X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=datapath%2Fdatapath.c;h=e4aa672f3c20dbb5c5778e3cbef160343fb06df2;hb=cc4c738e120a82dda70fe10c1619531a9f0b1249;hp=940120cfd3fd7f6fdde21cb0e32ed5bd9928125e;hpb=a097c0b230333cbc038be4ec67bf34b5b26c5112;p=sliver-openvswitch.git

diff --git a/datapath/datapath.c b/datapath/datapath.c
index 940120cfd..e4aa672f3 100644
--- a/datapath/datapath.c
+++ b/datapath/datapath.c
@@ -61,8 +61,6 @@
 #include "vport-internal_dev.h"
 #include "vport-netdev.h"
 
-#define REHASH_FLOW_INTERVAL (10 * 60 * HZ)
-
 int ovs_net_id __read_mostly;
 
 static void ovs_notify(struct sk_buff *skb, struct genl_info *info,
@@ -165,7 +163,7 @@ 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 +223,7 @@ 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 +236,7 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
 	}
 
 	/* Look up flow. */
-	flow = ovs_flow_tbl_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;
 
@@ -254,14 +253,15 @@ void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
 	OVS_CB(skb)->flow = flow;
 	OVS_CB(skb)->pkt_key = &key;
 
-	stats_counter = &stats->n_hit;
-	ovs_flow_used(OVS_CB(skb)->flow, skb);
+	ovs_flow_stats_update(OVS_CB(skb)->flow, skb);
 	ovs_execute_actions(dp, skb);
+	stats_counter = &stats->n_hit;
 
 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);
 }
 
@@ -456,31 +456,6 @@ 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;
-}
-
-static void clear_stats(struct sw_flow *flow)
-{
-	flow->used = 0;
-	flow->tcp_flags = 0;
-	flow->packet_count = 0;
-	flow->byte_count = 0;
-}
-
 static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info)
 {
 	struct ovs_header *ovs_header = info->userhdr;
@@ -585,15 +560,18 @@ static struct genl_ops dp_packet_genl_ops[] = {
 	}
 };
 
-static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats)
+static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats,
+			 struct ovs_dp_megaflow_stats *mega_stats)
 {
-	struct flow_table *table;
 	int i;
 
-	table = rcu_dereference_check(dp->table, lockdep_ovsl_is_held());
-	stats->n_flows = ovs_flow_tbl_count(table);
+	memset(mega_stats, 0, sizeof(*mega_stats));
+
+	stats->n_flows = ovs_flow_tbl_count(&dp->table);
+	mega_stats->n_masks = ovs_flow_tbl_num_masks(&dp->table);
 
 	stats->n_hit = stats->n_missed = stats->n_lost = 0;
+
 	for_each_possible_cpu(i) {
 		const struct dp_stats_percpu *percpu_stats;
 		struct dp_stats_percpu local_stats;
@@ -609,6 +587,7 @@ static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats)
 		stats->n_hit += local_stats.n_hit;
 		stats->n_missed += local_stats.n_missed;
 		stats->n_lost += local_stats.n_lost;
+		mega_stats->n_mask_hit += local_stats.n_mask_hit;
 	}
 }
 
@@ -650,11 +629,9 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp,
 {
 	const int skb_orig_len = skb->len;
 	struct nlattr *start;
-	struct ovs_flow_stats stats;
+	struct sw_flow_stats flow_stats;
 	struct ovs_header *ovs_header;
 	struct nlattr *nla;
-	unsigned long used;
-	u8 tcp_flags;
 	int err;
 
 	ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, flags, cmd);
@@ -683,24 +660,25 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp,
 
 	nla_nest_end(skb, nla);
 
-	spin_lock_bh(&flow->lock);
-	used = flow->used;
-	stats.n_packets = flow->packet_count;
-	stats.n_bytes = flow->byte_count;
-	tcp_flags = flow->tcp_flags;
-	spin_unlock_bh(&flow->lock);
-
-	if (used &&
-	    nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used)))
+	ovs_flow_stats_get(flow, &flow_stats);
+	if (flow_stats.used &&
+	    nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(flow_stats.used)))
 		goto nla_put_failure;
 
-	if (stats.n_packets &&
-	    nla_put(skb, OVS_FLOW_ATTR_STATS,
-		    sizeof(struct ovs_flow_stats), &stats))
-		goto nla_put_failure;
+	if (flow_stats.packet_count) {
+		struct ovs_flow_stats stats = {
+			.n_packets = flow_stats.packet_count,
+			.n_bytes = flow_stats.byte_count,
+		};
+
+		if (nla_put(skb, OVS_FLOW_ATTR_STATS,
+			    sizeof(struct ovs_flow_stats), &stats))
+			goto nla_put_failure;
+	}
 
