ath updates - please test

[Adrian Chadd]

On 27 December 2011 09:49, Martin Lesser
<ml-freebsd-wireless at bettercom.de> wrote:

> Unfortunately the "stuck beacons" appeared again - not as much as in STABLE
> in the past but probably too much to ignore:
>
> Dec 25 10:20:18 fs kernel: ath0: stuck beacon; resetting (bmiss count 4)
> Dec 25 10:20:27 fs last message repeated 29 times
> Dec 25 10:20:27 fs kernel: ath0: ath_raw_xmit: sc_inreset_cnt > 0; bailing
> Dec 25 10:20:40 fs kernel: ath0: stuck beacon; resetting (bmiss count 4)
> Dec 25 10:20:51 fs last message repeated 36 times
> Dec 25 10:20:51 fs kernel: ath0: ath_raw_xmit: sc_inreset_cnt > 0; bailing

Yup. :(

> I've set hw.ath.debug=2147483647 but I'm not sure whether that makes sense.
> Which debug-infos can I provide to help to get this bug out of the way? Are
> there plans to backport the actual /ath back to CURRENT?

"Actual ath" ? What do you mean? :)

>
> Additionally I ordered a D-Link DWA-547 [1] and hope not only to get rid of
> these stuck beacons but also to test 11n a little bit more.

Hi,

That's likely a bit much.

Also, the AR5416 isn't all that great a NIC, it's their first
generation 11n NIC. The newer stuff is a _lot_ nicer. So anything with
an AR9220/AR9280 or AR9227/AR9287 (PCI/PCIe) would be better.

That said, unless I find a mine of useful information about the older
pre-11n NICs, it's unlikely I'll find time/resources to track down and
fix bugs. The only thing I can think of is "overly noisy environment."
I need to write up some tools to make that easier to ascertain. Stay
tuned.

Thanks,



Adrian