Announcement

Collapse
No announcement yet.

Purism Might Develop An X11-Free Wayland Compositor Aligned With GNOME

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Purism Might Develop An X11-Free Wayland Compositor Aligned With GNOME

    Phoronix: Purism Might Develop An X11-Free Wayland Compositor Aligned With GNOME

    Yesterday we heard of Purism's plans to support desktop diversity but by default for their Librem 5 smartphone they will likely be using GNOME in order to maintain a unified experience across their devices. The latest now is they might develop a new Wayland compositor in line with GNOME...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    From my understanding a lot of the working going into shell & mutter atm is removing the x11 bits and clean-up old code since that's holding it back. Seems a bit weird to write an entire new compositor especially if the plan is to make PureOS the same on the desktop & mobile. Maintaining two compositors seems illogical.

    At least it's just talks now, I would like to see some more technical perspectives from both GNOME and Purism before they make a choice.

    Comment


    • #3
      not a single wayland client works well

      Comment


      • #4
        Question....would a phone OS NEED a full featured and possibly HEAVIER Wayland Compositor? What could be cleaved off of Mutter and still make a functional enough compositor for a phone OS ? Perhaps that is why they are looking at making a Wayland Compositor from scratch?

        Comment


        • #5
          Originally posted by Jumbotron View Post
          Question....would a phone OS NEED a full featured and possibly HEAVIER Wayland Compositor? What could be cleaved off of Mutter and still make a functional enough compositor for a phone OS ? Perhaps that is why they are looking at making a Wayland Compositor from scratch?
          You could trim off some with just a phone Wayland compositor, but as soon as you bring back convergence that is part of their expressed plans, you sort of need to bring all of it back. Seems like they really should be just trying to get Mutter cleaned up of any remaining X11/XCB bits to allow building that nicely without any optional X support.
          Michael Larabel
          https://www.michaellarabel.com/

          Comment


          • #6
            Gotcha'....thanks !

            Comment


            • #7
              TL; DR : Mutter is GNOME people's toy, we are not allowed to do the modifications we wanted, so we start a new project but still say it is for GNOME to appease the GNOME god.

              Comment


              • #8
                Well of course they "could" do these things... but "should they"

                Comment


                • #9
                  I wonder if Mir could be useful to them. Mir suppots Wayland clients, started out with convergence design goals and I think it's currently used in production for some IoT projects.
                  Mir is the fast, open and secure display server to unlock next-generation user experiences. Mir runs on a range of Linux powered devices including traditional desktops, IoT and embedded products.

                  Comment


                  • #10
                    Originally posted by Mr. Octus View Post
                    From my understanding a lot of the working going into shell & mutter atm is removing the x11 bits and clean-up old code since that's holding it back. Seems a bit weird to write an entire new compositor especially if the plan is to make PureOS the same on the desktop & mobile. Maintaining two compositors seems illogical.

                    At least it's just talks now, I would like to see some more technical perspectives from both GNOME and Purism before they make a choice.
                    Here is the reason:
                    "To sum it up, there are a number of problem areas that needs new solutions.
                    1. Low latency input forwarding
                    2. Low latency visual input event feedback (pointer cursor movement)
                    3. Low latency & zero copy client content forwarding (scan-out of client buffer)
                    4. Input methods in the shell UI
                    5. Stalls on the main thread stalls compositor frame redraws "
                    https://wiki.gnome.org/Initiatives/W...ll/GnomeShell4

                    Comment

                    Working...
                    X