2011-10-26 10:26:31 +08:00
|
|
|
/*
|
2014-09-16 10:37:25 +08:00
|
|
|
* Copyright (c) 2007-2014 Nicira, Inc.
|
2011-10-26 10:26:31 +08:00
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of version 2 of the GNU General Public
|
|
|
|
* License as published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful, but
|
|
|
|
* WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software
|
|
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
|
|
|
|
* 02110-1301, USA
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef DATAPATH_H
|
|
|
|
#define DATAPATH_H 1
|
|
|
|
|
|
|
|
#include <asm/page.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/mutex.h>
|
|
|
|
#include <linux/netdevice.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/u64_stats_sync.h>
|
2015-07-21 16:43:54 +08:00
|
|
|
#include <net/ip_tunnels.h>
|
2011-10-26 10:26:31 +08:00
|
|
|
|
2015-08-27 02:31:48 +08:00
|
|
|
#include "conntrack.h"
|
2011-10-26 10:26:31 +08:00
|
|
|
#include "flow.h"
|
2013-10-04 09:16:47 +08:00
|
|
|
#include "flow_table.h"
|
2011-10-26 10:26:31 +08:00
|
|
|
|
2012-08-24 03:40:54 +08:00
|
|
|
#define DP_MAX_PORTS USHRT_MAX
|
|
|
|
#define DP_VPORT_HASH_BUCKETS 1024
|
|
|
|
|
2011-10-26 10:26:31 +08:00
|
|
|
#define SAMPLE_ACTION_DEPTH 3
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct dp_stats_percpu - per-cpu packet processing statistics for a given
|
|
|
|
* datapath.
|
|
|
|
* @n_hit: Number of received packets for which a matching flow was found in
|
|
|
|
* the flow table.
|
|
|
|
* @n_miss: Number of received packets that had no matching flow in the flow
|
|
|
|
* table. The sum of @n_hit and @n_miss is the number of packets that have
|
|
|
|
* been received by the datapath.
|
|
|
|
* @n_lost: Number of received packets that had no matching flow in the flow
|
|
|
|
* table that could not be sent to userspace (normally due to an overflow in
|
|
|
|
* one of the datapath's queues).
|
2013-10-23 01:42:46 +08:00
|
|
|
* @n_mask_hit: Number of masks looked up for flow match.
|
|
|
|
* @n_mask_hit / (@n_hit + @n_missed) will be the average masks looked
|
|
|
|
* up per packet.
|
2011-10-26 10:26:31 +08:00
|
|
|
*/
|
|
|
|
struct dp_stats_percpu {
|
|
|
|
u64 n_hit;
|
|
|
|
u64 n_missed;
|
|
|
|
u64 n_lost;
|
2013-10-23 01:42:46 +08:00
|
|
|
u64 n_mask_hit;
|
2014-02-15 07:10:46 +08:00
|
|
|
struct u64_stats_sync syncp;
|
2011-10-26 10:26:31 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct datapath - datapath for flow-based packet switching
|
|
|
|
* @rcu: RCU callback head for deferred destruction.
|
|
|
|
* @list_node: Element in global 'dps' list.
|
2013-10-04 15:14:23 +08:00
|
|
|
* @table: flow table.
|
2012-08-24 03:40:54 +08:00
|
|
|
* @ports: Hash table for ports. %OVSP_LOCAL port always exists. Protected by
|
2013-04-16 04:23:03 +08:00
|
|
|
* ovs_mutex and RCU.
|
2011-10-26 10:26:31 +08:00
|
|
|
* @stats_percpu: Per-CPU datapath statistics.
|
2012-02-23 11:58:59 +08:00
|
|
|
* @net: Reference to net namespace.
|
2016-02-26 17:45:39 +08:00
|
|
|
* @max_headroom: the maximum headroom of all vports in this datapath; it will
|
|
|
|
* be used by all the internal vports in this dp.
|
2011-10-26 10:26:31 +08:00
|
|
|
*
|
|
|
|
* Context: See the comment on locking at the top of datapath.c for additional
|
|
|
|
* locking information.
|
|
|
|
*/
|
|
|
|
struct datapath {
|
|
|
|
struct rcu_head rcu;
|
|
|
|
struct list_head list_node;
|
|
|
|
|
|
|
|
/* Flow table. */
|
2013-10-04 15:14:23 +08:00
|
|
|
struct flow_table table;
|
2011-10-26 10:26:31 +08:00
|
|
|
|
|
|
|
/* Switch ports. */
|
2012-08-24 03:40:54 +08:00
|
|
|
struct hlist_head *ports;
|
2011-10-26 10:26:31 +08:00
|
|
|
|
|
|
|
/* Stats. */
|
|
|
|
struct dp_stats_percpu __percpu *stats_percpu;
|
2012-02-23 11:58:59 +08:00
|
|
|
|
|
|
|
/* Network namespace ref. */
|
2015-03-12 12:06:44 +08:00
|
|
|
possible_net_t net;
|
2013-12-13 22:22:18 +08:00
|
|
|
|
|
|
|
u32 user_features;
|
2016-02-26 17:45:39 +08:00
|
|
|
|
|
|
|
u32 max_headroom;
|
2011-10-26 10:26:31 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct ovs_skb_cb - OVS data in skb CB
|
2014-09-16 10:20:31 +08:00
|
|
|
* @input_vport: The original vport packet came in on. This value is cached
|
|
|
|
* when a packet is received by OVS.
|
2015-08-27 02:31:48 +08:00
|
|
|
* @mru: The maximum received fragement size; 0 if the packet is not
|
2016-06-11 02:49:33 +08:00
|
|
|
* @cutlen: The number of bytes from the packet end to be removed.
|
2015-08-27 02:31:48 +08:00
|
|
|
* fragmented.
|
2011-10-26 10:26:31 +08:00
|
|
|
*/
|
|
|
|
struct ovs_skb_cb {
|
2014-09-16 10:20:31 +08:00
|
|
|
struct vport *input_vport;
|
2015-08-27 02:31:48 +08:00
|
|
|
u16 mru;
|
2016-06-11 02:49:33 +08:00
|
|
|
u32 cutlen;
|
2011-10-26 10:26:31 +08:00
|
|
|
};
|
|
|
|
#define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb)
|
|
|
|
|
|
|
|
/**
|
|
|
|
* struct dp_upcall - metadata to include with a packet to send to userspace
|
|
|
|
* @cmd: One of %OVS_PACKET_CMD_*.
|
2013-02-16 09:29:22 +08:00
|
|
|
* @userdata: If nonnull, its variable-length value is passed to userspace as
|
2011-10-26 10:26:31 +08:00
|
|
|
* %OVS_PACKET_ATTR_USERDATA.
|
2014-11-06 22:57:27 +08:00
|
|
|
* @portid: Netlink portid to which packet should be sent. If @portid is 0
|
|
|
|
* then no packet is sent and the packet is accounted in the datapath's @n_lost
|
2011-10-26 10:26:31 +08:00
|
|
|
* counter.
|
2014-11-06 22:51:24 +08:00
|
|
|
* @egress_tun_info: If nonnull, becomes %OVS_PACKET_ATTR_EGRESS_TUN_KEY.
|
2015-08-27 02:31:48 +08:00
|
|
|
* @mru: If not zero, Maximum received IP fragment size.
|
2011-10-26 10:26:31 +08:00
|
|
|
*/
|
|
|
|
struct dp_upcall_info {
|
2015-08-31 09:09:38 +08:00
|
|
|
struct ip_tunnel_info *egress_tun_info;
|
2011-10-26 10:26:31 +08:00
|
|
|
const struct nlattr *userdata;
|
2015-05-27 11:59:43 +08:00
|
|
|
const struct nlattr *actions;
|
|
|
|
int actions_len;
|
2012-09-08 04:12:54 +08:00
|
|
|
u32 portid;
|
2014-11-06 22:57:27 +08:00
|
|
|
u8 cmd;
|
2015-08-27 02:31:48 +08:00
|
|
|
u16 mru;
|
2011-10-26 10:26:31 +08:00
|
|
|
};
|
|
|
|
|
2013-04-16 04:23:03 +08:00
|
|
|
/**
|
|
|
|
* struct ovs_net - Per net-namespace data for ovs.
|
|
|
|
* @dps: List of datapaths to enable dumping them all out.
|
|
|
|
* Protected by genl_mutex.
|
|
|
|
*/
|
|
|
|
struct ovs_net {
|
|
|
|
struct list_head dps;
|
|
|
|
struct work_struct dp_notify_work;
|
2015-08-27 02:31:52 +08:00
|
|
|
|
|
|
|
/* Module reference for configuring conntrack. */
|
|
|
|
bool xt_label;
|
2013-04-16 04:23:03 +08:00
|
|
|
};
|
|
|
|
|
netns: make struct pernet_operations::id unsigned int
Make struct pernet_operations::id unsigned.
