Re: hfsc cbq inaccuracy

[Michal Soltys]

Well, I have tested a bit more, including DFBSD. Although it handled the 
traffic with limits a bit better (otoh, the computer was much slower - p2 
333, so it might have had some influence), it still had problems. Option HZ 
bumped to 1000 (as is recommended in most cases, when altq is in question) 
solved the problem - accuracy was perfect.


I managed to compile OBSD 3.9 with HZ option (but it's not a documented one 
and I needed small patch posted not so long ago at openbsd-tech - 
http://marc.theaimsgroup.com/?l=openbsd-techm=113673508428056w=2 ). PF 
with hfsc is razor sharp now, and my OBSD seems perfectly fine.


Still - are there any reasons why HZ is left undocumented in case of OBSD ? 
Any potential issues that could show up after adjusting it ?

Re: hfsc cbq inaccuracy

[Michal Soltys]

Small correction - that table 1.5 vs. 12 was from some other test. This
is what I meant to post (hfsc, tbrsize 12k,
bandwidth/realtime/link/upper
set to the same value, although anything besides upperlimit
didn't matter much):

theoretical / actual
  8 /  8
 10 / 10
 12 / 11.95
 14 / 13.24
 16 / 14.13
 18 / 14.18
 20 / 14.27
 22 / 14.16
 24 / 14.50
 26 / 14.70
 28 / 14.60
 30 / 16.36
 32 / 16.48
 34 / 16.38
 36 / 16.39
 38 / 16.38
 40 / 17.28
 42 / 18.88
 44 / 19.46
 46 / 24.82
 48 / ~25.5
 50 / ~27