| Re: [hatari-devel] Re: RAM detection issue with VDI (was HDDriver issues with WinUAE CPU core) |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/hatari-devel Archives
]
- To: hatari-devel@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [hatari-devel] Re: RAM detection issue with VDI (was HDDriver issues with WinUAE CPU core)
- From: Uwe Seimet <Uwe.Seimet@xxxxxxxxx>
- Date: Sun, 11 Jan 2015 19:40:48 +0100
- Dkim-signature: v=1; a=rsa-sha1; c=relaxed/relaxed; t=1421001648; l=1750; s=domk; d=seimet.de; h=In-Reply-To:Content-Transfer-Encoding:Content-Disposition: Content-Type:MIME-Version:References:Subject:To:From:Date; bh=uGp72h0bvy2BKXQjuQWLZfIebU0=; b=iijdYWQs7B/xNGMyeKZjPmBN+qNK97ZesviwuvtbuuGqFHR9OrZZL2GLZEtmdi2Fbv/ z9ZzMJMp7m0Lij5aJ1fBYP2a8yrooJTGEE4zdl7szX1k2jLHgZwSB62ZS6dATM+TXhJHR BR/gYuPMj3WXuIzqsjXLAB/vzRzLBPimbOA=
Hi,
The memory test appears to work fine now on my machine, but a resolution of
1280x1168 (TT mode, monochrome) pixels does not work anymore now. There is
pixel garbage now at the bottom of the screen during the memory test,
but the more serious problem is that after the memory test the system
crashes with two bombs. Before there was neither pixel garbage nor a
crash.
Yes, I know that 1280x1168 pixels is not a solution supported by a real
TT, but it was working fine until the latest changes from a few minutes ago.
Any resolution with a screen height of more than 960 pixels does not seem to
work anymore.
Take care
Uwe
> Hi,
>
> On sunnuntai 11 tammikuu 2015, Nicolas Pomarède wrote:
> > Le 11/01/2015 17:53, Roger Burrows a écrit :
> > >> This is pretty bogus for a TT or a Falcon ... and I believe this issue
> > >> was related to running TOS 3.06, so must have been TT emulation ...
> > >
> > > Just to make it 100% clear what I was referring to: only ST-compatible
> > > resolutions use 32K. TT high, medium, low & all Falcon resolutions use
> > > more memory. So always setting phystop to memtop+32K does not seem
> > > right ...
> >
> > ok, but in that case that's not what the actual code does : it sets
> > memtop to *at least* 32k. If screensize is higher than 32k in vdi, then
> > it's taken into account.
>
> Roger was referring to phystop. But old TOS versions didn't work
> if gap between memtop -> phystop is larger.
>
>
> > Still, this doesn't seem to be enough to fix it.
>
> Please check the updated fix:
> http://hg.tuxfamily.org/mercurialroot/hatari/hatari/rev/036a41cfbc56
>
> What value works, seems both TOS & machine specific
> (in case of EmuTOS).
>
>
> - Eero
>
>