Commit 5b2941b1 authored by David S. Miller's avatar David S. Miller

Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch

Jesse Gross says:

====================
A number of significant new features and optimizations for net-next/3.12.
Highlights are:
 * "Megaflows", an optimization that allows userspace to specify which
   flow fields were used to compute the results of the flow lookup.
   This allows for a major reduction in flow setups (the major
   performance bottleneck in Open vSwitch) without reducing flexibility.
 * Converting netlink dump operations to use RCU, allowing for
   additional parallelism in userspace.
 * Matching and modifying SCTP protocol fields.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents b6750b40 5828cd9a
...@@ -91,6 +91,46 @@ Often we ellipsize arguments not important to the discussion, e.g.: ...@@ -91,6 +91,46 @@ Often we ellipsize arguments not important to the discussion, e.g.:
in_port(1), eth(...), eth_type(0x0800), ipv4(...), tcp(...) in_port(1), eth(...), eth_type(0x0800), ipv4(...), tcp(...)
Wildcarded flow key format
--------------------------
A wildcarded flow is described with two sequences of Netlink attributes
passed over the Netlink socket. A flow key, exactly as described above, and an
optional corresponding flow mask.
A wildcarded flow can represent a group of exact match flows. Each '1' bit
in the mask specifies a exact match with the corresponding bit in the flow key.
A '0' bit specifies a don't care bit, which will match either a '1' or '0' bit
of a incoming packet. Using wildcarded flow can improve the flow set up rate
by reduce the number of new flows need to be processed by the user space program.
Support for the mask Netlink attribute is optional for both the kernel and user
space program. The kernel can ignore the mask attribute, installing an exact
match flow, or reduce the number of don't care bits in the kernel to less than
what was specified by the user space program. In this case, variations in bits
that the kernel does not implement will simply result in additional flow setups.
The kernel module will also work with user space programs that neither support
nor supply flow mask attributes.
Since the kernel may ignore or modify wildcard bits, it can be difficult for
the userspace program to know exactly what matches are installed. There are
two possible approaches: reactively install flows as they miss the kernel
flow table (and therefore not attempt to determine wildcard changes at all)
or use the kernel's response messages to determine the installed wildcards.
When interacting with userspace, the kernel should maintain the match portion
of the key exactly as originally installed. This will provides a handle to
identify the flow for all future operations. However, when reporting the
mask of an installed flow, the mask should include any restrictions imposed
by the kernel.
The behavior when using overlapping wildcarded flows is undefined. It is the
responsibility of the user space program to ensure that any incoming packet
can match at most one flow, wildcarded or not. The current implementation
performs best-effort detection of overlapping wildcarded flows and may reject
some but not all of them. However, this behavior may change in future versions.
Basic rule for evolving flow keys Basic rule for evolving flow keys
--------------------------------- ---------------------------------
......
...@@ -41,6 +41,7 @@ ...@@ -41,6 +41,7 @@
#define NEXTHDR_ICMP 58 /* ICMP for IPv6. */ #define NEXTHDR_ICMP 58 /* ICMP for IPv6. */
#define NEXTHDR_NONE 59 /* No next header */ #define NEXTHDR_NONE 59 /* No next header */
#define NEXTHDR_DEST 60 /* Destination options header. */ #define NEXTHDR_DEST 60 /* Destination options header. */
#define NEXTHDR_SCTP 132 /* SCTP message. */
#define NEXTHDR_MOBILITY 135 /* Mobility header. */ #define NEXTHDR_MOBILITY 135 /* Mobility header. */
#define NEXTHDR_MAX 255 #define NEXTHDR_MAX 255
......
/* /*
* Copyright (c) 2007-2011 Nicira Networks. * Copyright (c) 2007-2013 Nicira, Inc.
