On Sun, Dec 20, 2015 at 08:52:37AM +0000, Peer, Ilan wrote: > Thanks. Let me know if there are issues with the last patch. It does increase the risk of failures with deployed APs. I'm not at all convinced that they can reply quickly enough to GAS comeback requests in all cases.. I had actually experimented with this some time earlier but with even a shorted timeout (100 ms). wpa_supplicant does not currently retry GAS comeback requests at higher layer and it may be necessary to add such a retry mechanism before dropping the wait timeout on each individual frame. Other than that, the change itself is clearly of significant benefit for the cases where the first wait cannot be extended and mac80211 ends up waiting for significant amount of time between each attempt. I just don't want to optimize this at the cost of reliability to already deployed use cases. -- Jouni Malinen PGP id EFC895FA _______________________________________________ Hostap mailing list Hostap@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/hostap
