2008-02-26 03:37:23

by Krzysztof Halasa

[permalink] [raw]
Subject: ath5k bf=xxx bf_skb=0 warnings

Hi,

Just wondering if "bf=xxx bf_skb=0" warnings printed to console
(ath5k/base.c:ath5k_beacon_send) are normal?

Linux ixp425 2.6.25-rc3 (ARM big-endian), AR5413

iwconfig wlan0 essid $ESSID mode ad-hoc channel $CH
iwconfig wlan0 key restricted $KEY
ifconfig wlan0 up

I understand sc->bbuf->skb is never set, the only possible place is
ath5k_beacon_update() but it's never called.
--
Krzysztof Halasa


2008-02-26 13:32:35

by Krzysztof Halasa

[permalink] [raw]
Subject: Re: ath5k bf=xxx bf_skb=0 warnings

bruno randolf <[email protected]> writes:

> yes, that's normal and harmless. i haven't got around to make a smarter
> solution yet.
>
> that's because we call ath5k_beacon_config() each time after a channel is set
> or the chip is reset, even if the beacon has not been configured yet.

I see, thanks.
--
Krzysztof Halasa

2008-02-26 06:11:42

by Bruno Randolf

[permalink] [raw]
Subject: Re: ath5k bf=xxx bf_skb=0 warnings

On Tuesday 26 February 2008 12:37:21 Krzysztof Halasa wrote:
> Hi,
>
> Just wondering if "bf=xxx bf_skb=0" warnings printed to console
> (ath5k/base.c:ath5k_beacon_send) are normal?
>
> Linux ixp425 2.6.25-rc3 (ARM big-endian), AR5413
>
> iwconfig wlan0 essid $ESSID mode ad-hoc channel $CH
> iwconfig wlan0 key restricted $KEY
> ifconfig wlan0 up
>
> I understand sc->bbuf->skb is never set, the only possible place is
> ath5k_beacon_update() but it's never called.

yes, that's normal and harmless. i haven't got around to make a smarter
solution yet.

that's because we call ath5k_beacon_config() each time after a channel is set
or the chip is reset, even if the beacon has not been configured yet.

we should fix that at some point.

bruno