Received-SPF: pass (google.com: domain of linux-bluetooth-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Luiz Augusto von Dentz <luiz.dentz@gmail.com>
To:     linux-bluetooth@vger.kernel.org
Subject: [RFC] Bluetooth: L2CAP: Fix general protection fault in l2cap_chan_timeout
Date:   Wed,  2 Jun 2021 16:23:48 -0700
Message-Id: <20210602232348.766496-1-luiz.dentz@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Even though l2cap_chan_del calls __clear_chan_timer that uses
cancel_delayed_work which doesn't prevent the work to run as it could be
already pending.

So instead this check if chan->conn has been set to NULL, which indicates
that l2cap_conn_del might have run, and proceed to release its reference
since the channel has already been cleanup.

Reported-by: syzbot+008cdbf7a9044c2c2f99@syzkaller.appspotmail.com
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
---
 net/bluetooth/l2cap_core.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
index 9ebb85df4db4..4408e07a00d1 100644
--- a/net/bluetooth/l2cap_core.c
+++ b/net/bluetooth/l2cap_core.c
@@ -414,11 +414,21 @@ static void l2cap_chan_timeout(struct work_struct *work)
 {
 	struct l2cap_chan *chan = container_of(work, struct l2cap_chan,
 					       chan_timer.work);
-	struct l2cap_conn *conn = chan->conn;
+	struct l2cap_conn *conn;
 	int reason;
 
 	BT_DBG("chan %p state %s", chan, state_to_string(chan->state));
 
+	conn = chan->conn;
+	if (!conn) {
+		/* Channel is no longer attached to a connection so
+		 * l2cap_conn_del might have run, just release reference
+		 * acquired via __set_chan_timer.
+		 */
+		l2cap_chan_put(chan);
+		return;
+	}
+
 	mutex_lock(&conn->chan_lock);
 	/* __set_chan_timer() calls l2cap_chan_hold(chan) while scheduling
 	 * this work. No need to call l2cap_chan_hold(chan) here again.
-- 
2.31.1