Re: [patch] Real-Time Preemption, -RT-2.6.10-rc1-mm3-V0.7.25-1

[Ingo Molnar <mingo@elte.hu>]

* Mark_H_Johnson@raytheon.com <Mark_H_Johnson@raytheon.com> wrote:

>  [1] major network delays while latencytest is running (ping drops
> packets or they get delayed by minutes). I did not see this on some
> previous tests where I made more of the /0 and /1 tasks RT. May have
> to do that again.

i think this is directly related to what priority the ksoftirqd threads
have.

>  [6] the latency trace may have some SMP race conditions where the
> entries displayed do not match the header. Examples are a 100 usec
> trace header followed by 8 entries that last about 4 usec.

i think i fixed a related bug in the latest kernel(s):
touch_preempt_timing() was mistakenly 'touching' a live user-triggered
trace and could interfere in a similar fashion. Please re-report if this
still happens with -V0.7.26-3-ish or later kernels.

>  [8] Some samples of /proc/loadavg during my big test showed some
> extremely large numbers. For example:
> 5.07 402.44 0.58 5/120 4448

i'm currently trying to track down this one. The rq->nr_uninterruptible
count got out of sync during one of the scheduler changes - and this
causes large negative task counts, messing up the load-average.

	Ingo