PDP-8 Unix was Re: Who wants a nice monitor (UK)?

On Tue, 7 Feb 2012, allison wrote: While not exactly C, I've done a bit of fiddling with a byte code derived from the venerable Small-C compiler. The compiler figures things out in terms of a simple single-stack machine. There is one module in the compiler that takes the single-stack primitives and outputs assembly code for the target processor. It's possible to replace that module with one that outputs the primitives, then make an back end to turn them into code for something else. By that means, I made a variant of Small C that could compile for a variety processors ranging from from Z80 to VAX. I reworked them a bit to come up with something that could describe a simple FORTH. *That* I made go on stuff ranging from PDP-8 to VAX. Since the low-level words were written in the primitives derived from Small C, the same source could be compiled for any of the processors I ran it on. Essentially, a byte code. Of course, the PDP-8 version used 12-bit bytes. In the PDP-8 variant, each bytecode consisted of a JMS I Z instruction. Page zero was loaded up with JMP instructions that transferred to the routines that implement each primitive. For what it's worth (which is not much), it can be found here: http://anachronda.homeunix.com:8000/~rivie/third/ I've only put up the PDP-8 and Z80 versions, but I've also done a handful of other processors; enough to wring out byte-order and word-size dependencies. The PDP-8 version was done as a degenerate case involving one-byte cells. I used long symbol names in the PDP-8 source, so you'll need a variant of Doug Jone's PAL assembler that I've tweaked to take them. -- roger ivie rivie at ridgenet.net