From tuhs at tuhs.org Sun Nov 1 03:36:12 2026 From: tuhs at tuhs.org (James Frew via TUHS) Date: Sat, 31 Oct 2026 10:36:12 -0700 Subject: [TUHS] C from 1980 binaries In-Reply-To: <20261030194928.C61F7125F8670@ary.qy> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> Message-ID: I love the memories this list dredges up: Back in The Day™ I was responsible for a CS student lab full of Teraks, and one of my first tasks was to get a serial line file transfer program running between the Teraks and the dept VAX. The program I inherited was written in PDP-11 assembler, needless to say completely uncommented. So I put my Lions-notes-level reading knowledge of assembler to work and commented it in C so I could understand it. The correspondence was almost 1:1 line-for-line. Pity the Terak didn't have a C compiler... So I'm seconding John's suggestion. Seems like a perfectly do-able task for an LLM! Cheers, /Frew On 2026-10-30 12:49 PM, John Levine via TUHS wrote: > I do not ever recall a PDP-11 to C decompiler. Compiling was painful enough. > On the other hand, the instruction set was small and the translation from C > to machine code was pretty straightforward so it might be interesting to set > an LLM loose on it and see what you get. > > On Fri, Oct 30, 2026 at 11:31 AM John Foust via TUHS wrote: >>> I'm working on a fun project with Prof. Tim Teitelbaum of Cornell, who >>> developed the Cornell Program Synthesizer, an early IDE that ran on the >>> Terak computer. We have an image of an 8" disk with the executables. >>> >>> Source has been lost, so I'd like to try to reconstruct it. He believes >>> the original was in an early C, circa 1979-1982, compiled on >>> (I presume Unix) on maybe an 11/30 and them copied down to the >>> Terak (like a PDP-11/03). >>> >>> Is there an existing tool chain that is good at reconstructing C from >>> binaries of the C compiler of this era? From tuhs at tuhs.org Sun Nov 1 13:30:22 2026 From: tuhs at tuhs.org (John Foust via TUHS) Date: Sat, 31 Oct 2026 22:30:22 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> Message-ID: <7.1.0.9.0.20261031223227.0989ee48@threedee.com> At 12:36 PM 10/31/2026, James Frew via TUHS wrote: >Back in The Day™ I was responsible for a CS student lab full of Teraks, Yes, that was the situation when I was an undergrad at UW-Madison in '81 and that's why I baby-duck imprinted on it and still have ten or so Teraks. >So I'm seconding John's suggestion. Seems like a perfectly do-able task for an LLM! Yes, Claude made short work of it. It studied the V7 'cc' code and its output, it re-used a PDP-11 disassembler toolchain I used for the Terak emulator work I did in another project, and it compared it against Reko's output, and found a few bugs but improved itself, it handled some sections of the binary that were straight assembler instead of C, it read Teitelbaum's paper on his CSP,it analyzed what the code was doing against the Terak hardware, it analyzed the functions and determined what they were doing, it reverse-eningeered the yacc for the PL/CS (PL/I-like) / language based on the decompiled code, and this interatively guided it to name variables and functions, so now I have code like that below. This is 100% Claude's output and creation. LLMs are fun as long as you can throw out everything you thought you knew about the software development process, but also retain all your grey-beard wisdom about what you want and how to get there. - John /* 074544..074722: Unlinks the subtree at the cursor's placeholder: a1 is the address of the subtree's last continuation link. In: a1 link address; curNode, curKind, kindList (list kinds), kindOptional (optional kinds), atListEnd. Out: if that link points back to curNode + 3 (the subtree was the only occupant) curNode's code byte is rewritten as empty: 0105 for a list placeholder, 26 for an optional element, 0116 otherwise; else setLink(curNode, the link's target) so the placeholder skips the removed part, parseLine, and atListHead = 1 (cursor at the list head) unless already at a list end. Calls: getLink, setLink, parseLine. */ unlinkSubtree(a1) int a1; { register int r3, r4; r4 = a1; r3 = getLink(r4); if (r3 == curNode + 3) { if (kindList[curKind] == 0) { if (kindOptional[curKind] == 0) { *(char *)curNode = 0116; } else { *(char *)curNode = 26; } } else { *(char *)curNode = 0105; } } else { setLink(curNode, getLink(r4)); parseLine(); if (atListEnd == 0) { atListHead = 1; } } } From tuhs at tuhs.org Sun Nov 1 23:08:23 2026 From: tuhs at tuhs.org (Ralph Corderoy via TUHS) Date: Sun, 01 Nov 2026 13:08:23 +0000 Subject: [TUHS] C from 1980 binaries In-Reply-To: <7.1.0.9.0.20261031223227.0989ee48@threedee.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> Message-ID: <20261101130823.51617C5440@hal.inputplus.co.uk> Hi John, > This is 100% Claude's output and creation. Assuming getLink() doesn't have side effects, line 18's getLink(r4) looks like it can be replaced with r3 given line 6. Is Claude deliberately rendering a duplicate call in the machine code? 1 unlinkSubtree(a1) 2 int a1; 3 { 4 register int r3, r4; 5 r4 = a1; 6 r3 = getLink(r4); 7 if (r3 == curNode + 3) { 8 if (kindList[curKind] == 0) { 9 if (kindOptional[curKind] == 0) { 10 *(char *)curNode = 0116; 11 } else { 12 *(char *)curNode = 26; 13 } 14 } else { 15 *(char *)curNode = 0105; 16 } 17 } else { 18 setLink(curNode, getLink(r4)); 19 parseLine(); 20 if (atListEnd == 0) { 21 atListHead = 1; 22 } 23 } 24 } -- Cheers, Ralph. From tuhs at tuhs.org Mon Nov 2 00:55:17 2026 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Sun, 1 Nov 2026 06:55:17 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: <7.1.0.9.0.20261031223227.0989ee48@threedee.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> Message-ID: <20261101145517.GK15589@mcvoy.com> On Sat, Oct 31, 2026 at 10:30:22PM -0500, John Foust via TUHS wrote: > At 12:36 PM 10/31/2026, James Frew via TUHS wrote: > >Back in The Day? I was responsible for a CS student lab full of Teraks, > > Yes, that was the situation when I was an undergrad at UW-Madison in '81 > and that's why I baby-duck imprinted on it and still have ten or so Teraks. I was there too but not in CS yet. I was undergrad '80-85, grad '86-87. It took me a while to get to CS, I started as an accounting major because my high school coach was also my accounting teacher, so I excelled in high school accounting because, well, coach. College accounting without my coach quickly taught me I hated accounting, still do. I then, much to my culture vulture father's surprise, drifted into Art History. Got good enough in Greek pottery that I pointed out a dating error in Janson. My dad was a physics prof there, he started taking art history classes with me. I became friends with the dean, we'd go out drinking together. He eventually told me "Larry, you are one of the best students I've had, but unless you are willing to do a PhD, there are no jobs, and even then, you'll be in a long line of people for every job". So I went to my dad and said "what do I do?" He gave me some very good advice. He said "It doesn't matter. Well, you can't pick something that goes against your grain like accounting, you have to pick something you like. When you do, learning becomes its own reward". So while I had been drifting around, I had taken a bunch of STEM classes, calc, diffie q, physics (all the easy stuff, I got to quantum mechanics and bowed out, just couldn't wrap my head around probability being the basis). And some CS classes. The data structures course in Pascal resonated with me and I got an A in that. So I picked that. It worked out well for me. All that is to say, John, I don't remember Teraks. I definately remember the terminal room with ~40 terminals hooked up to a VAX 780 (and I fondly remember slovax that had the BSD source on it, I got an account on there eventually and spent a lot of time reading. Ever since then, my main machine has been slovax. slovax is mcvoy.com). So where were the Teraks? I probably got serious in CS around 1982 or 83. Were they still somewhere? --lm From tuhs at tuhs.org Mon Nov 2 01:06:25 2026 From: tuhs at tuhs.org (Jeffrey Joshua Rollin via TUHS) Date: Sun, 1 Nov 2026 15:06:25 +0000 Subject: [TUHS] C from 1980 binaries In-Reply-To: <20261101145517.GK15589@mcvoy.