgradeable but FPU not
fitted).  There are also some with 18MHz clocks, these were sold as
KDJ11-BE, -BF, or higher.  Normally an 11/83 has a KDJ11-BE or higher
suffix.  Early 11/73 are 15MHz.  Just to add to the confusion, the -Bx
suffix actually refers to the EPROMs on the board, not the clock speed or
the FPU.  The *only* difference between a normal KDJ11-BE or -BF or -BH is
the firmware in the EPROMs.

However, the biggest difference between 11/83 and 11/73 is whether the
memory is used as QBus memory, or PMI memory, which is faster.  All of the
KDJ11-B boards can use PMI memory.  Beware, not all quad memory boards are
PMI-capable, but all the 1MB and bigger ones that I can think of are.

> do all 11/83 use PMI ?

Yes.  They will work with QBus memory instead (and if you put a PMI board
after the processor instead of before it, it will run as normal QBus
memory)
but then what you have is effectively an 11/73, not an 11/83.

> but the memory seems to work, or has
> the cpu board some memory on it?

No, there's no memory on the CPU board, but the memory you have is running
as QBus memory.

> when i picked up the box they booted it, i
> suppose this configuration was the
> way they used it there and should have
> worked.
> the RD54 controller has a 50pin ribboncable
> wich goes to a small board (wich
> was hanging on the backside) and a small
> frontpanel (from a 11/83 or 73) was
> hanging on it.

Literally "hanging"?  Not fixed to the front of the BA23?  Is this actually
a floor-standing (or possibly rack-mounting) BA23 with space for a TK50 and
a drive unit, or a rackmount BA11-S or BA11-N chassis with no space for
drives?

> it looks like
>
> *************
> *     *'''''*    O is a round hole (to hold a batch?)
> *  O  *'''''*    '' is a big hole (power switch i suppose)
> *     *'''''*
> *************
> *  X     B  *    X = run on/off ?(green led) B = reset ?
> *  X     X  *    X = write protect (red)  X = online(green) for disk 0 ?
> *  X     X  *    X = write protect (red)  X = online(green) for disk 1 ?
> *************
> X is a switch with led B is a button with led
> i never saw a pdp with this frontpanel

Neither have I.  DEC used pushbuttons for the disk controls on microPDP-11
panels.  Each section is separate, though; it sounds like someone has
replaced the pushbuttons or used third-party sub-panels.  The round hole
(if this is an original DEC panel) is for the badge that says whether it's
a microPDP-11/23, microPDP-11/73, microPDP-11/83, microPDP-11/53, etc.  The
rectangular hole is for the power switch in a BA23 or BA123 cabinet.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

AW: [pups] 2.11BSD disklabel-programm

[lothar felten]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3933 bytes --]

hi there!
hi pete!
> If I were to be really picky, I'd
> say you meant microPDP-11/23.  11/23-plus
> would mean a BA11-S enclosure with
> a different type of backplane :-)
mine is about 40 inch high, 19 inch wide and
as deep as
the rl02 is. the "microcomputer interfaces
1983-84" shows
some pictures of boxes: is might be the
BA11-S
>
> The backplane itself has a
> model number; for a BA23 it should
> be H9728-A.
>
on the backplane i found H9276-A.
i don´t know what -A means, but it should be
a Q22/CD
so no serpentine?
>
>
> I'm guessing you have a single
> memory board, probably an MSV11-Q (M7551),
> and an RLV12 (M8061, one quad
> board) rather than an RLV11 (two
> quad boards)
> -- if not, that makes a difference
> to the layout.  I'm also guessing at a
> DEQNA (M7504) rather than any
> other Ethernet controller, but it makes no
> difference to the placement, so
> long as it's a dual-height board.
>
now i´ve got the numbers here:
CPU:   M8190-AE      KDJ11-B
MEM:   M7551-CC      MSVC11-QC
RL02:  M8061
DELQA: M7516
RQDX3: M7555
DEQNA: M7546

