netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Kundrat <kundrat@kundrat.sk>
To: netdev@oss.sgi.com
Subject: cbq parent doesnt limit child
Date: Mon, 7 Oct 2002 00:22:31 +0200	[thread overview]
Message-ID: <20021006222231.GA7763@napri.sk> (raw)

Hello,

I experienced following unexpected behavior wrt. cbq - 
if child_rate > parent_rate .. child gets its full bandwidth,
so apparently parent_rate is not respected.

In http://lartc.org/howto/lartc.qdisc.classful.html is following
comment (section 9.5.2.2):

  In short, nested classes ONLY talk to their parent qdiscs, never to an
  interface. Only the root qdisc gets dequeued by the kernel!

  The upshot of this is that classes never get dequeued faster than their
  parents allow. And this is exactly what we want: this way we can have
  SFQ in an inner class, which doesn't do any shaping, only scheduling,
  and have a shaping outer qdisc, which does the shaping.

Observed behavior doesnt correspond to that description though.

I understand that configuring child_rate > parent_rate doesnt make much
sense (except for correctness test), so that behavivor could be considered 
a feature, not a bug.  In that case, maybe only documentation/howto should 
be corrected, correct?

Thanks,

Peter

PS: For the record, its kernel 2.4.19pre7 and this setup:
  tc qdisc add dev eth0  root  handle 2: cbq bandwidth 10mbit avpkt 1000 mpu 64
  # root class
  tc class add dev eth0 parent 2:0 classid 2:1 est 1sec 8sec cbq \
   bandwidth 10mbit rate 10mbit maxburst 1 weight 10mbit prio 1 \
   allot 1514 avpkt 1000
 
  tc class add dev eth0 parent 2:1 classid 2:10 est 1sec 8sec cbq \
    bandwidth 10mbit rate 8000 maxburst 1 weight 8000 prio 1 \
    allot 1514 avpkt 1000 bounded
  tc class add dev eth0 parent 2:10 classid 2:20 est 1sec 8sec cbq \
    bandwidth 10mbit rate 16000 maxburst 1 weight 16000 prio 1 \
    allot 1514 avpkt 1000 bounded

  tc filter add dev eth0 parent 2:0 prio 59900 protocol ip u32 \
    match ip dst 10.10.10.111/32 flowid 2:20


-- 
Peter Kundrat
peter@kundrat.sk

                 reply	other threads:[~2002-10-06 22:22 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20021006222231.GA7763@napri.sk \
    --to=kundrat@kundrat.sk \
    --cc=netdev@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).