Re: Re: 5.3-RELEASE: WARNING - WRITE_DMA interrupt timout - what does it mean?

From: Andrea Campi (andrea+freebsd_current_at_webcom.it)
Date: 11/10/04

  • Next message: Zoltan Frombach: "Re: Re: 5.3-RELEASE: WARNING - WRITE_DMA interrupt timout - what does it mean?"
    Date: Wed, 10 Nov 2004 10:25:29 +0100
    To: Robert Watson <rwatson@freebsd.org>
    
    

    Nitpicking mode on...

    On Wed, Nov 10, 2004 at 09:17:39AM +0000, Robert Watson wrote:
    > I'm still a bit skeptical that the task queue is at fault -- I run my
    > notebook with continuous measurement of the latency to schedule tasks,
    > generating a warning for any latency > .5 seconds, and the only time I
    [...]

    > @@ -187,8 +198,16 @@
    > pending = task->ta_pending;
    > task->ta_pending = 0;
    > task->ta_flags |= TAF_PENDING;
    > + tq_out++;
    > mtx_unlock(&queue->tq_mutex);
    >
    > + getnanotime(&tv);
    > + timespecsub(&tv, &task->ta_queuetime);
    > + if (tv.tv_nsec >= 0500000000) {

    Unless this is a typo while generating the patch for the list, and you
    are actually running with correct code, you're measuring latencies
    greater than 500M OCTAL, that is about 0.084 sec... Although that
    makes you argument even stronger I guess.

    Bye,
            Andrea

    -- 
          ...and that is how we know the Earth to be banana-shaped.
    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
    

  • Next message: Zoltan Frombach: "Re: Re: 5.3-RELEASE: WARNING - WRITE_DMA interrupt timout - what does it mean?"

    Relevant Pages

    • Re: RT patch acceptance (scheduler)
      ... > lock up the CPU in IRQ mode for human-perceptible time, ... For non-DMA IDE access data copies are CPU driven ... which can create tons of latency problems for that case. ... I suggest that you read the patch for the answer to softirq ...
      (Linux-Kernel)
    • RSDL development plans
      ... Can you try the attached patch please Al and Mike? ... This tends to make the max latency bound to a ... killing the latency of the non niced tasks. ... The rsdl patches queued up so far are stable and boot fine and are reasonably ...
      (Linux-Kernel)
    • Re: sky2 (was Re: 2.6.18-mm2)
      ... does not require the patch (in fact, ... Capabilities: Vital Product Data ... Device: Supported: MaxPayload 128 bytes, PhantFunc 0, ExtTag- ... Latency L0s unlimited, L1 unlimited ...
      (Linux-Kernel)
    • Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-Q5
      ... >regarding this particular latency, could you try the attached patch ... Alt-SysRq-T captured data, again everything I could look at in "S" mode. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [patch] voluntary-preempt-2.6.9-rc1-bk4-Q5
      ... >regarding this particular latency, could you try the attached patch ... Alt-SysRq-T captured data, again everything I could look at in "S" mode. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)