/*
- * Copyright (c) 2007-2013 Nicira, Inc.
+ * Copyright (c) 2007-2014 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
int ovs_net_id __read_mostly;
+static struct genl_family dp_packet_genl_family;
+static struct genl_family dp_flow_genl_family;
+static struct genl_family dp_datapath_genl_family;
+
+static struct genl_multicast_group ovs_dp_flow_multicast_group = {
+ .name = OVS_FLOW_MCGROUP
+};
+
+static struct genl_multicast_group ovs_dp_datapath_multicast_group = {
+ .name = OVS_DATAPATH_MCGROUP
+};
+
+struct genl_multicast_group ovs_dp_vport_multicast_group = {
+ .name = OVS_VPORT_MCGROUP
+};
+
/* Check if need to build a reply message.
* OVS userspace sets the NLM_F_ECHO flag if it needs the reply. */
static bool ovs_must_notify(struct genl_info *info,
const struct genl_multicast_group *grp)
{
return info->nlhdr->nlmsg_flags & NLM_F_ECHO ||
- netlink_has_listeners(genl_info_net(info)->genl_sock, grp->id);
+ netlink_has_listeners(genl_info_net(info)->genl_sock, GROUP_ID(grp));
}
-static void ovs_notify(struct sk_buff *skb, struct genl_info *info,
- struct genl_multicast_group *grp)
+static void ovs_notify(struct genl_family *family, struct genl_multicast_group *grp,
+ struct sk_buff *skb, struct genl_info *info)
{
- genl_notify(skb, genl_info_net(info), info->snd_portid,
- grp->id, info->nlhdr, GFP_KERNEL);
+ genl_notify(family, skb, genl_info_net(info),
+ info->snd_portid, GROUP_ID(grp), info->nlhdr, GFP_KERNEL);
}
/**
}
#endif
-static struct vport *new_vport(const struct vport_parms *);
static int queue_gso_packets(struct datapath *dp, struct sk_buff *,
const struct dp_upcall_info *);
static int queue_userspace_packet(struct datapath *dp, struct sk_buff *,
const struct dp_upcall_info *);
-/* Must be called with rcu_read_lock or ovs_mutex. */
-static struct datapath *get_dp(struct net *net, int dp_ifindex)
+/* Must be called with rcu_read_lock. */
+static struct datapath *get_dp_rcu(struct net *net, int dp_ifindex)
{
- struct datapath *dp = NULL;
- struct net_device *dev;
+ struct net_device *dev = dev_get_by_index_rcu(net, dp_ifindex);
- rcu_read_lock();
- dev = dev_get_by_index_rcu(net, dp_ifindex);
if (dev) {
struct vport *vport = ovs_internal_dev_get_vport(dev);
if (vport)
- dp = vport->dp;
+ return vport->dp;
}
+
+ return NULL;
+}
+
+/* The caller must hold either ovs_mutex or rcu_read_lock to keep the
+ * returned dp pointer valid. */
+static inline struct datapath *get_dp(struct net *net, int dp_ifindex)
+{
+ struct datapath *dp;
+
+ WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_ovsl_is_held());
+ rcu_read_lock();
+ dp = get_dp_rcu(net, dp_ifindex);
rcu_read_unlock();
return dp;
{
struct datapath *dp = container_of(rcu, struct datapath, rcu);
+ ovs_flow_tbl_destroy(&dp->table);
free_percpu(dp->stats_percpu);
release_net(ovs_dp_get_net(dp));
kfree(dp->ports);
ovs_vport_del(p);
}
-/* Must be called with rcu_read_lock. */
-void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
+void ovs_dp_process_packet_with_key(struct sk_buff *skb,
+ struct sw_flow_key *pkt_key,
+ bool recirc)
{
+ const struct vport *p = OVS_CB(skb)->input_vport;
struct datapath *dp = p->dp;
struct sw_flow *flow;
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);
-
- /* Extract flow from 'skb' into 'key'. */
- error = ovs_flow_extract(skb, p->port_no, &key);
- if (unlikely(error)) {
- kfree_skb(skb);
- return;
- }
+ OVS_CB(skb)->pkt_key = pkt_key;
/* Look up flow. */
- flow = ovs_flow_tbl_lookup_stats(&dp->table, &key, &n_mask_hit);
+ flow = ovs_flow_tbl_lookup_stats(&dp->table, pkt_key, skb_get_hash(skb),
+ &n_mask_hit);
if (unlikely(!flow)) {
struct dp_upcall_info upcall;
upcall.cmd = OVS_PACKET_CMD_MISS;
- upcall.key = &key;
upcall.userdata = NULL;
- upcall.portid = p->upcall_portid;
+ upcall.portid = ovs_vport_find_upcall_portid(p, skb);
ovs_dp_upcall(dp, skb, &upcall);
consume_skb(skb);
stats_counter = &stats->n_missed;
}
OVS_CB(skb)->flow = flow;
- OVS_CB(skb)->pkt_key = &key;
