On Thu, 2004-08-19 at 06:38, Lee Revell wrote:

On Thu, 2004-08-19 at 06:29, Lee Revell wrote:

quoted

Yes, this takes care of it.  Now the dominant latency is the 142us
latency from the via-rhine driver, which is fixed by using the driver
from -mm (specifically it's fixed in bk-netdev.patch).

OK, this is a new one:

This is the other common one:

preemption latency trace v1.0.1
-------------------------------
 latency: 74 us, entries: 9 (9)
    -----------------
    | task: XFree86/493, uid:0 nice:-10 policy:0 rt_prio:0
    -----------------
 => started at: schedule+0x5b/0x590
 => ended at:   schedule+0x2f0/0x590
=======>
 0.000ms (+0.000ms): schedule (work_resched)
 0.000ms (+0.000ms): sched_clock (schedule)
 0.001ms (+0.001ms): dequeue_task (schedule)
 0.001ms (+0.000ms): recalc_task_prio (schedule)
 0.002ms (+0.000ms): effective_prio (recalc_task_prio)
 0.002ms (+0.000ms): enqueue_task (schedule)
 0.004ms (+0.002ms): __switch_to (schedule)
 0.072ms (+0.067ms): finish_task_switch (schedule)
 0.073ms (+0.000ms): sub_preempt_count (schedule)

Not sure what to make of this one.

Lee