Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 19 Feb 2013 08:48:11 -0800
From:      Adrian Chadd <adrian@freebsd.org>
To:        Kamil Szczesny <k.s.mail@gmx.net>
Cc:        freebsd-wireless@freebsd.org
Subject:   Re: ath0: device timeout on 9.1-RELEASE
Message-ID:  <CAJ-Vmo=SRZMt5AsEzvomTWKLSyztP0u9e_TA6wiy8g6sviuh6w@mail.gmail.com>
In-Reply-To: <CAJ-Vmo=LZZcLo-mjBXzrkNYnqzTr-NpBFgeQBUBCk8D4X2w3JA@mail.gmail.com>
References:  <51175C69.6050003@gmx.net> <CAJ-Vmo=5ReDqivQakH9Yuv20_wvqCqHtsxoqpYKhh=-RvpUt%2Bg@mail.gmail.com> <51189E1E.70206@gmx.net> <CAJ-Vmo=HLRkXb6NyU=tR_GbPPdu7DFQMsFrXD2qXBzdC%2BMb%2BxQ@mail.gmail.com> <51194795.5070500@gmx.net> <CAJ-VmomB7t-tEb=LCUZkGqsv0YQqqQEFT-Zq8sndqAbJsNZJsA@mail.gmail.com> <511E910D.1080901@gmx.net> <CAJ-Vmon9FskQtUdZ_nnTer6n7cGznqWQ6c4MZTm4DqT9cNTTng@mail.gmail.com> <511E9446.4070708@gmx.net> <511E9615.3020007@gmx.net> <CAJ-VmokfSBqsOrGr7GRvyvjvA61mo9Dx5uTinPJyfs%2BuK4CoUg@mail.gmail.com> <512376AB.10003@gmx.net> <CAJ-Vmo=we%2Bx_8pohC=BVu0euEdg7VnHBABLNcJwpk6rNhfye_g@mail.gmail.com> <CAJ-Vmo=%2Bm=nLSBWYM0kdSWqP7vsZAg%2B52z0mtWvMmBZXPQvAtQ@mail.gmail.com> <5123A714.6050105@gmx.net> <CAJ-Vmo=LZZcLo-mjBXzrkNYnqzTr-NpBFgeQBUBCk8D4X2w3JA@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 19 February 2013 08:32, Adrian Chadd <adrian@freebsd.org> wrote:
> 0x2003 will make things appear. Just do an active scan (ifconfig wlanX
> scan) to ensure something happens.

Hi,

(Cc'ing the list so people can see the progress of debugging this.)

It looks like this happens:

* a channel change occurs;
* beacons are received;
* probe request frames are queued to the hardware;
* but before the TX completes (either fail or succeed) the queue gets
stopped and reset.

So either it's not going out, or the timers are all screwed up and
it's immediately resetting. It's hard to tell given the current
logging (ie, no microsecond resolution timestamps, sigh.)

Can you try changing the timecounter?

sysctl kern.timecounter

set it to something obviously inefficient, like "i8254":

sysctl kern.timecounter.hardware=i8254

.. and similarly the eventtimer:

sysctl kern.eventtimer.timer=i8254

and see if that changes things.

It looks like the scan times default to 200ms per channel in
ieee80211_scan_sta.c. It's tunable, but the scan code in ifconfig
doesn't have any option to change it. 200ms should be plenty of time
to finish calibrating and start sending frames though..

Thanks,


Adrian



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmo=SRZMt5AsEzvomTWKLSyztP0u9e_TA6wiy8g6sviuh6w>