[lkml]   [2006]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
    SubjectRe: rt20 scheduling latency testcase and failure data
    On Fri, 12 May 2006 19:24:53 -0700
    Darren Hart <> wrote:

    > The test case emulates a periodic thread that wakes up on time%PERIOD=0, so
    > rather than sleeping the same amount of time each round, it checks now
    > against the start of its next period and sleeps for that length of time.
    > Every so often it will miss it's period, I've captured that data and included
    > a few of the interesting bits below. The results are from a run with a
    > period of 5ms, although I have seen them with periods as high as 17ms. The
    > system was under heavy network load for some of the time, but not all.


    > I'd appreciate any feedback on the test case, and in particular suggestions on
    > how I can go about determining where this lost time is being spent.

    There's a multitude of ways how you can misconfigure your -rt system :)
    Tell us more about your setup. Hardware? Full preemption? High
    resolution timers? Priority setup? From your code i see you run at prio
    98. What about the IRQ handlers? And the softirq's, too? Other software?


    P.S.: I ran the test a few [20 or so] times and didn't get any failures
    of the sort you see. Even with a 1ms period:

    ~/downloads$ ./sched_latency_lkml
    Scheduling Latency

    Running 10000 iterations with a period of 1 ms
    Expected running time: 10 s

    --------- --------- ------------- --------
    10000 32 47 0

    Start Latency: 305 us: FAIL
    Min Latency: 16 us: PASS
    Avg Latency: 29 us: PASS
    Max Latency: 47 us: PASS
    Failed Iterations: 0

    ~/downloads$ uname -a
    Linux mango.fruits 2.6.16-rt20 #4 PREEMPT Wed May 10 12:53:39 CEST 2006 i686 GNU/Linux

    Ooops, i must admit i have the nvidia binary only kernel module loaded,
    but i suppose this wouldn't make a difference for the better ;)

    I got high resolution timers enabled and left the softirq threads at
    their defaults.

    Palimm Palimm!
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to
    More majordomo info at
    Please read the FAQ at

     \ /
      Last update: 2006-05-13 11:24    [W:0.023 / U:12.500 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site