Announcement

Collapse
No announcement yet.

Fixed: The Linux Desktop Responsiveness Problem?

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

  • #31
    backports to all the kernels up to now?

    "Cute", when is kernel.org going to start backports to all the old kernels?

    Comment


    • #32
      Originally posted by squirrl View Post
      "Cute", when is kernel.org going to start backports to all the old kernels?
      Why would they? It's up to individuals or distro's to backport if they deem necessary.

      Comment


      • #33
        Hey guys, 2.6.34 needed more backporting than I thought. Merge the branch available here to backport 2.6.35's memory management to 2.6.34: http://git.zen-kernel.org/?p=kernel/...eads/mm-2.6.34

        Comment


        • #34
          Another test: use hugin/panotools to stitch together a few dozen 8MP images into a panorama, or open up said panorama in Gimp.

          Comment


          • #35
            Are these the same patches some people tried?

            http://lkml.org/lkml/2010/8/5/38

            Comment


            • #36
              so, is it better/worse?

              Comment


              • #37
                Originally posted by nerdopolis View Post
                piotr I'm not exactly sure, but dd is probably pegging your CPU.

                see what happens to your performance when you are copying that 5GB file?
                better a little I think than dd. Also, I have here rare configuration. My rootfs and home are on LVM, where this LVM is on dmcrypt. so raw storage <> dmcrypt <> lvm <> filesystem. It may reduce performance. I forgot to do tests without this patch so I cant say that this is faster or not.

                Comment


                • #38
                  Originally posted by piotr View Post
                  better a little I think than dd.
                  Then I guess it's not better at all. :> I'll check myself if those patches make a difference. If the problem isn't fixed yet, then the best option is to report to lkml till somebody is trying to fix it.

                  Comment


                  • #39
                    Originally posted by piotr View Post
                    Well It's not working for me I think.

                    Code:
                    dd if=/dev/zero of=test bs=1M count=5024 && rm test -f
                    in while loop and firefox lag, mouse laggy from time to time, HARDCORE. :<

                    Also I think after fresh boot starting firefox took a way more time than normal.
                    I have an unpatched 2.6.35 kernel and my system is still responsive if I run that command. I have an Intel X-25M SSD. Could this issue predominantly affect mechanical hard drives?

                    Comment


                    • #40
                      I think I will provide more than just beer! This issue has been severely plaguing me in every ubuntu install on various types of hardware for years now. It's so bad that running virtual machines on ubuntu and linux is actually pointless. it does not work.

                      I have been following this bug: https://bugzilla.kernel.org/show_bug.cgi?id=12309 for a while now. They also mention some patches. Not sure if it's the same patches, but it's the same issue.

                      Man I will be happy. Biggest thing holding back linux on the desktop.

                      Comment

                      Working...
                      X