From: Eliad Peller Date: Tue, 5 Jan 2016 14:28:06 +0000 (+0200) Subject: mac80211: handle sched_scan_stopped vs. hw restart race X-Git-Url: http://drtracing.org/?a=commitdiff_plain;h=2bc533bd9dcf48eaf4af6fb89a338734a9e8f76e;p=deliverable%2Flinux.git mac80211: handle sched_scan_stopped vs. hw restart race On hw restart, mac80211 might try to reconfigure already stopped sched scan, if ieee80211_sched_scan_stopped_work() wasn't scheduled yet. This in turn will keep the device driver with scheduled scan configured, while both mac80211 and cfg80211 will clear their sched scan state once the work is scheduled. Fix it by ignoring ieee80211_sched_scan_stopped() calls while in hw restart, and flush the work before starting the reconfiguration. Signed-off-by: Eliad Peller Signed-off-by: Emmanuel Grumbach Signed-off-by: Johannes Berg --- diff --git a/net/mac80211/main.c b/net/mac80211/main.c index ed4c8e66b44a..8190bf27ebff 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -248,6 +248,7 @@ static void ieee80211_restart_work(struct work_struct *work) /* wait for scan work complete */ flush_workqueue(local->workqueue); + flush_work(&local->sched_scan_stopped_work); WARN(test_bit(SCAN_HW_SCANNING, &local->scanning), "%s called with hardware scan in progress\n", __func__); diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index a413e52f7691..8eb68ef42e8c 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -1213,6 +1213,14 @@ void ieee80211_sched_scan_stopped(struct ieee80211_hw *hw) trace_api_sched_scan_stopped(local); + /* + * this shouldn't really happen, so for simplicity + * simply ignore it, and let mac80211 reconfigure + * the sched scan later on. + */ + if (local->in_reconfig) + return; + schedule_work(&local->sched_scan_stopped_work); } EXPORT_SYMBOL(ieee80211_sched_scan_stopped);