mac80211: fix tx aggregation locking issue

Signed-off-by: Felix Fietkau <nbd@nbd.name>

1 files changed, 79 insertions, 0 deletions

diff --git a/package/kernel/mac80211/patches/subsys/310-mac80211-agg-tx-don-t-schedule_and_wake_txq-under-st.patch b/package/kernel/mac80211/patches/subsys/310-mac80211-agg-tx-don-t-schedule_and_wake_txq-under-st.patch
new file mode 100644
index 0000000000..008ee49cfb
--- /dev/null
+++ b/package/kernel/mac80211/patches/subsys/310-mac80211-agg-tx-don-t-schedule_and_wake_txq-under-st.patch
@@ -0,0 +1,79 @@
+From: Johannes Berg <johannes.berg@intel.com>
+Date: Mon, 29 Nov 2021 15:32:47 +0200
+Subject: [PATCH] mac80211: agg-tx: don't schedule_and_wake_txq() under
+ sta->lock
+
+When we call ieee80211_agg_start_txq(), that will in turn call
+schedule_and_wake_txq(). Called from ieee80211_stop_tx_ba_cb()
+this is done under sta->lock, which leads to certain circular
+lock dependencies, as reported by Chris Murphy:
+https://lore.kernel.org/r/CAJCQCtSXJ5qA4bqSPY=oLRMbv-irihVvP7A2uGutEbXQVkoNaw@mail.gmail.com
+
+In general, ieee80211_agg_start_txq() is usually not called
+with sta->lock held, only in this one place. But it's always
+called with sta->ampdu_mlme.mtx held, and that's therefore
+clearly sufficient.
+
+Change ieee80211_stop_tx_ba_cb() to also call it without the
+sta->lock held, by factoring it out of ieee80211_remove_tid_tx()
+(which is only called in this one place).
+
+This breaks the locking chain and makes it less likely that
+we'll have similar locking chain problems in the future.
+
+Reported-by: Chris Murphy <lists@colorremedies.com>
+Signed-off-by: Johannes Berg <johannes.berg@intel.com>
+Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
+---
+
+--- a/net/mac80211/agg-tx.c
++++ b/net/mac80211/agg-tx.c
+@@ -9,7 +9,7 @@
+  * Copyright 2007, Michael Wu <flamingice@sourmilk.net>
+  * Copyright 2007-2010, Intel Corporation
+  * Copyright(c) 2015-2017 Intel Deutschland GmbH
+- * Copyright (C) 2018 - 2020 Intel Corporation
++ * Copyright (C) 2018 - 2021 Intel Corporation
+  */
+ 
+ #include <linux/ieee80211.h>
+@@ -213,6 +213,8 @@ ieee80211_agg_start_txq(struct sta_info
+ 	struct ieee80211_txq *txq = sta->sta.txq[tid];
+ 	struct txq_info *txqi;
+ 
++	lockdep_assert_held(&sta->ampdu_mlme.mtx);
++
+ 	if (!txq)
+ 		return;
+ 
+@@ -290,7 +292,6 @@ static void ieee80211_remove_tid_tx(stru
+ 	ieee80211_assign_tid_tx(sta, tid, NULL);
+ 
+ 	ieee80211_agg_splice_finish(sta->sdata, tid);
+-	ieee80211_agg_start_txq(sta, tid, false);
+ 
+ 	kfree_rcu(tid_tx, rcu_head);
+ }
+@@ -889,6 +890,7 @@ void ieee80211_stop_tx_ba_cb(struct sta_
+ {
+ 	struct ieee80211_sub_if_data *sdata = sta->sdata;
+ 	bool send_delba = false;
++	bool start_txq = false;
+ 
+ 	ht_dbg(sdata, "Stopping Tx BA session for %pM tid %d\n",
+ 	       sta->sta.addr, tid);
+@@ -906,10 +908,14 @@ void ieee80211_stop_tx_ba_cb(struct sta_
+ 		send_delba = true;
+ 
+ 	ieee80211_remove_tid_tx(sta, tid);
++	start_txq = true;
+ 
+  unlock_sta:
+ 	spin_unlock_bh(&sta->lock);
+ 
++	if (start_txq)
++		ieee80211_agg_start_txq(sta, tid, false);
++
+ 	if (send_delba)
+ 		ieee80211_send_delba(sdata, sta->sta.addr, tid,
+ 			WLAN_BACK_INITIATOR, WLAN_REASON_QSTA_NOT_USE);