| Re: [hatari-devel] breakpoint on CALL_SUBRETURN condition doesn't work ? |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
Hi,
On maanantai 07 huhtikuu 2014, Nicolas Pomarède wrote:
> Le 06/04/2014 23:25, Eero Tamminen a écrit :
> > Does it work sometimes or never?
> >
> > (Sorry, had a busy week so I didn't have time to look at this yet.)
>
> From my tests, never :(
It seems to work fine for me:
---------------
> n subreturn
CPU=$e10804, VBL=99, FrameCycles=436, HBL=0, LineCycles=436, DSP=N/A
$e10804 : 4e75 rts
> trace cpu_disasm
> n subreturn
$e10804 : 4e75 rts
$e004ec : 2039 0000 045a move.l $45a,d0
$e004f2 : 6700 0032 beq $e00526
$e00526 : 2039 0000 045e move.l $45e,d0
$e0052c : 6700 0022 beq $e00550
$e00550 : 4eb9 00e0 8e7a jsr $e08e7a
$e08e7a : 3039 0000 043e move.w $43e,d0
$e08e80 : 665a bne.s $e08edc
CPU=$e08edc, VBL=99, FrameCycles=564, HBL=1, LineCycles=56, DSP=N/A
$e08edc : 4e75 rts
---------------
Can you provide an exact case that doesn't work?
It's easiest to do with CPU disassembly enabled.
>>> By the way, I think it could be useful to have a condition
>>> that breaks on any return (sub return or exception return).
>>
>> Additional command to the already existing ones?
>
> maybe another value for the parameter to the 'n' command.
Ok, I commited support for new "return" next option.
As there were no comments on my earlier question, I also
commited support for (by default) showing the next instruction
when (re)-entering the debugger.
- Eero