Re: ath device appears broken

4 msg	'kernel: rtfree: 0xc3c4bd98 has 2 refs' on rece...
8 msg	hard-lock with CPU spinning
3 msg	ZFS with USB 'SCSI'
3 msg	fdc0 doesn't attach on my current machine
7 msg	Best way to build a 'new' AMD64 current
9 msg	LOR in iwi with new 802.11 work
2 msg	panic in drm_ioctl
5 msg	Re: smb related problem
9 msg	Disappearing userland output with -current
4 msg	BIND 9.4.1 performance on FreeBSD 6.2 vs. 7.0
4 msg	i386/i386/busdma_machdep.c 1.88 breaks isp

ath device appears broken
\ Steve Kargl (14 Jun 2007)
. \ Andrew Thompson (14 Jun 2007)
. \ Sam Leffler (14 Jun 2007)
. . \ Steve Kargl (14 Jun 2007)
. . . \ Sam Leffler (14 Jun 2007)
. . . . \ Steve Kargl (14 Jun 2007)
. . . . . \ Sam Leffler (14 Jun 2007)

8 msg	KTR and SCTP
2 msg	old bugs
6 msg	ZFS tuning tips?
1 msg	(Delayed) HEADS UP: AUDIT in GENERIC
13 msg	HEADS UP: IPX over IP support removed
4 msg	RFC: diff(1) update
2 msg	Call for testers, amd64 only, new scheduler.
4 msg	HPS stack and ural

Subject:	Re: ath device appears broken
Group:	Freebsd-current
From:	Sam Leffler
Date:	14 Jun 2007

Steve Kargl wrote:
> On Wed, Jun 13, 2007 at 10:05:40PM -0700, Sam Leffler wrote:
>> Steve Kargl wrote:
>>> I can't find anything relevent at the moment. The 20070610 entry
>>> states:
>>> Users must also, for the moment, kldload the wlan_scan_sta and/or
>>> wlan_scan_ap modules if they use modules for wireless support.
>>>
>>> I don't use modules nor compile them. Perhaps, a sentence is
>>> missing that should direct people that don't use module to update
>>> their kernel config file?
>> Thanks, I'll add more explanation. Short answer is the scanning support
>> is now broken out into separate modules that you must either load or
>> specify in your kernel config file. wlan_scan_sta for sta mode
>> operation. wlan_sta_ap for ap mode operation. This is comparable to
>> how ciphers are split out into modules.
>
> OK. I've added all the wlan_* devices to my config
> and the kernel boots and my script now runs.
>
> Thanks for responding so quickly.
>
> PS: I seem to get a periodic "ath0: device timeout" message
> on the console. I doesn't seem to cause any problems (e.g.,
> long pauses).
>

No idea. I suspect there's a race in the tx done processing that makes
the driver think there is a timeout when one is not present.

Sam

_______________________________________________
freebsd-current mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe"