XWayland Aiming For Glamor Support, Merge Next X.Org Release 83
An anonymous reader writes that XWayland is nearly ready to be merged into the main X.org tree "X.Org Server 1.16 this summer should support XWayland, the means of allowing X11 applications to run atop Wayland-based compositors without the need for any application/game changes. With the revised design, XWayland has generic 2D acceleration over OpenGL and a cleaner design compared to earlier revisions. With GNOME 3.12 having better Wayland support and Plasma Next around the corner, it looks like 2014 could be the year of Wayland's take-off!"
The patch series emails have more details. The big news here is that XWayland is ditching its old DDX model for one based on Glamor. eliminating the need for any X.org drivers to be written to support X11 on Wayland: "Finally, the last patch adds the Xwayland DDX. Initially Xwayland was an
Xorg module that exposed an API for Xorg video drivers to hook into
so that we could reuse the native 2D acceleration. Now that glamor is
credible and still improving, a much better approach is to make Xwayland
its own DDX and use glamor for acceleration. A lot of the code in the Xorg
approach was busy preventing Xorg being Xorg, eg, preventing VT access,
preventing input driver loading, preventing drivers doing modesetting.
The new DDX in contrast is straight-forward, clean code, only 2500 lines of
code and neatly self-contained." It does not yet have direct rendering or any acceleration, but those patches should come soon.
HALLELUJAH! (Score:5, Funny)
Re:HALLELUJAH! (Score:4, Informative)
The Year of the Linux Desktop is upon us!
Again?
Re: (Score:2)
It'll never come, but the year of the Linux Palmtop is upon us.
Re: (Score:1)
Re: (Score:2)
Desktop computer explained (Score:2)
A desktop computer is a computer that hooks up to a TV and is controlled with a mouse instead of touch. It lets you do two things that most tablet computers can't do. One is split the screen so that you can have two or more things showing at once, so that the calculator app doesn't need to cover everything else up. The other is let you make apps. You know all these apps you run on your tablet? Someone made them on a desktop computer. There are also laptops, which look like a tablet with a keyboard but run
Awesome quote in TFS: (Score:2)
"It does not yet have direct rendering or any acceleration, but those patches should come soon."
How many projects are in the same state?
Re: (Score:1)
Re: (Score:1)
Re: (Score:1)
Re: (Score:1)
Re: (Score:2)
Re: (Score:1)
Main issue with proprietary software: Stuff always gets written even if the programmer is incompetent
Re: (Score:2)
I'm the opposite. I can't stand lacking the ability to dig in and change software when I don't like the way it works. It's rare that I actually do, but there's a huge freedom I get from knowing that when I need to extend the software, I can.
It's common for commercial software to not do what I want it to, either. I'd love to have a working amazon instant video client for my Android phone.
Re: (Score:1)
The proprietary cave has working software and a dessert bar.
Re: (Score:1)
Keep rewriting the compositor, keep compatibility (Score:2)
Working computers stop working eventually (Score:2)
Re: (Score:2)
Glamor acceleration (Score:1)
It does not yet have direct rendering or any acceleration, but those patches should come soon.
The patch series emails have more details. The big news here is that XWayland is ditching its old DDX model for one based on Glamor. eliminating the need for any X.org drivers to be written to support X11 on Wayland:
Glamor provides 2d acceleration using openGL.
Re: (Score:2)
Re: (Score:3)
The actual article where he says Plasma Next will not (at least initially) support Wayland is: http://blog.martin-graesslin.c... [martin-graesslin.com]
Remote display across network? (Score:3)
OK, so I need to buy a clue here... does this move the ball forward with respect to being able to run an X-Windows client application on one node, and set the display back to a Wayland-based display server running on another node elsewhere on the network?
Re:Remote display across network? (Score:5, Informative)
Yes, XWayland can be compared to an X Server running on Windows or over OS X, it translates X to the native display, in this case Wayland
Re: (Score:1)
Actually, you just want VNC: tightvnc, vnc4 or vino. It works so much better for any major application, especially if it uses any recent X extensions
Re:Remote display across network? (Score:4, Insightful)
Actually, I do *NOT* want VNC. My objective is to run an X app on a server that itself has no video capabilities at all and have that window appear fully integrated on my workstation desktop. I do not want a window that contains an entirely unneeded desktop that contains the app window that I actually want.
VNC does not work better for that application, at all.
Re:Remote display across network? (Score:4, Informative)
RDP can remote single applications and has been able to for years. No full desktop required.
Re: (Score:2)
RDP can remote single applications and has been able to for years. No full desktop required.
