On 28/01/14 14:36, Johannes Berg wrote: > On Tue, 2014-01-28 at 14:31 +0100, Antonio Quartulli wrote: >> On 28/01/14 14:01, Kalle Valo wrote: >>>> - Patch is getting resent because I did not CC'd the linux-wireless mailing list in >>>> my previous attempt. :) >>> >>> Oh, I missed that. Anyway, in previous version I commented about >>> documenting NULL usage. What do you think about that? >> >> Well documentation is always good :) If it is need I will add it. >> >> However, I'd rather ask you what ath6kl is trying to do in that case, >> since it's the only driver invoking cfg80211_ibss_joined() with a zero >> BSSID (and now a NULL channel) on IBSS leave. >> >> From the cfg80211 prospective, that call is seen like a notification of >> being connected to a new BSS having zero mac and NULL channel. I don't >> see how this can be useful at all, also because after leaving the IBSS I >> imagine that no other operation will take place on that vif unless it >> reconnects again. > > That seems completely pointless and should probably be removed? I totally agree. But we should do it with another patch targeting mac80211-next. Cheers, -- Antonio Quartulli
Attachment:
signature.asc
Description: OpenPGP digital signature