- ovs_flow_stats_update(OVS_CB(skb)->flow, skb);
- ovs_execute_actions(dp, skb);
+ ovs_flow_stats_update(OVS_CB(skb)->flow, pkt_key->tp.flags, skb);
+ ovs_execute_actions(dp, skb, recirc);
stats_counter = &stats->n_hit;
out:
u64_stats_update_end(&stats->sync);
}
-static struct genl_family dp_packet_genl_family = {
- .id = GENL_ID_GENERATE,
- .hdrsize = sizeof(struct ovs_header),
- .name = OVS_PACKET_FAMILY,
- .version = OVS_PACKET_VERSION,
- .maxattr = OVS_PACKET_ATTR_MAX,
- .netnsok = true,
- SET_PARALLEL_OPS
-};
+/* Must be called with rcu_read_lock. */
+void ovs_dp_process_received_packet(struct sk_buff *skb)
+{
+ int error;
+ struct sw_flow_key key;
+
+ /* Extract flow from 'skb' into 'key'. */
+ error = ovs_flow_extract(skb, OVS_CB(skb)->input_vport->port_no, &key);
+ if (unlikely(error)) {
+ kfree_skb(skb);
+ return;
+ }
+
+ ovs_dp_process_packet_with_key(skb, &key, false);
+}
int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb,
const struct dp_upcall_info *upcall_info)
struct dp_stats_percpu *stats;
int err;
+ BUG_ON(!OVS_CB(skb)->pkt_key);
+
if (upcall_info->portid == 0) {
err = -ENOTCONN;
goto err;
const struct dp_upcall_info *upcall_info)
{
unsigned short gso_type = skb_shinfo(skb)->gso_type;
- struct dp_upcall_info later_info;
struct sw_flow_key later_key;
struct sk_buff *segs, *nskb;
int err;
if (IS_ERR(segs))
return PTR_ERR(segs);
+ if (gso_type & SKB_GSO_UDP) {
+ /* The initial flow key extracted by ovs_flow_extract()
+ * in this case is for a first fragment, so we need to
+ * properly mark later fragments.
+ */
+ later_key = *OVS_CB(skb)->pkt_key;
+ later_key.ip.frag = OVS_FRAG_TYPE_LATER;
+ }
+
/* Queue all of the segments. */
skb = segs;
do {
+ if (gso_type & SKB_GSO_UDP && skb != segs)
+ OVS_CB(skb)->pkt_key = &later_key;
+
err = queue_userspace_packet(dp, skb, upcall_info);
if (err)
break;
- if (skb == segs && gso_type & SKB_GSO_UDP) {
- /* The initial flow key extracted by ovs_flow_extract()
- * in this case is for a first fragment, so we need to
- * properly mark later fragments.
- */
- later_key = *upcall_info->key;
- later_key.ip.frag = OVS_FRAG_TYPE_LATER;
-
- later_info = *upcall_info;
- later_info.key = &later_key;
- upcall_info = &later_info;
- }
} while ((skb = skb->next));
/* Free all of the segments. */
static size_t key_attr_size(void)
{
+ /* Whenever adding new OVS_KEY_ FIELDS, we should consider
+ * updating this function. */
+ BUILD_BUG_ON(OVS_KEY_ATTR_TUNNEL_INFO != 22);
+
return nla_total_size(4) /* OVS_KEY_ATTR_PRIORITY */
+ nla_total_size(0) /* OVS_KEY_ATTR_TUNNEL */
+ nla_total_size(8) /* OVS_TUNNEL_KEY_ATTR_ID */
+ nla_total_size(1) /* OVS_TUNNEL_KEY_ATTR_TTL */
+ nla_total_size(0) /* OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT */
+ nla_total_size(0) /* OVS_TUNNEL_KEY_ATTR_CSUM */
+ + nla_total_size(0) /* OVS_TUNNEL_KEY_ATTR_OAM */
+ + nla_total_size(256) /* OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS */
+ nla_total_size(4) /* OVS_KEY_ATTR_IN_PORT */
+ nla_total_size(4) /* OVS_KEY_ATTR_SKB_MARK */
+ + nla_total_size(4) /* OVS_KEY_ATTR_DP_HASH */
+ + nla_total_size(4) /* OVS_KEY_ATTR_RECIRC_ID */
+ nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */
+ nla_total_size(2) /* OVS_KEY_ATTR_ETHERTYPE */
+ nla_total_size(4) /* OVS_KEY_ATTR_8021Q */
struct ovs_header *upcall;
struct sk_buff *nskb = NULL;
struct sk_buff *user_skb; /* to be queued to userspace */
+ struct sw_flow_key *pkt_key = OVS_CB(skb)->pkt_key;
struct nlattr *nla;
struct genl_info info = {
#if LINUX_VERSION_CODE >= KERNEL_VERSION(3,14,0)
upcall->dp_ifindex = dp_ifindex;
nla = nla_nest_start(user_skb, OVS_PACKET_ATTR_KEY);
- ovs_nla_put_flow(upcall_info->key, upcall_info->key, user_skb);
+ err = ovs_nla_put_flow(dp, pkt_key, pkt_key, user_skb);
+ BUG_ON(err);
nla_nest_end(user_skb, nla);
if (upcall_info->userdata)
}
nla->nla_len = nla_attr_size(skb->len);
- skb_zerocopy(user_skb, skb, skb->len, hlen);
+ err = skb_zerocopy(user_skb, skb, skb->len, hlen);
+ if (err)
+ goto out;
