ath10k tx99-like test mode?

[Simon Wunderlich]

On Monday 16 March 2015 14:21:10 Ben Greear wrote:
> On 03/16/2015 01:51 PM, mrex at tranzeo.com wrote:
> > Ben,
> > 
> > The tx99 feature has the benefit of not needing any client or traffic to
> > be
> > passed for a power meter to measure.  So it’s incredibly fast to test
> > different rates and output power than having to wait for client to
> > re-associate, pass traffic and adjust power meter to duty cycle (depending
> > on how fancy the meter is).  This is more applicable for calibration and
> > evaluation than for adaption or protocol testing.
> > 
> > Tx99 is better suited for calibration and FCC certification testing and
> > not
> > for testing things like adaptive modulation or what happens when mixed
> > modulations are enabled.  It cuts down on lab certification test time and
> > complexity (lab engineers test a different wireless widget every day and
> > not familiar with webGUI’s and config settings, etc).  Tx99 often was
> > more ‘raw’ and didn’t behave the same as adaptive modulation algorithm,
> > probably more so with more streams than 1 and 11b/g/n.
> > 
> > So having a feature that can just output raw frames at full duty cycle at
> > a
> > specific rate and output power without client would be highly beneficial
> > for lab testing and reduce test time and complexity for radio validation.
> I tried to get ath10k firmware to send raw frames on a monitor interface,
> but it seems it requires a peer of some sort before it will send frames,
> and I had no real luck.
> 
> I'm imagine there is some way to make firmware able to do this, but I don't
> know how to do that at this time.

Thanks for all the input so far!

It is probably not too hard to implement a packet generating logic like in 
ath9k/tx99 for ath10k to remove the dependency on an external tool.

However the firmware question is a more tricky one, and we can't easily change 
that - we will not operate with a peer in the lab after all. We can try to go 
with mcast frames and see what duty cycle we can achieve.

 Ben, your firmware allows to set HT and VHT rates on mcast frames too, right?

Is the QCA driver supporting this test? I guess they could need it as well. :)

Thanks again,
    Simon
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.infradead.org/pipermail/ath10k/attachments/20150317/bd4f08e6/attachment.sig>