> -----Original Message----- > From: Igor Mitsyanko [mailto:igor.mitsyanko.os@xxxxxxxxxxxxx] > Sent: Saturday, June 29, 2019 00:29 > To: Otcheretianski, Andrei <andrei.otcheretianski@xxxxxxxxx>; > hostap@xxxxxxxxxxxxxxxxxxx > Subject: Re: [PATCH 5/6] AP: Add support for co-located AP configuration > > On 6/19/19 5:49 AM, Andrei Otcheretianski wrote: > > + if (!h || !h->started || !h->iface->current_mode || h == hapd) > > + continue; > > Hi Andrei, a few questions on when RnR element should be added: > - it should only be added to 2.4GHz and 5GHz BSS frames as far as I know. The spec mandates to add RNR element on 2.4GHz and 5GHz co-located AP's. I didn't find any restriction to add RNR on 6GHz AP, but I agree that it doesn't make much sense. > - shouldn't we only add RnR for 6Ghz interfaces that are beaconing? That is, > should we trace when interface start or stop beaconing (I believe !h->started is > not enough) > - should we track channel changes? Update RnR if interface switches away from > 6GHz channel, or switches to 6Ghz from other band. Yep. We should track both start/stop and channel switches. Need to update beacons in these cases. > - would it make sense for hostapd to automatically add RnR elements for any > beaconing 6GHz BSS to any 2.4GHz and 5GHz IEs so that user won't have to > specify co0located interfaces in configuration? What does spec says, is RnR > element mandatory in co-located configuration, or it is optional? According to 802.11ax/D4.1 "26.17.2.4 Out of band discovery of a 6 GHz BSS" it is indeed mandatory: "An AP that operates in the 2.4 GHz or 5 GHz bands and that is co-located with one or more APs that operate in the 6 GHz band *shall* include in Beacon and Probe Response frames that it transmits a Reduced Neighbor Report element with the Co-Located AP subfield in the TBTT Information Header subfield set to 1 to provide at least the operating channels and operating classes of the co-located APs in the 6 GHz band" So indeed, according to this, the configuration is not really needed. Andrei _______________________________________________ Hostap mailing list Hostap@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/hostap