Bug#348082: xserver-xorg: [ati/radeon] pinkness and interference pattern on Radeon RV250 If [Radeon 9000] rev 1 and TMDS-connected Samsung SyncMaster 213T

[Alex Deucher]

On 8/30/07, Michel Dänzer <daenzer at debian.org> wrote:
> On Thu, 2007-08-30 at 05:17 -0400, Branden Robinson wrote:
> >
> > The bit that concerns me is common to both:
> >
> >   (WW) RADEON(0): Option "UseFBDev" is not used
> >
> > Don't I kinda have to have that, since I'm on a PowerPC?
>
> No, hasn't been necessary (or even very useful) in a while.
>
>
> > Hmm, and it says ""MonitorLayout" is not used" when I try to use *that*.
> >
> > Anyway, with "MonitorLayout" I get ugly corruption for a second or two after
> > attempting to start the X server, before the monitor loses track of the video
> > signal from the adapter altogether.
> >
> > Without MonitorLayout, I'm just blank before the monitor loses it.
>
> This can't be related to MonitorLayout as the driver no longer knows
> that option, so it has no effect.
>
>
> Looking at the log file, the driver seems to misdetect the DVI
> connection as being VGA:
>
> > (II) RADEON(0): EDID data from the display on connector: VGA
> > ----------------------
> > (II) RADEON(0): Manufacturer: SAM  Model: 91  Serial#: 1312961073
> > (II) RADEON(0): Year: 2005  Week: 14
> > (II) RADEON(0): EDID Version: 1.3
> > (II) RADEON(0): Digital Display Input
>
> Note the 'Digital Display Input' (and Branden confirmed on IRC it's a
> real DVI connection).
>
> Alex Deucher can probably provide the Option "ConnectorTable" magic to
> work around this and then fix it properly. :)
>

Branden,  remind me again what card this is and what connectors your
card has.  Actually, the full log would be useful.  It's it a mac,
we'll need to add a quirk for it.  If it's a powerbook, I may have
already fixed it.

Alex