> There were some differences about whether
> an FPU could be fitted (due to an
> error on the original boards); those that
> would not take an FPU were only
> sold as KDJ11-BC and all had 15MHz clocks.
>  Others with 15MHz clocks were
> sold as KDJ11-BB (upgradeable but FPU not
> fitted).  There are also some with
> 18MHz clocks, these were sold as
> KDJ11-BE, -BF, or higher.
> Normally an 11/83 has a KDJ11-BE or higher
> suffix.  Early 11/73 are 15MHz.
> Just to add to the confusion, the -Bx
> suffix actually refers to the
> EPROMs on the board, not the clock speed or
> the FPU.  The *only* difference
> between a normal KDJ11-BE or -BF or -BH is
> the firmware in the EPROMs.

hmmm, this is really confusing, since i have
AE
can it take an FPU? maybe it has a fpu? what
does
the fpu look like?

>
> However, the biggest difference
> between 11/83 and 11/73 is whether the
> memory is used as QBus memory, or
> PMI memory, which is faster.  All of the
> KDJ11-B boards can use PMI memory.
>  Beware, not all quad memory boards are
> PMI-capable, but all the 1MB and
> bigger ones that I can think of are.

so i should put in first memory then cpu.
>
> > do all 11/83 use PMI ?
>
> Yes.  They will work with QBus
> memory instead (and if you put a PMI board
> after the processor instead of
> before it, it will run as normal QBus
> memory)
> but then what you have is
> effectively an 11/73, not an 11/83.
>
>
> Literally "hanging"?  Not fixed to
> the front of the BA23?  Is this actually
> a floor-standing (or possibly
> rack-mounting) BA23 with space for
> a TK50 and
> a drive unit, or a rackmount
> BA11-S or BA11-N chassis with no space for
> drives?
well it´s a BA11-S i suppose by now.
the panel and the pcb wich connects to the
two disks were literally "hanging"
when i got the box, it was a pdp-cabinet
and a second 19inch rack, containig old,
unused stuff and the tk50 and those 2 disks
there is a separate power supply for the
disks
now the tk50 and the disks are on a separate
table
>
> Neither have I.  DEC used
> pushbuttons for the disk controls
> on microPDP-11
> panels.  Each section is separate,
> though; it sounds like someone has
> replaced the pushbuttons or used
> third-party sub-panels.  The round hole
> (if this is an original DEC panel)

it´s a original dec panel and dec pcb.

> is for the badge that says whether it's
> a microPDP-11/23, microPDP-11/73,
> microPDP-11/83, microPDP-11/53, etc.  The
> rectangular hole is for the power
> switch in a BA23 or BA123 cabinet.

i suppose they changed CPU and memory from
the pdp11/23plus and put in half a 11/83.
the 3 switches at the frontbezel of the
BA11 work, the other "front"panel at the back
(from a 11/83) is only used to put the disks
online and write protect them. this would
also
explain why there is a connector "hanging"
at the rear-frontpanel.

a weird pdp.

--lothar

[pups] 2.11BSD disklabel-programm

[jkunz]

On 19 Jan, Pete Turnbull wrote:

> Not the 54- number, that's only the PCB part.  The backplane itself has a
> model number; for a BA23 it should be H9728-A.
"H9728-A"? The sticker on the two BA23 backplanes I have here says
"H9278-A". 

Lothar: If you wane look at the number and you have a BA23, you have to
remove the tin plate on top of the QBus card cage. 
-- 



tschuess,
          Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz

[pups] 2.11BSD disklabel-programm

[Pete Turnbull]

On Jan 19, 22:35, jkunz at unixag-kl.fh-kl.de wrote:

> Only the first 3 slots are Q/CD. The other 5 slots are Q/Q in serpentine
> bus grant wiring. Tony said that the memory should be in the first
> slot.

Lothar later said this is an H9276-A.  That's a straight backplane, all
Q22/CD.  It seems he has a BA11-S not a BA23 :-)

BTW, DEC normally recommended all comms and network cards go after the
memory, tapes next, then disks.