Do you happen to know how RDP handles windows that occur outside the root of the first window in single application mode? Does it "just work" with a new client-side window?
Re: (Score:1)
Re: (Score:3)
Vnc requires most of the things a full X server installation require and it's a bit clunky to use as a bonus. Compare to "ssh me@server -c MyXapp".
What do you recommend for forwarding a single app from Linux server to Linus workstation?
Re: (Score:2)
In that case what you want is xpra [xpra.org]. Each window is rendered off-screen and forwarded individually, as a compressed video stream (x264 if it's available). You can detach from the xpra server and reattach later, from the same client or a different one, with all your applications intact. A lot like how Wayland remoting will work, really, except that in Wayland it will be better integrated due to not needing to support all the legacy parts of X11.
Re: (Score:2)
I do use xpra sometimes when I need the extra functionality and it works well. My point ws that VNC is a non-starter for that need (it works well for other situations, especially for providing remote user support).
As for Wayland, the only thing I've seen there is experimental support for running the full blown Wayland server and compositor on the server and it will use RDP if you want to view it remotely. It's hard to tell though since it's all very hand wavey at this point.
Re: (Score:2)
As for Wayland, the only thing I've seen there is experimental support for running the full blown Wayland server and compositor on the server and it will use RDP if you want to view it remotely.
Well, you will need a Wayland compositor on the server, since Wayland is a local/shared-memory IPC protocol. The compositor will take the place of the xpra server, and communicate with a proxy (Wayland client) on the user's machine. It doesn't have to merge the windows into a single desktop, however. The current RDP backend in Weston is limited to the desktop mode, but if you can forward a complete desktop then there's nothing technically difficult about forwarding an individual window; it's just a matter o
Re: (Score:2)
It will be interesting to see how all of that plays out.
Re: (Score:2)
Because I can't use vaporware. Can you recommend these ready to install implementations that have the needed properties?
If you say write it yourself, I will reply why don't I just stick with X since it works right now.
If you believe X and VNC are the same, you must not know very much about either one.
Re:Remote display across network? (Score:4, Informative)
Re: (Score:2)
Essentially, it takes the Xorg server and adds a video driver that causes it to be a Wayland client.
Re: (Score:2)
Do you really care about remote X protocol, or do you want a remote window with the app on it? 'Cause Wayland checked in per-app RDP a year ago and making a chromeless RDP viewer ought to be pretty straightforward (if it doesn't exist already). ssh handles X specially - handling RDP specially could be something it adds.
For some people the distinction matters, but for others it's good enough (or better), depending on which needs more bandwidth, as X can sometimes be an unreasonable pig on the wire (see als
Re: (Score:2)
Really care. I want a headless box running legacy X applications to be able to display across the network without a noticeable increase in latency.
Re: (Score:2)
Yeah, Linux seems to be killing BSD and all flavors of UNIX on the desktop that don't have an apple logo on them. I'm not sure why Non-linux support is really needed. However, if it is needed, it kind of makes sense to get it working rock solid and tested on the largest of the three ( linux, BSD, UNIX), before starting a port to the others. When Gnome and KDE have made the transition, then I think it would probably be ready for BSD & Unix Ports.
Re: (Score:2)
They want this to win, they need it to run on UNIX & BSD.
Seriously? Linux is a very small fraction of the desktop market and (other than OSX) any other UNIX or BSD is a very small fraction of that.
They don't need BSD support to win anything other than friendship.
Is XWayland... (Score:3)
Re:Is XWayland... (Score:5, Informative)
XWayland is the X server for Wayland, so that you can run traditional X applications on Wayland (as opposed to Qt etc. applications, which will talk directly to Wayland). http://wayland.freedesktop.org... [freedesktop.org]
Re:Is XWayland... (Score:4, Informative)
Client X11 apps speak the X11 protocol to XWayland, XWayland speaks the Wayland protocol to Wayland so it's basically a big compatility shim. From Wayland's side it's just another client and if you use an X11 server you don't need it, it's not really part of either. Maybe the closest analogy is WINE, if you use Windows or run native Linux applications you don't need it. But if you want to run Windows applications on Linux you need WINE, likewise if you want to run X11 applications on Wayland you need XWayland. Basically you take an X11 server, stop it from talking to actual hardware and makes it draw to a Wayland window instead.
Re: (Score:1)
from TFS it sounds like it's part of x.org, it's all the same developers.
I assume it's a server that kicks to Wayland, similar to Xnest kicking to X.
