Cliff Harris

I've been looking at the BUA9340 code and found this bug:

LD295: LDX L3FEC ; B Counter last ref

LDAB L0033 ; Minor loop mode byte 1
ANDB #$F7 ; CLEAR b3 (1st DRP BIT (VALID PERIOD))
STAB L0033 ; Minor loop mode byte 1

INCB ; ???
INCB ; ???
STX L3FE4 ; B Counter, Begin next Dwell <-- why do this if there was no change to X???

INCB ; ???
LD2A4: MUL ; multiply A x B = D <-- why do this if the result is trashed by the next instruction???

LDD L3FFC ; MCU CSR (Control & Status Register)

Now I'm looking for the ARAP_hac to compare to this code to see if it has been fixed or changed. I can't find a copy anywhere. All the links I could find were dead. Can anyone help me with this?

Benny42

Try here ftp://diy-efi.org/pub/gmecm/bins/1227165/ . i hope that worked

tequilaboy

The lines in question are commented as shown (in my copy of $32B source code, not a source hac):

INCB * MORE ECU DELAY*

INCB * MORE ECU DELAY*
STX RISREF UPDATE RISREF WITH REF. TIME

INCB DELAY FOR ECU
ESTOUT70 MUL DELAY FOR ECU

This portion of code is consistent between $32 and $32B. Appears to be intentional.

Cliff Harris

I'm having a problem with the part where it increments the B register multiple times, then does the multiply but doesn't do anything with the result. Then it overwrites the A & B registers by LDD so the result of the multiply is lost.

The other problem I'm having with this code is that it loads the B register with the contents of L0033, which is a status byte, meaning that each bit has a meaning. Incrementing that doesn't make a lot of sense unless the author is doing something really tricky. Then the multiply also doesn't make any sense, as it's L0033, an incremented status byte (B register) multiplied by another version of L0033 (loaded into the A register earlier).

I'm still looking for the ARAP_hac. In other words, the commented disassembly. I'll disassemble the ARAP.bin and look for a pattern that's similar to the originally posted code.

Cliff Harris

So how would I get a copy of this source code?

I see that the labels in the quoted code are meaningful, rather than the Lnnnn you see in the disassemblies.

I have found a lot of bad guessing in the bua_hac.pdf file. Many inconsistencies and just plain incorrect info. It's been very tedious and painful to go through this trying to figure out what's going on. I've spent months on this. My hat's off to the MAJOR effort it took to get it to where it is, however flawed. I had a big advantage when I started this project because I'm an electronic engineer with a LOT of assembly language experience. I even wrote 68HC11 assembly code when I worked for E****n Electronics.

Cliff Harris

I have uploaded my BUA disassembly files with my comments to my 1227165 page. They are here:

http://home.earthlink.net/~cliff_har...5/BUA/BUA.html

Still seeking answers and information on this code.