Page 2 of 2

Re: Bus error handling problem

Posted: 19 May 2022 17:40
by exxos
sporniket wrote: 19 May 2022 17:29 It looks like accessing a1+2 does trigger the error handler then ?
Would seem so.

Re: Bus error handling problem

Posted: 19 May 2022 19:27
by mrbombermillzy
exxos wrote: 19 May 2022 16:33 There is still something odd going on :roll:
I can peek a byte or word in STOS and I get a number back, but a LONG I get BERR.
From that wording, I would hazard a guess that by using a long, your assembly variable space is stepping into an odd boundary (if compiling on a 68000, which I guess you are?) causing an odd address/bus error. This would require the STOS variables to be words/evenly aligned to avoid this.

It's possible that STOS only uses 16 bit words for its data handling...I wouldn't know.

Re: Bus error handling problem

Posted: 19 May 2022 20:54
by exxos
@mrbombermillzy I was only peeking byte or word in STOS. The assembly code was checking for a long. Indeed I confirmed that long failed in STOS as well. Flashyclock registers are 16bit so it wouldn't respond to long access. So now the asm code checks for words, it's all fine now :)

Re: Bus error handling problem

Posted: 19 May 2022 20:58
by mrbombermillzy
exxos wrote: 19 May 2022 20:54 it's all fine now :)
:thumbup: Glad you got it sorted.