OpenE2K/linux
6
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
eee2f04b80
linux/include/net/vxlan.h

99 lines
2.4 KiB
C
Raw Normal View History

vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
#ifndef __NET_VXLAN_H
#define __NET_VXLAN_H 1
net: Generalize ndo_gso_check to ndo_features_check GSO isn't the only offload feature with restrictions that potentially can't be expressed with the current features mechanism. Checksum is another although it's a general issue that could in theory apply to anything. Even if it may be possible to implement these restrictions in other ways, it can result in duplicate code or inefficient per-packet behavior. This generalizes ndo_gso_check so that drivers can remove any features that don't make sense for a given packet, similar to netif_skb_features(). It also converts existing driver restrictions to the new format, completing the work that was done to support tunnel protocols since the issues apply to checksums as well. By actually removing features from the set that are used to do offloading, it solves another problem with the existing interface. In these cases, GSO would run with the original set of features and not do anything because it appears that segmentation is not required. CC: Tom Herbert <therbert@google.com> CC: Joe Stringer <joestringer@nicira.com> CC: Eric Dumazet <edumazet@google.com> CC: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jesse Gross <jesse@nicira.com> Acked-by: Tom Herbert <therbert@google.com> Fixes: 04ffcb255f22 ("net: Add ndo_gso_check") Tested-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-24 07:37:26 +01:00
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/if_vlan.h>
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/udp.h>
#define VNI_HASH_BITS 10
#define VNI_HASH_SIZE (1<<VNI_HASH_BITS)
vxlan: Inline vxlan_gso_check(). Suggested-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-18 01:24:54 +01:00
/* VXLAN protocol header */
struct vxlanhdr {
__be32 vx_flags;
__be32 vx_vni;
};
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
struct vxlan_sock;
typedef void (vxlan_rcv_t)(struct vxlan_sock *vh, struct sk_buff *skb, __be32 key);
/* per UDP socket information */
struct vxlan_sock {
struct hlist_node hlist;
vxlan_rcv_t *rcv;
void *data;
struct work_struct del_work;
struct socket *sock;
struct rcu_head rcu;
struct hlist_head vni_list[VNI_HASH_SIZE];
atomic_t refcnt;
net: Add GRO support for vxlan traffic Add GRO handlers for vxlann, by using the UDP GRO infrastructure. For single TCP session that goes through vxlan tunneling I got nice improvement from 6.8Gbs to 11.5Gbs --> UDP/VXLAN GRO disabled $ netperf -H 192.168.52.147 -c -C $ netperf -t TCP_STREAM -H 192.168.52.147 -c -C MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.52.147 () port 0 AF_INET Recv Send Send Utilization Service Demand Socket Socket Message Elapsed Send Recv Send Recv Size Size Size Time Throughput local remote local remote bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB 87380 65536 65536 10.00 6799.75 12.54 24.79 0.604 1.195 --> UDP/VXLAN GRO enabled $ netperf -t TCP_STREAM -H 192.168.52.147 -c -C MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.52.147 () port 0 AF_INET Recv Send Send Utilization Service Demand Socket Socket Message Elapsed Send Recv Send Recv Size Size Size Time Throughput local remote local remote bytes bytes bytes secs. 10^6bits/s % S % S us/KB us/KB 87380 65536 65536 10.00 11562.72 24.90 20.34 0.706 0.577 Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-20 12:59:21 +01:00
struct udp_offload udp_offloads;
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
};
vxlan: Add support for UDP checksums (v4 sending, v6 zero csums) Added VXLAN link configuration for sending UDP checksums, and allowing TX and RX of UDP6 checksums. Also, call common iptunnel_handle_offloads and added GSO support for checksums. Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-06-05 02:20:29 +02:00
#define VXLAN_F_LEARN 0x01
#define VXLAN_F_PROXY 0x02
#define VXLAN_F_RSC 0x04
#define VXLAN_F_L2MISS 0x08
#define VXLAN_F_L3MISS 0x10
#define VXLAN_F_IPV6 0x20
#define VXLAN_F_UDP_CSUM 0x40
#define VXLAN_F_UDP_ZERO_CSUM6_TX 0x80
#define VXLAN_F_UDP_ZERO_CSUM6_RX 0x100
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
struct vxlan_sock *vxlan_sock_add(struct net *net, __be16 port,
vxlan_rcv_t *rcv, void *data,
vxlan: Add support for UDP checksums (v4 sending, v6 zero csums) Added VXLAN link configuration for sending UDP checksums, and allowing TX and RX of UDP6 checksums. Also, call common iptunnel_handle_offloads and added GSO support for checksums. Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-06-05 02:20:29 +02:00
bool no_share, u32 flags);
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
void vxlan_sock_release(struct vxlan_sock *vs);
vxlan: Factor out vxlan send api. Following patch allows more code sharing between vxlan and ovs-vxlan. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:17 +02:00
vxlan: remove net arg from vxlan[6]_xmit_skb() This argument is not used, let's remove it. Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-09-02 15:34:55 +02:00
int vxlan_xmit_skb(struct vxlan_sock *vs,
vxlan: Factor out vxlan send api. Following patch allows more code sharing between vxlan and ovs-vxlan. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:17 +02:00
