[PATCH 4.9 181/329] RDMA/qedr: Fix kernel panic when running fio over NFSoRDMA

[Greg Kroah-Hartman]

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: "Kalderon, Michal" <Michal.Kalderon@xxxxxxxxxx>

[ Upstream commit e3fd112cbf21d049faf64ba1471d72b93c22109a ]

Race in qedr_poll_cq, lastest_cqe wasn't protected by lock,
leading to a case where two context's accessing poll_cq at
the same time lead to one of them having a pointer to an old
latest_cqe and reading an invalid cqe element

Signed-off-by: Amit Radzi <Amit.Radzi@xxxxxxxxxx>
Signed-off-by: Michal Kalderon <Michal.Kalderon@xxxxxxxxxx>
Signed-off-by: Ariel Elior <Ariel.Elior@xxxxxxxxxx>
Signed-off-by: Jason Gunthorpe <jgg@xxxxxxxxxxxx>
Signed-off-by: Sasha Levin <alexander.levin@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
 drivers/infiniband/hw/qedr/verbs.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/infiniband/hw/qedr/verbs.c
+++ b/drivers/infiniband/hw/qedr/verbs.c
@@ -3467,7 +3467,7 @@ int qedr_poll_cq(struct ib_cq *ibcq, int
 {
 	struct qedr_dev *dev = get_qedr_dev(ibcq->device);
 	struct qedr_cq *cq = get_qedr_cq(ibcq);
-	union rdma_cqe *cqe = cq->latest_cqe;
+	union rdma_cqe *cqe;
 	u32 old_cons, new_cons;
 	unsigned long flags;
 	int update = 0;
@@ -3477,6 +3477,7 @@ int qedr_poll_cq(struct ib_cq *ibcq, int
 		return qedr_gsi_poll_cq(ibcq, num_entries, wc);
 
 	spin_lock_irqsave(&cq->cq_lock, flags);
+	cqe = cq->latest_cqe;
 	old_cons = qed_chain_get_cons_idx_u32(&cq->pbl);
 	while (num_entries && is_valid_cqe(cq, cqe)) {
 		struct qedr_qp *qp;