> AFAIK some 11/73 labeld boxen where sold with a 11/83 CPU in the first
> slot and the memory in the second.

It's the order of the boards (and the boot ROMs) that make it 11/73 or
11/83, not the circuit board.  Though original 11/73s are 15MHz and
original 11/83s are 18MHz.

> AFAIK a 11/83 CPU can use QBus and (or?) PMI memory. If it is
> configured with QBus memory it is calld a "11/73". But keep in mind
> that there is a "real" 11/73 CPU (M8192 = KDJ11-A).

That's a dual-height board, CPU only, with no boot ROMs, LTC, or SLUs.  It
was only sold as an OEM product or as an upgrade to 11/03 or 11/23 (not
11/23+ or microPDP-11/23) systems.  Whilst it is a "real 11/73", it's no
more "real" than any other :-)

> Hmm. Are there Q/CD only BA23 backplanes?

No.  There are straight Q/CD and serpentine Q/Q backplanes of the same size
but they're only used in other boxes (like BA11-N and BA11-S) or sold as
OEM units.

> I never saw a front panel like that, but your assumption sounds right.
> All my front panels have only one disk write protect / online switch.

The BA23 was only rated for one hard disk and either a TK50 or an RX50, but
the BA123 (which uses the same panels) was rated for up to 4 MSCP devices.
 That's why the WP and ONLINE switches and LEDs are on a subassembly, so
you can add another one.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

AW: [pups] 2.11BSD disklabel-programm

[Pete Turnbull]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2559 bytes --]

On Jan 20,  2:05, lothar felten wrote:

> mine is about 40 inch high, 19 inch wide and
> as deep as
> the rl02 is. the "microcomputer interfaces
> 1983-84" shows
> some pictures of boxes: is might be the
> BA11-S

Sounds like you have what was a standard PDP-11/23 (or maybe 11/23+) system
in an office-type cabinet (H9642 or equivalent), with an RL02 in the top, a
stiffener panel, BA11-N or BA11-S box, a second RL02, and two blanking
plates at the bottom.  Sometimes people moved the second RL02 down 3U and
put one blanking plate or an expansion box between the BA11 and the RL02.
 As far as I remember, the only significant difference between a BA11-N and
a BA11-S is an uprated power supply, and a 22-bit backplane instead of an
18-bit one (but you can upgrade the 18-bit ones).

> on the backplane i found H9276-A.
> i don´t know what -A means, but it should be
> a Q22/CD
> so no serpentine?

Correct.  There is a very similar backplane H9275-A which is all
serpentine, but not in a standard 11/23 box, as far as I remember.  H9276-A
is for a BA11-S.  BA11-N uses H9273.

> now i´ve got the numbers here:
> CPU:   M8190-AE      KDJ11-B
> MEM:   M7551-CC      MSVC11-QC
> RL02:  M8061
> DELQA: M7516
> RQDX3: M7555
> DEQNA: M7546

OK.  That should be fine, so long as there are no gaps in the A/B slots
(left side of backplane as you look into it from the back of the machine)
between the cards.

The original 11/23+ probably had an RQDX1 or RQDX2, and possibly an RLV11,
and certainly different memory.

> hmmm, this is really confusing, since i have
> AE can it take an FPU? maybe it has a fpu? what
> does the fpu look like?

The -AE means it's a later board, should be 18MHz, and should not only be
FPU-capable, it should actually have the FPJ-11 chip on it.  -AD is the
same thing without the FPJ-11 fitted (it still does floating point ops,
just more slowly).  If not, you'll probably find it easier to get a
replacement board with an FPU already on it, rather than get the FPU chip
on it's own.

> > However, the biggest difference between 11/83 and 11/73 is whether
> > memory is used as QBus memory, or PMI memory, which is faster. the
>
> so i should put in first memory then cpu.

