Other Do any of the classic X11 window managers run on XWayland?

I hope this isn't a duplicate. I've been reading about this Xwayland thing that is supposed to be an X-server running on top of wayland. I was wondering if any of the classic X11 window managers work when run on top of Xwayland. I'm talking about twm, fvwm, windowmaker, openbox, icewm, blackbox, xfce... and the rest. Does anyone have any experience? Does it work?

It strikes me that it should at the very least run twm, which after all is part of the X11 distribution, or Xwayland can hardly call itself an X-server...
 
please, do not read such highly-opinionated materials to make a view on subject. As for your question, just run them under X session and your are fine. Otherwise, there is substantial number of native wayland wm's.
 
OK, but what if I want to use windowmaker or xfce, or even just twm? Is it just not possible to run those on Xwayland? That's all I want to know. Has anyone managed to get it to work?


This seems to suggest that X11 clients, presumably including window managers, should be able to work on wayland by using Xwayland?
 
i don't understand your goal - why would you want to run wm's with xwayland? just to test? xwayland is a layer to run X applications on wayland, which, theoretically accepts running of wm's but it is an awkward approach to my eyes - this will be running wm in a wm ( aka nested ). as for windowmaker - i have no information whether it has wayland support, likely not. xfce4 has a major plan to be wayland ready in future but it is not yet ready.
 
So that I can keep using the window manager I know and love? And that I have invested lots of time in learning how to use? It used to be called "backwards compatibility"...
 
you can use your favorite wm's just fine. what is the reason of such a fear? ( perhaps, you reading this utterly non-sensical thread about X and Wayland - X.org is not going away ).
 

Redhat appears to be EOL'ing Xorg, which is the source of the freebsd Xorg port. AFAIK, no-one else is funding it. Or are you saying that Xorg will continue to be developed somehow?

That page says "Xwayland should be able to handle most X11 clients that won’t immediately be ported to Wayland", so I just wondered if any of the current X window managers work on Xwayland. That's all I wanted to know. Is there a viable backwards compatibility path? Or alternatively is Xorg development going to continue somehow?
 
i do not want this thread to be 10+ pages, so i will end up with this last reply.
my personal take - X.org will be alive for quite some time - by the time it will die out, xfce4 likely gain native wayland support and you can use it just as it was years before. I cannot speak of windowmaker, i never used it. Also, i would suggest trying some of wayland native wm's, see how it goes. I use wayland for many years and i absolutely love it - it is workstation dream, clean and lean. i do not use any extra-ordinary functionality many expressed opinion on to be functional with wayland. Perhaps they never found one or did not try. Main concerns, which also worrying me too, are: wayland is better designed protocol than X.org ( whatever die hard conservatives object about ) but many applications lack behind by years ( for porting ). The GNOME (and it is a Red Hat tie too) is particular bitch which producing, redesigning or violating concepts against wayland protocol too. Since, many low level libraries - glib, gtk+, cairo to name a few are also under their umbrella and many X applications are tied with such libraries you have an obstacles of various kinds. They are not wayland issues per se .
 
Fair enough. I agree, not wayland issues per se, at all; I'm not trying to bash wayland itself. The problem is with Xorg going EOL, which then forces the existing user base, with all the existing X11 WM's and apps, to try to switch to wayland / Xwayland. As you say, the problems are not just with WM's with but many existing apps and libraries as well.

Really it was an innocent question, I just wondered how good Xwayland (as opposed to wayland itself) is as a backwards compatibility path, whether it gives a way to transition transparently to a wayland base away from Xorg, as redhat claimed in their announcement. Cheers.
 
Xwayland is basically the same thing as Xephyr and Xnest (and Xming, XQuartz). It is a window containing a simple self contained Xserver.

So a Window Manager will run in there no problem but you will still be constrained to that nested window. Whether it is fullscreen or not, it won't be able to control any native Wayland windows.

Xweston, A modified Weston and Xwayland can almost provide a drop-in Xorg clone. And this will work fine until the Wayland guys damage X11 support in apps or Xwayland slowly becomes broken.
 
XFCE4 is going "slowly" toward Wayland, perhaps with WLRoots.
The other Window-Managers are a true unknown, I heard from the FFWM folks that they won't migrate to Wayland however if Xwayland is going to work fine problem solved then.

But I am concerned too about RH switching to Wayland, and after the recent direction taken, I am not even sure how long Wayland will be in the radar of IBM. We risk to end up with two critical projects dropped simultaneously.
 
Xwayland is basically the same thing as Xephyr and Xnest (and Xming, XQuartz). It is a window containing a simple self contained Xserver.

So a Window Manager will run in there no problem but you will still be constrained to that nested window. Whether it is fullscreen or not, it won't be able to control any native Wayland windows.

Xweston, A modified Weston and Xwayland can almost provide a drop-in Xorg clone. And this will work fine until the Wayland guys damage X11 support in apps or Xwayland slowly becomes broken.
Ah good, so there IS some kind of back compatibility path, even if it's not perfect. I am familiar with xephyr and xnest (use xephyr quite a lot) so I get your drift. Well, it's not perfect, but it's better than nothing at all. Presumably you can run it fullscreen. I wonder what the performance is like. I guess RH aren't too bothered if they're porting their stuff to wayland. Bah. XWeston sounds like it's worth a look.
 
XFCE4 is going "slowly" toward Wayland, perhaps with WLRoots.
The other Window-Managers are a true unknown, I heard from the FFWM folks that they won't migrate to Wayland however if Xwayland is going to work fine problem solved then.

But I am concerned too about RH switching to Wayland, and after the recent direction taken, I am not even sure how long Wayland will be in the radar of IBM. We risk to end up with two critical projects dropped simultaneously.
The real kicker is making Xorg end of life, of course. Still, it's their money, I guess it's fair enough if they don't want to spend more money on it, although I think it's a rather short-sighted move. Wayland may be great as a single-user PC display system, like windows has, eg for games, but you lose the power of the networked X11 protocol which is so valuable in development and test lab type environments, especially when distributed across differerent geographies. Things like vnc are a poor substitute, I suppose there's the MS rdesktop stuff. I wonder what nomachine are going to do as well. Oh well.

Regardless I will of course continue to use Xorg as long as it keeps working, but who knows how long that will be once RH sunsets it. And I guess another consequence is that everyone is likely to not do much more app development for X.

Of course if they drop both it will be a major bummer. It wouldn't be a first time, for good ol' ibm.
 
Ah good, so there IS some kind of back compatibility path, even if it's not perfect. I am familiar with xephyr and xnest (use xephyr quite a lot) so I get your drift. Well, it's not perfect, but it's better than nothing at all. Presumably you can run it fullscreen. I wonder what the performance is like. I guess RH aren't too bothered if they're porting their stuff to wayland. Bah. XWeston sounds like it's worth a look.
RH isn't concerned about the desktop. They get their $$$ from servers. Even then they've removed a lot of server stuff. They seem to be using the 80-20 rule. I worked as a developer at two companies that used the 80-20 rule to maximize profits. The 80-20 rule didn't serve them well. Both are defunct today.
 
I hope this isn't a duplicate.

Overlap with:

 
Back
Top