[wip] ar9300 hostap support

[Joshua Isom]

On 3/28/2013 3:15 PM, Adrian Chadd wrote:
> On 28 March 2013 13:11, Joshua Isom <jrisom at gmail.com> wrote:
>
>>> I'd like to see if any other TX queue has a frame hanging around in it
>>> that hasn't been completed.
>>
>>
>> I had syslogd pipe the output to a perl script to set the sysctl, not ideal
>> but effective I hope.  I've got two today.
>>
>>> Mar 28 07:03:01 jri kernel: ath0: ath_edma_tx_processq: Q1: empty?
>
> .. interesting.
>
>
> This is the interesting bit.
>
>>> Mar 28 07:03:02 jri kernel: HW TXQ 0: axq_depth=0, axq_aggr_depth=0,
>>> axq_fifo_depth=0, holdingbf=0
>>> Mar 28 07:03:02 jri kernel: HW TXQ 1: axq_depth=0, axq_aggr_depth=0,
>>> axq_fifo_depth=0, holdingbf=0
>>> Mar 28 07:03:02 jri kernel: HW TXQ 2: axq_depth=0, axq_aggr_depth=0,
>>> axq_fifo_depth=0, holdingbf=0
>>> Mar 28 07:03:02 jri kernel: HW TXQ 3: axq_depth=0, axq_aggr_depth=0,
>>> axq_fifo_depth=0, holdingbf=0
>>> Mar 28 07:03:02 jri kernel: HW TXQ 8: axq_depth=0, axq_aggr_depth=0,
>>> axq_fifo_depth=0, holdingbf=0
>
> Right. None of these have any frames stuck in the FIFO. Good. Now, I
> have to go chase down why we're seeing this.
>

I just noticed something odd, it counts from 0, 1, 2, 3, 8.  It skips 
4-7 and 9.  Why aren't the rest set up?

>
> oh I have attenuators. :-)
>
> Is it plausibly 7-9 dB in your environment? :-) I'm worried that I've
> not done the chip setup right and it's acting deaf-y.
>

Every device has poor reception in that room, but FreeBSD is the only 
one I can get numbers from.  That's why the antenna I'm doing is 
directional.  My laptop has an Intel chip, and even Intel's "Advanced 
Statistics" gives me no real numbers.  The best I could give you is a 
"bar" count.

>
> Adrian
>