* *
* This program is free software; you can redistribute it and/or * This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public * modify it under the terms of version 2 of the GNU General Public
...@@ -259,6 +259,7 @@ enum ovs_key_attr { ...@@ -259,6 +259,7 @@ enum ovs_key_attr {
OVS_KEY_ATTR_ND, /* struct ovs_key_nd */ OVS_KEY_ATTR_ND, /* struct ovs_key_nd */
OVS_KEY_ATTR_SKB_MARK, /* u32 skb mark */ OVS_KEY_ATTR_SKB_MARK, /* u32 skb mark */
OVS_KEY_ATTR_TUNNEL, /* Nested set of ovs_tunnel attributes */ OVS_KEY_ATTR_TUNNEL, /* Nested set of ovs_tunnel attributes */
OVS_KEY_ATTR_SCTP, /* struct ovs_key_sctp */
#ifdef __KERNEL__ #ifdef __KERNEL__
OVS_KEY_ATTR_IPV4_TUNNEL, /* struct ovs_key_ipv4_tunnel */ OVS_KEY_ATTR_IPV4_TUNNEL, /* struct ovs_key_ipv4_tunnel */
...@@ -333,6 +334,11 @@ struct ovs_key_udp { ...@@ -333,6 +334,11 @@ struct ovs_key_udp {
__be16 udp_dst; __be16 udp_dst;
}; };
struct ovs_key_sctp {
__be16 sctp_src;
__be16 sctp_dst;
};
struct ovs_key_icmp { struct ovs_key_icmp {
__u8 icmp_type; __u8 icmp_type;
__u8 icmp_code; __u8 icmp_code;
...@@ -379,6 +385,12 @@ struct ovs_key_nd { ...@@ -379,6 +385,12 @@ struct ovs_key_nd {
* @OVS_FLOW_ATTR_CLEAR: If present in a %OVS_FLOW_CMD_SET request, clears the * @OVS_FLOW_ATTR_CLEAR: If present in a %OVS_FLOW_CMD_SET request, clears the
* last-used time, accumulated TCP flags, and statistics for this flow. * last-used time, accumulated TCP flags, and statistics for this flow.
* Otherwise ignored in requests. Never present in notifications. * Otherwise ignored in requests. Never present in notifications.
* @OVS_FLOW_ATTR_MASK: Nested %OVS_KEY_ATTR_* attributes specifying the
* mask bits for wildcarded flow match. Mask bit value '1' specifies exact
* match with corresponding flow key bit, while mask bit value '0' specifies
* a wildcarded match. Omitting attribute is treated as wildcarding all
* corresponding fields. Optional for all requests. If not present,
* all flow key bits are exact match bits.
* *
* These attributes follow the &struct ovs_header within the Generic Netlink * These attributes follow the &struct ovs_header within the Generic Netlink
* payload for %OVS_FLOW_* commands. * payload for %OVS_FLOW_* commands.
...@@ -391,6 +403,7 @@ enum ovs_flow_attr { ...@@ -391,6 +403,7 @@ enum ovs_flow_attr {
OVS_FLOW_ATTR_TCP_FLAGS, /* 8-bit OR'd TCP flags. */ OVS_FLOW_ATTR_TCP_FLAGS, /* 8-bit OR'd TCP flags. */
OVS_FLOW_ATTR_USED, /* u64 msecs last used in monotonic time. */ OVS_FLOW_ATTR_USED, /* u64 msecs last used in monotonic time. */
OVS_FLOW_ATTR_CLEAR, /* Flag to clear stats, tcp_flags, used. */ OVS_FLOW_ATTR_CLEAR, /* Flag to clear stats, tcp_flags, used. */
OVS_FLOW_ATTR_MASK, /* Sequence of OVS_KEY_ATTR_* attributes. */
__OVS_FLOW_ATTR_MAX __OVS_FLOW_ATTR_MAX
}; };
......
...@@ -4,6 +4,7 @@ ...@@ -4,6 +4,7 @@
config OPENVSWITCH config OPENVSWITCH
tristate "Open vSwitch" tristate "Open vSwitch"
select LIBCRC32C
---help--- ---help---
Open vSwitch is a multilayer Ethernet switch targeted at virtualized Open vSwitch is a multilayer Ethernet switch targeted at virtualized
environments. In addition to supporting a variety of features environments. In addition to supporting a variety of features
......
