[Differential] [Commented On] D1231: Add Remote Access interface to KWayland

[graesslin (Martin Gräßlin)]

graesslin added a comment.


  In https://phabricator.kde.org/D1231#39191, @Kanedias wrote:
  
  > Can't we pass screen index along with all the invocations?
  
  
  what is a screen index? You mean the id of the global referencing the wl_output?
  
  > Krfb (and other recording tools) will know the screen configuration as they reside in same wayland session. They'll get the buffer and the screen index and will know exactly what to map. Am I missing something here?
  
  That might be a race condition. What if the output got removed? On the other hand if the events are queued, it might work.
  
  > Besides, I didn't find any mentions of multi-screen capabilities in Krfb at all. It currently works like this:
  > 
  >   d->framebufferImage = XGetImage(QX11Info::display(),
  >                                   id,
  >                                   0,
  >                                   0,
  >                                   QApplication::desktop()->width(),
  >                                   QApplication::desktop()->height(),
  >                                   AllPlanes,
  >                                   ZPixmap);
  >    
  > 
  > If that's the requirement, there will be huge amount of work to implement it from ground up.
  
  Right I see the problem. On X11 of course there is just one virtual screen for all outputs. Thus krfb just always get the whole screen. On Wayland we have the problem that the compositor is rendering to each output individually. So we end up with a buffer for each output. Argh that sucks. I'm not seeing a solution for it right now and would say just ignore it for the moment. Either we only support one output at the start or combine the image of all outputs to one.

REPOSITORY
  rKWAYLAND KWayland

REVISION DETAIL
  https://phabricator.kde.org/D1231

EMAIL PREFERENCES
  https://phabricator.kde.org/settings/panel/emailpreferences/

To: Kanedias, graesslin
Cc: plasma-devel, jensreuterberg, abetts, sebas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/plasma-devel/attachments/20160712/6c673990/attachment.html>