Mailinglist Archive: radeonhd (265 mails)

["Yang Zhao" <yang@xxxxxxxxxx>]

2009/1/5 Matthias Hopf <mhopf@xxxxxxx>:
> On Jan 05, 09 15:35:42 +0100, Matthias Hopf wrote:
> > > Currently, my approach is to verify that no rows of the input map has
> > > a value larger than 10 bits, otherwise assuming 16-bit input and
> > > truncating the LS 6 bits. This seems to work fine for xgamma, as the
>
> I don't see that in your patch... it's always assuming 16-bit AFAICS.
> Given that the previous rhdRRCrtcGammaSet() function assumed 8-bit data,
> I wonder whether that will have any unintentional side effects.

I ended up going through the code paths for all the various
gamma-related hooks in xorg-server, and noticed precision adjustment
is done automatically for the XF86-VidModeExtension paths; only
RandR's gamma hooks were causing issues.

Assuming 16-bit precision for the RandR gamma calls seems to be
working fine so far, but RRCrtcGetGamma doesn't seem to be hooked up.
Currently, gnome-screensaver is destroying the user set gamma ramp
going in and out of screen saver, but I haven't verified whether or
not the fix for this is in a release. I'm not seeing any code for it
in radeonhd, though, and haven't managed to find how/if this can be
done in the first place.

-- 
Yang Zhao
-- 
To unsubscribe, e-mail: radeonhd+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: radeonhd+help@xxxxxxxxxxxx