...@@ -10,10 +10,13 @@ openvswitch-y := \ ...@@ -10,10 +10,13 @@ openvswitch-y := \
dp_notify.o \ dp_notify.o \
flow.o \ flow.o \
vport.o \ vport.o \
vport-gre.o \
vport-internal_dev.o \ vport-internal_dev.o \
vport-netdev.o vport-netdev.o
ifneq ($(CONFIG_OPENVSWITCH_VXLAN),) ifneq ($(CONFIG_OPENVSWITCH_VXLAN),)
openvswitch-y += vport-vxlan.o openvswitch-y += vport-vxlan.o
endif endif
ifneq ($(CONFIG_OPENVSWITCH_GRE),)
openvswitch-y += vport-gre.o
endif
/* /*
* Copyright (c) 2007-2012 Nicira, Inc. * Copyright (c) 2007-2013 Nicira, Inc.
* *
* This program is free software; you can redistribute it and/or * This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public * modify it under the terms of version 2 of the GNU General Public
...@@ -22,6 +22,7 @@ ...@@ -22,6 +22,7 @@
#include <linux/in.h> #include <linux/in.h>
#include <linux/ip.h> #include <linux/ip.h>
#include <linux/openvswitch.h> #include <linux/openvswitch.h>
#include <linux/sctp.h>
#include <linux/tcp.h> #include <linux/tcp.h>
#include <linux/udp.h> #include <linux/udp.h>
#include <linux/in6.h> #include <linux/in6.h>
...@@ -31,6 +32,7 @@ ...@@ -31,6 +32,7 @@
#include <net/ipv6.h> #include <net/ipv6.h>
#include <net/checksum.h> #include <net/checksum.h>
#include <net/dsfield.h> #include <net/dsfield.h>
#include <net/sctp/checksum.h>
#include "datapath.h" #include "datapath.h"
#include "vport.h" #include "vport.h"
...@@ -352,6 +354,39 @@ static int set_tcp(struct sk_buff *skb, const struct ovs_key_tcp *tcp_port_key) ...@@ -352,6 +354,39 @@ static int set_tcp(struct sk_buff *skb, const struct ovs_key_tcp *tcp_port_key)
return 0; return 0;
} }
static int set_sctp(struct sk_buff *skb,
const struct ovs_key_sctp *sctp_port_key)
{
struct sctphdr *sh;
int err;
unsigned int sctphoff = skb_transport_offset(skb);
err = make_writable(skb, sctphoff + sizeof(struct sctphdr));
if (unlikely(err))
return err;
sh = sctp_hdr(skb);
if (sctp_port_key->sctp_src != sh->source ||
sctp_port_key->sctp_dst != sh->dest) {
__le32 old_correct_csum, new_csum, old_csum;
old_csum = sh->checksum;
old_correct_csum = sctp_compute_cksum(skb, sctphoff);
sh->source = sctp_port_key->sctp_src;
sh->dest = sctp_port_key->sctp_dst;
new_csum = sctp_compute_cksum(skb, sctphoff);
/* Carry any checksum errors through. */
sh->checksum = old_csum ^ old_correct_csum ^ new_csum;
skb->rxhash = 0;
}
return 0;
}
static int do_output(struct datapath *dp, struct sk_buff *skb, int out_port) static int do_output(struct datapath *dp, struct sk_buff *skb, int out_port)
{ {
struct vport *vport; struct vport *vport;
...@@ -376,8 +411,10 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb, ...@@ -376,8 +411,10 @@ static int output_userspace(struct datapath *dp, struct sk_buff *skb,
const struct nlattr *a; const struct nlattr *a;
int rem; int rem;
BUG_ON(!OVS_CB(skb)->pkt_key);
upcall.cmd = OVS_PACKET_CMD_ACTION; upcall.cmd = OVS_PACKET_CMD_ACTION;
upcall.key = &OVS_CB(skb)->flow->key; upcall.key = OVS_CB(skb)->pkt_key;
upcall.userdata = NULL; upcall.userdata = NULL;
upcall.portid = 0; upcall.portid = 0;
...@@ -459,6 +496,10 @@ static int execute_set_action(struct sk_buff *skb, ...@@ -459,6 +496,10 @@ static int execute_set_action(struct sk_buff *skb,
case OVS_KEY_ATTR_UDP: case OVS_KEY_ATTR_UDP:
err = set_udp(skb, nla_data(nested_attr)); err = set_udp(skb, nla_data(nested_attr));
break; break;
case OVS_KEY_ATTR_SCTP:
err = set_sctp(skb, nla_data(nested_attr));
break;
} }
return err; return err;
......