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: I have never heard of Teraks. Sounds like an alien species from Star Trek! (I'm of a similar vintage to Dan Cross and am familiar with such things as the separate ports to the 32-bit Interdata machines, although the closest I have gotten to a minicomputer is, depending on your perspective, either using the University library computer terminals which, I later found out, were connected to a VAX of some sort, or occasionally entertaining the idea of buying a PiDP.) Jeff On Sun, 1 Nov 2026, 14:55 Larry McVoy via TUHS, wrote: > On Sat, Oct 31, 2026 at 10:30:22PM -0500, John Foust via TUHS wrote: > > At 12:36 PM 10/31/2026, James Frew via TUHS wrote: > > >Back in The Day? I was responsible for a CS student lab full of Teraks, > > > > Yes, that was the situation when I was an undergrad at UW-Madison in '81 > > and that's why I baby-duck imprinted on it and still have ten or so > Teraks. > > I was there too but not in CS yet. I was undergrad '80-85, grad '86-87. > It took me a while to get to CS, I started as an accounting major because > my high school coach was also my accounting teacher, so I excelled in high > school accounting because, well, coach. College accounting without my > coach quickly taught me I hated accounting, still do. I then, much to my > culture vulture father's surprise, drifted into Art History. Got good > enough in Greek pottery that I pointed out a dating error in Janson. > My dad was a physics prof there, he started taking art history classes > with me. I became friends with the dean, we'd go out drinking together. > He eventually told me "Larry, you are one of the best students I've had, > but unless you are willing to do a PhD, there are no jobs, and even then, > you'll be in a long line of people for every job". > > So I went to my dad and said "what do I do?" He gave me some very good > advice. He said "It doesn't matter. Well, you can't pick something > that goes against your grain like accounting, you have to pick something > you like. When you do, learning becomes its own reward". So while I > had been drifting around, I had taken a bunch of STEM classes, calc, > diffie q, physics (all the easy stuff, I got to quantum mechanics and > bowed out, just couldn't wrap my head around probability being the basis). > And some CS classes. The data structures course in Pascal resonated with > me and I got an A in that. So I picked that. It worked out well for me. > > All that is to say, John, I don't remember Teraks. I definately remember > the terminal room with ~40 terminals hooked up to a VAX 780 (and I > fondly remember slovax that had the BSD source on it, I got an account > on there eventually and spent a lot of time reading. Ever since then, > my main machine has been slovax. slovax is mcvoy.com). > > So where were the Teraks? I probably got serious in CS around > 1982 or 83. Were they still somewhere? > > --lm > From tuhs at tuhs.org Mon Nov 2 01:57:41 2026 From: tuhs at tuhs.org (Jon Forrest via TUHS) Date: Sun, 1 Nov 2026 07:57:41 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: <07785eb9-5c57-48cf-9f34-88017e81fc5a@gmail.com> On 11/1/26 7:06 AM, Jeffrey Joshua Rollin via TUHS wrote: > I have never heard of Teraks. They were a version of LSI-11 (similar to a PDP-11/03) that has some special graphics hardware. It had less than 1MB of RAM. I also used the Teraks that Frew mentioned at UCSB to (badly) teach a course in PDP-11 assembler. Each student had their own 8" floppy on which they had a bootable version of RT-11. That version was sufficient to edit, assemble, run, and debug assembler programs. One of the reasons I taught the course so badly was that I had never actually programmed in PDP-11 assembler but I had been around a bunch of assembler programs where I had an internship. One of the professors in CS, Art Pyster, used UCSD Pascal on the Teraks to develop a compiler generator he later described in "Compiler Design and Construction", one of the early books about compilers. Cordially, Jon Forrest From tuhs at tuhs.org Mon Nov 2 02:00:06 2026 From: tuhs at tuhs.org (John Foust via TUHS) Date: Sun, 01 Nov 2026 10:00:06 -0600 Subject: [TUHS] C from 1980 binaries In-Reply-To: <20261101145517.GK15589@mcvoy.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: <7.1.0.9.0.20261101100014.0989e030@threedee.com> At 08:55 AM 11/1/2026, Larry McVoy wrote: >All that is to say, John, I don't remember Teraks. I definately remember >the terminal room with ~40 terminals hooked up to a VAX 780 (and I >fondly remember slovax that had the BSD source on it, I got an account >on there eventually and spent a lot of time reading. Ever since then, >my main machine has been slovax. slovax is mcvoy.com). >So where were the Teraks? I probably got serious in CS around >1982 or 83. Were they still somewhere? The Teraks were used to teach UCSD Pascal to entry-level CS classes and business school programming courses, but they were far more attractive to other CS students because they were $20,000 PDP-11/03 equivalents with mono bitmap graphics and one to four 8" floppies. To be fair, the Teraks were often used as terminals by people sharp enough to have a Terak disk that ran a terminal emulator and connected to their favorite mainframe over the Gandalf serial port switching network. Each Terak had a Gandalf. The Teraks were in their own room in the middle area of the CS basement. You could find them in the math department, too. Yes, there were terminal rooms full of glass TTYs. If the terminal room was full and the Teraks were empty, you could use a Terak as a terminal. There were a few students who were trying to craft their own Unix-alikes for it. You could run RT-11 if you had the disks. There was a more true Unix for the Terak, but I think it was developed in Minnesota and either hadn't made it to Wisconsin when I was there or it was still in development by the time I left. Prof. Teitelbaum's grad student was Tom Reps who went to UW-Madison to teach after the CSP project at Cornell. The CSP is regarded as a milestone on the road to IDEs, as one of the firsts. The CSP was developed using tools from both environments. Its disks are in UCSD p-System format, but the executables appear to have been cross-compiled from a V7 mainframe. But my memories of you are that you were in all those rooms, too. This might jog your memory: https://www.threedee.com/jcm/terak/ Paul Haeberli of SGI was cutting his teeth on Teraks at UW-Madison. Tom Christiansen of Perl was, too. There's a few other UW names in the fine print there. Prof. Bart Miller might've had something to do with the Teraks, as he was under Bowles at UCSD. - John From tuhs at tuhs.org Mon Nov 2 02:05:00 2026 From: tuhs at tuhs.org (John Foust via TUHS) Date: Sun, 01 Nov 2026 10:05:00 -0600 Subject: [TUHS] C from 1980 binaries In-Reply-To: <20261101130823.51617C5440@hal.inputplus.co.uk> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101130823.51617C5440@hal.inputplus.co.uk> Message-ID: <7.1.0.9.0.20261101101010.0989e2c0@threedee.com> At 07:08 AM 11/1/2026, Ralph Corderoy wrote: >Assuming getLink() doesn't have side effects, >line 18's getLink(r4) looks like it can be replaced with r3 given line 6. >Is Claude deliberately rendering a duplicate call in the machine code? I asked, and sent the longer reply in private email. The short answer is "The second getLink call is really in the machine code (two JSRs to getLink at 074556 and 074602 with the same argument, which V7 cc never merged), and the r3 is just the compiler parking the first result during the comparison, which the decompiler rendered as a register variable rather than folding it back into the if." The code appears to be full of horrid "goto", too. - John From tuhs at tuhs.org Mon Nov 2 03:02:33 2026 From: tuhs at tuhs.org (Gregg Levine via TUHS) Date: Sun, 1 Nov 2026 12:02:33 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: <07785eb9-5c57-48cf-9f34-88017e81fc5a@gmail.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <07785eb9-5c57-48cf-9f34-88017e81fc5a@gmail.com> Message-ID: Hello! I remember reading that book. So much so that I found the copy I did read online, and bought it. ---- Gregg C Levine gregg.drwho8 at gmail.com "This signature was once found posting rude messages in English in the Moscow subway." On Sun, Nov 1, 2026 at 10:58 AM Jon Forrest via TUHS wrote: > > > > On 11/1/26 7:06 AM, Jeffrey Joshua Rollin via TUHS wrote: > > I have never heard of Teraks. > They were a version of LSI-11 (similar to a PDP-11/03) > that has some special graphics hardware. It had less than > 1MB of RAM. > > I also used the Teraks that Frew mentioned at UCSB > to (badly) teach a course in PDP-11 assembler. Each > student had their own 8" floppy on which they had > a bootable version of RT-11. That version was sufficient > to edit, assemble, run, and debug assembler programs. > > One of the reasons I taught the course so badly was that > I had never actually programmed in PDP-11 assembler but > I had been around a bunch of assembler programs where > I had an internship. > > One of the professors in CS, Art Pyster, used UCSD Pascal on > the Teraks to develop a compiler generator he later > described in "Compiler Design and Construction", > one of the early books about compilers. > > Cordially, > Jon Forrest > From tuhs at tuhs.org Mon Nov 2 11:13:57 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Sun, 1 Nov 2026 20:13:57 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: The 8510/A was basically an LSI-11 with some graphics hardware, released in 1977. The Terak was arguably the world’s first true standalone graphics workstation. Terak OEM'ed LSI-11 (a KD11-F or maybe the KD11-HA) processor boards from DEC directly, and this used what DEC officially called the 'LSI-11 BUS '; but the world calls it by WD's name, the QBUS. So Terak designed a workstation around DEC's CPU. They built a custom, proprietary chassis, memory system, 8-inch floppy disk controller, and memory-mapped bitmap graphics framebuffer that sat directly in the PDP-11's I/O space. Because they kept the core processor a stock DEC LSI-11 and DEC's I/O layout, the Terak 8510/A can run RT-11, Mini-UNIX, and eventually the UCSD p-System. Theses URL have more details: https://www.threedee.com/jcm/terak/ and Wikipedia: h ttps://en.wikipedia.org/wiki/Terak_Corporation https://www.threedee.com/jcm/terak/ The LSI-11 and the Terak system share an interesting, interrelated history. I've forgotten many of the details and sales rules to get the chips and the ROM with the PDP-11-compatible microcode, but WD developed the MCP-1600 chip set and wrote the original microcode (by a real bad-ass programmer named Duane Dickhut, whom I met a few years later). Note that WD and DEC were reasonably tied. Gordon Bell/DEC had created the UART for the PDP-1, and in 1971 a small firm originally called General Digital made and released the first LSI version: the Western Digital WD1402A. Once it was available, it became the standard DEC UART on serial boards, and their business relationship began. As I understand it, in the early 1970s, DEC management had begun to realize it needed some LSI-based computers to protect the low end of its market, but lacked internal silicon fabrication capabilities (DEC Hudson, my future home, was not yet built). So DEC contracted Western Digital in 1974 to co-develop a chipset (DEC paid millions of dollars in NRE charges), planning to use it for a number of systems, starting with the by-then-popular PDP-11 line, but it was getting pressure in the market from other 16-bit players such as DG and HP who were selling at lower price points. At the time, WD was working on the chipset, and DEC had won the lawsuit it filed against California Data Processors (*a.k.a.