Hi, I've cleaned up and removed the txop-queue-limiting and scheduling from the patch (compared to my last RFC). It's still to early for the scheduling thing to go prime time. The fair queuing on the other hand does seem to work. In good RF conditions it seems to improve things (e.g. multiple TCP streams converge into a steady average). In bad RF conditions things look same grim as before (but not worse). I've done a few more experiments with naive DQL in ath10k and some flent tests prove the fair queuing in mac80211 works better than fq_codel qdisc as far as wake_tx_queue drivers are concerned. I'll be posting a separate thread after this. This is based on mac80211-next/master (0a87cadbb54e1595a5f64542adb4c63be914d290). Michal Kazior (2): mac80211: implement fair queuing per txq mac80211: expose some txq/fq internals and knobs via debugfs include/net/mac80211.h | 21 ++- net/mac80211/agg-tx.c | 8 +- net/mac80211/codel.h | 264 +++++++++++++++++++++++++++++ net/mac80211/codel_i.h | 89 ++++++++++ net/mac80211/debugfs.c | 86 ++++++++++ net/mac80211/debugfs_netdev.c | 29 +++- net/mac80211/debugfs_sta.c | 46 +++++ net/mac80211/ieee80211_i.h | 45 ++++- net/mac80211/iface.c | 24 ++- net/mac80211/main.c | 9 +- net/mac80211/rx.c | 2 +- net/mac80211/sta_info.c | 10 +- net/mac80211/sta_info.h | 27 +++ net/mac80211/tx.c | 384 +++++++++++++++++++++++++++++++++++++----- net/mac80211/util.c | 20 ++- 15 files changed, 983 insertions(+), 81 deletions(-) create mode 100644 net/mac80211/codel.h create mode 100644 net/mac80211/codel_i.h -- 2.1.4 -- 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