To get the best performance, yes.  It won't double the speed, or anything
like that, but it will go a bit faster.

> well it´s a BA11-S i suppose by now.

If it's an H9276 backplane and H7861 PSU, yes.

> a weird pdp.

Not quite factory-standard :-)   But none the less good.



-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] 2.11BSD disklabel-programm

[Pete Turnbull]

On Jan 20, 12:44, jkunz at unixag-kl.fh-kl.de wrote:
> On 19 Jan, Pete Turnbull wrote:
>
> > Not the 54- number, that's only the PCB part.  The backplane itself has
> >a  model number; for a BA23 it should be H9728-A.
> "H9728-A"? The sticker on the two BA23 backplanes I have here says
> "H9278-A".

Typo.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] 2.11BSD disklabel-programm

[Jochen Kunz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2045 bytes --]

On 2002.01.20 15:57 Pete Turnbull wrote:

> Lothar later said this is an H9276-A.  That's a straight backplane,
> all Q22/CD.  It seems he has a BA11-S not a BA23 :-)
Aha. OK. That is a different story. 

> BTW, DEC normally recommended all comms and network cards go after the
> memory, tapes next, then disks.
I usualy put comms after the CPU, then network, tape (or tape, network,
what fits best for the wiring) and disks last. 

> It's the order of the boards (and the boot ROMs) that make it 11/73 or
> 11/83, not the circuit board.  Though original 11/73s are 15MHz and
> original 11/83s are 18MHz.
[...]
> That's a dual-height board, CPU only, with no boot ROMs, LTC, or SLUs.
> It was only sold as an OEM product or as an upgrade to 11/03 or 11/23
Ahhhhh! My PDP 11 has this M8192 board with FPU but without ROMs...
I thought that this is the "original" 11/73 CPU. But on this PDP is
nothing "original". The BA23 was a MV II, the CPU was EPayed, the RAM
board was given to me by a friend. (Many PC/XTs had to donate there RAM
chips to fill it.) I found the console SLU at the scrap yard, the Dilog
ESDI controller was EPayd (in England BTW ;-) ) ... and I am still
looking for a ROM card... But I will get a /83 CPU card with ROMs in a
few months...

> > All my front panels have only one disk write protect / online
> switch.
> The BA23 was only rated for one hard disk and either a TK50 or an
RX50, 
I know. I once saw a MV II with a second RD54 in an external case. There
was a real mess of wiring to get it and the internal RD54 work together
on one RQDX3. Puting the TK50 out of the BA23 and mounting the second
RD54 in the BA23 would have been much simpler. But not the DEC way of
live. ;-)

> the BA123 (which uses the same panels) was rated for up to 4 MSCP
> devices.
>  That's why the WP and ONLINE switches and LEDs are on a subassembly,
> so you can add another one.
I know. My MV III lives in a BA123. (With 4 ESDI disks, TK50 and
floppy.)
-- 



tschüß,
         Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/

[pups] 2.11BSD disklabel-programm

[Pete Turnbull]

On Jan 20, 18:45, Jochen Kunz wrote:
>
> Ahhhhh! My PDP 11 has this M8192 board with FPU but without ROMs...
> I thought that this is the "original" 11/73 CPU. But on this PDP is
> nothing "original". The BA23 was a MV II, the CPU was EPayed, the RAM
> board was given to me by a friend. (Many PC/XTs had to donate there RAM
> chips to fill it.)

About the best use I can think of for a PC/XT :-)

> I found the console SLU at the scrap yard, the Dilog
> ESDI controller was EPayd (in England BTW ;-) ) ... and I am still
> looking for a ROM card...

If you can find an MRV11-D (to put MXV11-B ROMs into) or an MVX11-B, that
would be the best option (and the only ones DEC supported).  However, it
should be possible to put the code from MXV11-B ROMs into several 24-pin
2Kx8 EPROMs (2716 or equivalent), and put the EPROMs into a BDV11.
 However, you'd want to modify the BDV11 for 22-bit operation (that's ECO
005).