* CalData). The results of that lawsuit were fresh in the eyes of the team at WD [Ken O'Mohundro and Fred Cox's firm cloned a number of different ISAs but chose the UNIBUS, which got them in trouble. O'Mohundro had famously run the complete 11//40 diagnostics suite for Ken Olson at the National Computer show in CalData's booth]. As for what was normal for the industry, when Western Digital wanted to show off the power of their new chipset, they needed to create a reference design. I've been told some of the senior folks at WD didn't want to piss off DEC, which had been a good business partner for years and >>had<< funded much of the 1600 development. Using another firm's bus seemed unwise, so they created a simpler but 'Unibus-Like' multiplexed 16-bit address/data bus layout. DEC's engineers later took the original WD bus design, "refined" (I like to say peed on it to make it more DEC-like), and formally branded it the LSI-11 Bus. Roll the clock forward, and Terak comes around. So they built their devices for the CPU they had picked and created an early workstation. I'm not 100% sure if Ken Bowles and the UCSD team had a real DEC LSI-11 like the 11/03 as the first target, but at some point they discovered the Tek 8510/A, which, as I understand it, became the 'master system'/gold standard for the UCSD Pascal system. Their trick was to take the P4 Pascal reference compiler from Wirth and his students in Europe, reimplement its bytecode interpreter for a target ISA, expand it a bit, and wrap a development system around it. At some point they targeted their interpreter to the LSI-11 processor, Terak's I/O system, and in particular, its framebuffer. Since Pascal had by then become the standard teaching language in many universities, many of them purchased the 8510/A (which I suspect is how many people learned about them). You could also get the original p-System for free from UCSD on 8 floppies or on a distribution tape, and it would run on something like an LSI-11 with a VT-52 or VT-100 instead of the Terak if you configured it appropriately; I suspect many universities did that as well. In an interesting twist of fate, in 1979, because of the academic success of the UCSD p-System running on top of their own silicon (via DEC's LSI-11 p-code interpreter), Western Digital's engineers and management saw an opportunity, thinking: *“Why use an interpreter to translate p-code into PDP-11 code? Let's just burn the p-code interpreter directly into the microcode.” *The result was the WD/9000. I know it had new microcode, but I think they also updated and replaced the control-chip portion of their chipset family. Of course, by late 1979, Motorola had released the 68000 and the age of UNIX workstations had begun, and I think the Western Digital Pascal System was doomed and the 16-bit nature of the PDP-11 killed the Terak. On Sun, Nov 1, 2026 at 10:06 AM Jeffrey Joshua Rollin via TUHS < tuhs at tuhs.org> wrote: > I have never heard of Teraks. > > Sounds like an alien species from Star Trek! > > (I'm of a similar vintage to Dan Cross and am familiar with such things as > the separate ports to the 32-bit Interdata machines, although the closest I > have gotten to a minicomputer is, depending on your perspective, either > using the University library computer terminals which, I later found out, > were connected to a VAX of some sort, or occasionally entertaining the idea > of buying a PiDP.) > > Jeff > > On Sun, 1 Nov 2026, 14:55 Larry McVoy via TUHS, wrote: > > > On Sat, Oct 31, 2026 at 10:30:22PM -0500, John Foust via TUHS wrote: > > > At 12:36 PM 10/31/2026, James Frew via TUHS wrote: > > > >Back in The Day? I was responsible for a CS student lab full of > Teraks, > > > > > > Yes, that was the situation when I was an undergrad at UW-Madison in > '81 > > > and that's why I baby-duck imprinted on it and still have ten or so > > Teraks. > > > > I was there too but not in CS yet. I was undergrad '80-85, grad '86-87. > > It took me a while to get to CS, I started as an accounting major because > > my high school coach was also my accounting teacher, so I excelled in > high > > school accounting because, well, coach. College accounting without my > > coach quickly taught me I hated accounting, still do. I then, much to my > > culture vulture father's surprise, drifted into Art History. Got good > > enough in Greek pottery that I pointed out a dating error in Janson. > > My dad was a physics prof there, he started taking art history classes > > with me. I became friends with the dean, we'd go out drinking together. > > He eventually told me "Larry, you are one of the best students I've had, > > but unless you are willing to do a PhD, there are no jobs, and even then, > > you'll be in a long line of people for every job". > > > > So I went to my dad and said "what do I do?" He gave me some very good > > advice. He said "It doesn't matter. Well, you can't pick something > > that goes against your grain like accounting, you have to pick something > > you like. When you do, learning becomes its own reward". So while I > > had been drifting around, I had taken a bunch of STEM classes, calc, > > diffie q, physics (all the easy stuff, I got to quantum mechanics and > > bowed out, just couldn't wrap my head around probability being the > basis). > > And some CS classes. The data structures course in Pascal resonated with > > me and I got an A in that. So I picked that. It worked out well for me. > > > > All that is to say, John, I don't remember Teraks. I definately remember > > the terminal room with ~40 terminals hooked up to a VAX 780 (and I > > fondly remember slovax that had the BSD source on it, I got an account > > on there eventually and spent a lot of time reading. Ever since then, > > my main machine has been slovax. slovax is mcvoy.com). > > > > So where were the Teraks? I probably got serious in CS around > > 1982 or 83. Were they still somewhere? > > > > --lm > > > From tuhs at tuhs.org Mon Nov 2 11:34:24 2026 From: tuhs at tuhs.org (Bakul Shah via TUHS) Date: Sun, 1 Nov 2026 17:34:24 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> On Nov 1, 2026, at 5:13 PM, Clem Cole via TUHS wrote: > > The 8510/A was basically an LSI-11 with some graphics hardware, released in > 1977. The Terak was arguably the world’s first true standalone graphics > workstation. In 1976 or so I wrote some code to play with graphics. It was to simulate a fly-through a wire-frame dodecahedron to be displayed on a GT40 (this one also had the lunar lander program). I do not recall now if my code was cross-compiled on a TOPS-10 system or compiled right on the GT40 but it certainly seemed like a standalone graphics computer.... From tuhs at tuhs.org Mon Nov 2 12:10:03 2026 From: tuhs at tuhs.org (Jon Forrest via TUHS) Date: Sun, 1 Nov 2026 18:10:03 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: On 11/1/26 5:13 PM, Clem Cole via TUHS wrote: > At the time, WD was working on the chipset, and DEC had won the lawsuit it > filed against California Data Processors (*a.k.a.* CalData). The results of > that lawsuit were fresh in the eyes of the team at WD [Ken O'Mohundro and > Fred Cox's firm cloned a number of different ISAs but chose the UNIBUS, > which got them in trouble. O'Mohundro had famously run the complete 11//40 > diagnostics suite for Ken Olson at the National Computer show in CalData's > booth]. Ken O'Mohundro - there's a name I haven't heard in a while. In the early 1980s I ran the Santa Barbara VMS Users Group. At the time, there were a few vendors who were starting to make DEC-compatible peripherals, such as memory, and disk/communication/network controllers. One of the vendors was Able, in Orange County, CA. Somehow Ken contacted me to ask if he could come talk to our users group about DEC stuff. Ken was a self-described stutterer who enjoyed addressing groups of people as a form of speech therapy. So, he came all the way up to Santa Barbara from Irvine and gave a nice presentation. IIRC, Able Computer paid for lunch for everyone. He was a very nice guy. Cordially, Jon Forrest From tuhs at tuhs.org Mon Nov 2 12:16:55 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Sun, 1 Nov 2026 21:16:55 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> Message-ID: Fair. Did it have a disk of any type and was the local OS? Was it really standalone? I didn't think there was room in the case for anything else. I believe DEC created the GT44 which I believe could but I never saw one in the wild. FWIW. I also had a GT40 with an 11/05 CPU with a Unibus, but no local storage or an OS like DOS-11 or RT11. My experience was it had a 960 cps serial connection to its mothership, our Unix system