/* Pad OVS_PACKET_ATTR_PACKET if linear copy was performed */
if (!(dp->user_features & OVS_DP_F_UNALIGNED)) {
err = genlmsg_unicast(ovs_dp_get_net(dp), user_skb, upcall_info->portid);
out:
+ if (err)
+ skb_tx_error(skb);
kfree_skb(nskb);
return err;
}
struct sw_flow *flow;
struct datapath *dp;
struct ethhdr *eth;
+ struct vport *input_vport;
int len;
int err;
goto err_flow_free;
err = ovs_nla_copy_actions(a[OVS_PACKET_ATTR_ACTIONS],
- &flow->key, 0, &acts);
+ &flow->key, &acts);
rcu_assign_pointer(flow->sf_acts, acts);
if (err)
goto err_flow_free;
packet->mark = flow->key.phy.skb_mark;
rcu_read_lock();
- dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
+ dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex);
err = -ENODEV;
if (!dp)
goto err_unlock;
+ input_vport = ovs_vport_rcu(dp, flow->key.phy.in_port);
+ if (!input_vport)
+ input_vport = ovs_vport_rcu(dp, OVSP_LOCAL);
+
+ if (!input_vport)
+ goto err_unlock;
+
+ OVS_CB(packet)->input_vport = input_vport;
+
local_bh_disable();
- err = ovs_execute_actions(dp, packet);
+ err = ovs_execute_actions(dp, packet, false);
local_bh_enable();
rcu_read_unlock();
}
};
+static struct genl_family dp_packet_genl_family = {
+ .id = GENL_ID_GENERATE,
+ .hdrsize = sizeof(struct ovs_header),
+ .name = OVS_PACKET_FAMILY,
+ .version = OVS_PACKET_VERSION,
+ .maxattr = OVS_PACKET_ATTR_MAX,
+ .netnsok = true,
+ .parallel_ops = true,
+ .ops = dp_packet_genl_ops,
+ .n_ops = ARRAY_SIZE(dp_packet_genl_ops),
+};
+
static void get_dp_stats(struct datapath *dp, struct ovs_dp_stats *stats,
struct ovs_dp_megaflow_stats *mega_stats)
{
percpu_stats = per_cpu_ptr(dp->stats_percpu, i);
do {
- start = u64_stats_fetch_begin_bh(&percpu_stats->sync);
+ start = u64_stats_fetch_begin_irq(&percpu_stats->sync);
local_stats = *percpu_stats;
- } while (u64_stats_fetch_retry_bh(&percpu_stats->sync, start));
+ } while (u64_stats_fetch_retry_irq(&percpu_stats->sync, start));
stats->n_hit += local_stats.n_hit;
stats->n_missed += local_stats.n_missed;
}
}
-static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = {
- [OVS_FLOW_ATTR_KEY] = { .type = NLA_NESTED },
- [OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED },
- [OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG },
-};
-
-static struct genl_family dp_flow_genl_family = {
- .id = GENL_ID_GENERATE,
- .hdrsize = sizeof(struct ovs_header),
- .name = OVS_FLOW_FAMILY,
- .version = OVS_FLOW_VERSION,
- .maxattr = OVS_FLOW_ATTR_MAX,
- .netnsok = true,
- SET_PARALLEL_OPS
-};
-
-static struct genl_multicast_group ovs_dp_flow_multicast_group = {
- .name = OVS_FLOW_MCGROUP
-};
-
static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts)
{
return NLMSG_ALIGN(sizeof(struct ovs_header))
}
/* Called with ovs_mutex or RCU read lock. */
-static int ovs_flow_cmd_fill_info(const struct sw_flow *flow, int dp_ifindex,
- struct sk_buff *skb, u32 portid,
- u32 seq, u32 flags, u8 cmd)
+static int ovs_flow_cmd_fill_match(struct datapath *dp,
+ const struct sw_flow *flow,
+ struct sk_buff *skb)
{
- const int skb_orig_len = skb->len;
- struct nlattr *start;
- struct ovs_flow_stats stats;
- __be16 tcp_flags;
- unsigned long used;
- struct ovs_header *ovs_header;
struct nlattr *nla;
int err;
- ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, flags, cmd);
- if (!ovs_header)
- return -EMSGSIZE;
-
- ovs_header->dp_ifindex = dp_ifindex;
-
/* Fill flow key. */
nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY);
if (!nla)
- goto nla_put_failure;
+ return -EMSGSIZE;
- err = ovs_nla_put_flow(&flow->unmasked_key, &flow->unmasked_key, skb);
+ err = ovs_nla_put_flow(dp, &flow->unmasked_key,
+ &flow->unmasked_key, skb);
if (err)
- goto error;
+ return err;
nla_nest_end(skb, nla);
+ /* Fill flow mask. */
nla = nla_nest_start(skb, OVS_FLOW_ATTR_MASK);
if (!nla)
- goto nla_put_failure;
+ return -EMSGSIZE;
- err = ovs_nla_put_flow(&flow->key, &flow->mask->key, skb);
+ err = ovs_nla_put_flow(dp, &flow->key, &flow->mask->key, skb);
if (err)
- goto error;
-
