diff options
| author |   Alaa Hleihel <alaa@nvidia.com> | 2021-04-27 12:51:41 +0300 |
|---|---|---|
| committer |   Saeed Mahameed <saeedm@nvidia.com> | 2021-06-03 13:10:20 -0700 |
| commit | 39e8cc6d757af7ee5edf5826102c95e3f5bb374b (patch) | |
| tree | 7d6580c1d78bfc5439bdb0cd9f5b8f781f7a2d21 /drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c | |
| parent | net/mlx5e: Disable TX MPWQE in kdump mode (diff) | |
| download | wireguard-linux-39e8cc6d757af7ee5edf5826102c95e3f5bb374b.tar.xz wireguard-linux-39e8cc6d757af7ee5edf5826102c95e3f5bb374b.zip | |
net/mlx5e: Disable TLS device offload in kdump mode
Under kdump environment we want to use the smallest possible amount
of resources, that includes setting SQ size to minimum.
However, when running on a device that supports TLS device offload,
then the SQ stop room becomes larger than with non-capable device and
requires increasing the SQ size.
Since TLS device offload is not necessary in kdump mode, disable it to
reduce the memory requirements for capable devices.
With this change, the needed SQ stop room size drops by 33.
Signed-off-by: Alaa Hleihel <alaa@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c')
| -rw-r--r-- | drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c | 11 |
1 files changed, 7 insertions, 4 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c index 95293ee0d38d..d93aadbf10da 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c @@ -59,12 +59,15 @@ void mlx5e_ktls_build_netdev(struct mlx5e_priv *priv) struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; - if (mlx5_accel_is_ktls_tx(mdev)) { + if (!mlx5e_accel_is_ktls_tx(mdev) && !mlx5e_accel_is_ktls_rx(mdev)) + return; + + if (mlx5e_accel_is_ktls_tx(mdev)) { netdev->hw_features |= NETIF_F_HW_TLS_TX; netdev->features |= NETIF_F_HW_TLS_TX; } - if (mlx5_accel_is_ktls_rx(mdev)) + if (mlx5e_accel_is_ktls_rx(mdev)) netdev->hw_features |= NETIF_F_HW_TLS_RX; netdev->tlsdev_ops = &mlx5e_ktls_ops; @@ -89,7 +92,7 @@ int mlx5e_ktls_init_rx(struct mlx5e_priv *priv) { int err; - if (!mlx5_accel_is_ktls_rx(priv->mdev)) + if (!mlx5e_accel_is_ktls_rx(priv->mdev)) return 0; priv->tls->rx_wq = create_singlethread_workqueue("mlx5e_tls_rx"); @@ -109,7 +112,7 @@ int mlx5e_ktls_init_rx(struct mlx5e_priv *priv) void mlx5e_ktls_cleanup_rx(struct mlx5e_priv *priv) { - if (!mlx5_accel_is_ktls_rx(priv->mdev)) + if (!mlx5e_accel_is_ktls_rx(priv->mdev)) return; if (priv->netdev->features & NETIF_F_HW_TLS_RX) |