in my case or to an ASR33 with its papertape reader. In our case we had a runtime we used for the Graphics code which I wrote part of as part Bob Sproull's graphics course. It was similar to the ones for 3-Rivers GPD which had dedicated 11/20 and a connection to CS-Front End so they usually used one of the PDP-10s, C.mmp or one of the Unix systems as a mothership. In my case I used the Unix V6 C compiler, as and led. I wrote a loader for what GT40 ROM wanted, although I've completely forgotten and of those details. Also I know that when when Jack wrote Moonlander, he was using a PDP-10 for his development and downloaded to the GT40. I think he was using PPT. He told me once how he did the development, but I dont remember that detail, although I think he was using macro-11x and must have had some sort of load system. All editing, however, was done on the PDP-10. The Terack was sold as a complete 'PC' if will. Not needing anything else to support it. Sent from a handheld expect more typos than usual On Sun, Nov 1, 2026 at 8:34 PM Bakul Shah wrote: > On Nov 1, 2026, at 5:13 PM, Clem Cole via TUHS wrote: > > > > The 8510/A was basically an LSI-11 with some graphics hardware, released > in > > 1977. The Terak was arguably the world’s first true standalone graphics > > workstation. > > In 1976 or so I wrote some code to play with graphics. It was to simulate > a fly-through a wire-frame dodecahedron to be displayed on a GT40 (this > one also had the lunar lander program). I do not recall now if my code > was cross-compiled on a TOPS-10 system or compiled right on the GT40 but > it certainly seemed like a standalone graphics computer.... From tuhs at tuhs.org Mon Nov 2 13:15:04 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Sun, 1 Nov 2026 22:15:04 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: Indeed. Great guy. The Unix community stop solidly on his shoulders for many years. I haven't spoken to him in years. After the CalData lawsuit and when he created Able, he said he'd only make peripherals and swore off CPUs. His DMAX (DH/DM) became pretty much required if you were part of USENET, particularly if you had Telebit Modems. He made a lots money selling them to the Unix community. He knew it was very appreciative. He used to come to USENIX conferences regularly and warn us to watch out about Unibus implementations from a number of the 3rd party vendors. Which with him being one himself, was humorous. Besides his famous replacements for DEC's terminal mux products, he also made bus repeaters and caches. At an early USENIX shortly after he had released a cache for the 11/34, a couple of us were at dinner and I was sitting next to him. In those days, besides being a Unix hacker, I was also designing hardware and had just designed the MMU for the Tek Magnolia. So I asked Ken, since he was moving the Unibus memory down stream of his cache, why couldn't put a new MMU in it, and basically NOP the DEC MMU, but split 22-bit adresses out the other end. We sketched something and he said he would think about it. About a month or two later he calls me and asked if I wanted to try what he just finished. They ended up calling it the "Enable" and allowed 11/40 and 11/45 class systems to address up to 4M like the 11/70. If you google it, you'll find Noel's page on Gunkies. Noel also has links to the paper I wrote about it, as well as scan of the prepublication user manual and a sketch of the registers it supported whicj I had sent Noel a few years ago. The code is actually still 2.11BSD but disabled since it's no longer relevant as 2.11 pretty much requires an 11/70 and boards are hard to find these days. BTW, a few years later Mike O'Dell and I tried to get him to make a version on the DMAX for the ISA or EISA bus. We also suggested making an Ethernet version with a 68000 in it and we would wrote the code for it, since putting serial ports, particularly ones with modems, on PC hardware was such a losing proposition. But Ken told us that margins for PC peripherals was too low for him make any money on it. Soon there after we all bought Livingston boxes to do the job. I always felt bad, as I think would have done a great job if could have figured it out. BTW: he did eventually break his rule about CPUs slightly. He ended up creating a QBUS 68010 CPU board that he showed me when I visited him in Irvine at some point. I don't know how well it sold and lost track of him after that as I left the Vax/PDP-11 world and when I returned to it to work on Alpha, I never ran into him. Sent from a handheld expect more typos than usual On Sun, Nov 1, 2026 at 9:11 PM Jon Forrest via TUHS wrote: > > > On 11/1/26 5:13 PM, Clem Cole via TUHS wrote: > > > At the time, WD was working on the chipset, and DEC had won the lawsuit > it > > filed against California Data Processors (*a.k.a.* CalData). The results > of > > that lawsuit were fresh in the eyes of the team at WD [Ken O'Mohundro > and > > Fred Cox's firm cloned a number of different ISAs but chose the UNIBUS, > > which got them in trouble. O'Mohundro had famously run the complete > 11//40 > > diagnostics suite for Ken Olson at the National Computer show in > CalData's > > booth]. > Ken O'Mohundro - there's a name I haven't heard in a while. In the early > 1980s I ran the Santa Barbara VMS Users Group. At the time, there were a > few vendors who were starting to make DEC-compatible peripherals, such > as memory, and disk/communication/network controllers. One of the > vendors was Able, in Orange County, CA. Somehow Ken contacted me to > ask if he could come talk to our users group about DEC stuff. Ken > was a self-described stutterer who enjoyed addressing groups of > people as a form of speech therapy. So, he came all the way up > to Santa Barbara from Irvine and gave a nice presentation. IIRC, > Able Computer paid for lunch for everyone. > > He was a very nice guy. > > Cordially, > Jon Forrest > > From tuhs at tuhs.org Mon Nov 2 16:49:06 2026 From: tuhs at tuhs.org (Marc Donner via TUHS) Date: Mon, 2 Nov 2026 01:49:06 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: UNIBUS - that’s a term I haven’t heard in tears. At JPL as a young EE right out of college I designed a UNIBUS monitor that my boss used in various installations. That educated me about the bus, so years later I recognized that the Motorola 68000 had essentially implemented it faithfully, with the exception of making the Sync signal active-low instead of active-high (a true blessing in TTL days). ===== [My messages are from me, not from AI.] exploringaimatters.com nygeek.net On Sun, Nov 1, 2026 at 10:16 PM Clem Cole via TUHS wrote: > Indeed. Great guy. The Unix community stop solidly on his shoulders for > many years. I haven't spoken to him in years. After the CalData lawsuit > and when he created Able, he said he'd only make peripherals and swore off > CPUs. His DMAX (DH/DM) became pretty much required if you were part of > USENET, particularly if you had Telebit Modems. He made a lots money > selling them to the Unix community. He knew it was very appreciative. He > used to come to USENIX conferences regularly and warn us to watch out about > Unibus implementations from a number of the 3rd party vendors. Which with > him being one himself, was humorous. > > Besides his famous replacements for DEC's terminal mux products, he also > made bus repeaters and caches. At an early USENIX shortly after he had > released a cache for the 11/34, a couple of us were at dinner and I was > sitting next to him. In those days, besides being a Unix hacker, I was > also designing hardware and had just designed the MMU for the Tek > Magnolia. So I asked Ken, since he was moving the Unibus memory down > stream of his cache, why couldn't put a new MMU in it, and basically NOP > the DEC MMU, but split 22-bit adresses out the other end. We sketched > something and he said he would think about it. > > About a month or two later he calls me and asked if I wanted to try what he > just finished. They ended up calling it the "Enable" and allowed 11/40 and > 11/45 class systems to address up to 4M like the 11/70. If you google it, > you'll find Noel's page on Gunkies. Noel also has links to the paper I > wrote about it, as well as scan of the prepublication user manual and a > sketch of the registers it supported whicj I had sent Noel a few years > ago. The code is actually still 2.11BSD but disabled since it's no longer > relevant as 2.11 pretty much requires an 11/70 and boards are hard to find > these days. > > BTW, a few years later Mike O'Dell and I tried to get him to make a version > on the DMAX for the ISA or EISA bus. We also suggested making an Ethernet > version with a 68000 in it and we would wrote the code for it, since > putting serial ports, particularly ones with modems, on PC hardware was > such a losing proposition. But Ken told us that margins for PC peripherals > was too low for him make any money on it. > > Soon there after we all bought Livingston boxes to do the job. I always > felt bad, as I think would have done a great job if could have figured it > out. > > BTW: he did eventually break his rule about CPUs slightly. He ended up > creating a QBUS 68010 CPU board that he showed me when I visited him in > Irvine at some point. I don't know how well it sold and lost track of him > after that as I left the Vax/PDP-11 world and when I returned to it to work > on Alpha, I never ran into him. > > > Sent from a handheld expect more typos than usual > > On Sun, Nov 1, 2026 at 9:11 PM Jon Forrest via TUHS wrote: > > > > > > > On 11/1/26 5:13 PM, Clem Cole via TUHS wrote: > > > > > At the time, WD was working on the chipset, and DEC had won the lawsuit > > it > > > filed against California Data Processors (*a.k.a.