Re: [hatari-devel] Systematic launch tests |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
On 26 Mar 2017 at 22:53, Eero Tamminen wrote:
> On 03/26/2017 10:10 PM, Roger Burrows wrote:
> > On 26 Mar 2017 at 21:35, Eero Tamminen wrote:
> >> EmuTOS should AFAIK support both IDE & ACSI on any
> >> machine. It does work fine with ACSI on ST/STE/TT,
> >> but even latest snapshot doesn't work with Falcon,
> >> either with ACSI or IDE.
> >>
> >> That could be EmuTOS issue.
> >>
> > I'm not sure what you mean by this. In Hatari Falcon mode,
> > ACSI isn't present, so EmuTOS can't use it.
>
> On quick look at the Hatari's ACSI/FDC DMA register handling,
> they seem to be identical between different machine configurations,
> there's nothing Falcon specific. Thomas?
>
> And according to Compedium, the (0xff8604-0xff860D) regs indeed
> are same. (Only difference seems to be for TT, which has lots
> of additional SCSI registers.)
>
Yes, but those registers are used to access SCSI, not ACSI. There is no ACSI
on a Falcon.
>
> > If you add an IDE disk image, EmuTOS will access it fine
> > (and has done since at least 0.9.0). Don't forget that
> > the image has to be byte-swapped.
>
> I created another disk image, and EmuTOS seems to access fine
> that exactly same image as ACSI under TT and as IDE under Falcon.
>
> And I think that has actually been the case for long time,
> so I've been assuming that EmuTOS notices whether the disk
> needs to be byte-swapped or not?
>
All I can tell you is that I had to byte swap an ACSI image to get it to work
with IDE on Hatari Falcon. I assumed that was an artifact of Hatari.
>
> This is the disk image in question, it's partitioned & formatted
> with AHDI, I think v5:
> https://hatari.tuxfamily.org/hd80mb.image.gz
>
When I have time, I'll look at it.
Roger