Re: [hatari-devel] Debugger: instruction at PC

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi,

On torstai 13 maaliskuu 2014, Miro Kropáček wrote:
> Having said all of that, there is still something fishy going on. Look at
> 
> this:
> > r
> 
> [...]
> 0000d89e: 6602 4e68 43fa 0040 3018 BNE.B #$00000002 == 0000d8a2 (TRUE)
> 
> This, according to your/my observation should say that a jump is going to
> happen, right? Cool. It's true. But then explain why do I see this:
> 
> 0000d89e: 6602 4e68 43fa 0040 3018 BNE.B #$00000002 == 0000d8a2 (TRUE)
> next PC: 0000d8a0
> 
> > d
> 
> $00d8b4 : 0c50 0000                            cmpi.w    #0,(a0)
> $00d8b8 : 6706                                 beq.s     $d8c0
> 
> Shouldn't be 'd' showing the jump instruction? Interestingly, it's not
> always reproducible. I'm attaching complete log from this session (the
> example above is near the end) to prove that I'm not making this up ;-)

You used "lock disasm" to view disassembly on (re-)entering the debugger.

Further disassembly commands continue disassembling from the last address,
if you don't specify a new address.  Use of "r" command between "d" commands
does NOT reset disassembly address back to PC.  Only "d pc" (or re-entering
the debugger) does that.


	- Eero