* CalData). The > results > > of > > > that lawsuit were fresh in the eyes of the team at WD [Ken O'Mohundro > > and > > > Fred Cox's firm cloned a number of different ISAs but chose the UNIBUS, > > > which got them in trouble. O'Mohundro had famously run the complete > > 11//40 > > > diagnostics suite for Ken Olson at the National Computer show in > > CalData's > > > booth]. > > Ken O'Mohundro - there's a name I haven't heard in a while. In the early > > 1980s I ran the Santa Barbara VMS Users Group. At the time, there were a > > few vendors who were starting to make DEC-compatible peripherals, such > > as memory, and disk/communication/network controllers. One of the > > vendors was Able, in Orange County, CA. Somehow Ken contacted me to > > ask if he could come talk to our users group about DEC stuff. Ken > > was a self-described stutterer who enjoyed addressing groups of > > people as a form of speech therapy. So, he came all the way up > > to Santa Barbara from Irvine and gave a nice presentation. IIRC, > > Able Computer paid for lunch for everyone. > > > > He was a very nice guy. > > > > Cordially, > > Jon Forrest > > > > > From tuhs at tuhs.org Tue Nov 3 01:06:08 2026 From: tuhs at tuhs.org (al kossow via TUHS) Date: Mon, 2 Nov 2026 07:06:08 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> Message-ID: <784565bd-8734-4eae-aaea-eab2f9751acc@bitsavers.org> On 11/1/26 6:16 PM, Clem Cole via TUHS wrote: > I believe DEC created the GT44 which I believe could but I never saw one in > the wild. > GT44 was an 11/40 with RK05s and the VT11 display processor GT46 swapped the 11/40 for an 11/34. We had one at UW-Milwaukee I think it was the first computer there to run V7 Unix From tuhs at tuhs.org Tue Nov 3 02:00:49 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 2 Nov 2026 11:00:49 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: On Mon, Nov 2, 2026 at 1:49 AM Marc Donner wrote: > That educated me about the bus, so years later I recognized that the > Motorola 68000 had essentially implemented it faithfully, with the > exception of making the Sync signal active-low instead of active-high (a > true blessing in TTL days). > Yup, it is very close and was intended to be, as it was a model, but the designers were careful not to make it exactly like it. Remember, what became the 68000 was a skunkworks project and >>not<< an 'official' project at Moto. Les Crudele, Nick Tradenick, and a couple of others had joined Moto from Schlumberger, where they had been interfacing PDP-11 control systems for the oil drilling industry and were extremely familiar with both the PDP-11 ISA and the Unibus. But Les says he and the team were very careful. The bet they had with Tom Grunner was that they could basically build a PDP-11 on a single chip using Motorola's current process technology. Tom was interested enough to let them try in the back of the lab. But those folks (particularly Tom, as I understand it) were all extremely aware of the CalData lawsuit. Thus, they made sure the ISA and bus were 'different enough'. For instance, Nick made the instruction set use the same addressing modes and functions, but isn't the same, using address and data registers instead of general registers. I've long forgotten the specifics of the differences, but similarly Les said he spent a lot of time thinking about what we could use from the Unibus and QBUS, but still be different enough that they would not be taken to court as CalData had. The fact that, a few years later, Ken O'Mohundro put a 68010 on a QBUS fairly easily is living proof it wasn't that far off. The rest of the story of how 68000 went from being a skunkworks project to a real product is cool, and I am honored to have helped play a small part in that story with the TekLabs's Magnolia system; but it's not relevant here. From tuhs at tuhs.org Tue Nov 3 02:19:41 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Mon, 2 Nov 2026 11:19:41 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: <784565bd-8734-4eae-aaea-eab2f9751acc@bitsavers.org> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> <784565bd-8734-4eae-aaea-eab2f9751acc@bitsavers.org> Message-ID: below On Mon, Nov 2, 2026 at 10:06 AM al kossow via TUHS wrote: > > On 11/1/26 6:16 PM, Clem Cole via TUHS wrote: > > I believe DEC created the GT44 which I believe could but I never saw one > in > > the wild. > > > GT44 was an 11/40 with RK05s and the VT11 display processor > Running RT-11, I would suspect - do you know? > > GT46 swapped the 11/40 for an 11/34. We had one at UW-Milwaukee > Do you remember when they were released? > > I think it was the first computer there to run V7 Unix > Cool, so that was in 1979 or later, since that's when V7 was released. Do you remember how much memory it had? BTW: Neither V6 nor V7 had a VT-11 driver, nor did the DEC V7M tape. If memory serves me, there might have been one of the earlier USENIX tapes, or did you folks write one yourself? Clem From tuhs at tuhs.org Tue Nov 3 03:53:29 2026 From: tuhs at tuhs.org (Lawrence Stewart via TUHS) Date: Mon, 2 Nov 2026 12:53:29 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: <5FE4F696-6EA4-4410-8625-ABE899D42A9D@serissa.com> > On Nov 2, 2026, at 11:00, Clem Cole via TUHS wrote: > > On Mon, Nov 2, 2026 at 1:49 AM Marc Donner wrote: > >> That educated me about the bus, so years later I recognized that the >> Motorola 68000 had essentially implemented it faithfully, with the >> exception of making the Sync signal active-low instead of active-high (a >> true blessing in TTL days). >> > Yup, it is very close and was intended to be, as it was a model, but the > designers were careful not to make it exactly like it. Remember, what > became the 68000 was a skunkworks project and >>not<< an 'official' project > at Moto. Les Crudele, Nick Tradenick, and a couple of others had joined > Moto from Schlumberger, where they had been interfacing PDP-11 control > systems for the oil drilling industry and were extremely familiar with both > the PDP-11 ISA and the Unibus. But Les says he and the team were very > careful. The bet they had with Tom Grunner was that they could basically > build a PDP-11 on a single chip using Motorola's current process > technology. Tom was interested enough to let them try in the back of the > lab. But those folks (particularly Tom, as I understand it) were all > extremely aware of the CalData lawsuit. Thus, they made sure the ISA and > bus were 'different enough'. For instance, Nick made the instruction set > use the same addressing modes and functions, but isn't the same, using > address and data registers instead of general registers. I've long > forgotten the specifics of the differences, but similarly Les said he spent > a lot of time thinking about what we could use from the Unibus and QBUS, > but still be different enough that they would not be taken to court as > CalData had. The fact that, a few years later, Ken O'Mohundro put a 68010 > on a QBUS fairly easily is living proof it wasn't that far off. > > The rest of the story of how 68000 went from being a skunkworks project to > a real product is cool, and I am honored to have helped play a small part > in that story with the TekLabs's Magnolia system; but it's not relevant > here. It is not well known, but the first DEC Systems Research Center “Firefly” multiprocessor was actually a 68010. It drove a QBUS for I/O. We never went into production because we were able to get 78032 Microvax II chips in time. This would have been late 1984 or early 1985 IIRC. Chuck Thacker did the hardware design for the 68010 version, and I wrote the boot roms. https://en.wikipedia.org/wiki/DEC_Firefly Not exactly TUHS, but the connection is the Firefly OS would run standard Ultrix-32 binaries. From tuhs at tuhs.org Wed Nov 4 09:02:07 2026 From: tuhs at tuhs.org (Mary Ann Horton via TUHS) Date: Tue, 3 Nov 2026 15:02:07 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: <7.1.0.9.0.20261101100014.0989e030@threedee.com> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <7.1.0.9.0.20261101100014.0989e030@threedee.com> Message-ID: This is bringing back so many wonderful memories. I was at UW-Madison from 76 to 78 but we just had an 11/45 running V6 and an array of dumb terminals. In 19777, that was my intro to UNIX and I was soooo hooked. My main contribution was the hed editor - an enhanced ed. I was very loyal to it until Bill Joy gave me vi. After I transferred to Berkeley in 78, my PhD dissertation was what I called a "language editor". The Reps/Teitelbaum CPS system was one of my inspirations. I didn't realize they had it running on a Terak - I'd never heard of it until now. And I didn't know UW-Madison was using Terak workstations around 1980. Even when I went to Bell Labs, I didn't get a GUI until 88 when I got a BLIT. (IBM PCs didn't count.) CPS and many similar systems had you edit program parse trees rather than editing text. Instead of inserting and deleting text, you'd graft and prune trees. It was a creative idea, but I felt it was too restrictive on the user. Syntax errors were impossible, comments were part of the syntax and could only go in certain places. My approach was that you'd edit the tree, but with a textual user interface, so I built a mechanism for arbitrary text attached to a node in the tree. My tool, Babel, would perform incremental parsing and semantic error checking and highlight errors as soon as the text operation completed. (Eclipse and MS Visual Studio followed, but I wasn't involved.) Babel ran on Ernie (a VAX 11/780) and was horribly slow. A typical insertion took about 1 minute. It worked but wasn't a useful tool. Was CPS used by students on the Terak? How was the experience? I'm curious how CPS performance was on the Terak. Processors are just a bit faster now...  