RDMA/mlx5: Use correct device num_ports when modify DC

[ Upstream commit 746aa3c8cb1a650ff2583497ac646e505831b9b9 ]

Just like other QP types, when modify DC, the port_num should be compared
with dev->num_ports, instead of HCA_CAP.num_ports.  Otherwise Multi-port
vHCA on DC may not work.

Fixes: 776a3906b692 ("IB/mlx5: Add support for DC target QP")
Link: https://lore.kernel.org/r/20230420013906.1244185-1-markzhang@nvidia.com
Signed-off-by: Mark Zhang <markzhang@nvidia.com>
Reviewed-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>

diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index cf953d23d18dae8da53d51580e90e8f0f5eb8c9b..f7d3643b08f5076d35eed554bec0d23ea3ded4cf 100644 (file)
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -4408,7 +4408,7 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr,
                        return -EINVAL;
 
                if (attr->port_num == 0 ||
-                   attr->port_num > MLX5_CAP_GEN(dev->mdev, num_ports)) {
+                   attr->port_num > dev->num_ports) {
                        mlx5_ib_dbg(dev, "invalid port number %d. number of ports is %d\n",
                                    attr->port_num, dev->num_ports);
                        return -EINVAL;