> > The BA23 was only rated for one hard disk and either a TK50 or an
> RX50,
> I know. I once saw a MV II with a second RD54 in an external case. There
> was a real mess of wiring to get it and the internal RD54 work together
> on one RQDX3. Puting the TK50 out of the BA23 and mounting the second
> RD54 in the BA23 would have been much simpler. But not the DEC way of
> live. ;-)

Because old hard drives take a lot of current.  The PSU and wiring loom
won't take a full backplane and two hard drives.

I did one of mine a different way.  I have a BA11-N with the backplane
modified to be 22-bit.  In it is an RQDX2 (or an RQDX3, depending on what's
been shuffled around this month), with a 50-way ribbon cable going to a DEC
box (used to be a TKZ50) which has a PSU, a hard drive, and an RX50.  In
the box is also a small PCB I made to do the job of the distribution board
found in a BA123.  Also in the BA11-N backplane is a modified BDV11, with a
pair of 28-pin EPROM sockets which normally hold microPDP-11/23 boot ROMs.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] solution for the disklabel

[lothar felten]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 710 bytes --]

the pdp is running 2.11BSD !

the disk has now got a valid disklabel,
i got it on the disk by
setting all diskparameters without using
d (display) and wrote
it. after this i could use the display
option.

installation was no problem, but still i
have some questions:
my VT102 doesn´t do backspace, i only
get ^H. i tried the
terminal in ANSI and VT52 mode, no
difference.

i have some dec boards labeled M7513
does anyone know what this
is? i found:
M7513    - RQD   - RQDXE Q BUS drive
interface extension module
it is a double height card and it has 3
50pin connectors.

the RQDX3 has another connector, i
suppose for RX50 floppydrive.
can i hook up a 5,25" pc drive? maybe
with modifications?

--lothar

[pups] solution for the disklabel

[Pete Turnbull]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1630 bytes --]

On Jan 27, 23:42, lothar felten wrote:

> installation was no problem, but still i
> have some questions:
> my VT102 doesn´t do backspace, i only
> get ^H. i tried the
> terminal in ANSI and VT52 mode, no
> difference.

Maybe it wants a DEL character instead of backspace (backspace *is* ctrl-H,
shown as ^H or ^h).  Change it on the terminal by going into setup, or use
stty on the BSD system to change the delete character (stty del '^h').

> i have some dec boards labeled M7513
> does anyone know what this
> is? i found:
> M7513    - RQD   - RQDXE Q BUS drive
> interface extension module

That's exactly what it is.  The BA23 box only supports one hard drive; the
RQDXE is an adaptor for an RXDX2 or RXDX3 to permit use of additional
drives with a distribution board in a second enclosure.  One of the 50-pin
connectors goes to the RQDX3, one to the distribution board in the BA23,
and the third to a connector kit on the rear panel of the BA23.  There's a
different version for an RQDX1, called an RQDX1E.

> the RQDX3 has another connector, i
> suppose for RX50 floppydrive.

An RQDX3 has only one connector, the 50-pin one to go to the distribution
board.  Are you looking at the right thing?  Are you looking at a
distribution board?  That does have a 34-way connector for a floppy.

> can i hook up a 5,25" pc drive? maybe
> with modifications?

Not an ordinary PC floppy, no.  A TEAC FD55GFR is an 80-track double-sided
drive (not HD, though) that will work as an RX33.  Some other 80-track
5.25" drives may work, if you set the jumpers.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] solution for the disklabel

[Johnny Billquist]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 703 bytes --]

On Sun, 27 Jan 2002, lothar felten wrote:

> the pdp is running 2.11BSD !

Good news indeed.

> installation was no problem, but still i
> have some questions:
> my VT102 doesn´t do backspace, i only
> get ^H. i tried the
> terminal in ANSI and VT52 mode, no
> difference.