struct rtable *rt, struct sk_buff *skb,
__be32 src, __be32 dst, __u8 tos, __u8 ttl, __be16 df,
vxlan: add x-netns support This patch allows to switch the netns when packet is encapsulated or decapsulated. The vxlan socket is openned into the i/o netns, ie into the netns where encapsulated packets are received. The socket lookup is done into this netns to find the corresponding vxlan tunnel. After decapsulation, the packet is injecting into the corresponding interface which may stand to another netns. When one of the two netns is removed, the tunnel is destroyed. Configuration example: ip netns add netns1 ip netns exec netns1 ip link set lo up ip link add vxlan10 type vxlan id 10 group 239.0.0.10 dev eth0 dstport 0 ip link set vxlan10 netns netns1 ip netns exec netns1 ip addr add 192.168.0.249/24 broadcast 192.168.0.255 dev vxlan10 ip netns exec netns1 ip link set vxlan10 up Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-24 10:02:49 +02:00
__be16 src_port, __be16 dst_port, __be32 vni, bool xnet);
vxlan: Factor out vxlan send api. Following patch allows more code sharing between vxlan and ovs-vxlan. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:17 +02:00
net: Generalize ndo_gso_check to ndo_features_check GSO isn't the only offload feature with restrictions that potentially can't be expressed with the current features mechanism. Checksum is another although it's a general issue that could in theory apply to anything. Even if it may be possible to implement these restrictions in other ways, it can result in duplicate code or inefficient per-packet behavior. This generalizes ndo_gso_check so that drivers can remove any features that don't make sense for a given packet, similar to netif_skb_features(). It also converts existing driver restrictions to the new format, completing the work that was done to support tunnel protocols since the issues apply to checksums as well. By actually removing features from the set that are used to do offloading, it solves another problem with the existing interface. In these cases, GSO would run with the original set of features and not do anything because it appears that segmentation is not required. CC: Tom Herbert <therbert@google.com> CC: Joe Stringer <joestringer@nicira.com> CC: Eric Dumazet <edumazet@google.com> CC: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jesse Gross <jesse@nicira.com> Acked-by: Tom Herbert <therbert@google.com> Fixes: 04ffcb255f22 ("net: Add ndo_gso_check") Tested-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-24 07:37:26 +01:00
static inline netdev_features_t vxlan_features_check(struct sk_buff *skb,
netdev_features_t features)
vxlan: Inline vxlan_gso_check(). Suggested-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-18 01:24:54 +01:00
{
net: Generalize ndo_gso_check to ndo_features_check GSO isn't the only offload feature with restrictions that potentially can't be expressed with the current features mechanism. Checksum is another although it's a general issue that could in theory apply to anything. Even if it may be possible to implement these restrictions in other ways, it can result in duplicate code or inefficient per-packet behavior. This generalizes ndo_gso_check so that drivers can remove any features that don't make sense for a given packet, similar to netif_skb_features(). It also converts existing driver restrictions to the new format, completing the work that was done to support tunnel protocols since the issues apply to checksums as well. By actually removing features from the set that are used to do offloading, it solves another problem with the existing interface. In these cases, GSO would run with the original set of features and not do anything because it appears that segmentation is not required. CC: Tom Herbert <therbert@google.com> CC: Joe Stringer <joestringer@nicira.com> CC: Eric Dumazet <edumazet@google.com> CC: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jesse Gross <jesse@nicira.com> Acked-by: Tom Herbert <therbert@google.com> Fixes: 04ffcb255f22 ("net: Add ndo_gso_check") Tested-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-24 07:37:26 +01:00
u8 l4_hdr = 0;
if (!skb->encapsulation)
return features;
switch (vlan_get_protocol(skb)) {
case htons(ETH_P_IP):
l4_hdr = ip_hdr(skb)->protocol;
break;
case htons(ETH_P_IPV6):
l4_hdr = ipv6_hdr(skb)->nexthdr;
break;
default:
return features;;
}
if ((l4_hdr == IPPROTO_UDP) &&
vxlan: Inline vxlan_gso_check(). Suggested-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-18 01:24:54 +01:00
(skb->inner_protocol_type != ENCAP_TYPE_ETHER ||
skb->inner_protocol != htons(ETH_P_TEB) ||
(skb_inner_mac_header(skb) - skb_transport_header(skb) !=
sizeof(struct udphdr) + sizeof(struct vxlanhdr))))
