netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Patrick McHardy <kaber@trash.net>
To: "Waskiewicz Jr, Peter P" <peter.p.waskiewicz.jr@intel.com>
Cc: netdev@vger.kernel.org
Subject: Re: Tc filtering: broken 802_3 classifier?
Date: Thu, 26 Jul 2007 01:40:37 +0200	[thread overview]
Message-ID: <46A7DF75.5090609@trash.net> (raw)
In-Reply-To: <D5C1322C3E673F459512FB59E0DDC329034BFCD2@orsmsx414.amr.corp.intel.com>

Waskiewicz Jr, Peter P wrote:
>>In case of prio, if your manually installed filters don't 
>>match, it will fall back to the skb->priority based 
>>classification, which is based on tos and is probably 
>>responsible for what you're seeing. Feel free to investigate, 
>>but you could save us all some time by simply posting what 
>>you're doing, what you're expecting and what is actually 
>>happening, there's probably a good explanation.
> 
> 
> I thought I did that before, but I probably wasn't clear.  I'll try
> again (and if I'm still not clear, please pop me in the head).  I am
> aware that skb->priority is used if no filter matches, and that is
> derived from tos (and gets set in ipsockglue).
> 
> This is my setup.  8 bands with prio, with a priomap that is nice and
> simple:
> 
> # tc qdisc add dev eth0 root handle 1: prio bands 8 priomap 0 0 1 1 2 2
> 3 3 4 4 5 5 6 6 7 7
> 
> With this configuration, ICMP will default to flowid 1:1 (band 0), and
> ssh will default to flowid 1:4 (band 3) based on TOS.  I add this filter
> (802_3) and all traffic starts flowing into flowid 1:1 (including ssh),
> even though it should never match:
> 
> # tc filter add dev eth0 protocol 802_3 parent 1: prio 2 u32 match u32
> 0x00000800 0x0000ffff at 12 flowid 1:6
> 
> As soon as I remove the filter:
> 
> # tc filter del dev eth0 protocol 802_3 prio 2
> 
> ssh flows back into flowid 1:4.  No filters of protocol ip were added,
> only the 802.3 filter.
> 
> I hope this is more clear as to what I'm seeing.


It is .. now let me think about the good explanation, it doesn't
make sense at first :)


  reply	other threads:[~2007-07-25 23:41 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-07-25  0:19 Tc filtering: broken 802_3 classifier? Waskiewicz Jr, Peter P
2007-07-25  0:31 ` Patrick McHardy
2007-07-25 17:24   ` Waskiewicz Jr, Peter P
2007-07-25 17:28     ` Patrick McHardy
2007-07-25 17:34       ` Waskiewicz Jr, Peter P
2007-07-25 23:11         ` Patrick McHardy
2007-07-25 23:34           ` Waskiewicz Jr, Peter P
2007-07-25 23:40             ` Patrick McHardy [this message]
2007-07-25 23:58               ` Patrick McHardy
2007-07-26  0:10                 ` David Miller

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=46A7DF75.5090609@trash.net \
    --to=kaber@trash.net \
    --cc=netdev@vger.kernel.org \
    --cc=peter.p.waskiewicz.jr@intel.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).