Re: [hatari-devel] Hatari debugger question

[ Thread Index | Date Index | More lists.tuxfamily.org/hatari-devel Archives ]


Hi,

On lauantai 10 elokuu 2013, Nicolas Pomarède wrote:
> there's no generic way to intercept all RAM read/write. This is a
> tradeoff for emulation speed, since RAM accesses usually won't trigger
> anything,adding a check on each acces would slow down things (while I/O
> accesses need to be passed to the corresponding HW handler).

That doesn't require adding a check, just setting the address of
the memory being acessed to some internal Hatari variable,
e.g. LastMemWrite.

Access to that Hatari variable can then be provided to the debugger.
What Douglas asked would then just require a breakpoint like this:
	b  LastMemWrite < 141

(Default value for that value could be 0xffffffff, and that would
be set each time emulation is continued from the debugger.)


Would setting a variable in RAM write be too much overhead in
the CPU emulation code?


> I'm planning to add this kind of condition one day for RAM, but not at
> the moment. In the meantime, I think you can add a breakpoint on "$140
> != 140" for example, but this will work only for one memory address, if
> you want to do a large area, you would need to specify them all one
> after the other (could be made with a small script to generate the list
> of corresponding breakpoints' intruction)

And it doesn't catch writes which don't change the memory content...


	- Eero



Mail converted by MHonArc 2.6.19+ http://listengine.tuxfamily.org/