bridge.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: voxner <voxner.dev@gmail.com>
To: bridge@lists.linux-foundation.org
Subject: [Bridge] RSTP-Linux Bridge (2.6 Kernel) query
Date: Mon, 19 Apr 2010 15:05:35 +0530	[thread overview]
Message-ID: <i2xe5da85b41004190235t447ea22exa046c63fc3373971@mail.gmail.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 1934 bytes --]

Hi all,

I am fairly new to Bridging, and I have been trying to understand the
Bridge-RSTP communication. I have an RSTP implementation running in
user-space and I would like to use the existing Linux bridge data-structures
(like net_bridge, net_bridge_port used in Linux 2.6) so that I have a
Linux-Bridge/RSTP setup. For this I need to introduce some sort of
communication mechanism (maybe IOCTL calls) to alter these data-structures
by my RSTP.

My idea is to disable STP (thus making sure that routines like
br_configuration_update() are not called) and then use a communication
mechanism to alter bridge/bridge-port states(stored in net_bridge) of kernel
from my RSTP.

From my analysis, I found the following instances of communication

Bridge -> RSTP Communication
1. An RSTP BPDU (control) packet is received (like the routine br_stp_rcv in
existing linux)
2. When a bridge is added or removed
3. When a port in a bridge is added or removed

RSTP -> Bridge Communication
1.Setting a Bridge-Port state
2.Setting Bridge-Ports pirority
3.Getting number of ports.
4.Command to send a BPDU packet
etc .......

I would like to know if I am on the right track. To summarize I "disable"
all those STP routines, introduce a bunch of routines that set bridge/ports
based on communication received from the RSTP running in user-space.

I have looked at the differences between STP and RSTP but I am unsure as to
how this will play out in my solution. Should I go for another strategy? Is
there any similar work out there that does this? I would really appreciate
any input/pointers in this matter.

Thanks,
Vox

P.S. To get an insight into this problem I took a cursory look at another
RSTP implementation called RSTPLib (http://rstplib.sourceforge.net) but it
uses daemons, sysfs & crucially it is not "integrated" into the kernel. I am
also unclear if it's purely on a simulation tool. Also its code seems no
longer under development.

[-- Attachment #2: Type: text/html, Size: 2148 bytes --]

             reply	other threads:[~2010-04-19  9:35 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-19  9:35 voxner [this message]
2010-04-19 16:00 ` [Bridge] RSTP-Linux Bridge (2.6 Kernel) query Stephen Hemminger
2010-04-21  8:10   ` voxner

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=i2xe5da85b41004190235t447ea22exa046c63fc3373971@mail.gmail.com \
    --to=voxner.dev@gmail.com \
    --cc=bridge@lists.linux-foundation.org \
    /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).