There are 2 reasons to do so:
1)
This field is really an index into an zero based array and
thus is unsigned entity. Using negative value is out-of-bound
access by definition.
2)
On x86_64 unsigned 32-bit data which are mixed with pointers
via array indexing or offsets added or subtracted to pointers
are preffered to signed 32-bit data.
"int" being used as an array index needs to be sign-extended
to 64-bit before being used.
void f(long *p, int i)
{
g(p[i]);
}
roughly translates to
movsx rsi, esi
mov rdi, [rsi+...]
call g
MOVSX is 3 byte instruction which isn't necessary if the variable is
unsigned because x86_64 is zero extending by default.
Now, there is net_generic() function which, you guessed it right, uses
"int" as an array index:
static inline void *net_generic(const struct net *net, int id)
{
...
ptr = ng->ptr[id - 1];
...
}
And this function is used a lot, so those sign extensions add up.
Patch snipes ~1730 bytes on allyesconfig kernel (without all junk
messing with code generation):
add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730)
Unfortunately some functions actually grow bigger.
This is a semmingly random artefact of code generation with register
allocator being used differently. gcc decides that some variable
needs to live in new r8+ registers and every access now requires REX
prefix. Or it is shifted into r12, so [r12+0] addressing mode has to be
used which is longer than [r8]
However, overall balance is in negative direction:
add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730)
function old new delta
nfsd4_lock 3886 3959 +73
tipc_link_build_proto_msg 1096 1140 +44
mac80211_hwsim_new_radio 2776 2808 +32
tipc_mon_rcv 1032 1058 +26
svcauth_gss_legacy_init 1413 1429 +16
tipc_bcbase_select_primary 379 392 +13
nfsd4_exchange_id 1247 1260 +13
nfsd4_setclientid_confirm 782 793 +11
...
put_client_renew_locked 494 480 -14
ip_set_sockfn_get 730 716 -14
geneve_sock_add 829 813 -16
nfsd4_sequence_done 721 703 -18
nlmclnt_lookup_host 708 686 -22
nfsd4_lockt 1085 1063 -22
nfs_get_client 1077 1050 -27
tcf_bpf_init 1106 1076 -30
nfsd4_encode_fattr 5997 5930 -67
Total: Before=154856051, After=154854321, chg -0.00%
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2016-11-17 09:58:21 +08:00
|
|
|
extern unsigned int ovs_net_id;
|
2013-04-16 04:23:03 +08:00
|
|
|
void ovs_lock(void);
|
|
|
|
void ovs_unlock(void);
|
|
|
|
|
|
|
|
#ifdef CONFIG_LOCKDEP
|
|
|
|
int lockdep_ovsl_is_held(void);
|
|
|
|
#else
|
|
|
|
#define lockdep_ovsl_is_held() 1
|
|
|
|
#endif
|
|
|
|
|
2014-07-30 08:31:08 +08:00
|
|
|
#define ASSERT_OVSL() WARN_ON(!lockdep_ovsl_is_held())
|
2013-04-16 04:23:03 +08:00
|
|
|
#define ovsl_dereference(p) \
|
|
|
|
rcu_dereference_protected(p, lockdep_ovsl_is_held())
|
2013-12-04 02:58:53 +08:00
|
|
|
#define rcu_dereference_ovsl(p) \
|
|
|
|
rcu_dereference_check(p, lockdep_ovsl_is_held())
|
2013-04-16 04:23:03 +08:00
|
|
|
|
2014-11-06 22:58:52 +08:00
|
|
|
static inline struct net *ovs_dp_get_net(const struct datapath *dp)
