RDMA/mlx5: Don't set tx affinity when lag is in hash mode

In hash mode, without setting tx affinity explicitly, the port select
flow table decides which port is used for the traffic.
If port_select_flow_table_bypass capability is supported and tx affinity
is set explicitly for QP/TIS, they will be added into the explicit affinity
table in FW to check which port is used for the traffic.
1. The overloaded explicit affinity table may affect performance.
   To avoid this, do not set tx affinity explicitly by default.
2. The packets of the same flow need to be transmitted on the same port.
   Because the packets of the same flow use different QPs in slow & fast
   path, it shouldn't set tx affinity explicitly for these QPs.

Signed-off-by: Liu, Changcheng <jerrliu@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>

diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h
index 2e2ad39..c3e0142 100644 (file)
--- a/drivers/infiniband/hw/mlx5/mlx5_ib.h
+++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h
@@ -1540,6 +1540,18 @@ int mlx5_ib_test_wc(struct mlx5_ib_dev *dev);
 
 static inline bool mlx5_ib_lag_should_assign_affinity(struct mlx5_ib_dev *dev)
 {
+       /*
+        * If the driver is in hash mode and the port_select_flow_table_bypass cap
+        * is supported, it means that the driver no longer needs to assign the port
+        * affinity by default. If a user wants to set the port affinity explicitly,
+        * the user has a dedicated API to do that, so there is no need to assign
+        * the port affinity by default.
+        */
+       if (dev->lag_active &&
+           mlx5_lag_mode_is_hash(dev->mdev) &&
+           MLX5_CAP_PORT_SELECTION(dev->mdev, port_select_flow_table_bypass))
+               return 0;
+
        return dev->lag_active ||
                (MLX5_CAP_GEN(dev->mdev, num_lag_ports) > 1 &&
                 MLX5_CAP_GEN(dev->mdev, lag_tx_port_affinity));

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
index 0651022..4cc5fc3 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/lag.c
@@ -1275,6 +1275,22 @@ bool mlx5_lag_is_active(struct mlx5_core_dev *dev)
 }
 EXPORT_SYMBOL(mlx5_lag_is_active);
 
+bool mlx5_lag_mode_is_hash(struct mlx5_core_dev *dev)
+{
+       struct mlx5_lag *ldev;
+       unsigned long flags;
+       bool res = 0;
+
+       spin_lock_irqsave(&lag_lock, flags);
+       ldev = mlx5_lag_dev(dev);
+       if (ldev)
+               res = test_bit(MLX5_LAG_MODE_FLAG_HASH_BASED, &ldev->mode_flags);
+       spin_unlock_irqrestore(&lag_lock, flags);
+
+       return res;
+}
+EXPORT_SYMBOL(mlx5_lag_mode_is_hash);
+
 bool mlx5_lag_is_master(struct mlx5_core_dev *dev)
 {
        struct mlx5_lag *ldev;

diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 4815063..9114cac 100644 (file)
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1153,6 +1153,7 @@ int mlx5_cmd_destroy_vport_lag(struct mlx5_core_dev *dev);
 bool mlx5_lag_is_roce(struct mlx5_core_dev *dev);
 bool mlx5_lag_is_sriov(struct mlx5_core_dev *dev);
 bool mlx5_lag_is_active(struct mlx5_core_dev *dev);
+bool mlx5_lag_mode_is_hash(struct mlx5_core_dev *dev);
 bool mlx5_lag_is_master(struct mlx5_core_dev *dev);
 bool mlx5_lag_is_shared_fdb(struct mlx5_core_dev *dev);
 struct net_device *mlx5_lag_get_roce_netdev(struct mlx5_core_dev *dev);