From: Mark Bloch <redacted>
When in switchdev mode, there is no need to do self loopback checks
as we can't receive those packets, we insert steering rules to the
eswitch that make sure packets can't be looped back.
Signed-off-by: Mark Bloch <redacted>
Reviewed-by: Or Gerlitz <redacted>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Saeed Mahameed <redacted>
---
drivers/infiniband/hw/mlx5/qp.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 8aed091036c6..5663530ea5fd 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -1279,6 +1279,10 @@ static int create_raw_packet_qp_tir(struct mlx5_ib_dev *dev,
if (tunnel_offload_en)
MLX5_SET(tirc, tirc, tunneled_offload_en, 1);
+ if (dev->rep)
+ MLX5_SET(tirc, tirc, self_lb_block,
+ MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST_);
+
err = mlx5_core_create_tir(dev->mdev, in, inlen, &rq->tirn);
kvfree(in);
@@ -1570,6 +1574,10 @@ static int create_rss_raw_qp_tir(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *qp,
MLX5_SET(rx_hash_field_select, hfso, selected_fields, selected_fields);
create_tir:
+ if (dev->rep)
+ MLX5_SET(tirc, tirc, self_lb_block,
+ MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST_);
+
err = mlx5_core_create_tir(dev->mdev, in, inlen, &qp->rss_qp.tirn);
if (err)
--
2.14.3