+ return err;
nla_nest_end(skb, nla);
+ return 0;
+}
+
+/* Called with ovs_mutex or RCU read lock. */
+static int ovs_flow_cmd_fill_stats(const struct sw_flow *flow,
+ struct sk_buff *skb)
+{
+ struct ovs_flow_stats stats;
+ __be16 tcp_flags;
+ unsigned long used;
+
ovs_flow_stats_get(flow, &stats, &used, &tcp_flags);
if (used &&
nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used)))
- goto nla_put_failure;
+ return -EMSGSIZE;
if (stats.n_packets &&
nla_put(skb, OVS_FLOW_ATTR_STATS, sizeof(struct ovs_flow_stats), &stats))
- goto nla_put_failure;
+ return -EMSGSIZE;
if ((u8)ntohs(tcp_flags) &&
nla_put_u8(skb, OVS_FLOW_ATTR_TCP_FLAGS, (u8)ntohs(tcp_flags)))
- goto nla_put_failure;
+ return -EMSGSIZE;
+
+ return 0;
+}
+
+/* Called with ovs_mutex or RCU read lock. */
+static int ovs_flow_cmd_fill_actions(const struct sw_flow *flow,
+ struct sk_buff *skb, int skb_orig_len)
+{
+ struct nlattr *start;
+ int err;
/* If OVS_FLOW_ATTR_ACTIONS doesn't fit, skip dumping the actions if
* this is the first flow to be dumped into 'skb'. This is unusual for
nla_nest_end(skb, start);
else {
if (skb_orig_len)
- goto error;
+ return err;
nla_nest_cancel(skb, start);
}
- } else if (skb_orig_len)
- goto nla_put_failure;
+ } else if (skb_orig_len) {
+ return -EMSGSIZE;
+ }
+
+ return 0;
+}
+
+/* Called with ovs_mutex or RCU read lock. */
+static int ovs_flow_cmd_fill_info(struct datapath *dp,
+ const struct sw_flow *flow, int dp_ifindex,
+ struct sk_buff *skb, u32 portid,
+ u32 seq, u32 flags, u8 cmd)
+{
+ const int skb_orig_len = skb->len;
+ struct ovs_header *ovs_header;
+ int err;
+
+ ovs_header = genlmsg_put(skb, portid, seq, &dp_flow_genl_family, flags, cmd);
+ if (!ovs_header)
+ return -EMSGSIZE;
+ ovs_header->dp_ifindex = dp_ifindex;
+
+ err = ovs_flow_cmd_fill_match(dp, flow, skb);
+ if (err)
+ goto error;
+
+ err = ovs_flow_cmd_fill_stats(flow, skb);
+ if (err)
+ goto error;
+
+ err = ovs_flow_cmd_fill_actions(flow, skb, skb_orig_len);
+ if (err)
+ goto error;
return genlmsg_end(skb, ovs_header);
-nla_put_failure:
- err = -EMSGSIZE;
error:
genlmsg_cancel(skb, ovs_header);
return err;
}
/* Called with ovs_mutex. */
-static struct sk_buff *ovs_flow_cmd_build_info(const struct sw_flow *flow,
+static struct sk_buff *ovs_flow_cmd_build_info(struct datapath *dp,
+ const struct sw_flow *flow,
int dp_ifindex,
struct genl_info *info, u8 cmd,
bool always)
if (!skb || IS_ERR(skb))
return skb;
- retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb,
+ retval = ovs_flow_cmd_fill_info(dp, flow, dp_ifindex, skb,
info->snd_portid, info->snd_seq, 0,
cmd);
BUG_ON(retval < 0);
{
struct nlattr **a = info->attrs;
struct ovs_header *ovs_header = info->userhdr;
- struct sw_flow_key key, masked_key;
- struct sw_flow *flow;
+ struct sw_flow *flow, *new_flow;
struct sw_flow_mask mask;
struct sk_buff *reply;
struct datapath *dp;
struct sw_flow_match match;
int error;
- /* Extract key. */
+ /* Must have key and actions. */
error = -EINVAL;
- if (!a[OVS_FLOW_ATTR_KEY])
+ if (!a[OVS_FLOW_ATTR_KEY]) {
+ OVS_NLERR("Flow key attribute not present in new flow.\n");
goto error;
+ }
+ if (!a[OVS_FLOW_ATTR_ACTIONS]) {
+ OVS_NLERR("Flow actions attribute not present in new flow.\n");
+ goto error;
+ }
- ovs_match_init(&match, &key, &mask);
+ /* Most of the time we need to allocate a new flow, do it before
+ * locking. */
+ new_flow = ovs_flow_alloc();
+ if (IS_ERR(new_flow)) {
+ error = PTR_ERR(new_flow);
+ goto error;
+ }
+
+ /* Extract key. */
+ ovs_match_init(&match, &new_flow->unmasked_key, &mask);
error = ovs_nla_get_match(&match,
a[OVS_FLOW_ATTR_KEY], a[OVS_FLOW_ATTR_MASK]);
if (error)
- goto error;
+ goto err_kfree_flow;
- /* Validate actions. */
- error = -EINVAL;
- if (!a[OVS_FLOW_ATTR_ACTIONS])
- goto error;
+ ovs_flow_mask_key(&new_flow->key, &new_flow->unmasked_key, &mask);
+ /* Validate actions. */
acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS]));
error = PTR_ERR(acts);
if (IS_ERR(acts))
- goto error;
+ goto err_kfree_flow;
