The UCSD p-System, Apple Pascal, and a dream of cross-platform compatibility never quite realized
I’ve been thinking about the UCSD P-System a lot lately, and I thought I’d write about it a bit, and about bytecode systems more-generally. And along the way, I hope to write some interesting code.
What’s the p-System?
Way back in 1974-1978, computer scientists at the University of California, San Diego campus developed a new portable operating system, compiler, and tools to run on both the PDP-11 minicomputers and the increasingly-common microcomputers. They called this UCSD Pascal, and it was available for a wide variety of systems, in just a few years.
The secret to its wide portability was that the system sat “on top of” a very small kernel of machine-dependent code, which implemented a “virtual machine” called the “p-machine”, a kind of imaginary CPU specifically designed to be a good target for a Pascal Compiler. Because this was a relatively-small amount of code, it was possible to port the entire operating environment, including the operating system, the compiler, editor, filesystem and other tools, just by implementing the p-machine emulator for a new machine.
You could get a version of the p-System for a huge variety of computers. Besides DEC’s PDP-11 and Rainbow, there were versions for the Apple II, the IBM PC (one of 3 OS options for it at the start), for HP’s series 80 and series 9000 workstations, Texas Instruments’ TI-99/4a home computer, CPM machines, various mainframes, and on, and on. Western Digital even created a microprocessor that ran p-machine code as its native machine language.
The p-System was not the first portable byte-code interpreter and compiler system – that idea goes very far back, at least to the origins of the Pascal language itself. But it was arguably one of the most-successful early versions of the idea, and served as an inspiration for future portable software systems (including Java’s bytecode, and Infocom’s Z-machine).
Where did I first encounter the p-System?
In 1986, during my Junior year of High School, I took a Computer Programming class. That was primarily taught in BASIC, because of course it was. After completing that class, I was essentially let loose in the computer lab for a period of “Independent study” (thanks, Mr. Carlson!), and that was where I got to use Pascal for the first time. Apple Pascal, running on the Apple IIe. At that point, I was also using Pascal in my first after-school job, using UCSD Pascal on the HP 80 and 9000 series workstations, and also Turbo Pascal on IBM PC-compatibles (not the p-System, but definitely inspired by it).
My “final project” for my Senior year was building a spreadsheet program in Apple Pascal. I managed to get a formula editor, including basic arithmetic functions, the grid display, and printouts all working before I graduated. Unfortunately, that code is lost to time.
What am I going to do with the p-System?
I have a couple of goals I’m hoping to accomplish here over the next few months:
- Get Apple Pascal up and running in some kind of emulator on my Mac, so I can experience it again
- Build some tools to transfer files to/from disk images for the emulator
- These already exist, in various forms, but I really want to make my own, partly for the enjoyment of it, partly to have something simple to do in Rust to build familiarity with the language.
- Build a p-machine emulator, in Rust
- and probably a disassembler/assembler, and the rest of the tool chain
- maybe NOT a Pascal compiler, unless I get ambitious
- Port the p-machine emulator to something other than the Mac
- Ideally, something small – like an Arduino or Raspberry Pi Pico
Why build new tools for such an old system?
Partly because I can, and partly because we’re coming up on the 50th anniversary of the UCSD P-System. Development started around 1974, and UCSD Pascal was first released in 1978. I think it’d be fun to build something new with/for that old system, which was really important to me getting started in my career.
And partly because, while there are tools out there, they’re often badly-documented (see link rot lament, below), and they often were built decades ago, at this point, and require compilation with gcc 2.x, for instance. I think it’d be nice to bring back some old system software, using “new” tooling, to hopefully extend the legacy of this system a bit longer.
Where can you find out more about the p-System?
There is a lot of information available on the Internet, in but it’s scattered everywhere, and a lot of it is suffering link rot these days. One site I found had tantalizing links to a bunch of books, specifications, and historical documents, 80% of which were 404 NOT FOUND. I’m going to add links to other sites to these blog posts as I run across them.
Probably the best place to start digging is Hans Otten’s web site, Pascal for Small Machines, which in addition to p-System information, has a bunch of historical information about other versions of Pascal on “small computers”, meaning mini- and micro-computers.
I found some very interesting historical information about Apple Pascal and the p-machine at The Digital Antiquarian blog. This is part of a series of articles that goes from Pascal, to the p-System, to Sir-tech software and Wizardry, a roleplaying game for the Apple II which was built with Apple Pascal.
The Jefferson Computer Museum has a page about the UCSD p-System, and links to several interesting collections of software.