This diff is collapsed.
...@@ -88,11 +88,13 @@ struct datapath { ...@@ -88,11 +88,13 @@ struct datapath {
/** /**
* struct ovs_skb_cb - OVS data in skb CB * struct ovs_skb_cb - OVS data in skb CB
* @flow: The flow associated with this packet. May be %NULL if no flow. * @flow: The flow associated with this packet. May be %NULL if no flow.
* @pkt_key: The flow information extracted from the packet. Must be nonnull.
* @tun_key: Key for the tunnel that encapsulated this packet. NULL if the * @tun_key: Key for the tunnel that encapsulated this packet. NULL if the
* packet is not being tunneled. * packet is not being tunneled.
*/ */
struct ovs_skb_cb { struct ovs_skb_cb {
struct sw_flow *flow; struct sw_flow *flow;
struct sw_flow_key *pkt_key;
struct ovs_key_ipv4_tunnel *tun_key; struct ovs_key_ipv4_tunnel *tun_key;
}; };
#define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb) #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb)
...@@ -183,4 +185,8 @@ struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq, ...@@ -183,4 +185,8 @@ struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq,
int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb); int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb);
void ovs_dp_notify_wq(struct work_struct *work); void ovs_dp_notify_wq(struct work_struct *work);
#define OVS_NLERR(fmt, ...) \
pr_info_once("netlink: " fmt, ##__VA_ARGS__)
#endif /* datapath.h */ #endif /* datapath.h */
This diff is collapsed.
/* /*
* Copyright (c) 2007-2011 Nicira, Inc. * Copyright (c) 2007-2013 Nicira, Inc.
* *
* This program is free software; you can redistribute it and/or * This program is free software; you can redistribute it and/or
* modify it under the terms of version 2 of the GNU General Public * modify it under the terms of version 2 of the GNU General Public
...@@ -33,6 +33,8 @@ ...@@ -33,6 +33,8 @@
#include <net/inet_ecn.h> #include <net/inet_ecn.h>
struct sk_buff; struct sk_buff;
struct sw_flow_mask;
struct flow_table;
struct sw_flow_actions { struct sw_flow_actions {
struct rcu_head rcu; struct rcu_head rcu;
...@@ -97,8 +99,8 @@ struct sw_flow_key { ...@@ -97,8 +99,8 @@ struct sw_flow_key {
} addr; } addr;
union { union {
struct { struct {
__be16 src; /* TCP/UDP source port. */ __be16 src; /* TCP/UDP/SCTP source port. */
__be16 dst; /* TCP/UDP destination port. */ __be16 dst; /* TCP/UDP/SCTP destination port. */
} tp; } tp;
struct { struct {
u8 sha[ETH_ALEN]; /* ARP source hardware address. */ u8 sha[ETH_ALEN]; /* ARP source hardware address. */
...@@ -113,8 +115,8 @@ struct sw_flow_key { ...@@ -113,8 +115,8 @@ struct sw_flow_key {
} addr; } addr;
__be32 label; /* IPv6 flow label. */ __be32 label; /* IPv6 flow label. */
struct { struct {
__be16 src; /* TCP/UDP source port. */ __be16 src; /* TCP/UDP/SCTP source port. */
__be16 dst; /* TCP/UDP destination port. */ __be16 dst; /* TCP/UDP/SCTP destination port. */
} tp; } tp;
struct { struct {
struct in6_addr target; /* ND target address. */ struct in6_addr target; /* ND target address. */
...@@ -123,7 +125,7 @@ struct sw_flow_key { ...@@ -123,7 +125,7 @@ struct sw_flow_key {
} nd; } nd;
} ipv6; } ipv6;
}; };
}; } __aligned(__alignof__(long));
struct sw_flow { struct sw_flow {
struct rcu_head rcu; struct rcu_head rcu;
...@@ -131,6 +133,8 @@ struct sw_flow { ...@@ -131,6 +133,8 @@ struct sw_flow {
u32 hash; u32 hash;
struct sw_flow_key key; struct sw_flow_key key;
