flow_table.c

/*
 * Copyright (c) 2007-2013 Nicira, Inc.
 *
 * 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
 */

#include "flow.h"
#include "datapath.h"
#include <linux/uaccess.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
#include <net/llc_pdu.h>
#include <linux/kernel.h>
#include <linux/jhash.h>
#include <linux/jiffies.h>
#include <linux/llc.h>
#include <linux/module.h>
#include <linux/in.h>
#include <linux/rcupdate.h>
#include <linux/if_arp.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/sctp.h>
#include <linux/tcp.h>
#include <linux/udp.h>
#include <linux/icmp.h>
#include <linux/icmpv6.h>
#include <linux/rculist.h>
#include <net/ip.h>
#include <net/ipv6.h>
#include <net/ndisc.h>

static struct kmem_cache *flow_cache;

static u16 range_n_bytes(const struct sw_flow_key_range *range)
{
	return range->end - range->start;
}

void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src,
		       const struct sw_flow_mask *mask)
{
	const long *m = (long *)((u8 *)&mask->key + mask->range.start);
	const long *s = (long *)((u8 *)src + mask->range.start);
	long *d = (long *)((u8 *)dst + mask->range.start);
	int i;

	/* The memory outside of the 'mask->range' are not set since
	 * further operations on 'dst' only uses contents within
	 * 'mask->range'.
	 */
	for (i = 0; i < range_n_bytes(&mask->range); i += sizeof(long))
		*d++ = *s++ & *m++;
}

struct sw_flow *ovs_flow_alloc(void)
{
	struct sw_flow *flow;

	flow = kmem_cache_alloc(flow_cache, GFP_KERNEL);
	if (!flow)
		return ERR_PTR(-ENOMEM);

	spin_lock_init(&flow->lock);
	flow->sf_acts = NULL;
	flow->mask = NULL;

	return flow;
}

static struct flex_array *alloc_buckets(unsigned int n_buckets)
{
	struct flex_array *buckets;
	int i, err;

	buckets = flex_array_alloc(sizeof(struct hlist_head),
				   n_buckets, GFP_KERNEL);
	if (!buckets)
		return NULL;

	err = flex_array_prealloc(buckets, 0, n_buckets, GFP_KERNEL);
	if (err) {
		flex_array_free(buckets);
		return NULL;
	}

	for (i = 0; i < n_buckets; i++)
		INIT_HLIST_HEAD((struct hlist_head *)
					flex_array_get(buckets, i));

	return buckets;
}

static void flow_free(struct sw_flow *flow)
{
	kfree((struct sf_flow_acts __force *)flow->sf_acts);
	kmem_cache_free(flow_cache, flow);
}

static void rcu_free_flow_callback(struct rcu_head *rcu)
{
	struct sw_flow *flow = container_of(rcu, struct sw_flow, rcu);

	flow_free(flow);
}

void ovs_flow_free(struct sw_flow *flow, bool deferred)
{
	if (!flow)
		return;

	ovs_sw_flow_mask_del_ref(flow->mask, deferred);

	if (deferred)
		call_rcu(&flow->rcu, rcu_free_flow_callback);
	else
		flow_free(flow);
}

static void free_buckets(struct flex_array *buckets)
{
	flex_array_free(buckets);
}

static void __flow_tbl_destroy(struct flow_table *table)
{
	int i;

	if (table->keep_flows)
		goto skip_flows;

	for (i = 0; i < table->n_buckets; i++) {
		struct sw_flow *flow;
		struct hlist_head *head = flex_array_get(table->buckets, i);
		struct hlist_node *n;
		int ver = table->node_ver;

		hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) {
			hlist_del(&flow->hash_node[ver]);
			ovs_flow_free(flow, false);
		}
	}

	BUG_ON(!list_empty(table->mask_list));
	kfree(table->mask_list);

skip_flows:
	free_buckets(table->buckets);
	kfree(table);
}

static struct flow_table *__flow_tbl_alloc(int new_size)
{
	struct flow_table *table = kmalloc(sizeof(*table), GFP_KERNEL);

	if (!table)
		return NULL;

	table->buckets = alloc_buckets(new_size);

	if (!table->buckets) {
		kfree(table);
		return NULL;
	}
	table->n_buckets = new_size;
	table->count = 0;
	table->node_ver = 0;
	table->keep_flows = false;
	get_random_bytes(&table->hash_seed, sizeof(u32));
	table->mask_list = NULL;

	return table;
}

struct flow_table *ovs_flow_tbl_alloc(int new_size)
{
	struct flow_table *table = __flow_tbl_alloc(new_size);

	if (!table)
		return NULL;

	table->mask_list = kmalloc(sizeof(struct list_head), GFP_KERNEL);
	if (!table->mask_list) {
		table->keep_flows = true;
		__flow_tbl_destroy(table);
		return NULL;
	}
	INIT_LIST_HEAD(table->mask_list);

	return table;
}

static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu)
{
	struct flow_table *table = container_of(rcu, struct flow_table, rcu);

