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

Jesse Gross says:

====================
Open vSwitch

A set of updates for net-next/3.13. Major changes are:
 * Restructure flow handling code to be more logically organized and
   easier to read.
 * Rehashing of the flow table is moved from a workqueue to flow
   installation time. Before, heavy load could block the workqueue for
   excessive periods of time.
 * Additional debugging information is provided to help diagnose megaflows.
 * It's now possible to match on TCP flags.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

13 files changed, 2511 insertions, 2291 deletions

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index a74d375b439b..d120f9fe0017 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -63,15 +63,18 @@ enum ovs_datapath_cmd {
  * not be sent.
  * @OVS_DP_ATTR_STATS: Statistics about packets that have passed through the
  * datapath.  Always present in notifications.
+ * @OVS_DP_ATTR_MEGAFLOW_STATS: Statistics about mega flow masks usage for the
+ * datapath. Always present in notifications.
  *
  * These attributes follow the &struct ovs_header within the Generic Netlink
  * payload for %OVS_DP_* commands.
  */
 enum ovs_datapath_attr {
         OVS_DP_ATTR_UNSPEC,
-        OVS_DP_ATTR_NAME,       /* name of dp_ifindex netdev */
-        OVS_DP_ATTR_UPCALL_PID, /* Netlink PID to receive upcalls */
-        OVS_DP_ATTR_STATS,      /* struct ovs_dp_stats */
+        OVS_DP_ATTR_NAME,               /* name of dp_ifindex netdev */
+        OVS_DP_ATTR_UPCALL_PID,         /* Netlink PID to receive upcalls */
+        OVS_DP_ATTR_STATS,              /* struct ovs_dp_stats */
+        OVS_DP_ATTR_MEGAFLOW_STATS,     /* struct ovs_dp_megaflow_stats */
         __OVS_DP_ATTR_MAX
 };
 
@@ -84,6 +87,14 @@ struct ovs_dp_stats {
         __u64 n_flows;           /* Number of flows present */
 };
 
+struct ovs_dp_megaflow_stats {
+        __u64 n_mask_hit;        /* Number of masks used for flow lookups. */
+        __u32 n_masks;           /* Number of masks for the datapath. */
+        __u32 pad0;              /* Pad for future expension. */
+        __u64 pad1;              /* Pad for future expension. */
+        __u64 pad2;              /* Pad for future expension. */
+};
+
 struct ovs_vport_stats {
         __u64   rx_packets;             /* total packets received       */
         __u64   tx_packets;             /* total packets transmitted    */
@@ -260,6 +271,7 @@ enum ovs_key_attr {
         OVS_KEY_ATTR_SKB_MARK,  /* u32 skb mark */
         OVS_KEY_ATTR_TUNNEL,    /* Nested set of ovs_tunnel attributes */
         OVS_KEY_ATTR_SCTP,      /* struct ovs_key_sctp */
+        OVS_KEY_ATTR_TCP_FLAGS, /* be16 TCP flags. */
 
 #ifdef __KERNEL__
         OVS_KEY_ATTR_IPV4_TUNNEL,  /* struct ovs_key_ipv4_tunnel */
diff --git a/net/openvswitch/Makefile b/net/openvswitch/Makefile
index ea36e99089af..3591cb5dae91 100644
--- a/net/openvswitch/Makefile
+++ b/net/openvswitch/Makefile
@@ -9,6 +9,8 @@ openvswitch-y := \
         datapath.o \
         dp_notify.o \
         flow.o \
+        flow_netlink.o \
+        flow_table.o \
         vport.o \
         vport-internal_dev.o \
         vport-netdev.o
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 2aa13bd7f2b2..1408adc2a2a7 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -55,14 +55,10 @@
 
 #include "datapath.h"
 #include "flow.h"
+#include "flow_netlink.h"
 #include "vport-internal_dev.h"
 #include "vport-netdev.h"
 
-
-#define REHASH_FLOW_INTERVAL (10 * 60 * HZ)
-static void rehash_flow_table(struct work_struct *work);
-static DECLARE_DELAYED_WORK(rehash_flow_wq, rehash_flow_table);
-
 int ovs_net_id __read_mostly;
 
 static void ovs_notify(struct sk_buff *skb, struct genl_info *info,
@@ -165,7 +161,7 @@ static void destroy_dp_rcu(struct rcu_head *rcu)
 {
         struct datapath *dp = container_of(rcu, struct datapath, rcu);
 
-        ovs_flow_tbl_destroy((__force struct flow_table *)dp->table, false);
+        ovs_flow_tbl_destroy(&dp->table);
         free_percpu(dp->stats_percpu);
         release_net(ovs_dp_get_net(dp));
         kfree(dp->ports);
@@ -225,6 +221,7 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
         struct dp_stats_percpu *stats;
         struct sw_flow_key key;
         u64 *stats_counter;
+        u32 n_mask_hit;
         int error;
 
         stats = this_cpu_ptr(dp->stats_percpu);
@@ -237,7 +234,7 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
         }
 
         /* Look up flow. */
-        flow = ovs_flow_lookup(rcu_dereference(dp->table), &key);
+        flow = ovs_flow_tbl_lookup(&dp->table, &key, &n_mask_hit);
         if (unlikely(!flow)) {
                 struct dp_upcall_info upcall;
 
@@ -262,6 +259,7 @@ out:
         /* Update datapath statistics. */
         u64_stats_update_begin(&stats->sync);
         (*stats_counter)++;
+        stats->n_mask_hit += n_mask_hit;
         u64_stats_update_end(&stats->sync);
 }
 
@@ -435,7 +433,7 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex,
         upcall->dp_ifindex = dp_ifindex;
 
         nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY);
-        ovs_flow_to_nlattrs(upcall_info->key, upcall_info->key, user_skb);
+        ovs_nla_put_flow(upcall_info->key, upcall_info->key, user_skb);
         nla_nest_end(user_skb, nla);
 
         if (upcall_info->userdata)
@@ -455,398 +453,6 @@ out:
         return err;
 }
 
-/* Called with ovs_mutex. */
-static int flush_flows(struct datapath *dp)
-{
-        struct flow_table *old_table;
-        struct flow_table *new_table;
-
-        old_table = ovsl_dereference(dp->table);
-        new_table = ovs_flow_tbl_alloc(TBL_MIN_BUCKETS);
-        if (!new_table)
-                return -ENOMEM;
-
-        rcu_assign_pointer(dp->table, new_table);
-
-        ovs_flow_tbl_destroy(old_table, true);
-        return 0;
-}
-