One main difference between x11 and wayland is that in x11 all applications receive all input whereas in wayland the compositor receives input and decides which application may receive it.
This causes a major issue with application which need input grab for running or showing a nested desktop. This affects e.g. GUI virtual machines (SPICE clients), RDP clients, VNC clients, nested kwin/gnome-shell/whatever. See also: https://bugzilla.redhat.com/show_bug.cgi?id=1285770 . Games might also be affected, I don't know.
I'm posting this here to raise attention because 1. to fix this we need another protocol (extension) 2. this issue can't be fixed in a single application, it needs the whole stack (wayland library, compositors, affected applications) to change
Compared to the other bugs in "WaylandRelated" lists [1] [2], these bugs need architecture design, not just fixing bugs in one package.
I don't know whether this mailing list is the right place to discuss an issue like this. If you know better, please tell me.
[1] https://bugzilla.redhat.com/showdependencytree.cgi?id=1277927&hide_resol... [2] https://bugzilla.gnome.org/showdependencytree.cgi?id=757579&hide_resolve...
On Sun, Jul 17, 2016 at 12:42 PM, Christian Stadelmann genodeftest@fedoraproject.org wrote:
I don't know whether this mailing list is the right place to discuss an issue like this. If you know better, please tell me.
Those issues are addressed with the pointer constraints[1] and relative pointer[2] protocols which are still not declared final but pretty close to it. Discussion happens on fdo's wayland-devel mailing list.
Rui
[1] https://cgit.freedesktop.org/wayland/wayland-protocols/tree/unstable/pointer... [2] https://cgit.freedesktop.org/wayland/wayland-protocols/tree/unstable/relativ...
Oh, and how about keyboard input? these patches are just about mouse input as far as I can see.
On Tue, Jul 19, 2016 at 9:52 AM, Christian Stadelmann genodeftest@fedoraproject.org wrote:
Oh, and how about keyboard input?
What about it? Applications can't take exclusive control of keyboard input which is a good thing obviously. They won't be able to get events for key combos that are registered as global shortcuts in the compositor which I suppose is your worry.
We could address that with more protocol for apps to request a mode or mark their windows such that the compositor disables some (or even all but an "escape") global shortcuts when that window gets keyboard focus.
Rui
On Tue, Jul 19, 2016 at 9:52 AM, Christian Stadelmann <genodeftest(a)fedoraproject.org> wrote:
What about it? Applications can't take exclusive control of keyboard input which is a good thing obviously.
Some applications really have to. If they don't get this feature they will be broken on wayland.
They won't be able to get events for key combos that are registered as global shortcuts in the compositor which I suppose is your worry.
Correctly.
We could address that with more protocol for apps to request a mode or mark their windows such that the compositor disables some (or even all but an "escape") global shortcuts when that window gets keyboard focus.
Right. Virtual machines and remote desktop softare often needs to get every shortcut but an escape shortcut. That's what we need. In an ideal world the compositor could ask the user before doing this, so it cannot be abused by e.g. keyloggers.
Still, should I post this request to fdo's mailing list or bugzilla?
Any luck with a fix or updates for this? Having a hard time with my vmware view client and the same problems..
Hi,
Any luck with a fix or updates for this? Having a hard time with my vmware view client and the same problems..
Well, this is a longer term fix, because by design (and on purpose), Wayland doesn't allow clients to have an active grab so it's a bit complicated for Xwayland to translate these to the Wayland compositor.
A possibility would be to use some sort of a "private" Wayland protocol that only Xwayland would use to inform the compositor that an X11 application has an active grab. That means reaching an agreement on the protocol, then implementing it in Xwayland and the compositors (mutter/gnome-shell in our case).
Also, we wouldn't want just any Wayland client to be able to use it (otherwise we kinda lose the benefit of not having grabs), so as a bit of yak shaving, I've been trying to get an API to restrict the visible protocols to clients, https://patchwork.freedesktop.org/patch/105319/ - But not having the API in place is by no mean a show stopper.
I'll add this to the feature page.
Cheers, Olivier
Do you know what happened with this request? I am still having issues using virtualbox or boxes in wayland.
The original bug report on rhbz got some updates recently. I also think I've seen a patch on freedesktop.org mailing list, but I don't remember where. So don't expect this issue to be fixed in Fedora 26.
desktop@lists.fedoraproject.org
-
Alexandru Calinoiu -
Christian Stadelmann -
Olivier Fourdan -
p＠zbrba.com -
Rui Tiago Cação Matos