net: Generalize ndo_gso_check to ndo_features_check GSO isn't the only offload feature with restrictions that potentially can't be expressed with the current features mechanism. Checksum is another although it's a general issue that could in theory apply to anything. Even if it may be possible to implement these restrictions in other ways, it can result in duplicate code or inefficient per-packet behavior. This generalizes ndo_gso_check so that drivers can remove any features that don't make sense for a given packet, similar to netif_skb_features(). It also converts existing driver restrictions to the new format, completing the work that was done to support tunnel protocols since the issues apply to checksums as well. By actually removing features from the set that are used to do offloading, it solves another problem with the existing interface. In these cases, GSO would run with the original set of features and not do anything because it appears that segmentation is not required. CC: Tom Herbert <therbert@google.com> CC: Joe Stringer <joestringer@nicira.com> CC: Eric Dumazet <edumazet@google.com> CC: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jesse Gross <jesse@nicira.com> Acked-by: Tom Herbert <therbert@google.com> Fixes: 04ffcb255f22 ("net: Add ndo_gso_check") Tested-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-24 07:37:26 +01:00
return features & ~(NETIF_F_ALL_CSUM | NETIF_F_GSO_MASK);
vxlan: Inline vxlan_gso_check(). Suggested-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-18 01:24:54 +01:00
net: Generalize ndo_gso_check to ndo_features_check GSO isn't the only offload feature with restrictions that potentially can't be expressed with the current features mechanism. Checksum is another although it's a general issue that could in theory apply to anything. Even if it may be possible to implement these restrictions in other ways, it can result in duplicate code or inefficient per-packet behavior. This generalizes ndo_gso_check so that drivers can remove any features that don't make sense for a given packet, similar to netif_skb_features(). It also converts existing driver restrictions to the new format, completing the work that was done to support tunnel protocols since the issues apply to checksums as well. By actually removing features from the set that are used to do offloading, it solves another problem with the existing interface. In these cases, GSO would run with the original set of features and not do anything because it appears that segmentation is not required. CC: Tom Herbert <therbert@google.com> CC: Joe Stringer <joestringer@nicira.com> CC: Eric Dumazet <edumazet@google.com> CC: Hayes Wang <hayeswang@realtek.com> Signed-off-by: Jesse Gross <jesse@nicira.com> Acked-by: Tom Herbert <therbert@google.com> Fixes: 04ffcb255f22 ("net: Add ndo_gso_check") Tested-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-24 07:37:26 +01:00
return features;
vxlan: Inline vxlan_gso_check(). Suggested-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-18 01:24:54 +01:00
}
net: Add vxlan_gso_check() helper Most NICs that report NETIF_F_GSO_UDP_TUNNEL support VXLAN, and not other UDP-based encapsulation protocols where the format and size of the header differs. This patch implements a generic ndo_gso_check() for VXLAN which will only advertise GSO support when the skb looks like it contains VXLAN (or no UDP tunnelling at all). Implementation shamelessly stolen from Tom Herbert: http://thread.gmane.org/gmane.linux.network/332428/focus=333111 Signed-off-by: Joe Stringer <joestringer@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-14 01:38:12 +01:00
vxlan: Have the NIC drivers do less work for offloads This patch removes the burden from the NIC drivers to check if the vxlan driver is enabled in the kernel and also makes available the vxlan headrooms to them. Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-24 08:27:10 +02:00
/* IP header + UDP + VXLAN + Ethernet header */
#define VXLAN_HEADROOM (20 + 8 + 8 + 14)
/* IPv6 header + UDP + VXLAN + Ethernet header */
#define VXLAN6_HEADROOM (40 + 8 + 8 + 14)
#if IS_ENABLED(CONFIG_VXLAN)
vxlan: Notify drivers for listening UDP port changes This patch adds two more ndo ops: ndo_add_rx_vxlan_port() and ndo_del_rx_vxlan_port(). Drivers can get notifications through the above functions about changes of the UDP listening port of VXLAN. Also, when physical ports come up, now they can call vxlan_get_rx_port() in order to obtain the port number(s) of the existing VXLAN interface in case they already up before them. This information about the listening UDP port would be used for VXLAN related offloads. A big thank you to John Fastabend (john.r.fastabend@intel.com) for his input and his suggestions on this patch set. CC: John Fastabend <john.r.fastabend@intel.com> CC: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-09-04 11:13:38 +02:00
void vxlan_get_rx_port(struct net_device *netdev);
vxlan: Have the NIC drivers do less work for offloads This patch removes the burden from the NIC drivers to check if the vxlan driver is enabled in the kernel and also makes available the vxlan headrooms to them. Signed-off-by: Joseph Gasparakis <joseph.gasparakis@intel.com> Tested-by: Kavindya Deegala <kavindya.s.deegala@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2013-10-24 08:27:10 +02:00
#else
static inline void vxlan_get_rx_port(struct net_device *netdev)
{
}
#endif
vxlan: Extend vxlan handlers for openvswitch. Following patch adds data field to vxlan socket and export vxlan handler api. vh->data is required to store private data per vxlan handler. Signed-off-by: Pravin B Shelar <pshelar@nicira.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-19 20:23:07 +02:00
#endif
Reference in New Issue Copy Permalink