My PDP-11/73 has started misbehaving after several years of being very
stable. It is showing symptoms similar to when I've forgotten to enable
the LTC in the past, except this time, it is enable.
Power supply seems good - +5V, +12V, all seems there and appears to be
stable. I've probed the LTC going onto the backplane and I'm getting
50Hz. It looks a tad noisy, not sure if this is a problem. I will see if
I can get a picture of it, but it has very distinct rises and falls, so
hopefully that's ok. Looks to be about 50% duty cycle.
I've attempted several builds of the kernel, including the
out-of-the-box build from 2.11BSD distribution. I've also compiled a
shrunk down kernel without a bunch of devices I don't have, but it still
fails to work. I am using a SCSI2SD so it's easy for me to copy images
and test them in SIMH - they all boot fine in SIMH with a similar setup
(mscp, tmscp, 11/73, 1MB-ish RAM).
Does anyone have any ideas what might be causing this? I've run out of
ideas unfortunately. I'll stick a dump of the boot below, with debug
flag passed showing output from autoconfig.
I've also tried it without the tmscp card - just CPU, RAM and Emulex
UC07 in MSCP mode for scsi disk. I have tried it with less memory, and
I've tried removing the halt jumper from the 11/73 board (W5 if I
BOOT> DU 0
73Boot from ra(0,0,0) at 0172150
: ra(0,0,0)unix -D
Boot: bootdev=02400 bootcsr=0172150
2.11 BSD UNIX #116: Wed Dec 31 18:01:57 CST 1969
root at localhost.2bsd.com:/usr/src/sys/GENERIC
ra0: Ver 5 mod 13
ra0: RA81 size=1216601
phys mem = 1179648
avail mem = 959040
user mem = 307200
_hkprobe = 0
_hkattach = 143330
_hkVec = 0
hkintr = 120
_htprobe = 0
_htattach = 143550
_htVec = 0
htintr = 150
_raprobe = 0
_raattach = 34270
_raVec = 34200
raintr = 100
_rkprobe = 0
_rkattach = 144650
_rkVec = 0
rkintr = 130
_rlprobe = 0
_rlattach = 144770
_rlVec = 0
rlintr = 110
_tmprobe = 0
_tmattach = 143610
_tmVec = 0
tmintr = 160
_tmsprobe = 0
_tmsattach = 143740
_tmsVec = 143730
tmsintr = 200
_tsprobe = 0
_tsattach = 143660
_tsVec = 0
tsintr = 170
_xpprobe = 0
_xpattach = 0
_xpVec = 0
xpintr = 0
endvec = 1000
_conf_int = 16124
CGOOD = 14
CBAD = 24
_nextiv = 5500
trap = 4150
_version = 17400
KERN_NONSEP = 0
Grab 177440 = 36556hk ? csr 177440 vector 210 skipped: No CSR.
Grab 172440 = 36556ht ? csr 172440 vector 224 skipped: No CSR.
Grab 172150 = 0Grab 156 = 240Grab 154 = 100Stuff 14 @ 154
Stuff 340 @ 156
probe ra: return conf_int:Stuff 100 @ 154
Stuff 240 @ 156
ra ? csr 172150 vector 154 bad probe value 224.
Grab 177400 = 36556rk ? csr 177400 vector 220 skipped: No CSR.
Grab 174400 = 36556rl ? csr 174400 vector 160 skipped: No CSR.
Grab 172520 = 36556tm ? csr 172520 vector 224 skipped: No CSR.
Grab 174500 = 0Grab 262 = 0Grab 260 = 0tms ? csr 174500 vector 260 interrupt vector already in use.
Grab 172520 = 36556ts ? csr 172520 vector 224 skipped: No CSR.
xp ? csr 176700 vector 254 skipped: No autoconfig routines.
At this point it is stuck no matter how long I wait.
> From: Tony Duell <ard.p850ug1 at gmail.com>
> Were there 2 things called the KM11?
> The KM11 that I know is the maintenance unit
> From: Paul Birkel
> I think that we're all talking about the ML11-A, or at least are
> intending to ... although the Subject line has been erroneous from the
> get-go ...
Well, at least your brains are working, unlike mine! (Age starting to catch
up with me?) Yes, the ML11; I tried to correct the erroneous 'KL11' and
changed the wrong letter!
The person that refered me to my present job at a datacenter passed away
this past monday. He was a hardware / software engineer for modcomp
computers. He left me all of the computers and documents. there are too
many books to keep, stuff concerning the modcomp computers that is not
saved anywhere else that i can tell.
I have picked up storage containers for all the books, and i can scan it
all. after that, its all probally going in the recycle bin, as i dont know
where or how i would keep such a large pile of paper manuals on hand.
what is the prefered format to upload things to bitsavers in? is pdf
How can i create a pdf that is not too big on file size? Can the text be
recognized and be made searchable within the scanned pdf?
any input would be appreciated, Thanks.
> From: Mark Kahrs
There's a typo in your original Subject: line: the KL11 is a very early UNIBUS
(probably the very first UNIBUS device ever, looking at the board's Mxxx
number) asyn serial line interface:
> manx tells me that these documents were known to exist:
> But they are not online.
I couldn't find out anything about the KM11 with a Web search, but I did see
that it's in the DEC PDP-11 fiche set. My set does have the KM11 Tech Manual.
I've never heard of the KM11, and as I said, there's nothing about it online.
Is it worth doing a CHWiki page for it? (With the fiche, it would be pretty
easy to whip up one covering the basics: functionality, component boards, etc.
> So I can't say whether they are 18 bit compatible.
