The use of short type causes the value is truncated when
configured CQ ci, as a result, the hardware get the
different value and the CQ overflow.
Signed-off-by: Lijun Ou <oulijun@xxxxxxxxxx>
---
providers/hns/hns_roce_u_hw_v1.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/providers/hns/hns_roce_u_hw_v1.c b/providers/hns/hns_roce_u_hw_v1.c
index 013a2b4..1d256aa 100644
--- a/providers/hns/hns_roce_u_hw_v1.c
+++ b/providers/hns/hns_roce_u_hw_v1.c
@@ -411,7 +411,7 @@ static int hns_roce_u_v1_poll_cq(struct ibv_cq *ibvcq, int ne,
if (npolled) {
if (dev->hw_version == HNS_ROCE_HW_VER1) {
- *cq->set_ci_db = (unsigned short)(cq->cons_index &
+ *cq->set_ci_db = (cq->cons_index &
((cq->cq_depth << 1) - 1));
mmio_ordered_writes_hack();
}
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
