Re: Linux 2.6.0-test6

From: bill davidsen (davidsen_at_tmr.com)
Date: 10/09/03

  • Next message: Linus Torvalds: "Re: [RFC] disable_irq()/enable_irq() semantics and ide-probe.c"
    To: linux-kernel@vger.kernel.org
    Date:	9 Oct 2003 18:44:56 GMT
    
    

    In article <ddcbaa61f5ab6ec90c71a70bb3990b49@stdbev.com>,
    Jason Munro <jason@stdbev.com> wrote:
    | On October 2, 7:07 pm Nick Piggin <piggin@cyberone.com.au> wrote:
    | >
    | > Pedro Larroy wrote:
    | >
    | > > On Thu, Oct 02, 2003 at 01:05:36PM +1000, Nick Piggin wrote:
    | > > I'm afraid this selection criteria leads to a scheduler that isn't
    | > > predictable for situations that aren't the ones for which is tuned to
    | > > work. Of course I may be wrong, but to me, seems that saying
    | > > explicitly which tasks are interactive sounds better.
    | > >
    | >
    | > Have a look at my scheduler if you like. It won't estimate interactivity
    | > but it works quite well if you nice -10 your X server. Ie. explicitly
    | > state which process should be favoured.
    | > http://www.kerneltrap.org/~npiggin/v15a/
    |
    | I don't know much about kernel internals but of the 2.5 and 2.6 kernels I
    | have tried, 2.6.0-test6 is by far the best on the desktop for my use (xmms,
    | vmware, firebird, loads of other apps). With this patch it's better still.
    | Before patching simple things like ls or ps have an annoying slowness while
    | under a moderate/heavy load. For the most part things are fine but after
    | patching commands respond more quickly. This is the first time for me a
    | 2.5+ kernel has been responsive enough to use on a daily basis.

    I really like my "np15a" patch, but it doesn't seem to play well with
    preempt in terms of performance. Stability is fine so far.

    -- 
    bill davidsen <davidsen@tmr.com>
      CTO, TMR Associates, Inc
    Doing interesting things with little computers since 1979.
    -
    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: Linus Torvalds: "Re: [RFC] disable_irq()/enable_irq() semantics and ide-probe.c"

    Relevant Pages

    • 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)
    • Major 2.4 / 2.{5|6}-smp Scheduler Slow-ness
      ... Ok, first of all, I have never been able to run a 2.4 kernel on this ... much so that this is a scheduler issue. ... When i finally let the 2.6.0-test2-mm1 boot with smp support, ... 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)
    • Re: Bug: ll_rw_blk.c, elevator.c and displaying "default" IO Schedule r at boot-time (Cosm
      ... On Wed, Mar 09 2005, Roberts-Thomson, James wrote: ... > io scheduler", where XXX was one of the IO schedulers (cfq, anticipatory, ... etc) depending on kernel .config at compile time. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: call for more SD versus CFS comparisons (was: Re: [ck] Mainline plans)
      ... problems on my Amarok machine. ... usually test Amarok playback while kernel package make, ... Putting load onto the VM layer / block layer. ... also the behavior on a server for example - how does the scheduler behave ...
      (Linux-Kernel)