From: hristo (hristo_at_mail.galaxite.net)
Date: 2002-03-03 10:42:17 UTC
I would like to take a look at code too. Will you be so kind to post a link?
Thanks in advance
> We (a collegue and myself) modified the linux-wlan-ng
> driver to support 802.11b compliant WDS. It doesn't work.
> The firmware/card silently drops the transmit frames
> without generating a TxExc. Intersil's response for us
> was something like, "Yes it's broken and it might get
> fixed, kinda, maybe, probably not." In frustration, we
> wrote a non-standard WDS implementation using a "magic"
> mac address in the A3 header field of 00:00:de:ad:be:ef.
> At least I'm pretty sure it was the A3 field. Anyway, the
> driver would check for that mac address and if found,
> ignore the dest mac address in the 802.11 header and
> reconstruct the 802.3 frame from the payload only. The
> easiest way to do that was just force the "encapsulation"
> encoding method on all WDS stations' transmitted frames.
> Realizing that was a horrible kluge, we decided on a
> lesser kluge. We setup a proc configurable list of mac
> addresses to check for instead of the magic
> 00:00:de:ad:be:ef. So in that case, you just add the mac
> address for any "WDS" stations. Then when the AP receives
> a frame from a station in the list, it assumes it's an
> encapsulated frame and ignores the dest address on the
> 802.11 frame and netif_rx's the encapsulated frame instead
> of attempting to reconstruct the header.
> The scariest part was that both ways worked. Only we ran
> into interesting problems with broadcast frames but that
> would take a lot of explaining. If anyone isn't scared out
> of their mind and wants to look at the code, I'll post a
> link. We stopped without working out all the bugs and
> cleaning up the interface because I LEARNED HOW TO ROUTE
> and stop bridging like a wuss. Sorry, I had to. We
> realized we were going through all this trouble to avoid
> routing. It's not worth it so get over it. :P
> Kristian Hoffmann
> PC-InTOUCH/FireToWire System Administrator
> On Fri, 1 Mar 2002, Jerritt Collord wrote:
> > Well it's been a while since anyone mentioned this, so
> I'll > venture out cautiously... what's up with WDS
> implementations for > HostAP?
> > OpenAP came out and there looked to be interest and work
> towards > implementing their system, but concerns arose
> that it broke > 802.11 spec.
> > Jim Thompson questioned the Prism2 firmware support for
> WDS... > zero threadage on that.
> > Thanks,
> > Jerritt Collord