Date: Wed, 3 May 2006 18:55:24 -0700 From: John-Mark Gurney <gurney_j@resnet.uoregon.edu> To: tpeixoto@widesoft.com.br Cc: Lee Johnston <lee@wildcard.net.uk>, "."@babolo.ru, Julian Elischer <julian@elischer.org>, mihai@duras.ro, freebsd-net@freebsd.org Subject: Re: Packet loss with traffic shaper and routing Message-ID: <20060504015524.GV728@funkthat.com> In-Reply-To: <44595B76.9010901@widesoft.com.br> References: <1146645702.297895.80691.nullmailer@cicuta.babolo.ru> <44595B76.9010901@widesoft.com.br>
next in thread | previous in thread | raw e-mail | index | archive | help
tpeixoto@widesoft.com.br wrote this message on Wed, May 03, 2006 at 22:40 -0300:
> Anyway, I am very curious about the result of test 2. Why do the pipes
> have influence on system performance if there is nothing passing through
> them?
It looks like each tick all the pipes are scanned... In dummynet:
/* Sweep pipes trying to expire idle flow_queues. */
for (i = 0; i < HASHSIZE; i++)
SLIST_FOREACH(pipe, &pipehash[i], next)
That bit of code should probably be run less often...
--
John-Mark Gurney Voice: +1 415 225 5579
"All that I will do, has been done, All that I have, has not."
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060504015524.GV728>