On 09/28/2010 03:01 AM, Bruno Randolf wrote:
> On Tue September 28 2010 06:06:28 [email protected] wrote:

>>      /* configure operational mode */
>>      ath5k_hw_set_opmode(ah, sc->opmode);
>> @@ -698,13 +760,13 @@ ath5k_txbuf_setup(struct ath5k_softc *sc, struct
>> ath5k_buf *bf, flags |= AR5K_TXDESC_RTSENA;
>>              cts_rate = ieee80211_get_rts_cts_rate(sc->hw, info)->hw_value;
>>              duration = le16_to_cpu(ieee80211_rts_duration(sc->hw,
>> -                    sc->vif, pktlen, info));
>> +                    NULL, pktlen, info));
>
> hmm, this NULL means we don't handle short preamble and erp correctly. i don't
> know if we did before, but it would be better to use the corresponding vif - i
> think it can be found in ieee80211_tx_info *info.

This is nasty though...you can't trust that the vif still exists.
It could have been deleted right after pushing this packet, I think.

I'd rather not have to ask mac80211 to do a lookup here.

Perhaps if I forced a flush of all tx pkts when an interface
was deleted?

Thanks,
Ben

-- 
Ben Greear <[email protected]>
Candela Technologies Inc  http://www.candelatech.com

_______________________________________________
ath5k-devel mailing list
[email protected]
https://lists.ath5k.org/mailman/listinfo/ath5k-devel

Reply via email to