- ovs_flow_mask_key(&masked_key, &key, &mask);
- error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS],
- &masked_key, 0, &acts);
+ error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &new_flow->key,
+ &acts);
if (error) {
OVS_NLERR("Flow actions may not be safe on all matching packets.\n");
- goto err_kfree;
+ goto err_kfree_acts;
+ }
+
+ reply = ovs_flow_cmd_alloc_info(acts, info, false);
+ if (IS_ERR(reply)) {
+ error = PTR_ERR(reply);
+ goto err_kfree_acts;
}
ovs_lock();
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
- error = -ENODEV;
- if (!dp)
+ if (unlikely(!dp)) {
+ error = -ENODEV;
goto err_unlock_ovs;
-
+ }
/* Check if this is a duplicate flow */
- flow = ovs_flow_tbl_lookup(&dp->table, &key);
- if (!flow) {
- /* Allocate flow. */
- flow = ovs_flow_alloc();
- if (IS_ERR(flow)) {
- error = PTR_ERR(flow);
- goto err_unlock_ovs;
- }
-
- flow->key = masked_key;
- flow->unmasked_key = key;
- rcu_assign_pointer(flow->sf_acts, acts);
+ flow = ovs_flow_tbl_lookup(&dp->table, &new_flow->unmasked_key);
+ if (likely(!flow)) {
+ rcu_assign_pointer(new_flow->sf_acts, acts);
/* Put flow in bucket. */
- error = ovs_flow_tbl_insert(&dp->table, flow, &mask);
- if (error) {
+ error = ovs_flow_tbl_insert(&dp->table, new_flow, &mask);
+ if (unlikely(error)) {
acts = NULL;
- goto err_flow_free;
+ goto err_unlock_ovs;
}
+
+ if (unlikely(reply)) {
+ error = ovs_flow_cmd_fill_info(dp, new_flow,
+ ovs_header->dp_ifindex,
+ reply, info->snd_portid,
+ info->snd_seq, 0,
+ OVS_FLOW_CMD_NEW);
+ BUG_ON(error < 0);
+ }
+ ovs_unlock();
} else {
struct sw_flow_actions *old_acts;
* request. We also accept NLM_F_EXCL in case that bug ever
* gets fixed.
*/
- error = -EEXIST;
- if (info->nlhdr->nlmsg_flags & (NLM_F_CREATE | NLM_F_EXCL))
+ if (unlikely(info->nlhdr->nlmsg_flags & (NLM_F_CREATE
+ | NLM_F_EXCL))) {
+ error = -EEXIST;
goto err_unlock_ovs;
-
+ }
/* The unmasked key has to be the same for flow updates. */
- if (!ovs_flow_cmp_unmasked_key(flow, &match))
- goto err_unlock_ovs;
-
+ if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) {
+ /* Look for any overlapping flow. */
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (!flow) {
+ error = -ENOENT;
+ goto err_unlock_ovs;
+ }
+ }
/* Update actions. */
old_acts = ovsl_dereference(flow->sf_acts);
rcu_assign_pointer(flow->sf_acts, acts);
- ovs_nla_free_flow_actions(old_acts);
- }
- reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
- info, OVS_FLOW_CMD_NEW, false);
- ovs_unlock();
+ if (unlikely(reply)) {
+ error = ovs_flow_cmd_fill_info(dp, flow,
+ ovs_header->dp_ifindex,
+ reply, info->snd_portid,
+ info->snd_seq, 0,
+ OVS_FLOW_CMD_NEW);
+ BUG_ON(error < 0);
+ }
+ ovs_unlock();
- if (reply) {
- if (!IS_ERR(reply))
- ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
- else
- netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
- ovs_dp_flow_multicast_group.id,
- PTR_ERR(reply));
+ ovs_nla_free_flow_actions(old_acts);
+ ovs_flow_free(new_flow, false);
}
+
+ if (reply)
+ ovs_notify(&dp_flow_genl_family, &ovs_dp_flow_multicast_group, reply, info);
return 0;
-err_flow_free:
- ovs_flow_free(flow, false);
err_unlock_ovs:
ovs_unlock();
-err_kfree:
+ kfree_skb(reply);
+err_kfree_acts:
kfree(acts);
+err_kfree_flow:
+ ovs_flow_free(new_flow, false);
error:
return error;
}
+static struct sw_flow_actions *get_flow_actions(const struct nlattr *a,
+ const struct sw_flow_key *key,
+ const struct sw_flow_mask *mask)
+{
+ struct sw_flow_actions *acts;
+ struct sw_flow_key masked_key;
+ int error;
+
+ acts = ovs_nla_alloc_flow_actions(nla_len(a));
+ if (IS_ERR(acts))
+ return acts;
+
+ ovs_flow_mask_key(&masked_key, key, mask);
+ error = ovs_nla_copy_actions(a, &masked_key, &acts);
+ if (error) {
+ OVS_NLERR("Flow actions may not be safe on all matching packets.\n");
+ kfree(acts);
+ return ERR_PTR(error);
+ }
+
+ return acts;
+}
+
static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
{
struct nlattr **a = info->attrs;
