mac80211's sw scan doesn't play nice with buffered broadcast and multicast frames. Since drivers are required to keep track of the DTIM count this series provides some new basic functionality to let drivers help mac80211 make more prudent decisions for sw scan. Tested with an AP with a beacon interval of 100 and then two dtim intervals: 1, 2. I'll send out the PS fix separately, its an RFC because I am not aware of the impact of enabling PS after 2.6.34... did we *really* go on testing without PS since then? The good news is that at least with these patches my AR9003 device works well with PS enabled. Haven't tested the AR9002 family. You can force enable PS to test it by using iw: iw dev wlan0 set power_save on I believe we can expand on this to move more of the code ath9k does into mac80211 and just provide basic APIs for informating mac80211 what it should do. I'd appreciate if others would test, specially those without ath9k and using sw scan. Luis R. Rodriguez (3): ath9k: fix regression which disabled ps on ath9k on all cards mac80211: allow drivers to specify sw scan wait constraints ath9k: implement the sw scan wait constraints drivers/net/wireless/ath/ath9k/ath9k.h | 1 + drivers/net/wireless/ath/ath9k/init.c | 3 +- drivers/net/wireless/ath/ath9k/main.c | 69 ++++++++++++++++++++++++++++++++ include/net/mac80211.h | 12 ++++++ net/mac80211/driver-ops.h | 16 +++++++ net/mac80211/driver-trace.h | 18 ++++++++ net/mac80211/ieee80211_i.h | 2 + net/mac80211/scan.c | 42 +++++++++++++++++-- 8 files changed, 158 insertions(+), 5 deletions(-) -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
