aboutsummaryrefslogtreecommitdiffstats
path: root/net/mac80211/iface.c
diff options
context:
space:
mode:
authorToke Høiland-Jørgensen <toke@redhat.com>2021-06-23 15:47:55 +0200
committerJohannes Berg <johannes.berg@intel.com>2021-06-23 18:12:00 +0200
commit2433647bc8d983a543e7d31b41ca2de1c7e2c198 (patch)
treea171345b667b374056a290f946c12327e611207c /net/mac80211/iface.c
parentRevert "mac80211: HE STA disassoc due to QOS NULL not sent" (diff)
downloadlinux-dev-2433647bc8d983a543e7d31b41ca2de1c7e2c198.tar.xz
linux-dev-2433647bc8d983a543e7d31b41ca2de1c7e2c198.zip
mac80211: Switch to a virtual time-based airtime scheduler
This switches the airtime scheduler in mac80211 to use a virtual time-based scheduler instead of the round-robin scheduler used before. This has a couple of advantages: - No need to sync up the round-robin scheduler in firmware/hardware with the round-robin airtime scheduler. - If several stations are eligible for transmission we can schedule both of them; no need to hard-block the scheduling rotation until the head of the queue has used up its quantum. - The check of whether a station is eligible for transmission becomes simpler (in ieee80211_txq_may_transmit()). The drawback is that scheduling becomes slightly more expensive, as we need to maintain an rbtree of TXQs sorted by virtual time. This means that ieee80211_register_airtime() becomes O(logN) in the number of currently scheduled TXQs because it can change the order of the scheduled stations. We mitigate this overhead by only resorting when a station changes position in the tree, and hopefully N rarely grows too big (it's only TXQs currently backlogged, not all associated stations), so it shouldn't be too big of an issue. To prevent divisions in the fast path, we maintain both station sums and pre-computed reciprocals of the sums. This turns the fast-path operation into a multiplication, with divisions only happening as the number of active stations change (to re-compute the current sum of all active station weights). To prevent this re-computation of the reciprocal from happening too frequently, we use a time-based notion of station activity, instead of updating the weight every time a station gets scheduled or de-scheduled. As queues can oscillate between empty and occupied quite frequently, this can significantly cut down on the number of re-computations. It also has the added benefit of making the station airtime calculation independent on whether the queue happened to have drained at the time an airtime value was accounted. Co-developed-by: Yibo Zhao <yiboz@codeaurora.org> Signed-off-by: Yibo Zhao <yiboz@codeaurora.org> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com> Link: https://lore.kernel.org/r/20210623134755.235545-1-toke@redhat.com Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Diffstat (limited to 'net/mac80211/iface.c')
-rw-r--r--net/mac80211/iface.c3
1 files changed, 3 insertions, 0 deletions
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index 68375ef56b4a..1e5e9fc45523 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -1977,6 +1977,9 @@ int ieee80211_if_add(struct ieee80211_local *local, const char *name,
}
}
+ for (i = 0; i < IEEE80211_NUM_ACS; i++)
+ init_airtime_info(&sdata->airtime[i], &local->airtime[i]);
+
ieee80211_set_default_queues(sdata);
sdata->ap_power_level = IEEE80211_UNSET_POWER_LEVEL;