struct sw_flow_key unmasked_key;
struct sw_flow_mask *mask;
struct sw_flow_actions __rcu *sf_acts; struct sw_flow_actions __rcu *sf_acts;
spinlock_t lock; /* Lock for values below. */ spinlock_t lock; /* Lock for values below. */
...@@ -140,6 +144,20 @@ struct sw_flow { ...@@ -140,6 +144,20 @@ struct sw_flow {
u8 tcp_flags; /* Union of seen TCP flags. */ u8 tcp_flags; /* Union of seen TCP flags. */
}; };
struct sw_flow_key_range {
size_t start;
size_t end;
};
struct sw_flow_match {
struct sw_flow_key *key;
struct sw_flow_key_range range;
struct sw_flow_mask *mask;
};
void ovs_match_init(struct sw_flow_match *match,
struct sw_flow_key *key, struct sw_flow_mask *mask);
struct arp_eth_header { struct arp_eth_header {
__be16 ar_hrd; /* format of hardware address */ __be16 ar_hrd; /* format of hardware address */
__be16 ar_pro; /* format of protocol address */ __be16 ar_pro; /* format of protocol address */
...@@ -159,21 +177,21 @@ void ovs_flow_exit(void); ...@@ -159,21 +177,21 @@ void ovs_flow_exit(void);
struct sw_flow *ovs_flow_alloc(void); struct sw_flow *ovs_flow_alloc(void);
void ovs_flow_deferred_free(struct sw_flow *); void ovs_flow_deferred_free(struct sw_flow *);
void ovs_flow_free(struct sw_flow *flow); void ovs_flow_free(struct sw_flow *, bool deferred);
struct sw_flow_actions *ovs_flow_actions_alloc(int actions_len); struct sw_flow_actions *ovs_flow_actions_alloc(int actions_len);
void ovs_flow_deferred_free_acts(struct sw_flow_actions *); void ovs_flow_deferred_free_acts(struct sw_flow_actions *);
int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *, int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *);
int *key_lenp);
void ovs_flow_used(struct sw_flow *, struct sk_buff *); void ovs_flow_used(struct sw_flow *, struct sk_buff *);
u64 ovs_flow_used_time(unsigned long flow_jiffies); u64 ovs_flow_used_time(unsigned long flow_jiffies);
int ovs_flow_to_nlattrs(const struct sw_flow_key *,
int ovs_flow_to_nlattrs(const struct sw_flow_key *, struct sk_buff *); const struct sw_flow_key *, struct sk_buff *);
int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp, int ovs_match_from_nlattrs(struct sw_flow_match *match,
const struct nlattr *,
const struct nlattr *); const struct nlattr *);
int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow, int key_len, int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow,
const struct nlattr *attr); const struct nlattr *attr);
#define MAX_ACTIONS_BUFSIZE (32 * 1024) #define MAX_ACTIONS_BUFSIZE (32 * 1024)
#define TBL_MIN_BUCKETS 1024 #define TBL_MIN_BUCKETS 1024
...@@ -182,6 +200,7 @@ struct flow_table { ...@@ -182,6 +200,7 @@ struct flow_table {
struct flex_array *buckets; struct flex_array *buckets;
unsigned int count, n_buckets; unsigned int count, n_buckets;
struct rcu_head rcu; struct rcu_head rcu;
struct list_head *mask_list;
int node_ver; int node_ver;
u32 hash_seed; u32 hash_seed;
bool keep_flows; bool keep_flows;
...@@ -197,22 +216,44 @@ static inline int ovs_flow_tbl_need_to_expand(struct flow_table *table) ...@@ -197,22 +216,44 @@ static inline int ovs_flow_tbl_need_to_expand(struct flow_table *table)
return (table->count > table->n_buckets); return (table->count > table->n_buckets);
} }
struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *table, struct sw_flow *ovs_flow_lookup(struct flow_table *,
struct sw_flow_key *key, int len); const struct sw_flow_key *);