A few years ago I briefly brought up 3BSD on simh and ran the Babel binary. It was speedy, but I'd used the BREAK key to begin commands, and that doesn't translate today. One of these days I'll port it to modern C and play with it. Thanks, /Mary Ann Horton/ (she/her/ma'am)       Keynote Speaker on Inclusion and Innovation       Award Winning Author maryannhorton.com On 11/1/26 08:00, John Foust via TUHS wrote: > At 08:55 AM 11/1/2026, Larry McVoy wrote: >> All that is to say, John, I don't remember Teraks. I definately remember >> the terminal room with ~40 terminals hooked up to a VAX 780 (and I >> fondly remember slovax that had the BSD source on it, I got an account >> on there eventually and spent a lot of time reading. Ever since then, >> my main machine has been slovax. slovax is mcvoy.com). >> So where were the Teraks? I probably got serious in CS around >> 1982 or 83. Were they still somewhere? > The Teraks were used to teach UCSD Pascal to entry-level CS classes > and business school programming courses, but they were far more > attractive to other CS students because they were $20,000 > PDP-11/03 equivalents with mono bitmap graphics and one to > four 8" floppies. > > To be fair, the Teraks were often used as terminals by people > sharp enough to have a Terak disk that ran a terminal emulator > and connected to their favorite mainframe over the Gandalf > serial port switching network. Each Terak had a Gandalf. > > The Teraks were in their own room in the middle area of > the CS basement. You could find them in the math department, too. > Yes, there were terminal rooms full of glass TTYs. > If the terminal room was full and the Teraks were empty, > you could use a Terak as a terminal. > > There were a few students who were trying to craft their own > Unix-alikes for it. You could run RT-11 if you had the disks. > There was a more true Unix for the Terak, but I think it was > developed in Minnesota and either hadn't made it to Wisconsin > when I was there or it was still in development by the time I left. > > Prof. Teitelbaum's grad student was Tom Reps who went to UW-Madison > to teach after the CSP project at Cornell. The CSP is regarded as > a milestone on the road to IDEs, as one of the firsts. > > The CSP was developed using tools from both environments. Its > disks are in UCSD p-System format, but the executables appear > to have been cross-compiled from a V7 mainframe. > > But my memories of you are that you were in all those rooms, too. > > This might jog your memory: > > https://www.threedee.com/jcm/terak/ > > Paul Haeberli of SGI was cutting his teeth on Teraks at UW-Madison. > Tom Christiansen of Perl was, too. There's a few other UW names > in the fine print there. Prof. Bart Miller might've had something > to do with the Teraks, as he was under Bowles at UCSD. > > - John From tuhs at tuhs.org Wed Nov 4 09:41:52 2026 From: tuhs at tuhs.org (John Foust via TUHS) Date: Tue, 03 Nov 2026 17:41:52 -0600 Subject: [TUHS] Terak Mini-UNIX In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> <6F0E0FEF-7F61-4A30-8BA9-DEF2CB4229CA@iitbombay.org> Message-ID: <7.1.0.9.0.20261103174156.080c7838@threedee.com> Claude and I analyzed the Terak-Unix at Bitsavers https://bitsavers.trailing-edge.com/bits/Terak/mini-unix/. The Terak Unix on Bitsavers is based on Heinz Lycklama's MINI-UNIX (Bell Labs, 1977: UNIX Sixth Edition for PDP-11s without memory management), ported to the Terak 8510/a's QX floppy controller. Aado Perandi, a graduate student at the University of Minnesota, built the first Terak UNIX in spring 1979. It was described in the University Computing Center's newsletter in May 1979. The work was done on a PDP-11/40 running Unix V6 and copied to ten Terak 8" floppies, IBM 3740 SSSD. The C is V6, the libraries using the newer archive format from PWB V7. Perandi's modifications were lost in a disk failure that summer and he left; Marc G. Smith at SICL (the Special Interactive Computations Laboratory within the Space Science Center), redeveloped it from "mostly from scratch" and was still distributing floppies in April 1980. A troublesome flaw is that the swap area is kept on the boot floppy but the Terak hardware did not have any disk-changed notification. If you removed the boot floppy from the boot drive and inserted another and then the system swapped, it would write over whatever was on the second floppy. Many of the disks images on Bitsavers show this. Claude decompiled the binaries and restored them as C source, compared against the MINI-UNIX source, determined the changes that Smith made and the other Terak-unique parts, compiled it all again within my Terak emulator, rebuilding the kernel in a minute, and confirmed everything matches to the byte. I will pass my repairs to Bitsavers as soon as I figure out how to do that. Perandi and Smith are still around, I reached out but haven't had a response. >From my own perspective, I'm puzzled as to why this UNIX didn't make it to UW-Madison's Teraks as far as I remember. The interest was certainly there; I remember an elite hacker who crafted a Unix-alike under the UCSD p-System, with a shell that ran p-System executables with short names. Perhaps it was the cost of ten floppies and student programmers were broke back then. - John From tuhs at tuhs.org Wed Nov 4 10:12:34 2026 From: tuhs at tuhs.org (ROBERT MYERS via TUHS) Date: Tue, 3 Nov 2026 18:12:34 -0600 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? References: <759F09CA-499A-49E1-8078-8E11B3D72D4A.ref@yahoo.com> Message-ID: <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> I have a 11/83 with floating point and a 11/23+ that I haven’t determined if it does. Both have 22 bits of address space. Both will have either a QBone or QBone Duo module. Any issues with running Various Unixes on them? I know the 11/83 has 4Mb of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. Sent from my iPhone From tuhs at tuhs.org Wed Nov 4 10:23:02 2026 From: tuhs at tuhs.org (segaloco via TUHS) Date: Wed, 04 Nov 2026 00:23:02 +0000 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> References: <759F09CA-499A-49E1-8078-8E11B3D72D4A.ref@yahoo.com> <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> Message-ID: On Tuesday, November 3rd, 2026 at 16:12, ROBERT MYERS via TUHS wrote: > > I have a 11/83 with floating point and a 11/23+ that I haven’t determined if it does. Both have 22 bits of address space. Both will have either a QBone or QBone Duo module. > > Any issues with running Various Unixes on them? I know the 11/83 has 4Mb of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. > > Sent from my iPhone > There is a paper describing UNIX on the 11/23 here: https://www.tuhs.org/Archive/Documentation/TechReports/USG_Library/1588_UNIX_on_the_PDP11_23_and_PDP11_34_Computers.pdf That may be relevant to what you're trying to do. - Matt G. From tuhs at tuhs.org Wed Nov 4 10:54:57 2026 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Tue, 3 Nov 2026 16:54:57 -0800 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> References: <759F09CA-499A-49E1-8078-8E11B3D72D4A.ref@yahoo.com> <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> Message-ID: <20261104005457.GJ24641@mcvoy.com> On Tue, Nov 03, 2026 at 06:12:34PM -0600, ROBERT MYERS via TUHS wrote: > > I have a 11/83 with floating point and a 11/23+ that I haven???t determined if it does. Both have 22 bits of address space. Both will have either a QBone or QBone Duo module. > > Any issues with running Various Unixes on them? I know the 11/83 has 4Mb of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. Other than "just because it is cool", wouldn't running whatever on a simulator work just as well and run faster? Don't get me wrong, I learned assembler on a PDP-ll, Kendall Witte is who I'm pretty sure was my TA, the instruction set was so well thought out, he could come over, crack open a beer, look at the octal dump of what I was running and read it like it was the assembler. Made a big impression on me, that's a very clean instruction set. This crowd probably all knows, but the PDP-11 instruction set sort of did everything in a way that you could read it in an octal dump, there were patterns and the patterns didn't vary. I also loved the PDP-11 because it gave me a mental model of what a computer is that I use to this day. Yeah, computers today are way more complex, but the basic model still holds true. Can any of you imagine showing someone x86 and saying "see, that's a computer" and expect anyone to have a mental model they can reason about? I liked the National 320xx series as well, it seemed like it was trying to be a modern PDP-11 but they never worked the bugs out and it went nowhere. The 680xx seemed like it was actually a modern PDP-11 but I didn't get the sense the instruction set was as clean as the PDP-11, Clem or anyone care to comment? From tuhs at tuhs.org Wed Nov 4 11:15:24 2026 From: tuhs at tuhs.org (ROBERT MYERS via TUHS) Date: Tue, 3 Nov 2026 19:15:24 -0600 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <20261104005457.GJ24641@mcvoy.com> References: <20261104005457.GJ24641@mcvoy.com> Message-ID: My problem with this is that the build scripts for windows