-	if (tcp_flags &&
-	    nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, tcp_flags))
+	if ((u8)ntohs(flow_stats.tcp_flags) &&
+	    nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS,
+		       (u8)ntohs(flow_stats.tcp_flags)))
 		goto nla_put_failure;
 
 	/* If OVS_FLOW_ATTR_ACTIONS doesn't fit, skip dumping the actions if
@@ -767,6 +745,14 @@ static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow,
 	return skb;
 }
 
+static struct sw_flow *__ovs_flow_tbl_lookup(struct flow_table *tbl,
+					      const struct sw_flow_key *key)
+{
+	u32 __always_unused n_mask_hit;
+
+	return ovs_flow_tbl_lookup(tbl, key, &n_mask_hit);
+}
+
 static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
 {
 	struct nlattr **a = info->attrs;
@@ -776,7 +762,6 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
 	struct sw_flow_mask mask;
 	struct sk_buff *reply;
 	struct datapath *dp;
-	struct flow_table *table;
 	struct sw_flow_actions *acts = NULL;
 	struct sw_flow_match match;
 	int error;
@@ -817,61 +802,31 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
 	if (!dp)
 		goto err_unlock_ovs;
 
-	table = ovsl_dereference(dp->table);
-
 	/* Check if this is a duplicate flow */
-	flow = ovs_flow_tbl_lookup(table, &key);
+	flow = __ovs_flow_tbl_lookup(&dp->table, &key);
 	if (!flow) {
-		struct flow_table *new_table = NULL;
-		struct sw_flow_mask *mask_p;
-
 		/* Bail out if we're not allowed to create a new flow. */
 		error = -ENOENT;
 		if (info->genlhdr->cmd == OVS_FLOW_CMD_SET)
 			goto err_unlock_ovs;
 
-		/* Expand table, if necessary, to make room. */
-		if (ovs_flow_tbl_need_to_expand(table))
-			new_table = ovs_flow_tbl_expand(table);
-		else if (time_after(jiffies, dp->last_rehash + REHASH_FLOW_INTERVAL))
-			new_table = ovs_flow_tbl_rehash(table);
-
-		if (new_table && !IS_ERR(new_table)) {
-			rcu_assign_pointer(dp->table, new_table);
-			ovs_flow_tbl_destroy(table, true);
-			table = ovsl_dereference(dp->table);
-			dp->last_rehash = jiffies;
-		}
-
 		/* Allocate flow. */
 		flow = ovs_flow_alloc();
 		if (IS_ERR(flow)) {
 			error = PTR_ERR(flow);
 			goto err_unlock_ovs;
 		}
-		clear_stats(flow);
 
 		flow->key = masked_key;
 		flow->unmasked_key = key;
-
-		/* Make sure mask is unique in the system */
-		mask_p = ovs_sw_flow_mask_find(table, &mask);
-		if (!mask_p) {
-			/* Allocate a new mask if none exsits. */
-			mask_p = ovs_sw_flow_mask_alloc();
-			if (!mask_p)
-				goto err_flow_free;
-			mask_p->key = mask.key;
-			mask_p->range = mask.range;
-			ovs_sw_flow_mask_insert(table, mask_p);
-		}
-
-		ovs_sw_flow_mask_add_ref(mask_p);
-		flow->mask = mask_p;
 		rcu_assign_pointer(flow->sf_acts, acts);
 
 		/* Put flow in bucket. */
-		ovs_flow_tbl_insert(table, flow);
+		error = ovs_flow_tbl_insert(&dp->table, flow, &mask);
+		if (error) {
+			acts = NULL;
+			goto err_flow_free;
+		}
 
 		reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid,
 						info->snd_seq, OVS_FLOW_CMD_NEW);
@@ -906,11 +861,8 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info)
 					       info->snd_seq, OVS_FLOW_CMD_NEW);
 
 		/* Clear stats. */
-		if (a[OVS_FLOW_ATTR_CLEAR]) {
-			spin_lock_bh(&flow->lock);
-			clear_stats(flow);
-			spin_unlock_bh(&flow->lock);
-		}
+		if (a[OVS_FLOW_ATTR_CLEAR])
+			ovs_flow_stats_clear(flow);
 	}
 	ovs_unlock();
 
@@ -939,7 +891,6 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info)
 	struct sk_buff *reply;
 	struct sw_flow *flow;
 	struct datapath *dp;
-	struct flow_table *table;
 	struct sw_flow_match match;
 	int err;
 
@@ -960,8 +911,7 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info)
 		goto unlock;
 	}
 
