Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 18 Sep 2006 19:01:45 -0400
From:      Jung-uk Kim <jkim@FreeBSD.org>
To:        "Christian S.J. Peron" <csjp@FreeBSD.org>
Cc:        cvs-src@FreeBSD.org, src-committers@FreeBSD.org, cvs-all@FreeBSD.org
Subject:   Re: cvs commit: src/sys/dev/bge if_bge.c
Message-ID:  <200609181901.47414.jkim@FreeBSD.org>
In-Reply-To: <450F1EC1.4060106@FreeBSD.org>
References:  <200609182218.k8IMIMUT059300@repoman.freebsd.org> <450F1EC1.4060106@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 18 September 2006 06:33 pm, Christian S.J. Peron wrote:
> Jung-uk Kim wrote:
> > jkim        2006-09-18 22:18:22 UTC
> >
> >   FreeBSD src repository
> >
> >   Modified files:
> >     sys/dev/bge          if_bge.c
> >   Log:
> >   Do not strip VLAN tag in promiscuous mode.
> >
> >   Revision  Changes    Path
> >   1.148     +28 -15    src/sys/dev/bge/if_bge.c
>
> I don't think this is right. An interface does not have to be in
> promiscuous mode in order to have BPF peers attached to it. If this
> is being done to ensure that vlan info is preserved for things like
> tcpdump, the correct fix is to use bpf_peers_present() instead.

Yes, I know that.  This is just a stopgap until we have 'correct' fix.  
BTW, I am just following bad example, i.e., em(4). ;-)

> This issue has been discussed between myself, jhb, and andre for a
> couple of days now. The correct fix is to teach bpf to look at the
> mbuf and re-construct and insert the vlan tags before passing it
> through bpf_filter (and on to any peers), assuming the hardware is
> handling stripping and inserting vlan tags. I have agreed to work
> on this, time permitting.

It's good to hear that somebody is working on it.

Thanks!

Jung-uk Kim



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200609181901.47414.jkim>