#garfield Logs

Dec 06 2023

07:05 AM rue_mohr: morning
09:37 AM Felix is now known as OK9UWU
09:40 AM OK9UWU: I've managed to cook up an extremelly crude parallel ROM reader with two atmega328p MCUs (i dont have any 595 shift reg in THC package) and read the EPROM from the Z80 based radiomodem, got some bits and strings out of it but seems like i am still reading some addresses multiple times :/ back to the drawing board
12:07 PM polprog: what do you mean multiple times?
12:07 PM polprog: if they are repeated at the same offset they might just have burnt the ROM like that to avoid messing with unused address lines
12:09 PM polprog: what is that little board on the right? OK9UWU
12:12 PM OK9UWU: You mean in my tweet? Thats arduino nano, acting as address incrementor, i dont have any devboard or MCU that i could program that has enough GPIO.
12:13 PM OK9UWU: I am quite sure its the reader being all messed up and botched together, every run i get new stuff from the ROM, i can't get these two MCUs to sync together, sometimes the address one is running "in front" of the data reading one and other way around.
12:19 PM polprog: ah so the uno is used as a serial converter?
12:21 PM OK9UWU: Yes
12:23 PM OK9UWU: I am wondering if it would be possible to dump the EPROM using an old NIC, i know you could write to EEPROMs with them, should be possible the other way around
12:50 PM polprog: hm
12:51 PM polprog: not quite like that but it should be
12:51 PM polprog: you would have to insert it after the PC booted so it doesnt try to execute it as option ROM
12:51 PM polprog: but the option ROMs could be mapped into IO space.. somehow
12:51 PM polprog: Im not sure if they are mapped after the OS boots, so you might need DOS for that
12:51 PM polprog: but DEBUG.EXE on DOS can dump arbitrary memory anyway
12:52 PM polprog: so if the boot rom is mapped into the memory space (correction, not IO space) then DEBUG can dump it
12:52 PM polprog: lookup option roms on osdev wiki
12:53 PM polprog: that is actually a neat hack, though a lot of work needs to be done to make it work..
01:05 PM OK9UWU: Hmm i see, i looked thru my stash of NICs but none of them had the right formfactor of the EEPROM
01:38 PM OK9UWU: Welp, ordered some STM32s, guess i'll have to wait :)
01:38 PM OK9UWU: Those should have more than plenty of GPIO for this
06:30 PM rue_mohr: :)