Date: Thu, 5 Feb 2004 19:46:08 -0800 From: mrcpu@cdsnet.net To: Brooks Davis <brooks@one-eyed-alien.net> Cc: current@freebsd.org Subject: Re: VLAn support for atheros/Prism? Message-ID: <20040206034608.GV50677@backmaster.cdsnet.net> In-Reply-To: <20040206004924.GB20813@Odin.AC.HMC.Edu> References: <20040203213456.GT50677@backmaster.cdsnet.net> <200402051417.00117.sam@errno.com> <20040206003112.GU50677@backmaster.cdsnet.net> <20040206004924.GB20813@Odin.AC.HMC.Edu>
next in thread | previous in thread | raw e-mail | index | archive | help
I quote specifically from the FreeBSD vlan man page:
The rest of the ethernet NICs supported by FreeBSD can run VLANs using
software emulation in the vlan driver. However, most of them lack the
capability of transmitting and/or receiving oversized frames. Using such
a NIC as a parent interface implies a reduced MTU on the corresponding
vlan interfaces. In the modern Internet, this is likely to cause tcp(4)
connectivity problems due to massive, inadequate icmp(4) filtering that
breaks the Path MTU Discovery mechanism.
Nothing I said anywhere contradicts the above paragraph.
As I recall I tried creating a vlan dev attached to a wireless interface,
and it came up as 1480 for an MTU, rather than 1500.
So the original question still stands. Is the issue with VLAN support over
wireless a function of the hardware not supporting 1500 byte packets + the VLAN
header? Or just something in the software?
On Thu, Feb 05, 2004 at 04:49:26PM -0800, Brooks Davis wrote:
> On Thu, Feb 05, 2004 at 04:31:12PM -0800, mrcpu@cdsnet.net wrote:
> > Well, I'm just looking through the specs of the microtik stuff, and they
> > say they have vlan support over wireless now.
> >
> > I can fudge the VLAn support, at the cost of a reduced MTU using the
> > "software emulation" vlan tagging, or whatever it is that FreeBSD does,
> > and bridge those packets, but that's not the same as true hardware VLAN
> > tagging.
>
> You've got a serious misconception about how vlans work. The only
> hardware support that is required is support for a wire MTU of 1504
> bytes so the tag fits on the packet without reducing payload size. Some
> interfaces have support for removing the tag from the packet for you and
> indicating the tag out of band, but that's only an optimization.
>
> -- Brooks
>
> --
> Any statement of the form "X is the one, true Y" is FALSE.
> PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4
> !DSPAM:4022e49c607282046518854!
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040206034608.GV50677>