Huh? The KM11 doesn't plug into the UNIBUS (or QBUS); it's a MASSBUS device (a
solid-state storage device, actually), so it plugs into an RH11 or RH70 or
something like that. (I should work with the VAX MASSBUS controller, too.)
So the question 'is it 18 bit compatible' makes no sense.
I've read that there is circuitry in the expansion base (BA40A?) has circuitry . Does anyone know what the circuitry does? Is it required for SCSI operation? (I hope not, or I'll have to kludge one up to make use of pk2k SCSI boot-roms!)
> From: Paul Koning
> But the sector format is a different matter. If it's designed for
> PDP-11 and friends, presumably it has a 512 byte sector size. For
> PDP-10 or -20 use you'd presumably want a sector size consisting of
> some round number of 36 bit words.
Actually, the -10/-20 MASSBUS situation is even more complicated than that.
The MASSBUS can operate in 16 or 18 bit data width (for everyone else; this
is totally different from the Q16/Q18/Q22 of the QBUS, which is _address_
width), so it can support 36-bit words directly, using two extra data lines.
So for the RP04 and other disks, and their 'controllers' (at least, the part
that's in the device), they have to be able to turn the bit-stream from the
mass storage device into 18-bit wide words. (And they actually have different
sector formats depending on whether they are in 16- or 18-bit mode.)
What the KM11 does, I don't know (I'm too lazy to go look at the TM); I would
not at all be suprised to find that it can _only_ operate in 16-bit mode
(i.e. the array of memory chips is 16 bits wide, and it just ships a line at a
time from that out in parallel, so there's no way to even produce 18-bit wide
words). The name of the device (KM11) adds weight to that supposition.
Subject: Re: PDP-11/73 boot issues
References: <87ilytoikj.fsf at carbon.nat.rhwyd.co.uk> <CADBZjLYN9aUTDHJ6=XJwNnNefTUbzyZis4evRS1Coy2r9xcX5w at mail.gmail.com> <87fstxohuj.fsf at carbon.nat.rhwyd.co.uk> <edde8fc9-6e38-3215-970f-34b9f3a95ce0 at alembic.crystel.com> <21789e85-2aa4-3b61-db31-b21fd8c08a03 at dunnington.plus.com> <87czp1obv4.fsf at carbon.nat.rhwyd.co.uk>
User-agent: mu4e 0.9.18; emacs 27.2
In-reply-to: <87czp1obv4.fsf at carbon.nat.rhwyd.co.uk>
Aaron Jackson via cctalk writes:
> Pete Turnbull via cctalk writes:
>> On 21/09/2021 20:34, Chris Zach via cctalk wrote:
>>> Can an MXV11 be used in a 22 bit system? I thought it was an 18 bit
>> MXV11-B is 22-bit. MXV11-A is 18-bit but supposedly can be used in a
>> 22-bit system if the RAM is disabled.
> Yeah this is a 22 bit card. Josh sent me an xxdp image which I could
> easily boot from my scsi2sd (thanks!). Seems to be reporting an error
> with the CPU unfortunately:
> ]] .R ZKDJ??
> ]] ZKDJB2.BIC
> ]] ERROR WHILE TESTING BOARD FUNCTIONS
> ]] ERROR # =001166
> ]] ERROR PC =040662
> ]] 043632
> This happens regardless of whether W9 is installed or not (supposedly
> disables the LTC register on the CPU?)
> I'll see if I can borrow another CPU card form a friend this
> weekend. Unless anyone else has any ideas? Another suggestion on IRC
> was to disable the PSU LTC and enable the LTC on MXV11 but will need to
> look up some details on how to do this.
Had a nice cycling trip this evening to pick up a spare 11/73
card. Unfortunately it did not fix my issues so I'll have to do some
Has anyone disconnected the BEVENT line and used a signal generator to
provide the LTC? Curious to try this to figure out if mine is just being
noisy or something.
On Wed, 2021-09-22 at 12:00 -0500, cctalk-request at classiccmp.org wrote:
> Currently the Hobbyist Program covers Alpha and Itanium.? We?ve been
> told it will cover x86 at some point.? I for one can?t wait for x86,
> as I don?t really want to add an Itanium to run some of the newer
Not even that... looked at the VSI stuff and it is only the newest
Alphas that are supported. 21064/21164 and, I believe, 21264 are not
supported by any release licensed by VSI, and they changed the PRODUCER
key so you can't use those PAKs on DEC/CPQ/HP variants of VMS.
HP stopped issuing new hobbyist PAKs back very early in 2020. I put in
for a renew March or so and never got it. They were so disinterested in
the program that they didn't mention stopping it, so I guess it isn't
surprising that the page is still up.
Isn't VMS DCL pretty close to RSX? Never used RSX, but that is what I
was always told. Anyway, good system, pretty solid, expect to do a LOT
of typing for commands if you're used to UNIX, and don't put any VAX
with a public DCL account on the Internet because there is a huge
security hole in DCL that was never fixed for VAX.
Hopefully a few of the DEC/VMS fans here might be able to help!
I'm on a bit of a quest. I've been given some old VAX/VMS software -- a
cross compiler and some source code -- that I'd like to get running. My
goal is to get the source code building and experiment with the compiler
Problem is that I've never used VMS before, and don't have a clue how to
install or use it.
Can any point me to an idiot's guide to VMS, how to set it up and make
it possible to send files to it from my Linux box?
I'm thinking of using SIMH, unless there's a better emulator available.
I'm still waiting on a reply from HP with a hobbyist licence PAK (I've
filled out the form), but I figure I can get started on the learning
while I wait.
Where are you, and how long is that cable? I believe that I have a
spare that's around 18".
As luck would have it, I'm moving, and today I'm sorting and culling
the graphics cable tub anyway. If I have a spare I'll save it out.