net/mlx5e: Add support for PCI relaxed ordering

The concept of Relaxed Ordering in the PCI Express environment allows
switches in the path between the Requester and Completer to reorder some
transactions just received before others that were previously enqueued.

In ETH driver, there is no question of write integrity since each memory
segment is written only once per cycle. In addition, the driver doesn't
access the memory shared with the hardware until the corresponding CQE
arrives indicating all PCI transactions are done.

Running TCP single stream over ConnectX-4 LX, ARM CPU on remote-numa has
300% improvement in the bandwidth.

With relaxed ordering turned off: BW:10 [GB/s]
With relaxed ordering turned on: BW:40 [GB/s]

The driver turns relaxed ordering with respect to the firmware
capabilities and the return value from pcie_relaxed_ordering_enabled().

Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 0fb30fe93207b2dbfeb85c8b4c21652d363c0750..f2fa1307e90cbb15f7c8814f8b5b02546614a002 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -974,6 +974,7 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev);
 void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev);
 int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb,
                       bool enable_mc_lb);
+void mlx5e_mkey_set_relaxed_ordering(struct mlx5_core_dev *mdev, void *mkc);
 
 /* common netdev helpers */
 void mlx5e_create_q_counters(struct mlx5e_priv *priv);

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c
index 1e42c7ae621b9f14cfe602a8b827f2ea5bff2b49..a6cf008057b5f95ff53a80a8240544fdef42e048 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c
@@ -60,6 +60,16 @@ void mlx5e_destroy_tir(struct mlx5_core_dev *mdev,
        mutex_unlock(&mdev->mlx5e_res.td.list_lock);
 }
 
+void mlx5e_mkey_set_relaxed_ordering(struct mlx5_core_dev *mdev, void *mkc)
+{
+       bool ro_pci_enable = pcie_relaxed_ordering_enabled(mdev->pdev);
+       bool ro_write = MLX5_CAP_GEN(mdev, relaxed_ordering_write);
+       bool ro_read = MLX5_CAP_GEN(mdev, relaxed_ordering_read);
+
+       MLX5_SET(mkc, mkc, relaxed_ordering_read, ro_pci_enable && ro_read);
+       MLX5_SET(mkc, mkc, relaxed_ordering_write, ro_pci_enable && ro_write);
+}
+
 static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn,
                             struct mlx5_core_mkey *mkey)
 {
@@ -76,7 +86,7 @@ static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn,
        MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_PA);
        MLX5_SET(mkc, mkc, lw, 1);
        MLX5_SET(mkc, mkc, lr, 1);
-
+       mlx5e_mkey_set_relaxed_ordering(mdev, mkc);
        MLX5_SET(mkc, mkc, pd, pdn);
        MLX5_SET(mkc, mkc, length64, 1);
        MLX5_SET(mkc, mkc, qpn, 0xffffff);

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index ac91504be8e8bc0e720387284d79bfe4bb1c2d24..f374348fa810030ce82502b6f5a4e595895ad030 100644 (file)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -274,7 +274,7 @@ static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev,
        MLX5_SET(mkc, mkc, lw, 1);
        MLX5_SET(mkc, mkc, lr, 1);
        MLX5_SET(mkc, mkc, access_mode_1_0, MLX5_MKC_ACCESS_MODE_MTT);
-
+       mlx5e_mkey_set_relaxed_ordering(mdev, mkc);
        MLX5_SET(mkc, mkc, qpn, 0xffffff);
        MLX5_SET(mkc, mkc, pd, mdev->mlx5e_res.pdn);
        MLX5_SET64(mkc, mkc, len, npages << page_shift);