1 |
Just porting to a toolkit that "supports" Wayland won't be much help to |
2 |
window managers. A WM has to implement the wayland server side (compositor) |
3 |
while applications are clients. The toolkits abstract away the X/wayland |
4 |
client API calls (E.G. Qt platform plugins) so you simply create your |
5 |
widgets, setup your (toolkit native) callbacks and are done. But as soon as |
6 |
you call X-specific functions in your applications things will get harder. |
7 |
Qt now also implements a wayland-compostior which HELPS creating a wayland |
8 |
server. But still you need to do quite some work. Porting to gtk3 (xfce... |
9 |
) will have a similar impact: It won't allow XFCE to automatically run on X |
10 |
and Wayland. Probably it makes some things easier but in the end there is |
11 |
not so much the toolkit can abstract away in terms of creating a |
12 |
compositor/server/WM. |
13 |
|
14 |
Personally I tried to get running Plasma on wayland several times and while |
15 |
I finally got it started there were so many crashes (e.g. some applications |
16 |
esp. those having to run on XWayland, but also systemsettings) and issues |
17 |
with managing windows that I gave up on it for the moment. Gnome might be a |
18 |
different thing as they go wayland exclusively and have a working |
19 |
implementation for a longer time than kde. I also tried enlightenment with |
20 |
wayland which didn't run more stable than with X :( |
21 |
|
22 |
2017-07-11 17:25 GMT+02:00 Rich Freeman <rich0@g.o>: |
23 |
|
24 |
> On Tue, Jul 11, 2017 at 10:51 AM, Ian Zimmerman <itz@××××××××××××.org> |
25 |
> wrote: |
26 |
> > On 2017-07-11 09:02, Rich Freeman wrote: |
27 |
> > |
28 |
> >> > I use GNOME with Wayland for some time and I actually didn't notice |
29 |
> >> > that I switched until I tried to get synergy working ( mouse sharing |
30 |
> >> > software, which only works on X ), seems like GDM automatically |
31 |
> >> > chose Wayland since some upgrade. XWayland works pretty seamlessly |
32 |
> >> > as well, so I'll just stay with Wayland for now, but it might be |
33 |
> >> > more annoying to use it with other DEs/WMs. |
34 |
> > |
35 |
> >> > However, I have less screen tearing with fullscreen applications |
36 |
> >> > with Wayland than I had with X ( with radeon + mesa ). |
37 |
> > |
38 |
> >> My sense is that this is probably what people would see. It will |
39 |
> >> probably work fine for any of the major DEs, but you'll find these |
40 |
> >> little cases of tools that aren't ported. One BIG area that will be |
41 |
> >> affected is X11 forwarding. I'm not sure if that works over ssh or |
42 |
> >> not with wayland, but wayland in general doesn't support network |
43 |
> >> sockets. |
44 |
> > |
45 |
> > What about "3rd party" window managers like openbox? From my limited |
46 |
> > understanding of wayland, that functionality just goes out of the window |
47 |
> > (OOPS, sorry); window management becomes a responsibility of the toolkit |
48 |
> > and there is no way to plug in a different one. |
49 |
> |
50 |
> I'm going out on a limb a bit here, but my understanding is not so |
51 |
> much that it is impossible for arbitrary applications to talk to |
52 |
> wayland (that seems silly - it is just an API). Rather, the major |
53 |
> toolkits simply have already done all the hard work so that if you use |
54 |
> one of those toolkits then your application will work. |
55 |
> |
56 |
> I'm sure there is no reason an application that doesn't use qt/gtk/etc |
57 |
> couldn't just make direct calls to wayland. However, it will require |
58 |
> a lot more porting work on the part of upstream, and so it probably |
59 |
> won't happen quickly. |
60 |
> |
61 |
> In the same way an application written to use QT probably can be made |
62 |
> to work on OSX or Windows with very little additional work, because |
63 |
> the toolkits provide a single API across all the platforms. You could |
64 |
> write an application that works on all these platforms without using a |
65 |
> toolkit, but then the developer needs to maintain all the API |
66 |
> abstraction. |
67 |
> |
68 |
> Getting back to openbox/etc, I suspect that you have a couple of extremes |
69 |
> here: |
70 |
> |
71 |
> * Full-fledged DEs like Gnome/KDE. They have a ton of functionality |
72 |
> that would be impacted by Wayland, but they also use toolkits that |
73 |
> have probably already taken care of this. |
74 |
> * Very minimal window managers (think fvwm/twm/etc). They may not use |
75 |
> a toolkit that was ported, but on the other hand their functionality |
76 |
> is minimal and porting might not be so hard. Also, there seems to be |
77 |
> some effort to port more minimal toolkits like motif to wayland. |
78 |
> * In-between environments (think xfce, openstep, etc). They don't |
79 |
> benefit from the toolkit but still have a lot of functionality to |
80 |
> port. I heard that xfce is being ported to gtk for just this reason. |
81 |
> |
82 |
> I suspect that Wayland is going to drive adoption of gtk/qt much more |
83 |
> widely. For the effort of directly porting to Wayland you could just |
84 |
> port to gtk and then get coverage on other platforms as well. |
85 |
> |
86 |
> > |
87 |
> > Or does xwayland help with that? I'll be grateful for an explanation of |
88 |
> > this area, as I'm worried about the future of the X server but I'm also |
89 |
> > married to openbox. |
90 |
> > |
91 |
> |
92 |
> I suspect that xwayland would cover some of this, but I haven't messed |
93 |
> with either. |
94 |
> |
95 |
> -- |
96 |
> Rich |
97 |
> |
98 |
> |