And? ^H is backspace... Are you trying to delete characters?
Typical PC confusion perhaps. For some reason the PC world have decided to
use backspace to delete characters...

	Johnny

Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt at update.uu.se           ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol

[pups] solution for the disklabel

[Johnny Billquist]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 978 bytes --]

On Wed, 30 Jan 2002, Pete Turnbull wrote:

> On Jan 27, 23:42, lothar felten wrote:
> 
> > installation was no problem, but still i
> > have some questions:
> > my VT102 doesn´t do backspace, i only
> > get ^H. i tried the
> > terminal in ANSI and VT52 mode, no
> > difference.
> 
> Maybe it wants a DEL character instead of backspace (backspace *is* ctrl-H,
> shown as ^H or ^h).  Change it on the terminal by going into setup, or use
> stty on the BSD system to change the delete character (stty del '^h').

On a real VT100 you cannot get the delete key to generate a backspace. He
must be pressing the backspace key, or he's not using a VT100 at all, but
instead some emulator, which isn't doing things the VT100 way...

	Johnny

Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt at update.uu.se           ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol

[pups] solution for the disklabel

[Pete Turnbull]

On Jan 30, 16:03, Johnny Billquist wrote:
> On Wed, 30 Jan 2002, Pete Turnbull wrote:

> > Maybe it wants a DEL character instead of backspace (backspace *is*
ctrl-H,
> > shown as ^H or ^h).  Change it on the terminal by going into setup, or
use
> > stty on the BSD system to change the delete character (stty del '^h').
>
> On a real VT100 you cannot get the delete key to generate a backspace. He
> must be pressing the backspace key, or he's not using a VT100 at all, but
> instead some emulator, which isn't doing things the VT100 way...

Of course.  I've spent too long using my VT420.  A real VT102 has separate
delete and backspace keys.  Probably he's using some not-really-VT102
emulation in an xterm window or some emulator.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] VT102 and other hardware.... was: solution for the disklabel

[lothar felten]

> >
> > Maybe it wants a DEL
> character instead of
> backspace (backspace *is* ctrl-H,
> > shown as ^H or ^h).  Change
> it on the terminal by going
> into setup, or use
> > stty on the BSD system to
> change the delete character
> (stty del '^h').
>
> On a real VT100 you cannot
> get the delete key to
> generate a backspace. He
> must be pressing the
> backspace key, or he's not
> using a VT100 at all, but
> instead some emulator, which
> isn't doing things the VT100 way...

thanks for helping me!

i'm using a "real" vt102 box. now it works, problem was just i thought
"backspace" should behave pc-like. the del-key does erase my letters and
the backspace displays ^H. sorry but i'm just no used to this "behaviour"
of the keys.
this weekend i put a qbus serial card into the pdp, but i dont get a login
prompt on this ports: i edited /etc/ttys, the special files in /dev are
present. i'm still using the generic kernel (still not brave enough to
start configuring & compiling a new one, how long would this approx. take
on a 11/83 with 4mb? is it worth it?). on the console connected to the
cpu-board i always get a #prompt without login, is this normal?
i have a DEQNA-nic, but with the MAKEDEV script i can't make the qe0
device. i would make it myself with mknod, but wich major/minor number do i
need?
there are devices for my RL02 disks, but everytime i try to access them by
disklabel the system stops (the run light goes off). is there a way to
check if the controller is working ?
on my RQDX3-distribution panel i found a 34 pin floppy connector. i
connected two 5,25" diskdrives (pc drives, TEAC, jumpered different) they
work fine.

