Re: [hatari-devel] Strange behaviour with ROL.B ?

[Laurent <laurent.sallafranque@xxxxxxx>]

Hi,


I will test on the real falcon this afternoon, but when tested under 
hatari only, the result seems inconsistant.

I can't explain it for now.

Regards

Laurent


Le 28/06/2020 à 10:15, Nicolas Pomarède a écrit :
> Le 28/06/2020 à 10:04, Laurent a écrit :
> > Hi everybody,
> >
> >
> > I encounter a situation I can't explain for now with ROL.B and the 
> > Carry flag in the 68030 state register.
> >
> >
> > My example program :
> >
> > Clear d7
> >
> > Set d0 to 9
> >
> > rol.b d0 2 times with values $10 and $40 and check the C flag of the 
> > state register after each rol.
> >
> > rol.b d0 by $10   (if carry bit is set, add 1 to d7)
> >
> > rol.b d0 by $40   (if carry bit is set, add 2 to d7)
> >
> > Then I display the value of d7 on screen (0, 1, 2 or 3) depending of 
> > the state of the Carry bit.
> >
> >
> > When I run the program, I can read on screen  "3"; which means that 
> > the carry bit was "1" for the 2 rol instructions.
> >
> > When I trace the program under hatari debugger (which is supposed to be 
>
> Hi
>
> do you mean you see a difference between a real falcon and Hatari in 
> 68030 mode ?