and visual studio haven’t been brought up to date for Windows 11 and Visual Studio 2026. I also have to come up to speed with GitHub, cmake, and any other requirements to build under this environment. Since I have the 11/83 and 11/23+ machines, there is I believe Unix v6 targeting a 11/73 emulation (going from memory; had to remove a hard drive & controller as my 11/83 is in a questionable state right now). Sent from my iPhone On Nov 3, 2026, at 6:55 PM, Larry McVoy wrote: On Tue, Nov 03, 2026 at 06:12:34PM -0600, ROBERT MYERS via TUHS wrote: > > I have a 11/83 with floating point and a 11/23+ that I haven???t determined if it does. Both have 22 bits of address space. Both will have either a QBone or QBone Duo module. > > Any issues with running Various Unixes on them? I know the 11/83 has 4Mb of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. Other than "just because it is cool", wouldn't running whatever on a simulator work just as well and run faster? Don't get me wrong, I learned assembler on a PDP-ll, Kendall Witte is who I'm pretty sure was my TA, the instruction set was so well thought out, he could come over, crack open a beer, look at the octal dump of what I was running and read it like it was the assembler. Made a big impression on me, that's a very clean instruction set. This crowd probably all knows, but the PDP-11 instruction set sort of did everything in a way that you could read it in an octal dump, there were patterns and the patterns didn't vary. I also loved the PDP-11 because it gave me a mental model of what a computer is that I use to this day. Yeah, computers today are way more complex, but the basic model still holds true. Can any of you imagine showing someone x86 and saying "see, that's a computer" and expect anyone to have a mental model they can reason about? I liked the National 320xx series as well, it seemed like it was trying to be a modern PDP-11 but they never worked the bugs out and it went nowhere. The 680xx seemed like it was actually a modern PDP-11 but I didn't get the sense the instruction set was as clean as the PDP-11, Clem or anyone care to comment? From tuhs at tuhs.org Wed Nov 4 12:23:40 2026 From: tuhs at tuhs.org (Larry McVoy via TUHS) Date: Tue, 3 Nov 2026 18:23:40 -0800 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: References: <20261104005457.GJ24641@mcvoy.com> Message-ID: <20261104022340.GL24641@mcvoy.com> Well then go for it. I wouldn't mind being in front of an actual PDP-11 all these years later. I'm 64, my PDP-11 days are 40 years ago but I remember those days very fondly. It was a great machine, big enough and small enough. I fish on the ocean on a boat that is described the same way, I have a big small boat or small big boat. It's good enough in much the same way the PDP-11 was good enough, you could run a multi user Unix on a 64KB (and banks and I think you got to 256KB?), I fish on the ocean in my boat, got a 157 pound bluefin this year (been chasing that for 3 years, finally got one). It's a small but capable boat, pretty much like the PDP-11. The fact that Unix worked, multi user, on such a small machine is amazing. Huge credit to Dennis and Ken and team making things small, that is a lost art, my browser, one tab, has 37GB virtual, 5GB real. That's one tab. Don't get me started about slack, it's just as bad. We are a long way from PDP-11 unix, you want to go run that, go you. On Tue, Nov 03, 2026 at 07:15:24PM -0600, ROBERT MYERS wrote: > My problem with this is that the build scripts for windows and visual studio haven???t been brought up to date for Windows 11 and Visual Studio 2026. I also have to come up to speed with GitHub, cmake, and any other requirements to build under this environment. > > Since I have the 11/83 and 11/23+ machines, there is I believe Unix v6 targeting a 11/73 emulation (going from memory; had to remove a hard drive & controller as my 11/83 is in a questionable state right now). > > > Sent from my iPhone > > On Nov 3, 2026, at 6:55???PM, Larry McVoy wrote: > > ???On Tue, Nov 03, 2026 at 06:12:34PM -0600, ROBERT MYERS via TUHS wrote: > > > > I have a 11/83 with floating point and a 11/23+ that I haven???t determined if it does. Both have 22 bits of address space. Both will have either a QBone or QBone Duo module. > > > > Any issues with running Various Unixes on them? I know the 11/83 has 4Mb of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. > > Other than "just because it is cool", wouldn't running whatever on a > simulator work just as well and run faster? > > Don't get me wrong, I learned assembler on a PDP-ll, Kendall Witte is who > I'm pretty sure was my TA, the instruction set was so well thought out, > he could come over, crack open a beer, look at the octal dump of what I > was running and read it like it was the assembler. Made a big impression > on me, that's a very clean instruction set. This crowd probably all > knows, but the PDP-11 instruction set sort of did everything in a way > that you could read it in an octal dump, there were patterns and the > patterns didn't vary. > > I also loved the PDP-11 because it gave me a mental model of what a > computer is that I use to this day. Yeah, computers today are way more > complex, but the basic model still holds true. Can any of you imagine > showing someone x86 and saying "see, that's a computer" and expect anyone > to have a mental model they can reason about? > > I liked the National 320xx series as well, it seemed like it was trying to > be a modern PDP-11 but they never worked the bugs out and it went nowhere. > The 680xx seemed like it was actually a modern PDP-11 but I didn't get > the sense the instruction set was as clean as the PDP-11, Clem or anyone > care to comment? -- --- Larry McVoy Retired to fishing http://www.mcvoy.com/lm/boat From tuhs at tuhs.org Wed Nov 4 13:06:56 2026 From: tuhs at tuhs.org (ROBERT MYERS via TUHS) Date: Tue, 3 Nov 2026 21:06:56 -0600 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <20261104022340.GL24641@mcvoy.com> References: <20261104005457.GJ24641@mcvoy.com> <20261104022340.GL24641@mcvoy.com> Message-ID: <17BEAF0F-5BDF-402F-905B-4FD42DCE160B@yahoo.com> I’m actually 69 and I went to C_MU with Clem a number of years back. I’m trying to understand more about emulation of hardware and software but have run into a few stumbling blocks with Open-Simh (PDP-11 portion). I’m hoping to get the PDP-11 code running by the end of the year, since I fear maintenance of the repository has not been kept up to date with current OS and development software distributions. From tuhs at tuhs.org Wed Nov 4 13:07:55 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Tue, 3 Nov 2026 22:07:55 -0500 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> References: <759F09CA-499A-49E1-8078-8E11B3D72D4A.ref@yahoo.com> <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> Message-ID: 2.11BSD will boot fine. https://minnie.tuhs.org/PUPS/Setup/2.11bsd_setup.html has an older version of Steven's steup guide. https://www.tuhs.org/cgi-bin/utree.pl?file=2.11BSD/doc/2.10/setup.2.11 has the troff for a version from a few years later; but I'm pretty sure you an find a pdf of it also. On Tue, Nov 3, 2026 at 7:12 PM ROBERT MYERS via TUHS wrote: > > I have a 11/83 with floating point and a 11/23+ that I haven’t determined > if it does. Both have 22 bits of address space. Both will have either a > QBone or QBone Duo module. > > Any issues with running Various Unixes on them? I know the 11/83 has 4Mb > of PMI memory and the 11/23+ will have emulated memory of up to 4Mb. > > Sent from my iPhone > From tuhs at tuhs.org Wed Nov 4 13:23:27 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Tue, 3 Nov 2026 22:23:27 -0500 Subject: [TUHS] Are various DEC UNIXes targeted to address space and/or Floating Point implemented? In-Reply-To: <20261104005457.GJ24641@mcvoy.com> References: <759F09CA-499A-49E1-8078-8E11B3D72D4A.ref@yahoo.com> <759F09CA-499A-49E1-8078-8E11B3D72D4A@yahoo.com> <20261104005457.GJ24641@mcvoy.com> Message-ID: On Tue, Nov 3, 2026 at 7:55 PM Larry McVoy via TUHS wrote > > I liked the National 320xx series as well, it seemed like it was trying to > be a modern PDP-11 but they never worked the bugs out and it went nowhere. > Actually, the model was the VAX; but you are right. I never understood why they never quite get it together, and they lost their window to the 68020/68030 > The 680xx seemed like it was actually a modern PDP-11 but I didn't get > the sense the instruction set was as clean as the PDP-11, Clem or anyone > care to comment? > I discussed this a few days ago in the [TUHS] C from 1980 binaries thread, replying to Marc's comment about the 68000's bus being close to the Unibus. Given the goals, I think it was very clean. It used all the same addressing modes. Even the instruction-word bit layout is extremely recognizable from the PDP-11. The real difference was that it had address and data registers instead of general registers, which I suspect is the source of your gripe. But blame DEC for suing Ken O'Mohundro and Fred Cox @ CalData for making a true PDP-11 clone for the choice Nick and Les made with the 68000 ISA. copy/pasted from that message: On Mon, Nov 2, 2026 at 1:49 AM Marc Donner wrote: > That educated me about the bus, so years later I recognized that the > Motorola 68000 had essentially implemented it faithfully, with the > exception of making the Sync signal active-low instead of active-high (a > true blessing in TTL days). > Yup, it is very close and was intended to be, as it was a model, but the designers were careful not to