struct ovs_header *ovs_header = info->userhdr;
- struct sw_flow_key key, masked_key;
+ struct sw_flow_key key;
struct sw_flow *flow;
struct sw_flow_mask mask;
struct sk_buff *reply = NULL;
struct datapath *dp;
- struct sw_flow_actions *acts = NULL;
+ struct sw_flow_actions *old_acts = NULL, *acts = NULL;
struct sw_flow_match match;
int error;
/* Extract key. */
error = -EINVAL;
- if (!a[OVS_FLOW_ATTR_KEY])
+ if (!a[OVS_FLOW_ATTR_KEY]) {
+ OVS_NLERR("Flow key attribute not present in set flow.\n");
goto error;
+ }
ovs_match_init(&match, &key, &mask);
error = ovs_nla_get_match(&match,
/* Validate actions. */
if (a[OVS_FLOW_ATTR_ACTIONS]) {
- acts = ovs_nla_alloc_flow_actions(nla_len(a[OVS_FLOW_ATTR_ACTIONS]));
- error = PTR_ERR(acts);
- if (IS_ERR(acts))
+ acts = get_flow_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, &mask);
+ if (IS_ERR(acts)) {
+ error = PTR_ERR(acts);
goto error;
+ }
+ }
- ovs_flow_mask_key(&masked_key, &key, &mask);
- error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS],
- &masked_key, 0, &acts);
- if (error) {
- OVS_NLERR("Flow actions may not be safe on all matching packets.\n");
- goto err_kfree;
+ /* Can allocate before locking if have acts. */
+ if (acts) {
+ reply = ovs_flow_cmd_alloc_info(acts, info, false);
+ if (IS_ERR(reply)) {
+ error = PTR_ERR(reply);
+ goto err_kfree_acts;
}
}
ovs_lock();
dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
- error = -ENODEV;
- if (!dp)
+ if (unlikely(!dp)) {
+ error = -ENODEV;
goto err_unlock_ovs;
-
+ }
/* Check that the flow exists. */
- flow = ovs_flow_tbl_lookup(&dp->table, &key);
- error = -ENOENT;
- if (!flow)
- goto err_unlock_ovs;
-
- /* The unmasked key has to be the same for flow updates. */
- error = -EEXIST;
- if (!ovs_flow_cmp_unmasked_key(flow, &match))
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (unlikely(!flow)) {
+ error = -ENOENT;
goto err_unlock_ovs;
+ }
/* Update actions, if present. */
- if (acts) {
- struct sw_flow_actions *old_acts;
-
+ if (likely(acts)) {
old_acts = ovsl_dereference(flow->sf_acts);
rcu_assign_pointer(flow->sf_acts, acts);
- ovs_nla_free_flow_actions(old_acts);
+
+ if (unlikely(reply)) {
+ error = ovs_flow_cmd_fill_info(dp, flow,
+ ovs_header->dp_ifindex,
+ reply, info->snd_portid,
+ info->snd_seq, 0,
+ OVS_FLOW_CMD_NEW);
+ BUG_ON(error < 0);
+ }
+ } else {
+ /* Could not alloc without acts before locking. */
+ reply = ovs_flow_cmd_build_info(dp, flow,
+ ovs_header->dp_ifindex,
+ info, OVS_FLOW_CMD_NEW, false);
+ if (unlikely(IS_ERR(reply))) {
+ error = PTR_ERR(reply);
+ goto err_unlock_ovs;
+ }
}
- reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
- info, OVS_FLOW_CMD_NEW, false);
/* Clear stats. */
if (a[OVS_FLOW_ATTR_CLEAR])
ovs_flow_stats_clear(flow);
ovs_unlock();
- if (reply) {
- if (!IS_ERR(reply))
- ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
- else
- netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
- ovs_dp_flow_multicast_group.id,
- PTR_ERR(reply));
- }
+ if (reply)
+ ovs_notify(&dp_flow_genl_family, &ovs_dp_flow_multicast_group, reply, info);
+ if (old_acts)
+ ovs_nla_free_flow_actions(old_acts);
return 0;
err_unlock_ovs:
ovs_unlock();
-err_kfree:
+ kfree_skb(reply);
+err_kfree_acts:
kfree(acts);
error:
return error;
goto unlock;
}
- flow = ovs_flow_tbl_lookup(&dp->table, &key);
- if (!flow || !ovs_flow_cmp_unmasked_key(flow, &match)) {
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (!flow) {
err = -ENOENT;
goto unlock;
}
- reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex, info,
+ reply = ovs_flow_cmd_build_info(dp, flow, ovs_header->dp_ifindex, info,
OVS_FLOW_CMD_NEW, true);
if (IS_ERR(reply)) {
err = PTR_ERR(reply);
err = ovs_flow_tbl_flush(&dp->table);
goto unlock;
}
- flow = ovs_flow_tbl_lookup(&dp->table, &key);
- if (unlikely(!flow || !ovs_flow_cmp_unmasked_key(flow, &match))) {
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (unlikely(!flow)) {
err = -ENOENT;
goto unlock;
}
if (likely(reply)) {
if (likely(!IS_ERR(reply))) {