-	table = ovsl_dereference(dp->table);
-	flow = ovs_flow_tbl_lookup(table, &key);
+	flow = __ovs_flow_tbl_lookup(&dp->table, &key);
 	if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) {
 		err = -ENOENT;
 		goto unlock;
@@ -989,7 +939,6 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	struct sk_buff *reply;
 	struct sw_flow *flow;
 	struct datapath *dp;
-	struct flow_table *table;
 	struct sw_flow_match match;
 	int err;
 
@@ -1001,7 +950,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	}
 
 	if (!a[OVS_FLOW_ATTR_KEY]) {
-		err = flush_flows(dp);
+		err = ovs_flow_tbl_flush(&dp->table);
 		goto unlock;
 	}
 
@@ -1010,8 +959,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 	if (err)
 		goto unlock;
 
-	table = ovsl_dereference(dp->table);
-	flow = ovs_flow_tbl_lookup(table, &key);
+	flow = __ovs_flow_tbl_lookup(&dp->table, &key);
 	if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) {
 		err = -ENOENT;
 		goto unlock;
@@ -1023,7 +971,7 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
 		goto unlock;
 	}
 
-	ovs_flow_tbl_remove(table, flow);
+	ovs_flow_tbl_remove(&dp->table, flow);
 
 	err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid,
 				     info->snd_seq, 0, OVS_FLOW_CMD_DEL);
@@ -1042,8 +990,8 @@ unlock:
 static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
 {
 	struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh));
+	struct table_instance *ti;
 	struct datapath *dp;
-	struct flow_table *table;
 
 	rcu_read_lock();
 	dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
@@ -1052,14 +1000,14 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
 		return -ENODEV;
 	}
 
-	table = rcu_dereference(dp->table);
+	ti = rcu_dereference(dp->table.ti);
 	for (;;) {
 		struct sw_flow *flow;
 		u32 bucket, obj;
 
 		bucket = cb->args[0];
 		obj = cb->args[1];
-		flow = ovs_flow_tbl_dump_next(table, &bucket, &obj);
+		flow = ovs_flow_tbl_dump_next(ti, &bucket, &obj);
 		if (!flow)
 			break;
 
@@ -1125,6 +1073,7 @@ static size_t ovs_dp_cmd_msg_size(void)
 
 	msgsize += nla_total_size(IFNAMSIZ);
 	msgsize += nla_total_size(sizeof(struct ovs_dp_stats));
+	msgsize += nla_total_size(sizeof(struct ovs_dp_megaflow_stats));
 
 	return msgsize;
 }
@@ -1134,6 +1083,7 @@ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb,
 {
 	struct ovs_header *ovs_header;
 	struct ovs_dp_stats dp_stats;
+	struct ovs_dp_megaflow_stats dp_megaflow_stats;
 	int err;
 
 	ovs_header = genlmsg_put(skb, portid, seq, &dp_datapath_genl_family,
@@ -1149,8 +1099,14 @@ static int ovs_dp_cmd_fill_info(struct datapath *dp, struct sk_buff *skb,
 	if (err)
 		goto nla_put_failure;
 
-	get_dp_stats(dp, &dp_stats);
-	if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats), &dp_stats))
+	get_dp_stats(dp, &dp_stats, &dp_megaflow_stats);
+	if (nla_put(skb, OVS_DP_ATTR_STATS, sizeof(struct ovs_dp_stats),
+			&dp_stats))
+		goto nla_put_failure;
+
+	if (nla_put(skb, OVS_DP_ATTR_MEGAFLOW_STATS,
+			sizeof(struct ovs_dp_megaflow_stats),
+			&dp_megaflow_stats))
 		goto nla_put_failure;
 
 	return genlmsg_end(skb, ovs_header);
@@ -1223,9 +1179,8 @@ static int ovs_dp_cmd_new(struct sk_buff *skb, struct genl_info *info)
 	ovs_dp_set_net(dp, hold_net(sock_net(skb->sk)));
 
 	/* Allocate table. */
-	err = -ENOMEM;
-	rcu_assign_pointer(dp->table, ovs_flow_tbl_alloc(TBL_MIN_BUCKETS));
-	if (!dp->table)
+	err = ovs_flow_tbl_init(&dp->table);
+	if (err)
 		goto err_free_dp;
 
 	dp->stats_percpu = alloc_percpu(struct dp_stats_percpu);
@@ -1282,7 +1237,7 @@ err_destroy_ports_array:
 err_destroy_percpu:
 	free_percpu(dp->stats_percpu);
 err_destroy_table:
-	ovs_flow_tbl_destroy(ovsl_dereference(dp->table), false);
+	ovs_flow_tbl_destroy(&dp->table);
 err_free_dp:
 	release_net(ovs_dp_get_net(dp));
 	kfree(dp);