* Re: [Battlemesh] wifibroadcast
[not found] ` <20170422153703.GE1681@makrotopia.org>
@ 2017-04-22 22:01 ` Simon Wunderlich
2017-04-25 14:37 ` Johannes Berg
0 siblings, 1 reply; 2+ messages in thread
From: Simon Wunderlich @ 2017-04-22 22:01 UTC (permalink / raw)
To: Daniel Golle; +Cc: battlemesh, gnunet-developers, Johannes Berg, linux-wireless
[-- Attachment #1: Type: text/plain, Size: 2023 bytes --]
Hi Daniel,
On Saturday, April 22, 2017 5:37:03 PM CEST Daniel Golle wrote:
> Hi Simon,
>
> On Fri, Apr 21, 2017 at 09:55:03PM +0200, Simon Wunderlich wrote:
> > Hi Daniel,
> >
> > > mac80211: Parse legacy and HT rate in injected frames
> >
> > Yup, Sven was so kind and helped with this. :D
> >
> > I'm experimenting with some opportunistic routing ideas ...
>
> I just realized that the radiotap headers are still not part of the
> UAPI exported by the kernel. How is one supposed to go about this
> issue? Have lots of diverging local copies of ieee80211_radiotap.h?
Propose a patch and send it to linux wireless to move it? :)
> And change all the kernel-specific types (u16 and such) into their
> corresponding userland types (u_int16)...?
Hmm, uapi is full with __u16, __u32, ... see linux/types.h
>
> I'd be glad if you can share your idea about proper and/or intentional
> use of the kernel's radiotap capabilities...
Well, there is a radiotap parser in the kernel which is quite useful for
reading (net/wireless/radiotap.c). And there is a radiotap library which also
does parsing [1] (seems it's based on the kernel implementation), referenced
from radiotap.org [2]. What is missing (or at least I didn't find it and
implemented something on my own) is a library/function to build radiotap
headers (i.e. for writing). There are quite a few things to consider (padding,
alignment, etc). To make it useful, you would also need to build ieee80211
headers at the same time ...
I have some rough working C code for all of that (integrated into something
bigger), not sure if it would be useful to make a library out of it.
I don't have an idea about "proper" usage though. Since Johannes seems to
maintain most of that radiotap stuff, I'm CC'ing him and linux-wireless. Maybe
there is already more than what we know. :)
Cheers,
Simon
[1] https://github.com/radiotap/radiotap-library/blob/master/radiotap.c
[2] http://www.radiotap.org/
[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [Battlemesh] wifibroadcast
2017-04-22 22:01 ` [Battlemesh] wifibroadcast Simon Wunderlich
@ 2017-04-25 14:37 ` Johannes Berg
0 siblings, 0 replies; 2+ messages in thread
From: Johannes Berg @ 2017-04-25 14:37 UTC (permalink / raw)
To: Simon Wunderlich, Daniel Golle
Cc: battlemesh, gnunet-developers, linux-wireless
On Sun, 2017-04-23 at 00:01 +0200, Simon Wunderlich wrote:
> What is missing (or at least I didn't find it
> and implemented something on my own) is a library/function to build
> radiotap headers (i.e. for writing). There are quite a few things to
> consider (padding, alignment, etc).
Yeah, that could be done - but generally it's actually quite easy since
a given application will only have a given subset of radiotap
information it generates... wrapping all of that with some kind of
helper function could result in pretty bloated code.
> To make it useful, you would also need to build ieee80211 headers at
> the same time ...
I don't really agree - these can and should be separate IMHO.
Anyway, there isn't any such thing today.
johannes
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2017-04-25 14:37 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <CANjd3nfeiGgaHMcgOyKOKhRRKEwnn43VgC4GRcw-fuCkCdg2zQ@mail.gmail.com>
[not found] ` <2945991.ay5rM9MeUZ@prime>
[not found] ` <20170422153703.GE1681@makrotopia.org>
2017-04-22 22:01 ` [Battlemesh] wifibroadcast Simon Wunderlich
2017-04-25 14:37 ` Johannes Berg
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).