rcu_read_lock(); /* Keep RCU checker happy. */
- err = ovs_flow_cmd_fill_info(flow,
+ err = ovs_flow_cmd_fill_info(dp, flow,
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
OVS_FLOW_CMD_DEL);
rcu_read_unlock();
BUG_ON(err < 0);
- ovs_notify(reply, info, &ovs_dp_flow_multicast_group);
+ ovs_notify(&dp_flow_genl_family, &ovs_dp_flow_multicast_group, reply, info);
} else {
- netlink_set_err(sock_net(skb->sk)->genl_sock, 0,
- ovs_dp_flow_multicast_group.id,
- PTR_ERR(reply));
+ genl_set_err(&dp_flow_genl_family, sock_net(skb->sk), 0,
+ GROUP_ID(&ovs_dp_flow_multicast_group), PTR_ERR(reply));
+
}
}
struct datapath *dp;
rcu_read_lock();
- dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
+ dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex);
if (!dp) {
rcu_read_unlock();
return -ENODEV;
if (!flow)
break;
- if (ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex, skb,
+ if (ovs_flow_cmd_fill_info(dp, flow, ovs_header->dp_ifindex, skb,
NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, NLM_F_MULTI,
OVS_FLOW_CMD_NEW) < 0)
return skb->len;
}
+static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = {
+ [OVS_FLOW_ATTR_KEY] = { .type = NLA_NESTED },
+ [OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED },
+ [OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG },
+};
+
static struct genl_ops dp_flow_genl_ops[] = {
{ .cmd = OVS_FLOW_CMD_NEW,
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
},
};
-static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = {
- [OVS_DP_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
- [OVS_DP_ATTR_UPCALL_PID] = { .type = NLA_U32 },
- [OVS_DP_ATTR_USER_FEATURES] = { .type = NLA_U32 },
-};
-
-static struct genl_family dp_datapath_genl_family = {
+static struct genl_family dp_flow_genl_family = {
.id = GENL_ID_GENERATE,
.hdrsize = sizeof(struct ovs_header),
- .name = OVS_DATAPATH_FAMILY,
- .version = OVS_DATAPATH_VERSION,
- .maxattr = OVS_DP_ATTR_MAX,
+ .name = OVS_FLOW_FAMILY,
+ .version = OVS_FLOW_VERSION,
+ .maxattr = OVS_FLOW_ATTR_MAX,
.netnsok = true,
- SET_PARALLEL_OPS
-};
-
-static struct genl_multicast_group ovs_dp_datapath_multicast_group = {
- .name = OVS_DATAPATH_MCGROUP
+ .parallel_ops = true,
+ .ops = dp_flow_genl_ops,
+ .n_ops = ARRAY_SIZE(dp_flow_genl_ops),
+ .mcgrps = &ovs_dp_flow_multicast_group,
+ .n_mcgrps = 1,
};
static size_t ovs_dp_cmd_msg_size(void)
goto err_destroy_table;
}
+ for_each_possible_cpu(i) {
+ struct dp_stats_percpu *dpath_stats;
+ dpath_stats = per_cpu_ptr(dp->stats_percpu, i);
+ u64_stats_init(&dpath_stats->sync);
+ }
+
dp->ports = kmalloc(DP_VPORT_HASH_BUCKETS * sizeof(struct hlist_head),
GFP_KERNEL);
if (!dp->ports) {
parms.options = NULL;
parms.dp = dp;
parms.port_no = OVSP_LOCAL;
- parms.upcall_portid = nla_get_u32(a[OVS_DP_ATTR_UPCALL_PID]);
+ parms.upcall_portids = a[OVS_DP_ATTR_UPCALL_PID];
ovs_dp_change(dp, a);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+ ovs_notify(&dp_datapath_genl_family, &ovs_dp_datapath_multicast_group, reply, info);
return 0;
err_destroy_ports_array:
err_destroy_percpu:
free_percpu(dp->stats_percpu);
err_destroy_table:
- ovs_flow_tbl_destroy(&dp->table, false);
+ ovs_flow_tbl_destroy(&dp->table);
err_free_dp:
release_net(ovs_dp_get_net(dp));
kfree(dp);
ovs_dp_detach_port(ovs_vport_ovsl(dp, OVSP_LOCAL));
/* RCU destroy the flow table */
- ovs_flow_tbl_destroy(&dp->table, true);
-
call_rcu(&dp->rcu, destroy_dp_rcu);
}
__dp_destroy(dp);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+ ovs_notify(&dp_datapath_genl_family, &ovs_dp_datapath_multicast_group, reply, info);
return 0;
err_unlock_free:
BUG_ON(err < 0);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_datapath_multicast_group);
+ ovs_notify(&dp_datapath_genl_family, &ovs_dp_datapath_multicast_group, reply, info);
return 0;
err_unlock_free:
return skb->len;
}
+static const struct nla_policy datapath_policy[OVS_DP_ATTR_MAX + 1] = {
+ [OVS_DP_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
