Announcement

Collapse
No announcement yet.

Systemd 202 Starts Playing With D-Bus In The Kernel

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

  • Systemd 202 Starts Playing With D-Bus In The Kernel

    Phoronix: Systemd 202 Starts Playing With D-Bus In The Kernel

    Systemd 202 has been released and it begin experimental work on supporting kdbus, the implementation of D-Bus within the Linux kernel. There's also other fixes and features to this new systemd release...

    http://www.phoronix.com/vr.php?view=MTM1NTE

  • #2
    This is probably going to go over about as well as a fart in a spacesuit, with the kernel guys.

    Can you say "NAK?" I know you can.

    Comment


    • #3
      Originally posted by Grogan View Post
      This is probably going to go over about as well as a fart in a spacesuit, with the kernel guys.

      Can you say "NAK?" I know you can.
      Gregkh is developing the kernel side code. I'd say it has a good chance of getting merged.

      Comment


      • #4
        Originally posted by Grogan View Post
        This is probably going to go over about as well as a fart in a spacesuit, with the kernel guys.

        Can you say "NAK?" I know you can.
        Over all attitude of the kernel developers so far as been positive. Greg KH is the one doing it and he's got a good reputation backing his code, which helps, So far the over all attitude of a revamped IPC kernel mechanism is "Does it work? Is it good quality code? Does it do its job? Yes? Good. Merge it." Because the arguments of "WHY?" have already been answered

        Comment


        • #5
          Performance?

          I'd be interested when they switch this if it's faster. Presumably that's the point.

          Comment


          • #6
            Originally posted by snadrus View Post
            I'd be interested when they switch this if it's faster. Presumably that's the point.
            Here's the plan "Moving things onto kdbus/libsystemd-bus is going to be a step-by-step process, happening over the coming year or so".

            Comment


            • #7
              No way, after 50 years of UNIX we will finally get an IPC system that doesn't suck?! And GH is developing it? Awesome!

              Comment


              • #8
                Originally posted by tomato View Post
                No way, after 50 years of UNIX we will finally get an IPC system that doesn't suck?! And GH is developing it? Awesome!
                It sounds like a good time to revamp the traditional UNIX IPC protocol to improve performance and reliability, so perhaps kdbus and libsystemd dbus implementations could make great new backbones for a new and improved IPC protocol. Also would the changes be transparent enough so that these dbus implementations don't break existing code or require major library changes?

                Comment


                • #9
                  Originally posted by DeepDayze View Post
                  It sounds like a good time to revamp the traditional UNIX IPC protocol to improve performance and reliability, so perhaps kdbus and libsystemd dbus implementations could make great new backbones for a new and improved IPC protocol. Also would the changes be transparent enough so that these dbus implementations don't break existing code or require major library changes?

                  The goal is for it to be completly transparent to end users and apps, no breakage. but as Greg and Lennart both said "We'll see if that ends up being possible."

                  Comment

                  Working...
                  X