Bug#467105: ATI driver: use of Xv causes one monitor to be repeatedly blanked

[Alex Deucher]

On Tue, Feb 26, 2008 at 4:50 PM, Darren Salt
<linux at youmustbejoking.demon.co.uk> wrote:
> I demand that Alex Deucher may or may not have written...
>
>
> > On Tue, Feb 26, 2008 at 12:07 PM, Darren Salt
>  > <linux at youmustbejoking.demon.co.uk> wrote:
>
> >> I demand that Alex Deucher may or may not have written...
>  [snip]
>
> >>> Why is gxine calling the DPMS hooks for the monitors?
>  >> It's designed to allow blanking when nothing is being played or the
>  >> stream is paused; on starting or resuming playback, it is intended that
>  >> gxine will unblank the display automatically.
>  [snip]
>
> >> Also, I don't want it to ever be able to fail to restore the DPMS
>  >> settings – by not altering them, there's nothing to restore. This does
>  >> require ensuring that the blank timer never reaches 0 (unless it's set
>  >> ludicrously low).
>
>  > But why does it do it repeatedly?
>
>  See above.
>
>
>  > that's what causes the blinking.  it should do it once, or am I missing
>  > something?
>
>  If crtc1 is already on, why touch crtc0 at all? (Also, if crtc0 is in some
>  state other than 'on', why switch it off?)
>

This is a bug in certain radeons where you have to enable crtc1 before
crtc0 or you get a blank screen.  I'll see if I can dig up some more
info on which ones are affect and maybe I can narrow the range.

>  (Tweaking gxine to take more notice of DPMSInfo()'s output should fix one
>  side of this bug...)
>
>
>  >>> Do other video players have the same problem?
>  >> Any which use the same methods will, since "xset dpms force on" is
>  >> sufficient to trigger the problem.
>
>  > when you run "xset dpms force on" does the blinking continue after you run
>  > it or does it happen once?
>
>  Once.

Ok.  thanks.

Alex