Announcement

Collapse
No announcement yet.

The Linux Desktop Responsiveness Patches Are Feeling Good

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

  • The Linux Desktop Responsiveness Patches Are Feeling Good

    Phoronix: The Linux Desktop Responsiveness Patches Are Feeling Good

    As was reported on Phoronix yesterday, the Linux desktop responsiveness problem may be fixed. This is the issue that has affected many Linux desktop users for numerous months where when dealing with large file transfers or other disk operations, the desktop interface (regardless of whether its GNOME, KDE, Xfce, etc) would become unresponsive and it could be a good number of seconds before a simple action like clicking a menu item would be processed...

    http://www.phoronix.com/vr.php?view=ODQ3OQ

  • #2
    Your assumption is wrong. KernelofTruth's backport only contains 2 out of 7 of the patches. There are 5 additional that he did not backport:

    http://forums.gentoo.org/viewtopic-p...0.html#6377520

    Please try to do fact checks on your assumptions in the future before reporting them.

    Comment


    • #3
      Oh there's 7 patches that all together fix the issue?

      Comment


      • #4
        Originally posted by DeepDayze View Post
        Oh there's 7 patches that all together fix the issue?
        indeed, but they are more tricky to backpart

        at least starting with the 3rd patch I tried it and parts of the code are spread all over the file

        since I'm not that experienced I'm waiting for someone else to backport (most favorably the zen-kernel devs )

        Comment


        • #5
          since I'm pretty busy - I unfortunately can't do the further backporting

          anyways: the results already are pretty impressive considering that it's only a small amount of code that's been added / removed (those 2 patches)

          Comment


          • #6
            I thought this was FUD, but it was true? Linux could not handle things without lagging sometimes? And now, is the situation better or does it still occur?

            Comment


            • #7
              Originally posted by kebabbert View Post
              I thought this was FUD, but it was true? Linux could not handle things without lagging sometimes? And now, is the situation better or does it still occur?
              it was true for some time in the past (for me)

              from my observation it most probably got worse / introduced after 2.6.34 since that kernel was working excellent for me back then - even under heavy traffic


              luckily with those lots of changes and the other improvements coming (e.g. reducing barrier writing, a unified slab (v3), and more) the future looks bright

              Comment


              • #8
                2.6.22 (?) with Con's SD or RSD cpu scheduler was another masterpiece in terms of performance & responsibility

                Comment


                • #9
                  Originally posted by kebabbert View Post
                  I thought this was FUD, but it was true? Linux could not handle things without lagging sometimes? And now, is the situation better or does it still occur?
                  Sometimes it happened and sometimes it didn't.
                  For example yesterday I put a 2.8 gb file on my local lighttp and downloaded it with aria2. This has to be the most I/O one would normally produce and my desktop was perfectely responsive...

                  Comment


                  • #10
                    Originally posted by kernelOfTruth View Post
                    indeed, but they are more tricky to backpart

                    at least starting with the 3rd patch I tried it and parts of the code are spread all over the file

                    since I'm not that experienced I'm waiting for someone else to backport (most favorably the zen-kernel devs )
                    Damentz (he's one of the zen devs) has some experience with backporting patches so he might take a crack at backporting all 7 patches to 2.6.35 (and maybe 2.6.34)

                    Comment

                    Working...
                    X