softmac: Fix deadlock of wx_set_essid with assoc work
PATCH 1/1 softmac: Fix deadlock of wx_set_essid with assoc work
The essid wireless extension does deadlock against the assoc mutex, as we don't unlock the assoc mutex when flushing the workqueue, which also holds the lock.
--
David, please stresstest this.
John, please apply this, if David reports success from his tests. John, please notify me when you applied this, so I can immediately send it to -stable.
net/ieee80211/softmac/ieee80211softmac_wx.c +8 -3
@@ -74,8 +74,8 @@ ieee80211softmac_wx_set_essid(struct netstruct ieee80211softmac_auth_queue_item *authptr;int length =0 ; +check_assoc_again:mutex_lock (&sm->associnfo.mutex); -/* Check if we're already associating to this or another network * If it's another network, cancel and start over with our new network * If it's our network, ignore the change, we're already doing it!
@@ -98,13 +98,18 @@ ieee80211softmac_wx_set_essid(struct netcancel_delayed_work (&authptr->work); sm->associnfo.bssvalid =0 ; sm->associnfo.bssfixed =0 ; -flush_scheduled_work (); sm->associnfo.associating =0 ; sm->associnfo.associated =0 ; +/* We must unlock to avoid deadlocks with the assoc workqueue +* on the associnfo.mutex */ +mutex_unlock (&sm->associnfo.mutex); +flush_scheduled_work (); +/* Avoid race! Check assoc status again. Maybe someone started an +* association while we flushed. */ + goto check_assoc_again; } } - sm->associnfo.static_essid =0 ; sm->associnfo.assoc_wait =0 ;
Quoted 1 line · Michael Buesch
David, please stresstest this.
It seemed to survive being left with NetworkManager overnight, which the 2.6.22.1 kernel was consistently failing to cope with beforehand.
Signature · 1 line
dwmw2
Quoted 5 lines · David Woodhouse
Quoted 1 line · Michael Buesch
David, please stresstest this.
It seemed to survive being left with NetworkManager overnight, which the 2.6.22.1 kernel was consistently failing to cope with beforehand.
So John, please apply this.
Quoted 7 lines · Michael Buesch
Quoted 4 lines · David Woodhouse
Quoted 1 line · Michael Buesch
David, please stresstest this.
It seemed to survive being left with NetworkManager overnight, which the 2.6.22.1 kernel was consistently failing to cope with beforehand.
So John, please apply this.
ACK.
Larry