CT60 patch MonTT v.3.10

[hakim]

Hi there,

If you have CT60 for Falcon, then you've probably noticed that basically no 030 debugger/monitor software doesn't work. Sure there is this internal debugging info about d-, a- and other registers, but this just isn't enough...

Tried everything from Devpac to Turbo Assembler, Pure Debugger etc... The result was always the same, it didn't work on CT60... Well, actually one did work, MonST 2, but the problem with this debugger was, that it only supported 68000, 010 and 020 instruction sets. The other downside was, that it didn't support TT-RAM...

Also tried looking from the forums, what should be done to make them work. Only found some messages, that somebody has managed to fix Pure Debugger to run on Milan 060 or such, but unfortunately no info on how that was made possible...

Based on the following document (http://e-www.motorola.com/files/32bit/d ... 8060UM.pdf) found out, that the 68060 processor doesn't support certain registers or instructions. In MonTT v.3.10 case it was:

Unsupported registers: psr, mmusr, isp, msp, caar

Unsupported instructions: ptest, pmove

With this patch file these unsupported registers and instructions are removed (hopefully) and you will be able to use MonTT v.3.10 even on your 060 machine. Haven't tested though, if 030 version still works...

To run this patch you should place the MON_P.PRG in the same directory with 030 version of Mon v.3.10. Run the program, and the MON.PRG will be the 060 patched version and MON.BAK will be the original 030 version...

Word of warning... Although a back-up file is being created before the file is patched, it doesn't hurt to make a back-up before running this patch, just to be sure...

Another word of warning... This is only a patch for 030 version of MonTT v.3.10 (and that version only!), which means that it doesn't support 060 registers or instructions, but in most cases it can run them if it can figure out what is the length of the instructions in words.

Do NOT expect a fully operational 060 debugger that works like 030 debugger. In fact I've found out, that in some cases the program doesn't stop (!) on breakpoints and single stepping the trap can produce some unexpected results. Single stepping also causes some illegal instructions even if there are no obvious reasons for those. This is due to incomplete support of 060 processor. For a better support of 060 processor, illegal and trace vector handlers, among others, should be reprogrammed...

Do expect though to be able to search'n'browse opcodes/hex (up till 040) and modify hex/asc and locate those unsupported registers and instructions, especially in the applications. With games and demos the chances of locating unsupported registers and instructions while running the program are from slim to none...

Anyway here is the patch file, version 0.12, that makes MonTT v.3.10 to run on CT60 Falcon accelerator, maybe on Hades and Milan 060 as well? This haven't been tested though...

Regards,
hakim

[[ProToS]]

hi,

I'm not a coder so I can't test this patch
but I'm a CT060 owner and I say many thanks for your work

[[MiKRO]]

yep, nice work pal! only thing I don't understand is why you atached Easyrider's reassemble instead of original one ;-) Btw tested under magic, worked well, too. Thanks for this patch!!

[Zorro 2]

Hi mister Hakim,

I like your job !

I prevented sceners for your patch.

I'm not a Falcon user and c'est bien dommage !

[Coda]

Hi.

Based on the MON_P patch for the 030's MON.PRG, I have created the equivalent patch for the resident version - AMON030.PRG. I needed something memory resident and I havent tested it much, but it seems to work OK. If anyone wants the binary let me know.

regards,
Coda.

[[ProToS]]

you can attach it to the forum if you whant

[Coda]

Here is the Binary AMON030.PRG.

[[ProToS]]

thanks Coda