void ovs_flow_tbl_destroy(struct flow_table *table); struct sw_flow *ovs_flow_lookup_unmasked_key(struct flow_table *table,
void ovs_flow_tbl_deferred_destroy(struct flow_table *table); struct sw_flow_match *match);
void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred);
struct flow_table *ovs_flow_tbl_alloc(int new_size); struct flow_table *ovs_flow_tbl_alloc(int new_size);
struct flow_table *ovs_flow_tbl_expand(struct flow_table *table); struct flow_table *ovs_flow_tbl_expand(struct flow_table *table);
struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table); struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table);
void ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow,
struct sw_flow_key *key, int key_len);
void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow);
struct sw_flow *ovs_flow_tbl_next(struct flow_table *table, u32 *bucket, u32 *idx); void ovs_flow_insert(struct flow_table *table, struct sw_flow *flow);
void ovs_flow_remove(struct flow_table *table, struct sw_flow *flow);
struct sw_flow *ovs_flow_dump_next(struct flow_table *table, u32 *bucket, u32 *idx);
extern const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1]; extern const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1];
int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr, int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr,
struct ovs_key_ipv4_tunnel *tun_key); struct sw_flow_match *match, bool is_mask);
int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb, int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb,
const struct ovs_key_ipv4_tunnel *tun_key); const struct ovs_key_ipv4_tunnel *tun_key,
const struct ovs_key_ipv4_tunnel *output);
bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
const struct sw_flow_key *key, int key_end);
struct sw_flow_mask {
int ref_count;
struct rcu_head rcu;
struct list_head list;
struct sw_flow_key_range range;
struct sw_flow_key key;
};
struct sw_flow_mask *ovs_sw_flow_mask_alloc(void);
void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *);
void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *, bool deferred);
void ovs_sw_flow_mask_insert(struct flow_table *, struct sw_flow_mask *);
struct sw_flow_mask *ovs_sw_flow_mask_find(const struct flow_table *,
const struct sw_flow_mask *);
void ovs_flow_key_mask(struct sw_flow_key *dst, const struct sw_flow_key *src,
const struct sw_flow_mask *mask);
#endif /* flow.h */ #endif /* flow.h */
...@@ -16,7 +16,6 @@ ...@@ -16,7 +16,6 @@
* 02110-1301, USA * 02110-1301, USA
*/ */
#ifdef CONFIG_OPENVSWITCH_GRE
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/if.h> #include <linux/if.h>
...@@ -271,5 +270,3 @@ const struct vport_ops ovs_gre_vport_ops = { ...@@ -271,5 +270,3 @@ const struct vport_ops ovs_gre_vport_ops = {
.get_name = gre_get_name, .get_name = gre_get_name,
.send = gre_tnl_send, .send = gre_tnl_send,
}; };
#endif /* OPENVSWITCH_GRE */
...@@ -25,6 +25,7 @@ ...@@ -25,6 +25,7 @@
#include <linux/llc.h> #include <linux/llc.h>
#include <linux/rtnetlink.h> #include <linux/rtnetlink.h>
#include <linux/skbuff.h> #include <linux/skbuff.h>
#include <linux/openvswitch.h>
#include <net/llc.h> #include <net/llc.h>
...@@ -74,6 +75,15 @@ static rx_handler_result_t netdev_frame_hook(struct sk_buff **pskb) ...@@ -74,6 +75,15 @@ static rx_handler_result_t netdev_frame_hook(struct sk_buff **pskb)
return RX_HANDLER_CONSUMED; return RX_HANDLER_CONSUMED;
} }