-- lothar

[pups] 2.11BSD disklabel-programm

[emanuel stiebler]

Wilko Bulte wrote:
> 
> I think MDM can also do it for you on a mVAX with RQDX3

Common error ;-)

MDM can only deal with a pefect formatted HD drive, which has no new
bad blocks since the last format. 
Otherwise, see the hints about the MV2000 & pdp11 ;-)

cheers

[pups] 2.11BSD disklabel-programm

[Wilko Bulte]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 802 bytes --]

On Sat, Jan 26, 2002 at 06:39:42PM +0100, jkunz at unixag-kl.fh-kl.de wrote:
> On 26 Jan, lothar felten wrote:
> 
> > i do have some "winchester" controllers for
> > PC/ISA, i could check the disks on
> > a linux pc, no problem, but i donŽt want to
> > overwrite something "digital"-specific
> > that i could not restore. those RD54-disks
> > are regular ST506-mfm, right?
> Do not use RDxx disks in a PeeCee! You will damage the RQDX format and
> the disks will be unusable on the RQDX. If you destroyed the RQDX
> format you have to get a VAXstation 2000 to reformat the disks, or you
> have to get XXDP onto your PDP 11 to do the job. 

I think MDM can also do it for you on a mVAX with RQDX3

-- 
|   / o / /_  _   		email: 	wilko at FreeBSD.org
|/|/ / / /(  (_)  Bulte		Arnhem, the Netherlands

[pups] 2.11BSD disklabel-programm

[jkunz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 624 bytes --]

On 26 Jan, lothar felten wrote:

> i do have some "winchester" controllers for
> PC/ISA, i could check the disks on
> a linux pc, no problem, but i don´t want to
> overwrite something "digital"-specific
> that i could not restore. those RD54-disks
> are regular ST506-mfm, right?
Do not use RDxx disks in a PeeCee! You will damage the RQDX format and
the disks will be unusable on the RQDX. If you destroyed the RQDX
format you have to get a VAXstation 2000 to reformat the disks, or you
have to get XXDP onto your PDP 11 to do the job. 
-- 



tschuess,
          Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz

[pups] 2.11BSD disklabel-programm

[lothar felten]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2036 bytes --]

about the pdp-11/83:

swapping cpu and memoryboard does work, but
the performance stays the same:
none, because i still don´t have any system
by now. there is a small difference:
the run led stays off all the time, but the
system ODT works fine. i suppose this
is a problem of passing the signals, but this
is only a vague guess.
when i boot the TK50 2.11BSD tape, i can get
the disklabel program loaded.
it asks: Disk?
when i try to access the disks, the system
hangs. when i access rl(0,0) (RL02
disk drive0) the system just hangs, no
response. when i access ra(0,0) (RD54 disk
drive0) i get a "menu": d(isplay) D(efault)
m(odify) w(rite) q(uit)?
i can enter D, then it should write the
defaults to the disk. after this i get:
d(isplay) D(efault) m(odify) w(rite) q(uit)?
i enter d (to see what the program did), and
i get:
type:MSCP
disk:RD54
flags:
bytes/sector:512
sectors/track:17
tr
(it stops right after tr) i suppose the
program is asking the controller about
drive parameters, and that´s where it fails.
i can wait for hours, i´ll get no
response.
i can use all the menu, except d=display.
i thought that maybe one of the controlles is
broken and causes trouble on the
qbus (maybe the RL02 controller). is there a
way to check all this stuff?
i do have some "winchester" controllers for
PC/ISA, i could check the disks on
a linux pc, no problem, but i don´t want to
overwrite something "digital"-specific
that i could not restore. those RD54-disks
are regular ST506-mfm, right?
maybe i should "downsize" the system to the
basic elements i need to get it running
maybe there really is a bus problem.
i´d try this configuration:
mem - cpu - rqdx3 - tqk50 (top-bottom)
would this be ok?
or, instead of using RD54 disks, should i try
to use the rl02 as "pair" (one swap,
one systemdisk)?
i think step by step checking the hardware is
the only thing i can do to get the
pdp up and running.

have a nice weekend

-- lothar

btw: does anyone know a good book/link about
system-architecture, specially harvard-
architecture ?

