All of lore.kernel.org
 help / color / mirror / Atom feed
From: Huy Nguyen <huyn@nvidia.com>
To: <netdev@vger.kernel.org>
Cc: <steffen.klassert@secunet.com>, <saeedm@nvidia.com>,
	<borisp@nvidia.com>, <raeds@nvidia.com>, <danielj@nvidia.com>,
	<yossiku@nvidia.com>, <kuba@kernel.org>, <huyn@nvidia.com>
Subject: [PATCH net-next v5 1/3] net/mlx5: Optimize mlx5e_feature_checks for non IPsec packet
Date: Mon, 14 Jun 2021 17:33:47 +0300	[thread overview]
Message-ID: <20210614143349.74866-2-huyn@nvidia.com> (raw)
In-Reply-To: <20210614143349.74866-1-huyn@nvidia.com>

mlx5e_ipsec_feature_check belongs to mlx5e_tunnel_features_check.
Also, IPsec is not the default configuration so it should be
checked at the end instead of the beginning of mlx5e_features_check.

Signed-off-by: Raed Salem <raeds@nvidia.com>
Signed-off-by: Huy Nguyen <huyn@nvidia.com>
---
 .../mellanox/mlx5/core/en_accel/ipsec_rxtx.h      | 15 +++++++++------
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |  8 +++++---
 2 files changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
index 3e80742a3caf..cfa98272e4a9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h
@@ -93,8 +93,8 @@ static inline bool mlx5e_ipsec_eseg_meta(struct mlx5_wqe_eth_seg *eseg)
 void mlx5e_ipsec_tx_build_eseg(struct mlx5e_priv *priv, struct sk_buff *skb,
 			       struct mlx5_wqe_eth_seg *eseg);
 
-static inline bool mlx5e_ipsec_feature_check(struct sk_buff *skb, struct net_device *netdev,
-					     netdev_features_t features)
+static inline netdev_features_t
+mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t features)
 {
 	struct sec_path *sp = skb_sec_path(skb);
 
@@ -102,9 +102,11 @@ static inline bool mlx5e_ipsec_feature_check(struct sk_buff *skb, struct net_dev
 		struct xfrm_state *x = sp->xvec[0];
 
 		if (x && x->xso.offload_handle)
-			return true;
+			return features;
 	}
-	return false;
+
+	/* Disable CSUM and GSO for software IPsec */
+	return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK);
 }
 
 #else
@@ -120,8 +122,9 @@ static inline bool mlx5e_ipsec_eseg_meta(struct mlx5_wqe_eth_seg *eseg)
 }
 
 static inline bool mlx5_ipsec_is_rx_flow(struct mlx5_cqe64 *cqe) { return false; }
-static inline bool mlx5e_ipsec_feature_check(struct sk_buff *skb, struct net_device *netdev,
-					     netdev_features_t features) { return false; }
+static inline netdev_features_t
+mlx5e_ipsec_feature_check(struct sk_buff *skb, netdev_features_t features)
+{ return features & ~(NETIF_F_CSUM_MASK | NETIF_F_GSO_MASK); }
 #endif /* CONFIG_MLX5_EN_IPSEC */
 
 #endif /* __MLX5E_IPSEC_RXTX_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index bca832cdc4cb..43c0a473cc9a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -4278,6 +4278,11 @@ static netdev_features_t mlx5e_tunnel_features_check(struct mlx5e_priv *priv,
 		/* Support Geneve offload for default UDP port */
 		if (port == GENEVE_UDP_PORT && mlx5_geneve_tx_allowed(priv->mdev))
 			return features;
+#endif
+		break;
+#ifdef CONFIG_MLX5_EN_IPSEC
+	case IPPROTO_ESP:
+		return mlx5e_ipsec_feature_check(skb, features);
 #endif
 	}
 
@@ -4295,9 +4300,6 @@ netdev_features_t mlx5e_features_check(struct sk_buff *skb,
 	features = vlan_features_check(skb, features);
 	features = vxlan_features_check(skb, features);
 
-	if (mlx5e_ipsec_feature_check(skb, netdev, features))
-		return features;
-
 	/* Validate if the tunneled packet is being offloaded by HW */
 	if (skb->encapsulation &&
 	    (features & NETIF_F_CSUM_MASK || features & NETIF_F_GSO_MASK))
-- 
2.24.1


  reply	other threads:[~2021-06-14 14:34 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-14 14:33 [PATCH net-next v5 0/3] Fix IPsec crypto offloads with vxlan tunnel Huy Nguyen
2021-06-14 14:33 ` Huy Nguyen [this message]
2021-06-14 14:33 ` [PATCH net-next v5 2/3] net/xfrm: Add inner_ipproto into sec_path Huy Nguyen
2021-06-16  5:44   ` Steffen Klassert
2021-06-14 14:33 ` [PATCH net-next v5 3/3] net/mlx5: Fix checksum issue of VXLAN and IPsec crypto offload Huy Nguyen
2021-06-22 21:53 ` [PATCH net-next v5 0/3] Fix IPsec crypto offloads with vxlan tunnel Saeed Mahameed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210614143349.74866-2-huyn@nvidia.com \
    --to=huyn@nvidia.com \
    --cc=borisp@nvidia.com \
    --cc=danielj@nvidia.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=raeds@nvidia.com \
    --cc=saeedm@nvidia.com \
    --cc=steffen.klassert@secunet.com \
    --cc=yossiku@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.