Peter Stark's Humbug monitor for the Motorola 68xx. Copied from the Kilobaud/Micromputing articles of July, August and September of 1980. Copies can be found in the docs directory.
And copied from a disassembly of the Tandy MC10 Humbug+ corrupted tape cassette image (hb7500).
And a copy of the 6809 Humbug source. This is a little harder to use as the 6800 has a subset of the instructions, a subset of the addressing, a subset of the Registers and many techniques can't be used with the 6800.
If anyone has a copy of the Humbug ROMs for the 6800, please share! We can disassemble them and finally get a working set for the 6800, the 6801, the 6809 & 68K family.
The usual, trying capture all the ROM bugs for the Motorola boards. This one is one of those and I beleive it to be a very popular ROM bug. Peter Stark wrote this for the Motorola 6800, 6802, 6803, 6809 and 68k. I've found the original articles, the cassette tapes for the Tandy MC10 (6801) the Tandy CoCo (6809) and the PT68K-4 (68K).
While the articles give plenty of detail they are not complete and seem to be made up of parts of different packages. Parts appear to be missing so it's very confusing. In addition to the articles I have a mostly working MC10 cassette version. This is really an odd image as it seems to overwrite itself. I think it is a damaged version. But between the two and the Tandy CoCo version (which I hope is intact) I hope to build a working Humbug ROM for the 6800, 6802 & 6801.
Right now this is just a mess of files. It now assembles but is a mess. I've had to resort to abusing all the memory I can use as I whittle away at the wayward RAM usage. I also need to keep an eye on any code that is unintentionally being self modifiying code. But progress is being made on both subjects.
April 24, 2024
I took a different route to solve the problems I've been running into. I took the version 2 MIKBUG and started adding the Humbug features. I've successfully got the Break and Single Step working. I call the monitor modbug. At the moment there is no logic to its layout. I'm cleaning it up, fixing a few bugs and I'll continue to add more features. When done I'll clean up 6800 Humbug and MC10 Humbug. The modbug code hasn't made it over to here yet. I just didn't want folks to think I've given up on the Humbug code. It's sprint, I like to ride my bike, my house needs my attention and I still like to code. So I'm a little busy (of course).
April 06, 2024
Well I'm totally confused. 2 steps forward and 20 steps back. No clude what I did, what I fixed and what I broke.
April 2, 2024
Well the jokes on me, SS kinda works. Still that's something. Now I need to figure out the kinda part and make it just work.
April 1, 2024
Yeah, April 1 but no joke. I think I have SS working now. Yea! I went through the articles, clean up a bunch of variables and snuck a peak at Smithbug (which is broken). Still lots more work to do but Humbug may now be usable. :-)
March 31, 2024
Found a few bugs I let creep in, Hopefully I've fixed those. I Still a lot of variable to correct. Also I've managed to get Humbug to run on the board from the ROM emulator. While BPs work the SS and ST and anything to do with the SWI handling and stack are messed up. This comes down to the variables.
March 30, 2024
I'm now at the point where I can't do much more testing in the simulator. So I broke out the memSIM2. The USB to ROM interfaces make it much quicker to update code. Now things get weird. I've tested Mikbug, Smithbug and Humbug. Mikbug works, very limitted but works. Now when I start working on the RT68MX code I think I'll need the breakpoints and a few other niceties. So I've tested with Smithbug and it mostly works. Breakpoint setting gos off into lala-land. Will look inot that. Now Humbug just doesn't want to work. It boots, spits out a hello message, and accepts the AD command (and address input). No other commands work. Odd part is that it works fine in the simulator. I must be messing up the X register as it traverses the table.
March 27, 2024
I think I finally fixed the RC (register change). Turns out I had problems with my abx macro. I've patched it up but I still need more testing on that macro. The reason I say I think is that I'm not yet ready to test with user code running in the monitor. Still have a few more things to work on like step (ST/SS) and go (GO).
March 22, 2024
I'm overly pleased with myself. I couldn't find a PUnch routine so I wrote my own. While it's only a few bytes long it's all mine. I did get bit by bsr/jsr to various routines. They clobbered the registers I was using. Still working on getting all the code working but I'm now working on the SS, JU and BR. I think BR is working but JU is wonky.
March 21, 2024
Fixed up AI (ASCII Input) and AO (ASCII Output)
March 20, 2024
Fixed up AD (ASCII Dump) so it display 16 octets at a time. HD only does 8 at a time. I've stolen LO (Load S19 file) from Mikbug. I've added PU (Punch S19) but this is a work in progress.Found that SAVEX is being used by something while testing PU. Switched to USAVEX. Should expect that as the 6800 doesn't have a pshx/pulx. Some more of the varaibles are getting comments. Hope to remove redundancies soon.
March 19, 2024
Code assembles and some functions work. The code is still questionable at this time.
Started March 9, 2024
Current status is that the code assembles but that doesn't mean it works. This code is a mash of the 6801/MC10 code which has a keyboard and video output. The 6800 needs to deal with a serial port (9600 8,N,1). Additionally the RAM usage is a mess. There's variables in $7300, variables in $7400. The variables in $7400 should be okay. But I stuffed the mystery variable from the MC10 (remember the code overwrote itself) at the end of $7300. I know these variables are either the starting $7300 or $7400 variables. Also the current hb7500.asm is writing to $4XXX which is OS RAM on the MC10. So the related code needs to be fixed.
Anything that has been checked (X) is running okay but not heavily tested. Everthing else either hasn't been tested or is broken. Assume broken.
- AD - ASCII Dump
- AI - ASCII Input
- AO - ASMTAPE1CII Output (Oh not Zero)
- AT - Analyze Tape (MC10)
- BA - Change Baud
- BP - Print Break points
- BR - set/reset Breakpoints
- CO - Continue (after a break)
- CS - Checksum
- DE - Desemble (not Disassemble, only bytes)
- EX - Exit to BASIC (MC10)
- FI - Find 1, 2 or 3 bytes
- FM - Fill Memory
- GO - Go to address pointed to by PC/A048
- HD - Hex Dump
- HE - Help
- JU - Jump (actually JSR)
- LO - LOAD S1
- MC - Memory Compare
- ME - Memory Examine/Edit
- MM - Memory Move
- MT - Memory Test
- PC - Printe contents of PC/A048 (used with GO command)
- PU - Punch S1
- RC - Register Change
- RE - Register Examine
- SA - CSAVEM to cassette (MC10)
- SS - Single Step (lots of limitations, careful)
- ST - Start SS (if no Breakpoint)
- !! - Monitor Reset (Cold start)
- UA - User A
- UB - User B
- FL - Boot Flex (not written)
- OS - Boot other OS (not written)
Not written and MC10 code are not include yet.
| Filename | description |
|---|---|
| MC6800.inc | Macros - do not use |
| Makefile | Makefile to build ROM |
| README.md | This file |
| asl.inc | asl.inc - macros, might not need |
| begin.py | Python code to filter a file until the 'begin string' |
| filter.py | Python code to filter a symbol table |
| hb7500.asm | WIP, 6800 version of Peter Stark's MC10 Humbug ROM monitor |
| hb7500.lst | ASL file listing |
| hb7500d.asm | Peter Stark's MC10 Humbug+ ROM monitor (bin mostly works) |
| hb7500d.lst | ASL file listing |
| hb7500d.s19 | ASL S19 dumo of MC10 Humbug+ ROM monitor |
| mikbug.asm | Motorla MIKBUG ROM monitor |
| minibug.asm | |
| minibug.lst | |
| minipro-help.txt | Minipro, for the TL866+ help file |
| minipro-list.txt | Minipro EPROM list |
| miniprohex-help.txt | miniprohex, for the TL866+ help file |
| motorola.inc | |
| s0.sh | Script to generate S0 header |
| smithbug.inc |
srecord ASL Macro assembler ( http://john.ccac.rwth-aachen.de:8000/as/ )
If you are going to assemble the code manually.
VAR='_E000'
asl -i . -D ${VAR} -L modbug.asm
mv modbug.lst mb${VAR}.lst
p2hex +5 -F Moto -r \$-\$ modbug.p mb${VAR}.s19
srec_cat mb${VAR}.s19 -o hmb${VAR}.s19
echo "hmb${VAR}.s19"
memsim2 hmb${VAR}.s19
VAR='_E000';asl -i . -D ${VAR} -L modbug.asm; mv modbug.lst mb${VAR}.lst;p2hex +5 -F Moto -r \$-\$ modbug.p mb${VAR}.s19
srec_cat mb${VAR}.s19 -o hmb${VAR}.s19;echo "hmb${VAR}.s19";memsim2 hmb${VAR}.s19
This is a Work In Progress, the base files (humbug-mc.asm & hb7500d.asm) are incomplete files. I am managing to put the files together into a workable file. At the moment the humbug-mc.asm still contains a few 6801 op code that I need to rework with 6800 code. Replacing the adx OP code has been interesting but I think I have the equivalent code from Peter's articles.
The MC6800.inc contains a few macros I'm experimenting with. Do be careful with them as I haven't documented the flags and the macros don't work with all addressing modes.
I will clean this up when done. I am trying to do that as I go along. I seem to be adding to the mess. ;-)
VAR='_E000';asl -i . -D ${VAR} -L hb7500.asm; mv hb7500.lst hb7500${VAR}.lst;p2hex +5 -F Moto -r \$-\$ hb7500.p hb7500${VAR}.s19
srec_cat hb7500${VAR}.s19 -o hb${VAR}.s19;echo "hb${VAR}.s19";memsim2 hb${VAR}.s19
VAR='_SIM';asl -i . -D ${VAR} -L hb7500.asm; mv hb7500.lst hb7500${VAR}.lst;p2hex +5 -F Moto -r \$-\$ hb7500.p hb7500${VAR}.s19
sim6800 hb7500_SIM.s19
Note: I'm currently using all of E000 thru FFFF. yes I'm being sloppy but my EPROM emulator works nicely with this so I'll continue this. My ACIA is at $E000 and the PIA at $E004(?) or $E008. Code actually starts at $E100 and unused sections are filled with $FF.
- Peter Stark (RIP), original author
- Peter Stark (RIP), original author