Re: [hatari-devel] Suspected Hatari 2.3.0 regression (prefetch?)

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi,

On 12/17/20 12:44 AM, Eero Tamminen wrote:
> On 12/16/20 10:00 PM, Nicolas Pomarède wrote:
> > Please check that delmpaint is working better now
>
> Besides it now working again without MMU, this
> also seems to fix its ST/STE compat palette mode
> colors with EmuTOS, both without *and* with MMU.
>
> How latter can happen, wasn't the bug fixed only
> for code path that is used without MMU?

Any comments on this?

E.g. "Sidetracked" (GFA musicdisk) that earlier
didn't work with TOS4 *even with MMU*, starts up
fine now.


> Besides Delmpaint, this change fixes also
> Confusion demo starting to need MMU with
> Hatari v2.3.  I think there may actually be
> quite a few other programs impacted by this...

There are a many of programs that work again
without MMU (as they should, as they're not using
MMU instructions).

I've already tested and updated emutos.txt on
things requiring MMU with it (but not TOS4) + few
things that didn't work earlier even with MMU.

After more testing I'll finish updates also to
compatibility.html & release-notes.txt.


I've noticed also something working less
than earlier, but so far that has happened
only with EmuTOS (so it could potentially
be some change in EmuTOS side too):

* Escape's "_" demo will now exit in first
  screen with EmuTOS git version

* Confusion game demo 1 & 2 are now again
  stuck at startup


	- Eero