Looking for Unix-11 for PDP-11


I am coming into several PDP-11/34A CPUs and related peripherals this weekend, including RX02 and RL02 disk drives, but no large hard disks and no 45/70/93-style Unix-supported MMU. I understand that there are copies of Unix-11 floating around out there that will run on the 11/34 and 11/35 machines. Does anyone have a line on where I might find such a distribution?

(I also don’t (yet) have a tape drive, so getting an image onto the machine may be yet another journey!)

1 Like

There are some tar files here:

And possibly might be worth asking on the PiDP-11 list:


Awesome, thanks! I checked tuhs, but didn’t find anything there, and web searches for “unix-11” were … less than fruitful.


And v7 has a 10 MB RL image!


Ethan, Bob Supnick’s Computer History site has tonnes of resources on DEC and non-DEC classic systems. His SimH simulator supports PDP-11. I recall there are UNIX tar files for the machine. This site is worth browsing.


Looks like those are the same images as @EdS listed on classiccmp.org. I didn’t know SIMH had maintained images, although I should have guessed.

I’ve been unable to successfully boot that image in SIMH (the kernel traps on every model of PDP-11 I’ve tried thus far). In particular, it doesn’t seem to run on the 11/34 w/ 256 kB of RAM, which is what I’m picking up this weekend. I’ll have to dig deeper to see if I can make it run. The description says it has sources, so if nothing else, if I can get it running in SIMH, maybe I can build a working kernel.

I don’t expect this to be an overnight project. :slight_smile:

1 Like
PDP-11/70 looking for a new home (in NZ most likely)

Although I’m a fellow DEC admirer, my experience with the PDP is rather limited. My college had only one PDP-11/70 remaining, by the time I arrived there in the early 1980s—the heyday of the VAX-11/780. So, I grew up on BSD-powered VAX-11s. Indeed, I wrote a book about running a 4.3BSD on SimH’s VAX-11/780, and published it on the 40th anniversary of the VAX-11. I posted about the book on Ed’s G+ page, at the time.

By the way, http://www.bitsavers.org/ has a tonne of software, manuals, and other resources on the entire DEC family. I believe Bob Supnick runs this site, too. I’m sure you’ll find something of use there, when you recompile the kernel.

Incidentally, on a typical MacBook, recompiling a 4.3BSD kernel on the SimH VAX-11/780 equipped with 32 MB of RAM took mere minutes, a task that used to take an entire night on my school’s various VAX-11s. But when you reconfigure SimH to use 4 MB, a more realistic amount for that era, compilation took several hours, just like back in the day—swapping, swapping, swapping.

Good luck, this weekend, mate; I envy you.



I’ve used a bunch of stuff from bitsavers, and I have a PiDP-11 and have played with 2.11 BSD and a few revisions of Research Unix there. It’s a bit bewildering, though, as while many things are the same, many things have changed enough that there’s a lot of reading to do – and some of the documentation has proven hard to find (because it was run off on paper at the time, and it’s not all been OCR’d so the interwebs can find it!). I look forward to being able to poke at the real hardware.

Thanks for the well wishes, it’ll be a 12 hour round trip.


I’ll be checking this forum, over the weekend. :slight_smile:

By the way, how on earth did you manage to find a running piece of kit? That’s quite astonishing.


It’s more a “running when last shut off” situation than “running”, and we all know how that goes.

I actually found 4x PDP-11/34As, a bevy of terminals and printing terminals, a pile of RL01/02s and RX01/02s, some disks, and some documentation. I’m very, very excited.


That’s a veritable mother lode!


I am growing more convinced that the images from classiccmp/bitsavers above are just regular Unix v5/v6/v7 images, and not UNIX-11 for the smaller PDPs. I’ve been unable to boot them in SIMH. I have also been unable to boot them on simulated 11/45 or 11/70 machines, though, so perhaps I’m dong something wrong.

More study is warranted.

In the meantime, though, as seen in this other post,. I have many other things to occupy my time. I’ll revisit booting UNIX-11 once I have anything at all booting!

1 Like

By Unix-11, it seems we mean an early Unix for a PDP-11 that lacks an MMU… I still don’t have an answer but I’ve found a number of interesting links:

The earliest mention of Unix-11 I found is in The UNIX Time-Sharing System by Ritchie from 1970 or so:

There are two versions of UNIX. The first, which has been in existence about a year, runs on the PDP-7 and -9 computers; a more modern version, a few months old, uses the PDP-11. This document describes UNIX-11…

UNIX-7 also has a version of the compiler writing language TMGL contributed by M. D. McIlroy, and besides its own assembler, there is a PDP-11 assembler which was used to write UNIX-11.

In the later and prettier paper from 1974, The UNIX TimeSharing System we find there are now three versions of UNIX, and

The second version ran on the unprotected PDP-11/20 computer.

Here’s a dissertation, by John Gilmour, aiming to run this version in emulation, in 2018:

The Restoration of UNIX: Emulating UNIX version 1.0 on a 16-
bit DEC PDP 11/20

In 1995 two engineers, Paul Vixie and Keith Bostic, dug deep enough to find several DEC tapes “under the floor of the computer room [at Bell Labs]” (UNIX Past) which held several original binary files for UNIX. With much help from several individuals, these two engineers were able to reverse engineer the binary tapes they found and 15 years later, put up a repository for this ancient system on GitHub. There it has stayed, waiting for someone to use it like it was meant to be used, on a DEC PDP 11/20, the computer UNIX v1 was initially written for.

the repository of UNIX v1 that can be found on GitHub is at its core the original version of UNIX, but alas, has some makeup of v2. This is where their research ends, and where ours begins.

Here’s Nick Janetakis doing something similar in 2017 using a Docker image of a suitably equipped SIMH:

and there are some related links and pointers in this stackexchange post:

Here’s Warren Toomey in 2010, with First edition Unix: Its creation and restoration

Until recently, the earliest versions of the Unix operating system were believed to have been lost completely. In 2008, however, a restoration team from the Unix Heritage Society completed an effort to resurrect and restore the first edition Unix to a running and usable state from a newly discovered listing of the system’s assembly source code.

… because the PDP-11/20 provided no memory protection against kernel corruption by the running process, it was considered a courtesy for a programmer to yell ‘‘a.out?’’—the name of the Unix assembler’s default output file—before running a new executable for the first time. This gave the other users a chance to save any files they were editing.

(Just for fun, another anecdote from Ritchie’s oddities paper:

at some point several were sitting around working away. Bob Morris asked, almost conversationally, “what are the arguments to ld?” Someone told him. We continued typing for the next minute, as a thought began to percolate, not quite to the top of the brain-- in other words, not quite fast enough. The terminal stopped echoing before anyone could stop and say “Hold on Bob, what is it you’re trying to do?”


Enough time has passed and enough technology invented that some people now can wonder how a multi user OS is even possible without memory protection. Some discussion and links here and here.

Ritchie’s 1979 paper The Evolution of the Unix Time-sharing System is an interesting read.


Note that Paul Vixie can be found in Twitter, in case one wants to ask for directions/recollections.