|
2012-02-23 11:58:59 +08:00
|
|
|
{
|
|
|
|
return read_pnet(&dp->net);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline void ovs_dp_set_net(struct datapath *dp, struct net *net)
|
|
|
|
{
|
|
|
|
write_pnet(&dp->net, net);
|
|
|
|
}
|
|
|
|
|
2013-04-16 04:23:03 +08:00
|
|
|
struct vport *ovs_lookup_vport(const struct datapath *dp, u16 port_no);
|
|
|
|
|
|
|
|
static inline struct vport *ovs_vport_rcu(const struct datapath *dp, int port_no)
|
|
|
|
{
|
|
|
|
WARN_ON_ONCE(!rcu_read_lock_held());
|
|
|
|
return ovs_lookup_vport(dp, port_no);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline struct vport *ovs_vport_ovsl_rcu(const struct datapath *dp, int port_no)
|
|
|
|
{
|
|
|
|
WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_ovsl_is_held());
|
|
|
|
return ovs_lookup_vport(dp, port_no);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline struct vport *ovs_vport_ovsl(const struct datapath *dp, int port_no)
|
|
|
|
{
|
|
|
|
ASSERT_OVSL();
|
|
|
|
return ovs_lookup_vport(dp, port_no);
|
|
|
|
}
|
|
|
|
|
2011-10-26 10:26:31 +08:00
|
|
|
extern struct notifier_block ovs_dp_device_notifier;
|
2013-11-19 22:19:38 +08:00
|
|
|
extern struct genl_family dp_vport_genl_family;
|
2011-10-26 10:26:31 +08:00
|
|
|
|
2014-09-16 10:28:44 +08:00
|
|
|
void ovs_dp_process_packet(struct sk_buff *skb, struct sw_flow_key *key);
|
2011-10-26 10:26:31 +08:00
|
|
|
void ovs_dp_detach_port(struct vport *);
|
|
|
|
int ovs_dp_upcall(struct datapath *, struct sk_buff *,
|
2016-06-11 02:49:33 +08:00
|
|
|
const struct sw_flow_key *, const struct dp_upcall_info *,
|
|
|
|
uint32_t cutlen);
|
2011-10-26 10:26:31 +08:00
|
|
|
|
2014-09-16 10:37:25 +08:00
|
|
|
const char *ovs_dp_name(const struct datapath *dp);
|
2011-10-26 10:26:31 +08:00
|
|
|
struct sk_buff *ovs_vport_cmd_build_info(struct vport *, u32 pid, u32 seq,
|
|
|
|
u8 cmd);
|
|
|
|
|
2014-09-16 10:15:28 +08:00
|
|
|
int ovs_execute_actions(struct datapath *dp, struct sk_buff *skb,
|
2014-11-06 22:58:52 +08:00
|
|
|
const struct sw_flow_actions *, struct sw_flow_key *);
|
2014-09-16 10:37:25 +08:00
|
|
|
|
2013-04-16 04:23:03 +08:00
|
|
|
void ovs_dp_notify_wq(struct work_struct *work);
|
2013-08-08 11:01:00 +08:00
|
|
|
|
2014-09-16 10:37:25 +08:00
|
|
|
int action_fifos_init(void);
|
|
|
|
void action_fifos_exit(void);
|
|
|
|
|
2015-08-27 02:31:45 +08:00
|
|
|
/* 'KEY' must not have any bits set outside of the 'MASK' */
|
|
|
|
#define OVS_MASKED(OLD, KEY, MASK) ((KEY) | ((OLD) & ~(MASK)))
|
|
|
|
#define OVS_SET_MASKED(OLD, KEY, MASK) ((OLD) = OVS_MASKED(OLD, KEY, MASK))
|
|
|
|
|
2014-11-06 23:03:05 +08:00
|
|
|
#define OVS_NLERR(logging_allowed, fmt, ...) \
|
2014-02-04 09:06:46 +08:00
|
|
|
do { \
|
2014-11-06 23:03:05 +08:00
|
|
|
if (logging_allowed && net_ratelimit()) \
|
|
|
|
pr_info("netlink: " fmt "\n", ##__VA_ARGS__); \
|
2014-02-04 09:06:46 +08:00
|
|
|
} while (0)
|
2011-10-26 10:26:31 +08:00
|
|
|
#endif /* datapath.h */
|