realtime-preempt-2.6.12-final-V0.7.51-11 glitches

From: Rui Nuno Capela
Date: Thu Jul 07 2005 - 13:51:16 EST

Hi all,

These are one of my latest consolidated results while using (my)
jack_test4.2 suite, against a couple of 2.6.12 kernels patched for
PREEMPT_RT, on my P4@xxxxxx/UP laptop.

See anything funny?

As it seems, the kernel latency performance is in some unfortunate
regression, and I'm experiencing this unsatisfactory behavior ever since
the latest RT-V0.7.50-xx patch series.

------------------------------ ------------- -------------
RT-V0.7.51-11 RT-V0.7.49-01
------------------------------ ------------- -------------
Total seconds ran . . . . . . : 900 900
Number of clients . . . . . . : 14 14
Ports per client . . . . . . : 4 4
Frames per buffer . . . . . . : 64 64
Number of runs . . . . . . . : 1 1
------------------------------ ------------- -------------
Failure Rate . . . . . . . . : ( 0.0 ) ( 0.0) /hour
XRUN Rate . . . . . . . . . . : 373.3 0.0 /hour
Delay Rate (>spare time) . . : 220.0 0.0 /hour
Delay Rate (>1000 usecs) . . : 0.0 0.0 /hour
Delay Maximum . . . . . . . . : 7853 295 usecs
Cycle Maximum . . . . . . . . : 852 943 usecs
Average DSP Load. . . . . . . : 41.8 44.4 %
Average CPU System Load . . . : 6.8 16.3 %
Average CPU User Load . . . . : 28.8 30.1 %
Average CPU Nice Load . . . . : 0.0 0.0 %
Average CPU I/O Wait Load . . : 0.0 0.1 %
Average CPU IRQ Load . . . . : 0.0 0.0 %
Average CPU Soft-IRQ Load . . : 0.0 0.0 %
Average Interrupt Rate . . . : 1679.3 1680.4 /sec
Average Context-Switch Rate . : 12508.6 14463.2 /sec
------------------------------ ------------- -------------

JFYI respective kernel configs are also attached.

Cheers.
--
rncbc aka Rui Nuno Capela
rncbc@xxxxxxxxx

Attachment: jack_test4.2.tar.gz
Description: application/gzip-compressed

Attachment: config-2.6.12-RT-V0.7.49-01.0.gz
Description: Binary data

Attachment: config-2.6.12-RT-V0.7.51-11.0.gz
Description: Binary data