static struct net_device *get_dpdev(struct datapath *dp)
{
struct vport *local;
local = ovs_vport_ovsl(dp, OVSP_LOCAL);
BUG_ON(!local);
return netdev_vport_priv(local)->dev;
}
static struct vport *netdev_create(const struct vport_parms *parms) static struct vport *netdev_create(const struct vport_parms *parms)
{ {
struct vport *vport; struct vport *vport;
...@@ -103,10 +113,15 @@ static struct vport *netdev_create(const struct vport_parms *parms) ...@@ -103,10 +113,15 @@ static struct vport *netdev_create(const struct vport_parms *parms)
} }
rtnl_lock(); rtnl_lock();
err = netdev_master_upper_dev_link(netdev_vport->dev,
get_dpdev(vport->dp));
if (err)
goto error_unlock;
err = netdev_rx_handler_register(netdev_vport->dev, netdev_frame_hook, err = netdev_rx_handler_register(netdev_vport->dev, netdev_frame_hook,
vport); vport);
if (err) if (err)
goto error_unlock; goto error_master_upper_dev_unlink;
dev_set_promiscuity(netdev_vport->dev, 1); dev_set_promiscuity(netdev_vport->dev, 1);
netdev_vport->dev->priv_flags |= IFF_OVS_DATAPATH; netdev_vport->dev->priv_flags |= IFF_OVS_DATAPATH;
...@@ -114,6 +129,8 @@ static struct vport *netdev_create(const struct vport_parms *parms) ...@@ -114,6 +129,8 @@ static struct vport *netdev_create(const struct vport_parms *parms)
return vport; return vport;
error_master_upper_dev_unlink:
netdev_upper_dev_unlink(netdev_vport->dev, get_dpdev(vport->dp));
error_unlock: error_unlock:
rtnl_unlock(); rtnl_unlock();
error_put: error_put:
...@@ -140,6 +157,7 @@ static void netdev_destroy(struct vport *vport) ...@@ -140,6 +157,7 @@ static void netdev_destroy(struct vport *vport)
rtnl_lock(); rtnl_lock();
netdev_vport->dev->priv_flags &= ~IFF_OVS_DATAPATH; netdev_vport->dev->priv_flags &= ~IFF_OVS_DATAPATH;
netdev_rx_handler_unregister(netdev_vport->dev); netdev_rx_handler_unregister(netdev_vport->dev);
netdev_upper_dev_unlink(netdev_vport->dev, get_dpdev(vport->dp));
dev_set_promiscuity(netdev_vport->dev, -1); dev_set_promiscuity(netdev_vport->dev, -1);
rtnl_unlock(); rtnl_unlock();
......
...@@ -203,7 +203,7 @@ struct vport *ovs_vport_add(const struct vport_parms *parms) ...@@ -203,7 +203,7 @@ struct vport *ovs_vport_add(const struct vport_parms *parms)
* ovs_vport_set_options - modify existing vport device (for kernel callers) * ovs_vport_set_options - modify existing vport device (for kernel callers)
* *
* @vport: vport to modify. * @vport: vport to modify.
* @port: New configuration. * @options: New configuration.
* *
* Modifies an existing device with the specified configuration (which is * Modifies an existing device with the specified configuration (which is
* dependent on device type). ovs_mutex must be held. * dependent on device type). ovs_mutex must be held.
...@@ -328,6 +328,7 @@ int ovs_vport_get_options(const struct vport *vport, struct sk_buff *skb) ...@@ -328,6 +328,7 @@ int ovs_vport_get_options(const struct vport *vport, struct sk_buff *skb)
* *
* @vport: vport that received the packet * @vport: vport that received the packet
* @skb: skb that was received * @skb: skb that was received
* @tun_key: tunnel (if any) that carried packet
* *
* Must be called with rcu_read_lock. The packet cannot be shared and * Must be called with rcu_read_lock. The packet cannot be shared and
* skb->data should point to the Ethernet header. * skb->data should point to the Ethernet header.
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment