Bug#785448: xserver-xorg-video-ati: Screen is badly tinged with green when using the open source driver

Owen Riddy
Email: owen.riddy@gmail.com
Mobile: 040 163 2663

--

On 20 May 2015 at 19:26, Owen Riddy <owen.riddy@gmail.com> wrote:

Thanks for that. So my problem has narrowed down a bit (it still happens but I can fix it by changing the input colour format using the buttons on the side of the screen). The screen switches to use YCbCr input format:

** When the computer is booted up
** When I log in using kdm
** When the screen switches off after a period of inactivity

I've used this screen for more than a year and it hasn't happened before. Is there any reason that the driver might set the input colour format to YCbCr, or is it possible for a Radeon card to set a hint for input colour format that could persist across a reboot?

I'll try with the 4.1 kernel and see if anything changes.

Owen Riddy
Email: owen.riddy@gmail.com
Mobile: 040 163 2663

--

On 19 May 2015 at 23:21, Alex Deucher <alexdeucher@gmail.com> wrote:
On Mon, May 18, 2015 at 4:32 AM, Owen Riddy <owen.riddy@gmail.com> wrote:
> No worries.
>
> It turns out that the problem was the "Color Format" of my screen. Dunno
> what that is, but it was configuring poorly and the open source driver seems
> happier when it is set to RGB rather than YCbCr. Maybe Catalyst is a bit
> more forgiving or maybe there is some configuration issue here I don't
> understand.
>
> My problem is solved; there may still be a minor bug here in how the
> auto-configuration of the driver works.

The open source driver does not support YCbCr at the moment.

Alex

>
> Owen Riddy
> Email: owen.riddy@gmail.com
> Mobile: 040 163 2663
>
> --
>
> On 18 May 2015 at 17:41, Michel Dänzer <michel@daenzer.net> wrote:
>>
>> On 18.05.2015 16:28, Owen Riddy wrote:
>> > I have not, but if I reboot the computer to an install of Jessie using
>> > fglrx the tinge is not present; and it appeared shortly after updating
>> > to Jessie + unstable. I'll try a few things with the cable & report back
>> > if any of them have an impact, but the fglrx test suggests this problem
>> > is 100% fixable in software.
>>
>> I guess I misunderstood the comments about fglrx in your original
>> report. I agree it's probably a software bug then, though it's more
>> likely in the kernel driver than in the Xorg driver.
>>
>>
>> --
>> Earthling Michel Dänzer | http://www.amd.com
>> Libre software enthusiast | Mesa and X developer
>
>
>

> _______________________________________________
> xorg-driver-ati mailing list
> xorg-driver-ati@lists.x.org
> http://lists.x.org/mailman/listinfo/xorg-driver-ati
>