ath0: device timeout with 802.11n client

[Lev Serebryakov]

Hello, Adrian.
You wrote 29 июля 2012 г., 15:23:30:

AC> I saw, thanks.
AC> Please enable reset debugging too - sysctl dev.ath.X.debug=0x20.
  Done :) I'll post new debug messages as soon as they'll appear.

  BTW,  I've  tried  association  debugging  and  got  a  TONS of such
  messages when client (Blackberry PlayBook) goes to sleep (but it is
  still associated):

Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x11
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x13
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x11
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x13
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x11
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x13
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x11
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x13
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x11
Jul 29 15:29:05 gateway kernel: wlan0: [00:0c:42:64:69:92] HT bss occupancy change: 1 sta, 1 ht, 0 ht40, non-HT sta present, HT protmode now 0x13

  I've turned assotiation debugging off (and posted folloup to PR).

AC> Jul 27 04:24:09 gateway kernel: ath0: ath_bar_response: called;
AC> tap=0xc48c32b8, atid=0xc48c4df4, txa_tid=0, atid->tid=0, status=1,
AC> attempts=50
AC> Jul 27 04:24:09 gateway kernel: ath0: ath_tx_tid_bar_unsuspend:
AC> tid=0xc48c4df4, called
AC> Jul 27 04:24:09 gateway kernel: ath0: ath_bar_response: called;
AC> tap=0xc48c32b8, atid=0xc48c4df4, txa_tid=0, atid->tid=0, status=1,
AC> attempts=50
AC> Jul 27 04:24:09 gateway kernel: ath0: ath_tx_tid_bar_unsuspend:
AC> tid=0xc48c4df4, called
AC> Jul 27 04:24:09 gateway kernel: ath0: ath_tx_tid_bar_unsuspend:
AC> bar_tx=0, bar_wait=0: ?
AC> That's very odd. I wonder if I broke that specific handling @ attempt=50.
 As side note: I'm sure, that client was sleeping at this time, as I
 was sleeping for sure :)

-- 
// Black Lion AKA Lev Serebryakov <lev at FreeBSD.org>