Re: [hatari-devel] No address error when accessing odd addresses?

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi,

On keskiviikko 11 syyskuu 2013, Thomas Huth wrote:
> > Yes, I'm using the WinUAE core.

For now, I think we can consider it completely broken for ST emulation,
it cannot even boot some (most?) TOS versions.  And even for Falcon,
there are still couple of programs than work better with oldUAE CPU core
than with WinUAE CPU core.

Any ST/STe testing should be done with oldUAE CPU core.


>> Well, somebody has to use it in order
>> find what's probably still has to be fixed.

IMHO that makes sense only after WinUAE CPU core passes testing done
by tests/tosboot/tos_tester.py i.e. basic TOS functionality is working
under all TOS versions with it.


> > > But first ... what did you exactly change in the GUI? The "Slower
> > > but more compatible CPU" option is named there "Prefetch mode,
> > > slower" instead, if I remember correctly.
> > 
> > First I enabled "More compatible, slower", but there was still not
> > (expected) address error. Next I enabled "Cycle exact, slower", and
> > then the IDE emulation appeared to cause problems.
> 
> "More compatible, slower" is an FPU option, that's why it did not work.
> Not sure what could be wrong with the IDE emulation, though.

Only TOS version that works on ST and supports IDE is EmuTOS...


> > > Also, do you see a line starting with "Building CPU" on the console
> > > when you change that option and return the emulation? If yes, what
> > > does it exactly print?
> > 
> > Building CPU, 45989 opcodes (3 -1 1)
> > CPU=68030, FPU=68882, MMU=0, JIT=0.
> > CPU cycleunit: 1 (0.500)
> 
> Well, you did not change to 68000 mode here. In 68030 mode, you can
> access uneven addresses for sure.


	- Eero