make it exactly like it. Remember, what became the 68000 was a skunkworks project and >>not<< an 'official' project at Moto. Les Crudele, Nick Tradenick, and a couple of others had joined Moto from Schlumberger, where they had been interfacing PDP-11 control systems for the oil drilling industry and were extremely familiar with both the PDP-11 ISA and the Unibus. But Les says he and the team were very careful. The bet they had with Tom Grunner was that they could basically build a PDP-11 on a single chip using Motorola's current process technology. Tom was interested enough to let them try in the back of the lab. But those folks (particularly Tom, as I understand it) were all extremely aware of the CalData lawsuit. Thus, they made sure the ISA and bus were 'different enough'. For instance, Nick made the instruction set use the same addressing modes and functions, but isn't the same, using address and data registers instead of general registers. I've long forgotten the specifics of the differences, but similarly Les said he spent a lot of time thinking about what we could use from the Unibus and QBUS, but still be different enough that they would not be taken to court as CalData had. The fact that, a few years later, Ken O'Mohundro put a 68010 on a QBUS fairly easily is living proof it wasn't that far off. The rest of the story of how 68000 went from being a skunkworks project to a real product is cool, and I am honored to have helped play a small part in that story with the TekLabs's Magnolia system; but it's not relevant here. From tuhs at tuhs.org Thu Nov 5 04:49:01 2026 From: tuhs at tuhs.org (Arnold Robbins via TUHS) Date: Wed, 04 Nov 2026 11:49:01 -0700 Subject: [TUHS] What does `ld` stand for? Message-ID: <202611041849.6A4In13I045763@freefriends.org> Hi All. Douglas McIlroy via TUHS wrote: > Fortran came with source code. I stayed up with that one night > debugging a bizarre error: failure on computed gotos with 74 mod 75 > branches. I wouldn't be surprised if our unlucky customer was the only > user ever to stumble on that one. I asked for more stories like this. Here's Doug's reply (by permission). Followups should probably go to COFF. > There's actually more to the Fortran-bug story. The bug involved a > programming trick as bizarre as the symptom. > > The instruction set for the IBM 700 series had an index-register "tag" > field.. A loop instruction (TIX) counted an index down to zero, with a > conditional transfer on non-zero. A store-tag instruction was > introduced in the 709. I never saw an STT in use--except for > this bug. > > A loop that counted blocks of 75 computed-goto > addresses for spilling to secondary storage was controlled by > TIX, but in an extraordinary way. The fall-through was controlled > by zeroing out the tag rather than by reaching zero. Since the > condition was sensed elsewhere, it had to be detected on > every path that led to the TIX, and the programmer had missed > one. > > What, I wonder, inspired the creation of this baroque scheme? From tuhs at tuhs.org Thu Nov 5 06:20:23 2026 From: tuhs at tuhs.org (Lyndon Nerenberg (VE7TFX/VE6BBM) via TUHS) Date: Wed, 04 Nov 2026 13:20:23 -0700 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <7.1.0.9.0.20261030133125.085d7e50@threedee.com> <56b6be1a-8253-4b30-b5b6-36177f45edfa@xavax.com> <20261026233611.1B47D1253528F@ary.qy> <20261030194928.C61F7125F8670@ary.qy> <7.1.0.9.0.20261031223227.0989ee48@threedee.com> <20261101145517.GK15589@mcvoy.com> Message-ID: <9f28c38cf2d4fa5c@orthanc.ca> > The fact that, a few years later, Ken O'Mohundro put a 68010 > on a QBUS fairly easily is living proof it wasn't that far off. Was this the precursor to the NBI 68010 QBUS Unix system (I forget the marketing model name or number). I owned on of those at home back in the day[1], complete with 10 Mb/s ethernet board and a DH-11. It ran an off-the-tape compilation of 4.2 BSD, and it was absolutely covinced it was a VAX :-) I was running a small usenet hub on it, and when I connected the Telebit, the machine started to panic in the DH11 driver under 9600+ baud serial load when there was also any ethernet traffic present. I called a friend and said "please send me dh11.c and .h from the 4.2 release tape." Compiling them required changing a single #define (compile for QBUS, not UNIBUS). I re-linked the kernel, rebooted, and the problem vanished! Oh how I miss that machine. It was like I had a VAX in the 2nd bedroom, but with a proper CPU :-) --lyndon [1] I first ran into one circa 1987 at a lawyer's office. I acquired mine in 1991. From tuhs at tuhs.org Thu Nov 5 12:22:31 2026 From: tuhs at tuhs.org (segaloco via TUHS) Date: Thu, 05 Nov 2026 02:22:31 +0000 Subject: [TUHS] C from 1980 binaries In-Reply-To: <9f28c38cf2d4fa5c@orthanc.ca> References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <20261101145517.GK15589@mcvoy.com> <9f28c38cf2d4fa5c@orthanc.ca> Message-ID: On Wednesday, November 4th, 2026 at 12:20, Lyndon Nerenberg (VE7TFX/VE6BBM) via TUHS wrote: > > The fact that, a few years later, Ken O'Mohundro put a 68010 > > on a QBUS fairly easily is living proof it wasn't that far off. > > Was this the precursor to the NBI 68010 QBUS Unix system (I forget > the marketing model name or number). > > I owned on of those at home back in the day[1], complete with > 10 Mb/s ethernet board and a DH-11. It ran an off-the-tape > compilation of 4.2 BSD, and it was absolutely covinced it was > a VAX :-) > > I was running a small usenet hub on it, and when I connected the > Telebit, the machine started to panic in the DH11 driver under 9600+ > baud serial load when there was also any ethernet traffic present. > I called a friend and said "please send me dh11.c and .h from the > 4.2 release tape." Compiling them required changing a single #define > (compile for QBUS, not UNIBUS). I re-linked the kernel, rebooted, > and the problem vanished! > > Oh how I miss that machine. It was like I had a VAX in the 2nd > bedroom, but with a proper CPU :-) > > --lyndon > > [1] I first ran into one circa 1987 at a lawyer's office. I > acquired mine in 1991. > All this UNIBUS/QBUS talk has me wondering, what with the clear influence of the PDP-11 on the M68000, was there also a line connecting VERSAbus and later VMEbus back to the DEC tradition of system buses? For the record, some of the early non-DEC System V targets are M68K systems based on VERSA and VME bus systems. VME then goes onto feature in numerous other UNIX targets, among them the WE321SB single board and Sun machines. - Matt G. From tuhs at tuhs.org Thu Nov 5 12:49:50 2026 From: tuhs at tuhs.org (Clem Cole via TUHS) Date: Wed, 4 Nov 2026 21:49:50 -0500 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <20261101145517.GK15589@mcvoy.com> <9f28c38cf2d4fa5c@orthanc.ca> Message-ID: On Wed, Nov 4, 2026 at 9:22 PM segaloco via TUHS wrote: > > All this UNIBUS/QBUS talk has me wondering, what with the clear > influence of the PDP-11 on the M68000, was there also a line connecting > VERSAbus and later VMEbus back to the DEC tradition of system buses? > As best I can tell, only indirectly. VERSAbus and VMEbus competed with Intel's Multibus and Multibus-II. When Intel originally created Multibus, they were trying to address the issues with the S-100 bus. DEC's PDP-8 Ominibus and Unibus had been around for a long time by then. I don't think the LSI-11/QBUS was yet official. Multibus certainly looked more like the DEC bus offerings than the S-100 bus, and like DEC when they started, Intel tried to lock them up the same way DEC had done. By Multibus-II, they certainly succeeded. But Multibus, like the S-100 bus and the later PC buses, ended up being 'open.' For example, the Stanford University Network 68000 board was Multibus-based, and I don't believe any of their licensees such as VLSI Systems (ney Sun Micro) needed a license of any type from Intel. The Apollo and Masscomp systems also used Multibus as their I/O bus. But I know that in the latter case, Masscomp made some extensions on the undefined connector part of the board edge. All of their CPUs worked with COTS Multibus I/O boards, but when Dave Cain created the memory system for the MC-500, he was fresh off leading both the Comet (750) and the BI projects at DEC, and Masscomp's memory bus looked more like the BI than anything else. FWIW: IIRC, Moto tried to lock up the VMEbus a bit, and I don't know how well that worked out. ISTR that Sun switched to it at some point, but somewhat abandoned it in their own design by the time of the SPARC, although the larger systems that need more demanding controllers like the ones Xylogic's built, I think stayed using VME as the I/O bus for a good bit. Someone like Larry is likely to know much more about that. Clem From tuhs at tuhs.org Fri Nov 6 01:04:45 2026 From: tuhs at tuhs.org (al kossow via TUHS) Date: Thu, 5 Nov 2026 07:04:45 -0800 Subject: [TUHS] C from 1980 binaries In-Reply-To: References: <188ef5e9-d1c4-4acd-a6f4-e97a018abf24@xavax.com> <20261101145517.GK15589@mcvoy.com> <9f28c38cf2d4fa5c@orthanc.ca> Message-ID: <461c1fea-5e03-496b-8b2f-66376b31cb46@bitsavers.org> No, that was Integrated Solutions, with people who came from Zilog (the reason all their peripheral controllers used Z8000s) http://bitsavers.org/pdf/isi Cadmus also sold a Qbus 68K system On 11/4/26 6:22 PM, segaloco via TUHS wrote: >>> The fact that, a few years later, Ken O'Mohundro put a 68010 >>> on a QBUS fairly easily is living proof it wasn't that far off. >> Was this the precursor to the NBI 68010 QBUS Unix system (I forget >> the marketing model name or number). >>