Re: [Bluez-devel] Bypass for L2CAP packets

[Marcel Holtmann <marcel@holtmann.org>]

Hi Matthias,

> sorry for a question to the same problem again...
> I have a (problem) with the L2CAP layer from BlueZ.
> I wrote a program that communicates with a mobile phone by using its ow=
n L2CAP=20
> layer (the program builds its own L2CAP packets for protocol testing).
> For example I send a valid "connection request" to that phone and the p=
hone=20
> answers with a valid "connection response".
> So far so good.=20
> But then the L2CAP layer from BlueZ sends a "command reject", because t=
his=20
> layer doesn=B4t know about the "connection request" and cannot handle=20
> the "connection response".
> Ok, i understand that.=20
> But has anybody an idea how i can communicate with my own L2CAP layer a=
nd the=20
> phone only without using the BlueZ L2CAP layer, because the Bluez layer=
 always=20
> prevents a further communication?
> Hmm... is in this case the L2CAP RAW socket the solution?
> I=B4m stucked with that problem and i hope anybody as an idea.

as Xavier mentioned the easiest solution is to unload the L2CAP kernel
module. However I looked at the code and it seems that the the kernel
respond to packets with all idents. It should not do that, because they
are shared between the kernel and the userspace. I think a patch for
that is needed so that for commands from the RAW socket no responses are
send.

Regards

Marcel




-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel