2010-11-21 10:47:59

by Roberto Riggio

[permalink] [raw]
Subject: Use case for the radiotap rate patch

In the following thread Pavel was sacking for an use case for his patch to
support the radiotap rate field;

http://thread.gmane.org/gmane.linux.kernel.wireless.general/52229/focus=52265

I'm currently using the patch in a custom routing protocol for wireless mesh
networking (http://www.wing-project.org). The protocol uses broadcast frames sent
at rates different from the baserate in order to implement its routing
metric
(WCETT). Without the patch I would be forced to sample each neighbor using
unicast frames making the metric implementation not scalable with the number
of nodes.

Also, which would be the way to support the multi-rate retry chain from
user-space?
I'm currently playing with some custom RC algorithms in users space
using the
click modular router, but I cannot take advantage of the mrr.

thanks
R.


2010-11-22 15:39:45

by Johannes Berg

[permalink] [raw]
Subject: Re: Use case for the radiotap rate patch

On Sun, 2010-11-21 at 11:47 +0100, Roberto Riggio wrote:

> Also, which would be the way to support the multi-rate retry chain from
> user-space?
> I'm currently playing with some custom RC algorithms in users space
> using the
> click modular router, but I cannot take advantage of the mrr.

radiotap now has a way to specify one attribute multiple times, so that
should be used for MRR

johannes


2010-11-22 18:00:21

by Roberto Riggio

[permalink] [raw]
Subject: Re: Use case for the radiotap rate patch

Il 22/11/2010 16:39, Johannes Berg ha scritto:
> radiotap now has a way to specify one attribute multiple times, so that
> should be used for MRR
Is there a patch available for the linux kernel that supports this
new approach to specify the MRR? I can easly modify the click
radiotap header generation code, but without the support in
the kernel it would be rather pointless.

R.
> johannes
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html


2010-11-22 18:01:08

by Johannes Berg

[permalink] [raw]
Subject: Re: Use case for the radiotap rate patch

On Mon, 2010-11-22 at 19:00 +0100, Roberto Riggio wrote:
> Il 22/11/2010 16:39, Johannes Berg ha scritto:
> > radiotap now has a way to specify one attribute multiple times, so that
> > should be used for MRR
> Is there a patch available for the linux kernel that supports this
> new approach to specify the MRR? I can easly modify the click
> radiotap header generation code, but without the support in
> the kernel it would be rather pointless.

The kernel parser will support this, but you'd have to provide the code
to actually use multiple rates as MRR

johannes