MP1022: I'm not confident that the ROM data is good, but I tried it in MAME anyway. The entrypoint is $80(hex data in patent at $3C0 is not ambigious), that is BRANCH 00. So it gets stuck in an infinite loop rightaway. I tried several things, like swapping the chapter bit, but can't get it working. Either TI tampered with the ROM data, or the bit/byte order is not what we expect.
That's really odd. I disassembled the hex file and it doesn't look right to me, so I reordered it assuming that it was in execution order, but the disassembly still doesn't look good. That infinite loop would be in the same place, anyway. I don't think it looks wonky enough that it is in the raw physical die layout, though.
I noticed the hysteresis source code uses the ALEC opcode, which was removed from the TMS1100. They might have a macro set up to translate that, though. It's also possible that they played with the instruction PLA, but I'd expect BRANCH 00 to stay the same.
Hmm, just found US4298946A with a different layout of the MP1022 ROM listing, and US4442972A with object code for TMS1670. Actually, it looks like there are many TI thermostat patents.... maybe one of them shows the OPLA.