	__flow_tbl_destroy(table);
}

void ovs_flow_tbl_destroy(struct flow_table *table, bool deferred)
{
	if (!table)
		return;

	if (deferred)
		call_rcu(&table->rcu, flow_tbl_destroy_rcu_cb);
	else
		__flow_tbl_destroy(table);
}

struct sw_flow *ovs_flow_tbl_dump_next(struct flow_table *table,
				       u32 *bucket, u32 *last)
{
	struct sw_flow *flow;
	struct hlist_head *head;
	int ver;
	int i;

	ver = table->node_ver;
	while (*bucket < table->n_buckets) {
		i = 0;
		head = flex_array_get(table->buckets, *bucket);
		hlist_for_each_entry_rcu(flow, head, hash_node[ver]) {
			if (i < *last) {
				i++;
				continue;
			}
			*last = i + 1;
			return flow;
		}
		(*bucket)++;
		*last = 0;
	}

	return NULL;
}

static struct hlist_head *find_bucket(struct flow_table *table, u32 hash)
{
	hash = jhash_1word(hash, table->hash_seed);
	return flex_array_get(table->buckets,
				(hash & (table->n_buckets - 1)));
}

static void __tbl_insert(struct flow_table *table, struct sw_flow *flow)
{
	struct hlist_head *head;

	head = find_bucket(table, flow->hash);
	hlist_add_head_rcu(&flow->hash_node[table->node_ver], head);

	table->count++;
}

static void flow_table_copy_flows(struct flow_table *old,
				  struct flow_table *new)
{
	int old_ver;
	int i;

	old_ver = old->node_ver;
	new->node_ver = !old_ver;

	/* Insert in new table. */
	for (i = 0; i < old->n_buckets; i++) {
		struct sw_flow *flow;
		struct hlist_head *head;

		head = flex_array_get(old->buckets, i);

		hlist_for_each_entry(flow, head, hash_node[old_ver])
			__tbl_insert(new, flow);
	}

	new->mask_list = old->mask_list;
	old->keep_flows = true;
}

static struct flow_table *__flow_tbl_rehash(struct flow_table *table,
					    int n_buckets)
{
	struct flow_table *new_table;

	new_table = __flow_tbl_alloc(n_buckets);
	if (!new_table)
		return ERR_PTR(-ENOMEM);

	flow_table_copy_flows(table, new_table);

	return new_table;
}

struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table)
{
	return __flow_tbl_rehash(table, table->n_buckets);
}

struct flow_table *ovs_flow_tbl_expand(struct flow_table *table)
{
	return __flow_tbl_rehash(table, table->n_buckets * 2);
}

static u32 flow_hash(const struct sw_flow_key *key, int key_start,
		     int key_end)
{
	u32 *hash_key = (u32 *)((u8 *)key + key_start);
	int hash_u32s = (key_end - key_start) >> 2;

	/* Make sure number of hash bytes are multiple of u32. */
	BUILD_BUG_ON(sizeof(long) % sizeof(u32));

	return jhash2(hash_key, hash_u32s, 0);
}

static int flow_key_start(const struct sw_flow_key *key)
{
	if (key->tun_key.ipv4_dst)
		return 0;
	else
		return rounddown(offsetof(struct sw_flow_key, phy),
					  sizeof(long));
}

static bool cmp_key(const struct sw_flow_key *key1,
		    const struct sw_flow_key *key2,
		    int key_start, int key_end)
{
	const long *cp1 = (long *)((u8 *)key1 + key_start);
	const long *cp2 = (long *)((u8 *)key2 + key_start);
	long diffs = 0;
	int i;

	for (i = key_start; i < key_end;  i += sizeof(long))
		diffs |= *cp1++ ^ *cp2++;

	return diffs == 0;
}

static bool flow_cmp_masked_key(const struct sw_flow *flow,
				const struct sw_flow_key *key,
				int key_start, int key_end)
{
	return cmp_key(&flow->key, key, key_start, key_end);
}

bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
			       struct sw_flow_match *match)
{
	struct sw_flow_key *key = match->key;
	int key_start = flow_key_start(key);
	int key_end = match->range.end;

	return cmp_key(&flow->unmasked_key, key, key_start, key_end);
}

static struct sw_flow *masked_flow_lookup(struct flow_table *table,
					  const struct sw_flow_key *unmasked,
					  struct sw_flow_mask *mask)
{
	struct sw_flow *flow;
	struct hlist_head *head;
	int key_start = mask->range.start;
	int key_end = mask->range.end;
	u32 hash;
	struct sw_flow_key masked_key;

	ovs_flow_mask_key(&masked_key, unmasked, mask);
	hash = flow_hash(&masked_key, key_start, key_end);
	head = find_bucket(table, hash);
	hlist_for_each_entry_rcu(flow, head, hash_node[table->node_ver]) {
		if (flow->mask == mask &&
		    flow_cmp_masked_key(flow, &masked_key,
					  key_start, key_end))
			return flow;
	}
	return NULL;
}

struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *tbl,
				    const struct sw_flow_key *key)
{
	struct sw_flow *flow = NULL;
	struct sw_flow_mask *mask;

	list_for_each_entry_rcu(mask, tbl->mask_list, list) {
		flow = masked_flow_lookup(tbl, key, mask);
		if (flow)  /* Found */
			break;
	}

	return flow;
}

void ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow)
{
	flow->hash = flow_hash(&flow->key, flow->mask->range.start,
			flow->mask->range.end);
	__tbl_insert(table, flow);
}

void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow)
{
	BUG_ON(table->count == 0);
	hlist_del_rcu(&flow->hash_node[table->node_ver]);
	table->count--;
}

struct sw_flow_mask *ovs_sw_flow_mask_alloc(void)
{
	struct sw_flow_mask *mask;

	mask = kmalloc(sizeof(*mask), GFP_KERNEL);
	if (mask)
		mask->ref_count = 0;

	return mask;
}

void ovs_sw_flow_mask_add_ref(struct sw_flow_mask *mask)
{
	mask->ref_count++;
}

static void rcu_free_sw_flow_mask_cb(struct rcu_head *rcu)
{
	struct sw_flow_mask *mask = container_of(rcu, struct sw_flow_mask, rcu);

	kfree(mask);
}

void ovs_sw_flow_mask_del_ref(struct sw_flow_mask *mask, bool deferred)
{
	if (!mask)
		return;

	BUG_ON(!mask->ref_count);
	mask->ref_count--;

	if (!mask->ref_count) {
		list_del_rcu(&mask->list);
		if (deferred)
			call_rcu(&mask->rcu, rcu_free_sw_flow_mask_cb);
		else
			kfree(mask);
	}
}

static bool mask_equal(const struct sw_flow_mask *a,
		       const struct sw_flow_mask *b)
{
	u8 *a_ = (u8 *)&a->key + a->range.start;
	u8 *b_ = (u8 *)&b->key + b->range.start;

	return  (a->range.end == b->range.end)
		&& (a->range.start == b->range.start)
		&& (memcmp(a_, b_, range_n_bytes(&a->range)) == 0);
}

struct sw_flow_mask *ovs_sw_flow_mask_find(const struct flow_table *tbl,
					   const struct sw_flow_mask *mask)
{
	struct list_head *ml;

	list_for_each(ml, tbl->mask_list) {
		struct sw_flow_mask *m;
		m = container_of(ml, struct sw_flow_mask, list);
		if (mask_equal(mask, m))
			return m;
	}

	return NULL;
}

/**
 * add a new mask into the mask list.
 * The caller needs to make sure that 'mask' is not the same
 * as any masks that are already on the list.
 */
void ovs_sw_flow_mask_insert(struct flow_table *tbl, struct sw_flow_mask *mask)
{
	list_add_rcu(&mask->list, tbl->mask_list);
}

/* Initializes the flow module.
 * Returns zero if successful or a negative error code. */
int ovs_flow_init(void)
{
	BUILD_BUG_ON(__alignof__(struct sw_flow_key) % __alignof__(long));
	BUILD_BUG_ON(sizeof(struct sw_flow_key) % sizeof(long));

	flow_cache = kmem_cache_create("sw_flow", sizeof(struct sw_flow), 0,
					0, NULL);
	if (flow_cache == NULL)
		return -ENOMEM;

	return 0;
}

/* Uninitializes the flow module. */
void ovs_flow_exit(void)
{
	kmem_cache_destroy(flow_cache);
}