[PATCH 5.4 38/49] RDMA/addr: Fix race with netevent_callback()/rdma_addr_cancel()

From: Greg Kroah-Hartman
Date: Sat Oct 31 2020 - 07:37:59 EST


From: Jason Gunthorpe <jgg@xxxxxxxxxx>

commit 2ee9bf346fbfd1dad0933b9eb3a4c2c0979b633e upstream.

This three thread race can result in the work being run once the callback
becomes NULL:

CPU1 CPU2 CPU3
netevent_callback()
process_one_req() rdma_addr_cancel()
[..]
spin_lock_bh()
set_timeout()
spin_unlock_bh()

spin_lock_bh()
list_del_init(&req->list);
spin_unlock_bh()

req->callback = NULL
spin_lock_bh()
if (!list_empty(&req->list))
// Skipped!
// cancel_delayed_work(&req->work);
spin_unlock_bh()

process_one_req() // again
req->callback() // BOOM
cancel_delayed_work_sync()

The solution is to always cancel the work once it is completed so any
in between set_timeout() does not result in it running again.

Cc: stable@xxxxxxxxxxxxxxx
Fixes: 44e75052bc2a ("RDMA/rdma_cm: Make rdma_addr_cancel into a fence")
Link: https://lore.kernel.org/r/20200930072007.1009692-1-leon@xxxxxxxxxx
Reported-by: Dan Aloni <dan@xxxxxxxxxxxx>
Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxx>
Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>

---
drivers/infiniband/core/addr.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)

--- a/drivers/infiniband/core/addr.c
+++ b/drivers/infiniband/core/addr.c
@@ -645,13 +645,12 @@ static void process_one_req(struct work_
req->callback = NULL;

spin_lock_bh(&lock);
+ /*
+ * Although the work will normally have been canceled by the workqueue,
+ * it can still be requeued as long as it is on the req_list.
+ */
+ cancel_delayed_work(&req->work);
if (!list_empty(&req->list)) {
- /*
- * Although the work will normally have been canceled by the
- * workqueue, it can still be requeued as long as it is on the
- * req_list.
- */
- cancel_delayed_work(&req->work);
list_del_init(&req->list);
kfree(req);
}