[pups] 2.11BSD disklabel-programm

[Pete Turnbull]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1596 bytes --]

On Jan 19, 12:43, lothar felten wrote:

> i don´t know if this is a hardware (bus?)
> error, or maybe the tape is bad.
> the tapes i use are unused original dec TK50
> tapes, i made several ones, because i thought
> it might be a tape error, but
> all tapes are the same.
>
> hardware:
> PDP-11/83, 4megs of ram, TK50, two RD54
> (maxtor), two RL02 disks.
> qbus cards (top to bottom):
> *cpu (quad)
> *memory (quad)
> *controller for RL02 disks (quad)
> *controller for RD54 disks (double)
> *controller for TK50 (double)
> *network controller (double)

FWIW, I don't know about the tape error, but that layout looks OK apart
from the fact that if it's an 11/83, the memory shold be in the first slot
and the CPU in the second.  The essential difference between an 11/73 and
an 11/83 is that the 11/83 uses PMI memory.  Assuming your backplane is the
right one, in a BA23 or BA123 box, and that your memory is a single 4MB
board, you should swap them round, otherwise what you actually have is an
11/73.

I assume your RD54 controller is a genuine DEC RQDX3, so it's in the right
place.  It's possible you have an old version of the firmware on it, but it
should still work even if you do.

> i didn´t find a kind of terminator, but i
> didn´t change the order of the
> cards since i picked the box up.
> there are no empty slots between the cards,
> and i´m not sure if the Qbus
> need a special terminator.

There normally isn't an extra terminator in an 11/73 or 11/83, unless you
add an expansion backplane.

-- 
Pete						Peter Turnbull
						Network Manager
						University of York

[pups] 2.11BSD disklabel-programm

[jkunz]

On 19 Jan, lothar felten wrote:

> hardware:
> PDP-11/83, 4megs of ram, TK50, two RD54
> (maxtor), two RL02 disks.
> qbus cards (top to bottom):
> *cpu (quad)
> *memory (quad)
> *controller for RL02 disks (quad)
> *controller for RD54 disks (double)
> *controller for TK50 (double)
> *network controller (double)
What enclosure? BA23? If yes you have empty slots between the cards as
this box has 3 Q/CD slots on top and 5 Q/Q slots below. Have a look at
the QBus HOWTO at http://vaxarchive.sevensages.org/
-- 



tschuess,
          Jochen

Homepage: http://www.unixag-kl.fh-kl.de/~jkunz

[pups] 2.11BSD disklabel-programm

[lothar felten]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1477 bytes --]

hi there,
maybe someone can help me installing 2.11BSD
on a PDP-11/83.

my problem:
the standalone disklabel-programm stops when
displaying the MSCP disk
information.

in the console ODT i write/see:
BOOT MU0
Starting system from mu0
83Boot from tms(0,0,0) at 0174500
:tms(0,1)
disklabel
Disk? ra(0,0)
d(isplay) D(efault) m(odify) w(rite) q(uit)?
d
type:MSCP
disk:RD54
flags:
bytes/sector: 512
sectors/reack: 17
tr

and then it just stops, right after "tr", but
the RUN led stays on.
when i try to disklabel a RL02 disk i get
invalid disk (for rl(0,1)) or a
system stop (for rl(0,0) with RUN led off).

i don´t know if this is a hardware (bus?)
error, or maybe the tape is bad.
the tapes i use are unused original dec TK50
tapes, i made several ones, because i thought
it might be a tape error, but
all tapes are the same.

hardware:
PDP-11/83, 4megs of ram, TK50, two RD54
(maxtor), two RL02 disks.
qbus cards (top to bottom):
*cpu (quad)
*memory (quad)
*controller for RL02 disks (quad)
*controller for RD54 disks (double)
*controller for TK50 (double)
*network controller (double)

i didn´t find a kind of terminator, but i
didn´t change the order of the
cards since i picked the box up.
there are no empty slots between the cards,
and i´m not sure if the Qbus
need a special terminator.

i made a TK50 boottape on my DECstation
5000/200. i got the software from
the pups archive, and made the tape with the
"maketape" program.

any idea welcome.

regards,

--lothar