Major 2.4 / 2.{5|6}-smp Scheduler Slow-ness

From: root (root_at_rachel.trevorj.com)
Date: 07/31/03

  • Next message: Marc Heckmann: "UP IO-APIC fix in 2.4.22-pre?"
    Date:	Wed, 30 Jul 2003 20:27:27 -0400
    To: linux-kernel@vger.kernel.org
    
    
    

    Ok, first of all, I have never been able to run a 2.4 kernel on this
    machine, it runs _very_ slow. By slow, i mean it takes literally HOURS
    to boot. I have tried 2.4.20,2.4.21, and also both with the openmosix
    patches.

    Now, the 2.5/6 kernel runs beautifully, so i have been running both
    vanilla, and vanilla+mm patches, both run fine. Although, using the smp
    scheduler makes it do the exact same thing as 2.4, albeit a little
    faster. So this leads me into thinking *very*
    much so that this is a scheduler issue.

    When i finally let the 2.6.0-test2-mm1 boot with smp support, the cpu
    usage is ALWAYS at 75-80% System. And this is with all tasks sleeping
    except for top.

    I have brought this up many times with people on irc and such, and we
    are all stumped. So, i am bringing this up to your attention. i have
    attached the dmesg output ( 2.6.0-test2-mm1+smp ), lspci output, my
    .config, and my /proc/cpuinfo output. If you need anymore information, I
    will be happy to provide =).

    Also -- another thing to bring up with the scheduler problems. I have an
    onboard ethernet port (8139too), which only pukes with an error something like
    SCIO*: function not found ( replace wildcard ), if you want the full
    output of it, i can boot into it again and copy/paste it. Now, it is
    very important to realize that it only fails at the same times as i am
    having the issue above, otherwise it works fine.

    I am ready to try anything out to fix this issue, so if you have any
    suggestions, I shall try them.

    Thanks.

    >> Trevor Joynson <admin@trevorj.com>

    
    
    
    
    

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/






  • Next message: Marc Heckmann: "UP IO-APIC fix in 2.4.22-pre?"

    Relevant Pages

    • Re: RFC: [2.6 patch] better i386 CPU selection
      ... If we boot a 386 kernel on a ppro with that bug, ... For the handful of bytes saved in the mtrr driver, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: The system works (2.6.14-rc2): functional k8n-dl
      ... I was able to install full FC4 and boot without a problem ... > archives you will see that the nVidia SATA controller is something I ... > The kernel that came with FC4 worked just fine. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Linux 2.6.0-test6
      ... |> Have a look at my scheduler if you like. ... | I don't know much about kernel internals but of the 2.5 and 2.6 kernels I ... | patching commands respond more quickly. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • how to build a working 2.6 kernel with SATA
      ... I am trying to build a bootable 2.6 kernel for a Dell 8300, ... which has both the onboard ICH5 SATA controller and a PCI ... I have seen a variety of boot problems, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Linux 2.6.0-test6
      ... |> Have a look at my scheduler if you like. ... | I don't know much about kernel internals but of the 2.5 and 2.6 kernels I ... | patching commands respond more quickly. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)