Re: [hatari-devel] MFP Interrupt Regression

[Nicolas Pomarède <npomarede@xxxxxxxxxxxx>]

On 19/06/2013 01:20, Nicolas Pomarède wrote:
> On 18/06/2013 20:00, David Savinkoff wrote:
> > Hi,
> >
> > Printing must be enabled.
> > Runs on real ST and all TOS I've tried.
>
> Hi
>
> I can see the same problem between 1.6.2 and 1.7, no output in 1.7.
>
> I will have a look to the code of LBOS, but having 2 tasks doing their
> own job is making the disasm trace less easier to read.
>
> At least, that's an interesting small "multi task" stack, even if
> limited :)

Hi, this is now fixed.

This was a really stupid bug / typo, printer uses MFP's interrupt 0, but 
the code was :

      NewInt = MFP_CheckPendingInterrupts ();
      if ( NewInt > 0 )

So, only interrupts 1 to 15 could create an IRQ :(

Thanks for pointing this one.

Nicolas