5.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mark Zhang markzhang@nvidia.com
[ Upstream commit 12d044770e12c4205fa69535b4fa8a9981fea98f ]
When a DCT QP is created on an active lag, it's dctc.port is assigned in a round-robin way, which is from 1 to dev->lag_port. In this case when querying this QP, we may get qp_attr.port_num > 2. Fix this by setting qp->port when modifying a DCT QP, and read port_num from qp->port instead of dctc.port when querying it.
Fixes: 7c4b1ab9f167 ("IB/mlx5: Add DCT RoCE LAG support") Signed-off-by: Mark Zhang markzhang@nvidia.com Reviewed-by: Maher Sanalla msanalla@nvidia.com Link: https://patch.msgid.link/94c76bf0adbea997f87ffa27674e0a7118ad92a9.1737290358... Signed-off-by: Leon Romanovsky leon@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/infiniband/hw/mlx5/qp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index e0db91d1e7492..a00dde2f3fd39 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -4219,6 +4219,8 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
set_id = mlx5_ib_get_counters_id(dev, attr->port_num - 1); MLX5_SET(dctc, dctc, counter_set_id, set_id); + + qp->port = attr->port_num; } else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_RTR) { struct mlx5_ib_modify_qp_resp resp = {}; u32 out[MLX5_ST_SZ_DW(create_dct_out)] = {}; @@ -4714,7 +4716,7 @@ static int mlx5_ib_dct_query_qp(struct mlx5_ib_dev *dev, struct mlx5_ib_qp *mqp, }
if (qp_attr_mask & IB_QP_PORT) - qp_attr->port_num = MLX5_GET(dctc, dctc, port); + qp_attr->port_num = mqp->port; if (qp_attr_mask & IB_QP_MIN_RNR_TIMER) qp_attr->min_rnr_timer = MLX5_GET(dctc, dctc, min_rnr_nak); if (qp_attr_mask & IB_QP_AV) {