netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: James Chapman <jchapman@katalix.com>
To: "David S. Miller" <davem@davemloft.net>, bcrl@kvack.org
Cc: netdev@oss.sgi.com, kleptog@svana.org, mostrows@styx.uwaterloo.ca
Subject: Re: PPP-over-L2TP kernel support, new patch for review
Date: Tue, 21 Sep 2004 10:55:25 +0100	[thread overview]
Message-ID: <1095760525.414ffa8d111a1@www.katalix.com> (raw)
In-Reply-To: <20040920141704.16085067.davem@davemloft.net>

Quoting "David S. Miller" <davem@davemloft.net>:

> On Mon, 20 Sep 2004 22:11:44 +0100
> James Chapman <jchapman@katalix.com> wrote:
>
> > Attached is a revised version of the new PPP over L2TP support for
> > review. Thanks DaveM and Herbert for comments so far. The following
> > comments have been addressed in this new version:
>
> What relation does your work have to the L2TP implementation
> being worked on by Ben LaHaise?  See:
>
> http://marc.theaimsgroup.com/?l=linux-netdev&m=109375044707414&w=2
>
> Do we have two people working on this thing. :-/
>
> Ben didn't post any pointers to his work so I couldn't do the
> comparison myself.
>

Ben and I are working on separate projects. I was unaware of his work
until I saw his netdev post a few weeks ago and mailed him privately to find
out more. He's using the old Babylon (Spellcaster) proprietary PPP
stack that has now been GPL'd.

Unfortunately I haven't seen Ben's code yet either so I can't give a
direct comparison. Ben? I did have a look at the Babylon stuff
(1.6-pre3), although I've no idea how much of it Ben has
changed. Here's a summary, fyi.

Babylon:-

- Architecture seems to be using char devices for communication with
  the kernel and all the PPP datapath is handled by custom virtual
  net_devices; the generic PPP kernel code isn't used as far as I can
  tell. Unfortunately it is very old (linux-2.0 era I think) but Ben
  has probably updated it.

- Some form of L2TP support is there but it is very basic. Userspace
  sends data through char devices (read()/write() which the kernel
  char driver converts to skbs and passes on. Nasty.

- PPP stack supports multiple PPP sessions in one daemon (unlike pppd).

- Unlikely to integrate with the new native IPSEC stuff.

OpenL2TP:-

- Communication with kernel is through a new PPPoL2TP socket family.
  There's one socket per L2TP session so MAX_FILES limits max
  sessions. Works with the new native IPSEC kernel code.

- Comprehensive userspace L2TP protocol implementation written from
  scratch, targetted specifically for enterprise VPN and embedded
  networking products. Efficient kernel datapath was deemed essential
  for this environment.

- Plugin architecture allows different PPP implementations to be used.
  Only pppd supported so far (limits max sessions still further due
  to process overhead) but I'm working on a daemon to support multiple
  sessions -- still early stages, evaluating alternatives. Babylon or
  hacking pppd or start again...

rp-l2tp:-

- No kernel datapath (all data copied into userspace through ptys).
  However, it could be modified to use the socket based kernel driver.

I think for general Linux L2TP support, a socket architecture makes
more sense. But maybe I'm biased... :)

If you want to find out more about OpenL2TP, checkout the
online man pages at http://openl2tp.sourceforge.net/.

BTW, I asked on linux-ppp if anyone was working on a single daemon PPP
to handle multiple sessions but got zero response.  Anyone on this
list know of any work in this area?

I hope this was useful.

/james

  reply	other threads:[~2004-09-21  9:55 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-09-20 21:11 PPP-over-L2TP kernel support, new patch for review James Chapman
2004-09-20 21:17 ` David S. Miller
2004-09-21  9:55   ` James Chapman [this message]
2004-09-21 21:04     ` Benjamin LaHaise
2004-09-21 23:07       ` Herbert Xu
2004-09-22  0:00         ` Michael Richardson
2004-09-22  1:14         ` Benjamin LaHaise
2004-09-22  2:42           ` David S. Miller
2004-09-22  3:03           ` jamal
2004-09-22  9:58       ` James Chapman
2004-09-22 10:53         ` Herbert Xu
2004-09-22 21:28           ` James Chapman
2004-09-22 22:05             ` Herbert Xu
2004-09-23 18:34             ` Martijn van Oosterhout
2004-09-21 21:11   ` Benjamin LaHaise

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=1095760525.414ffa8d111a1@www.katalix.com \
    --to=jchapman@katalix.com \
    --cc=bcrl@kvack.org \
    --cc=davem@davemloft.net \
    --cc=kleptog@svana.org \
    --cc=mostrows@styx.uwaterloo.ca \
    --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).