Skip to contentSkip to conversation

/net/9p/trans_fd.c: fix race-condition by flushing workqueue before the kfree()

PATCH In review Tomas Bortoli 2 participants 2 messages Jul 19, 2018 – Jul 20, 2018

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 /net/9p/trans_fd.c: fix race-condition by flushing workqueue before the kfree()

The patch adds the flush in p9_mux_poll_stop() as it the function used by p9_conn_destroy(), in turn called by p9_fd_close() to stop the async polling associated with the data regarding the connection.

---
As shown by Syzbot, it is possible to provoke a race between p9_fd_close() and p9_poll_workfn() that is called to take care of the async read/write work to do. To make sure p9_fd_close() frees "trans" when it is not used anymore, it has to explicitly call flush_scheduled_work() before the kfree().

 net/9p/trans_fd.c | 2 ++
 1 file changed, 2 insertions(+)
net/9p/trans_fd.c +2 -0
index bf459ee0feab..76ae134c05d9 100644
@@ -185,6 +185,8 @@ static void p9_mux_poll_stop(struct p9_conn *m)
 	spin_lock_irqsave(&p9_poll_lock, flags);
 	list_del_init(&m->poll_pending_link);
 	spin_unlock_irqrestore(&p9_poll_lock, flags);
+
+	flush_scheduled_work();
 }
 
 /**
Signature · 1 line
2.11.0
#1 lore raw
Quoted 20 lines · Tomas Bortoli

The patch adds the flush in p9_mux_poll_stop() as it the function used by p9_conn_destroy(), in turn called by p9_fd_close() to stop the async polling associated with the data regarding the connection.

---
As shown by Syzbot, it is possible to provoke a race between p9_fd_close() and p9_poll_workfn() that is called to take care of the async read/write work to do. To make sure p9_fd_close() frees "trans" when it is not used anymore, it has to explicitly call flush_scheduled_work() before the kfree().

 net/9p/trans_fd.c | 2 ++
 1 file changed, 2 insertions(+)
net/9p/trans_fd.c +0 -0
index bf459ee0feab..76ae134c05d9 100644

net/9p/trans_fd.c static void p9_mux_poll_stop(struct p9_conn *m) · line 189

Quoted 5 lines
net/9p/trans_fd.c +2 -0
@@ -185,6 +185,8 @@ static void p9_mux_poll_stop(struct p9_conn *m)
 	spin_lock_irqsave(&p9_poll_lock, flags);
 	list_del_init(&m->poll_pending_link);
 	spin_unlock_irqrestore(&p9_poll_lock, flags);
+
+	flush_scheduled_work();

I suggest we should use flush_work(&p9_poll_work), flush_scheduled_work is not suitable.

Thanks.

Quoted 4 lines
net/9p/trans_fd.c +0 -0
 }
 
 /**