Netdev List
 help / color / mirror / Atom feed
From: Saeed Mahameed <saeedm@mellanox.com>
To: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org, Dave Watson <davejwatson@fb.com>,
	Boris Pismenny <borisp@mellanox.com>,
	Ilya Lesokhin <ilyal@mellanox.com>,
	Saeed Mahameed <saeedm@mellanox.com>
Subject: [PATCH V5 net-next 03/14] net: Add Software fallback infrastructure for socket dependent offloads
Date: Tue, 27 Mar 2018 16:56:08 -0700	[thread overview]
Message-ID: <20180327235619.31308-4-saeedm@mellanox.com> (raw)
In-Reply-To: <20180327235619.31308-1-saeedm@mellanox.com>

From: Ilya Lesokhin <ilyal@mellanox.com>

With socket dependent offloads we rely on the netdev to transform
the transmitted packets before sending them to the wire.
When a packet from an offloaded socket is rerouted to a different
device we need to detect it and do the transformation in software.

Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
---
 include/net/sock.h | 21 +++++++++++++++++++++
 net/Kconfig        |  4 ++++
 net/core/dev.c     |  4 ++++
 3 files changed, 29 insertions(+)

diff --git a/include/net/sock.h b/include/net/sock.h
index 709311132d4c..7607eeed6be2 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -481,6 +481,11 @@ struct sock {
 	void			(*sk_error_report)(struct sock *sk);
 	int			(*sk_backlog_rcv)(struct sock *sk,
 						  struct sk_buff *skb);
+#ifdef CONFIG_SOCK_VALIDATE_XMIT
+	struct sk_buff*		(*sk_validate_xmit_skb)(struct sock *sk,
+							struct net_device *dev,
+							struct sk_buff *skb);
+#endif
 	void                    (*sk_destruct)(struct sock *sk);
 	struct sock_reuseport __rcu	*sk_reuseport_cb;
 	struct rcu_head		sk_rcu;
@@ -2328,6 +2333,22 @@ static inline bool sk_fullsock(const struct sock *sk)
 	return (1 << sk->sk_state) & ~(TCPF_TIME_WAIT | TCPF_NEW_SYN_RECV);
 }
 
+/* Checks if this SKB belongs to an HW offloaded socket
+ * and whether any SW fallbacks are required based on dev.
+ */
+static inline struct sk_buff *sk_validate_xmit_skb(struct sk_buff *skb,
+						   struct net_device *dev)
+{
+#ifdef CONFIG_SOCK_VALIDATE_XMIT
+	struct sock *sk = skb->sk;
+
+	if (sk && sk_fullsock(sk) && sk->sk_validate_xmit_skb)
+		skb = sk->sk_validate_xmit_skb(sk, dev, skb);
+#endif
+
+	return skb;
+}
+
 /* This helper checks if a socket is a LISTEN or NEW_SYN_RECV
  * SYNACK messages can be attached to either ones (depending on SYNCOOKIE)
  */
diff --git a/net/Kconfig b/net/Kconfig
index 0428f12c25c2..fe84cfe3260e 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -407,6 +407,10 @@ config GRO_CELLS
 	bool
 	default n
 
+config SOCK_VALIDATE_XMIT
+	bool
+	default n
+
 config NET_DEVLINK
 	tristate "Network physical/parent device Netlink interface"
 	help
diff --git a/net/core/dev.c b/net/core/dev.c
index e13807b5c84d..e8a126a09d28 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3105,6 +3105,10 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
 	if (unlikely(!skb))
 		goto out_null;
 
+	skb = sk_validate_xmit_skb(skb, dev);
+	if (unlikely(!skb))
+		goto out_null;
+
 	if (netif_needs_gso(skb, features)) {
 		struct sk_buff *segs;
 
-- 
2.14.3

  parent reply	other threads:[~2018-03-27 23:56 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-03-27 23:56 [PATCH V5 net-next 00/14] TLS offload, netdev & MLX5 support Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 01/14] tcp: Add clean acked data hook Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 02/14] net: Rename and export copy_skb_header Saeed Mahameed
2018-03-27 23:56 ` Saeed Mahameed [this message]
2018-03-27 23:56 ` [PATCH V5 net-next 04/14] net: Add TLS offload netdev ops Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 05/14] net: Add TLS TX offload features Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 06/14] net/tls: Add generic NIC offload infrastructure Saeed Mahameed
2018-03-28  4:22   ` Shannon Nelson
2018-03-28 15:11   ` Kirill Tkhai
2018-04-02 10:37     ` Boris Pismenny
2018-03-27 23:56 ` [PATCH V5 net-next 07/14] net/tls: Support TLS device offload with IPv6 Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 08/14] net/mlx5e: Move defines out of ipsec code Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 09/14] net/mlx5: Accel, Add TLS tx offload interface Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 10/14] net/mlx5e: TLS, Add Innova TLS TX support Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 11/14] net/mlx5e: TLS, Add Innova TLS TX offload data path Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 12/14] net/mlx5e: TLS, Add error statistics Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 13/14] MAINTAINERS: Update mlx5 innova driver maintainers Saeed Mahameed
2018-03-27 23:56 ` [PATCH V5 net-next 14/14] MAINTAINERS: Update TLS maintainers 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=20180327235619.31308-4-saeedm@mellanox.com \
    --to=saeedm@mellanox.com \
    --cc=borisp@mellanox.com \
    --cc=davejwatson@fb.com \
    --cc=davem@davemloft.net \
    --cc=ilyal@mellanox.com \
    --cc=netdev@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox