On Fri, Aug 26, 2005 at 02:11:07PM +0200, Jiri Benc wrote: > On Fri, 26 Aug 2005 11:04:15 +0200, Stefan Rompf wrote: > > there has been a discussion about AVS or radiotap header on the ipw list > > Does it have an archive on web? If so, could you post an address of that > thread?
It's archived off the sourceforge page (http://ipw2200.sourceforge.net or http://sourceforge.net/mailarchive/forum.php?forum_id=38938) The radiotap patches are also @ http://www.kismetwireless.net/code/linux-radiotap along with a short explanation of why I thought they were a good idea. The libpcap patch will be going into the next libpcap version, we just need the arphrd defined if/when this goes mainline (803 being the one that makes sense, and if I recall the one madwifi uses for their raidotap emit). Basically radiotap is just a set of bitfields that indicate the included attributes, with room for expansion and a dynamic total header size. Drivers include the link attributes that they're able to provide. Ethereal and tcpdump already had support, kismet had support on *bsd and I just flipped some config bits to turn that on in Linux as well. One of the first static header fields is the total header size, so any apps that don't care about signal headers (other 80211 stuff like aircrack, airsnort, etc) can just skip them like they'd skip AVS with minimal recoding. Radiotap has a few things AVS doesn't currently have, like FCS byte indication in the header (critical for sniffer apps like Kismet and anything else that decodes the frame and needs to know if 4 trailing bytes are tacked on). If you already emit prism2 or avs headers, switching to emitting radiotap is about 10 minutes worth of code. -m -- Mike Kershaw/Dragorn <[EMAIL PROTECTED]> GPG Fingerprint: 3546 89DF 3C9D ED80 3381 A661 D7B2 8822 738B BDB1 "Mommy, is that Mother Nature?" "No, dear. That's a ninja." -- Overheard in a comic store.
pgpko9glTp9FL.pgp
Description: PGP signature
