Ticket #858 (closed defect: fixed)

Opened 5 years ago

Last modified 5 years ago

[patch] Using a raw socket in 802.11 Radiotap mode to transmit does not work correctly

Reported by: abpost@mpi-sws.mpg.de Assigned to:
Priority: major Milestone: version 0.9.3
Component: madwifi: 802.11 stack Version: trunk
Keywords: radiotap transmit raw packets Cc:
Patch is attached: 1 Pending:

Description

Currently sending to a 802.11 radiotap mode raw socket with prepended tx_radiotap structure does not work correctly. Currently this is not handled at all and goes to the default case of the switch statement. Thus the tx_radiotap header is not processed correctly. In this patch, it is now handled correctly. It reads the correct values and sets the physical parameters for the transmission. The tx_radiotap header is removed, and the packet is handled correctly. This allows the user to send packets with prepeneded tx_radiotap headers and adjust the power/rate of these packets.

Signed-off-by: Ansley Post <abpost@mpi-sws.mpg.de>

Attachments

madwifi.patch (0.7 kB) - added by abpost@mpi-sws.mpg.de on 09/05/06 17:47:04.
patch for transmitting using 802.11 radiotap
010-true_radiotap_parser.diff (4.3 kB) - added by xmxwx@asn.pl on 10/19/06 01:55:14.

Change History

09/05/06 17:47:04 changed by abpost@mpi-sws.mpg.de

  • attachment madwifi.patch added.

patch for transmitting using 802.11 radiotap

09/05/06 18:01:57 changed by mentor

Looks good.

09/05/06 20:18:43 changed by mrenzmann

  • milestone set to version 0.9.3.

Only one small thing to mention from my side: the block that encloses the instructions of the whole "case" is unnecessary and should be removed before the patch is submitted, and there should be a space between the if and the bracket. Other than that it looks ready to be committed.

09/06/06 04:06:34 changed by mentor

That's the style in the rest of the file; if you want it changed, you'll have to do it yourself :P

09/06/06 04:24:58 changed by mentor

  • status changed from new to closed.
  • resolution set to fixed.

Commited in r1708

09/06/06 06:14:19 changed by mrenzmann

@mentor: Indeed, I already started such work months ago, but didn't finish it yet. Guess I have to pick that up when I have a bit more time left over (wonder if that ever will be the case, though).

Thanks for committing this patch, and thanks for the OP for the contribution.

09/06/06 06:22:44 changed by mentor

/me threatens to run madwifi through indent

09/06/06 07:47:25 changed by mrenzmann

In fact, this is the reason why I stopped my work on that task... to look up the parameters for indent and experimenting a bit. But something else then asked for my attention and made me suspend the indent idea for a while. However, you're welcome to take over that part if you like ;)

10/18/06 19:57:32 changed by xmxwx@asn.pl

  • status changed from closed to reopened.
  • resolution deleted.

IMHO, fixing the input to use only madwifi's tx_radiotap appears to be inconvenient when writing software which uses Monitor VAP as raw radio interface. Developer would rather expect from the driver to conform to the radiotap standard (i.e. to provide proper parsing of the header instead of using its limited subset). My suggestion is to provide such a parser so that maximum compatibility with userspace programs can be achieved. This has been already done and applied to Lintrack's madwifi patchset:

http: //dev.lintrack.org/browser/trunk/packages/madwifi/patches/010-true_radiotap_parser.diff.bz2?format=raw

Signed-off-by: Michal Wrobel <xmxwx@asn.pl>

10/19/06 01:55:14 changed by xmxwx@asn.pl

  • attachment 010-true_radiotap_parser.diff added.

10/23/06 11:15:06 changed by kelmo

  • status changed from reopened to closed.
  • resolution set to fixed.

Applied: r1761.

Add/Change #858 ([patch] Using a raw socket in 802.11 Radiotap mode to transmit does not work correctly)