Comment removed (Score:4, Insightful)
Re: (Score:2, Insightful)
Re: (Score:1)
At this point, it's mostly just you.
That's just not the case. I have been amused at the proliferation of Linux graphics stack diagrams that have emerged on Wikipedia as contributors try to explain this stuff. There are six (nice) distinct diagrams on this page [wikipedia.org], for instance. The fact is that rendering graphics is hard and lengthy pipelines tend to emerge and unless you have your head in it for some reason it appears "insanely complex."
This [wikipedia.org] diagram actually illustrates where Glamor fits into the graphics stack. In simple terms, Glamor uses
Re: acceleration (Score:3)
Anything without acceleration is an experiment. It doesn't matter how many lines of code you've written, or how efficient it seems. 100% of the required functionality is acceleration.
Acceleration is why X is being replaced by Wayland. 2D X11 requires a separate driver for every different type of hardware. 3D X11, from what I read by the Wayland people themselves, has three different APIs. For a long time, the only drivers with good 3D acceleration were proprietary drivers from AMD and nVidia.
I want Wayland to succeed, but I feel that it's still a long way off. The devil is in the acceleration. Think about the time spent by XFree86 developers over the decades writing acceleration code versus everything else, and that's the part we're missing right now. I'm not very clear on just where the acceleration is missing, but it sounds like it's missing in a foundational piece.
Re: (Score:2)
Think about the time spent by XFree86 developers over the decades writing acceleration code versus everything else, and that's the part we're missing right now.
The Wayland developers are, for the most part, the X developers, so they not only have access to all that existing X driver code that took so long to write, they're the folks who best understand that code, and know how to adapt it to a new environment. They're standing on the shoulders of giants (and in some cases, are the giants).
The biggest changes we should expect to see are in the API. Under the hood, I expect to see a whole lot of code that's identical to the current Xserver, or nearly so. As I underst
Re: acceleration (Score:4, Insightful)
Re: (Score:2)
Not letting it slide when some fanboy that's never run Wayland but likes the idea says "X sux because of some lie I made up" is not an anti-wayland agenda.
Re: (Score:2)
And where did I say anything like "X sux..."? I said Wayland is going to have the benefit of the work that was done on X to create drivers, and the benefit of people on its team who generally understand those drivers. And you attacked like some sort of rabid mongoose.
I like X. I use its remote features regularly. And I'm quite satisfied with its performance. I'm going to be reluctant to switch to Wayland until it supports (directly or through XWayland) all the features I need. Nevertheless, I think I'm prob
Re: (Score:3)
It's not quite as bad as it sounds, the actual hardware drivers are still accelerated and exposed as OpenGL, it's just that XWayland doesn't make use of it. If you look at this diagram [wikipedia.org] it's the line between the X-server and libDRM that's broken when you use XWayland instead because Wayland can't talk directly down to that level. XWayland needs to be rewritten to accelerate graphics using OpenGL instead, then it'll hook into the green box above libDRM and all will be well. Luckily for the Wayland project so
Re: (Score:2)
Yup. Beat around the bush. Yadda Yadda about compatibility layer upon compatibility layer to keep features before acceleration. Just start anew & solid and the rest will follow.
X from user space (Score:2)
Re:X from user space (Score:5, Informative)
X has always been in userspace.
Re: (Score:2, Informative)
X has always been in userspace.
That's a serious overstatement.
X.org runs as root and directly maps parts of the PCI address space into its address space. It has the same amount of raw metal access as the kernel; it's a kernel which runs on top of another kernel.
Re: (Score:3)
If it runs as a user (even root), it runs in userspace. It is a privileged user, but that's not actually the same thing.
While not typically done, there is no reason it can't be split into a part that maps the PCI device and a part that requests specific access to those parts.
If you can put up with a performance loss, Xorg CAN be run on top of the framebuffer device as non-root [debian.org]. This ha been doable for years.
Wrong way round (Score:1)
Complexity? (Score:2)
Wayland is supposed to reduce code complexity.
Is XWayland, running on top of Wayland actually less complex than X on its own?
Re: (Score:1)
Wayland is a fraud.
It's advertised as a way to make things less complex. But it lacks critical functionality from day 1, it's already due a rewrite to support OpenGL properly (which I'd say is PRETTY BASIC FUNCTIONALITY), and certain other feature requests, like remote apps, get contradictory and often ridiculous responses, from "You don't need that" ("But I use it!) "Then you're just wierd, fuck off weirdo" to "Oh, in Wayland we'll be turning each window into a live H.264 video stream! That'll have no l