XWayland Updated Against X.Org Server 1.12

[phoronix]

Phoronix: XWayland Updated Against X.Org Server 1.12

Kristian Høgsberg has pushed updated patches this morning for XWayland in the X.Org Server that re-base this work to the X.Org Server 1.12 series. He's also updated the XWayland support for the xf86-video-intel graphics driver. XWayland is the effort for allowing an X.Org Server to run as a Wayland client...

http://www.phoronix.com/vr.php?view=MTA3Nzg

[asdx]

This is all nice and good but wake me up when something like KDE and Gnome is able to run on Wayland natively.

[phred14]

This is good. Much better than calling me and people like me 'tards, just because we regularly export DISPLAY across the network. I'm very happy with this report, much more so than with the PC-desktop-centric view espoused by many here.

[uid313]

This is all nice and good but wake me up when something like KDE and Gnome is able to run on Wayland natively.

Why natively, why not just under XWayland?
As I understand, that would bring benefits too.

[curaga]

Someone guessed there may be a benefit due to simpler buffer swap handling. While on the other hand it's certain that it's one extra daemon, and one extra level of indirection.

Make of that what you will.

[snadrus]

Someone guessed there may be a benefit due to simpler buffer swap handling. While on the other hand it's certain that it's one extra daemon, and one extra level of indirection.

Make of that what you will.

Few complain that keyboard or mouse input is too slow. The concern is X's output. In that case, for fullscreen use of XWayland it actually writes directly to Wayland's output buffer preventing any additional copy. This combination makes it easy to test & improve Wayland

[crazycheese]

Can Wayland actually display the gfx application running under different user output on the screen of current user session.

For example, I, from my current session, want to play opentransporttycoon as user "games". Xorg can handle that.

[magoo]

Few complain that keyboard or mouse input is too slow. The concern is X's output. In that case, for fullscreen use of XWayland it actually writes directly to Wayland's output buffer preventing any additional copy. This combination makes it easy to test & improve Wayland

OK, let me complain about the mouse input being too slow. I run an OpenGL application that updates at 60 FPS and drag an object with the mouse. Result: obviously visible lag between the position of the mouse cursor and the object tracking it. This is not screen update lag, as I say it is 60 FPS, it is lag in receiving the mouse position change notification.