Re: [hatari-devel] Emulation of the TT second MFP

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi Uwe,

On 5/6/19 9:23 PM, Nicolas Pomarède wrote:
> Le 06/05/2019 à 19:14, Uwe Seimet a écrit :
> > By the way, before the crash Hatari reports this:
> >
> > WARN : Write to unimplemented RTC/NVRAM interrupt enable bits 0x40
>
> This might give a hint on possible cause, because RTC IRQ signal is 
> indeed connected to the 2nd TT MFP and for now this is not "connected" 
> into Hatari, only ncr5380's IRQ is.
>
> Maybe let's wait a little that I connect other IRQ signals to the TT 
> MFP, then you can try again to see if this fixes the issue.

Linux assumes that if setting SCU interrupt mask doesn't generate
a fault, it and VME are both present.  To get TT booting (using IDE)
with Linux, I had to change its access to cause a fault.

Have you checked whether ASV tries to set SCU system or VME interrupt
mask registers 0xff8e01 & 0xff8e0d?


	- Eero