Thanks for the report. I will see if I can see the large
delta-t values with my domapp test, domapptest.c (in dor-driver/driver,
if you want to try it yourself). Unless this is a high priority
item I would prefer to wait until all the newpkt communications
problems are solved.
Hello icebuggers, particularly John J,
I've implemented a new test in my datacollector output
verification routine which examines the delta-T between hit
records. Looking at CPU triggered events, I see two distinctly
different 'dead' periods in data between the usual ~400us spaced CPU
events. First, I observe regular pauses of ~40ms which are
presumably caused by buffer readout, and we know this time should be
dead. Also, I observe much larger irregular pauses. These
pauses are infrequent ( roughly 5% of 40ms pause rate ) and can range
to slightly over a second in duration.
Here is an example:
( Units in 10^-10 seconds )
d893596f5ab1: Significant lag: 415169330 time: 5443370105623697
d893596f5ab1: Significant lag: 417093828 time: 5443370795107249
d893596f5ab1: Significant lag: 413047083 time: 5443371480680556
d893596f5ab1: Significant lag: 412592583 time: 5443372164431364
d893596f5ab1: Significant lag: 416565078 time: 5443372852289668
d893596f5ab1: Significant lag: 416736329 time: 5443373541751471
d893596f5ab1: Significant lag: 412734332 time: 5443374226352528
d893596f5ab1: Significant lag: 422602822 time: 5443374921198325
d893596f5ab1: Significant lag: 412013333 time: 5443375605539383
d893596f5ab1: Significant lag: 410899584 time: 5443376288425191
*d893596f5ab1: Significant lag: 7027859640 time:
5443383588635806*
d893596f5ab1: Significant lag: 414512831 time: 5443384275996861
d893596f5ab1: Significant lag: 418089827 time: 5443384965876163
Thanks,
Jim B