+ [OVS_DP_ATTR_UPCALL_PID] = { .type = NLA_U32 },
+ [OVS_DP_ATTR_USER_FEATURES] = { .type = NLA_U32 },
+};
+
static struct genl_ops dp_datapath_genl_ops[] = {
{ .cmd = OVS_DP_CMD_NEW,
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
},
};
-static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = {
- [OVS_VPORT_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
- [OVS_VPORT_ATTR_STATS] = { .len = sizeof(struct ovs_vport_stats) },
- [OVS_VPORT_ATTR_PORT_NO] = { .type = NLA_U32 },
- [OVS_VPORT_ATTR_TYPE] = { .type = NLA_U32 },
- [OVS_VPORT_ATTR_UPCALL_PID] = { .type = NLA_U32 },
- [OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED },
-};
-
-static struct genl_family dp_vport_genl_family = {
+static struct genl_family dp_datapath_genl_family = {
.id = GENL_ID_GENERATE,
.hdrsize = sizeof(struct ovs_header),
- .name = OVS_VPORT_FAMILY,
- .version = OVS_VPORT_VERSION,
- .maxattr = OVS_VPORT_ATTR_MAX,
+ .name = OVS_DATAPATH_FAMILY,
+ .version = OVS_DATAPATH_VERSION,
+ .maxattr = OVS_DP_ATTR_MAX,
.netnsok = true,
- SET_PARALLEL_OPS
-};
-
-struct genl_multicast_group ovs_dp_vport_multicast_group = {
- .name = OVS_VPORT_MCGROUP
+ .parallel_ops = true,
+ .ops = dp_datapath_genl_ops,
+ .n_ops = ARRAY_SIZE(dp_datapath_genl_ops),
+ .mcgrps = &ovs_dp_datapath_multicast_group,
+ .n_mcgrps = 1,
};
/* Called with ovs_mutex or RCU read lock. */
if (nla_put_u32(skb, OVS_VPORT_ATTR_PORT_NO, vport->port_no) ||
nla_put_u32(skb, OVS_VPORT_ATTR_TYPE, vport->ops->type) ||
- nla_put_string(skb, OVS_VPORT_ATTR_NAME, vport->ops->get_name(vport)) ||
- nla_put_u32(skb, OVS_VPORT_ATTR_UPCALL_PID, vport->upcall_portid))
+ nla_put_string(skb, OVS_VPORT_ATTR_NAME, vport->ops->get_name(vport)))
goto nla_put_failure;
ovs_vport_get_stats(vport, &vport_stats);
&vport_stats))
goto nla_put_failure;
+ if (ovs_vport_get_upcall_portids(vport, skb))
+ goto nla_put_failure;
+
err = ovs_vport_get_options(vport, skb);
if (err == -EMSGSIZE)
goto error;
parms.options = a[OVS_VPORT_ATTR_OPTIONS];
parms.dp = dp;
parms.port_no = port_no;
- parms.upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]);
+ parms.upcall_portids = a[OVS_VPORT_ATTR_UPCALL_PID];
vport = new_vport(&parms);
err = PTR_ERR(vport);
BUG_ON(err < 0);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+ ovs_notify(&dp_vport_genl_family, &ovs_dp_vport_multicast_group, reply, info);
return 0;
exit_unlock_free:
if (a[OVS_VPORT_ATTR_STATS])
ovs_vport_set_stats(vport, nla_data(a[OVS_VPORT_ATTR_STATS]));
- if (a[OVS_VPORT_ATTR_UPCALL_PID])
- vport->upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]);
+
+ if (a[OVS_VPORT_ATTR_UPCALL_PID]) {
+ err = ovs_vport_set_upcall_portids(vport,
+ a[OVS_VPORT_ATTR_UPCALL_PID]);
+ if (err)
+ goto exit_unlock_free;
+ }
err = ovs_vport_cmd_fill_info(vport, reply, info->snd_portid,
info->snd_seq, 0, OVS_VPORT_CMD_NEW);
BUG_ON(err < 0);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+ ovs_notify(&dp_vport_genl_family, &ovs_dp_vport_multicast_group, reply, info);
return 0;
exit_unlock_free:
ovs_dp_detach_port(vport);
ovs_unlock();
- ovs_notify(reply, info, &ovs_dp_vport_multicast_group);
+ ovs_notify(&dp_vport_genl_family, &ovs_dp_vport_multicast_group, reply, info);
return 0;
exit_unlock_free:
int i, j = 0;
rcu_read_lock();
- dp = get_dp(sock_net(skb->sk), ovs_header->dp_ifindex);
+ dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex);
if (!dp) {
rcu_read_unlock();
return -ENODEV;
return skb->len;
}
+static const struct nla_policy vport_policy[OVS_VPORT_ATTR_MAX + 1] = {
+ [OVS_VPORT_ATTR_NAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ - 1 },
+ [OVS_VPORT_ATTR_STATS] = { .len = sizeof(struct ovs_vport_stats) },
+ [OVS_VPORT_ATTR_PORT_NO] = { .type = NLA_U32 },
+ [OVS_VPORT_ATTR_TYPE] = { .type = NLA_U32 },
+ [OVS_VPORT_ATTR_UPCALL_PID] = { .type = NLA_U32 },
+ [OVS_VPORT_ATTR_OPTIONS] = { .type = NLA_NESTED },
+};
+
static struct genl_ops dp_vport_genl_ops[] = {
{ .cmd = OVS_VPORT_CMD_NEW,
.flags = GENL_ADMIN_PERM, /* Requires CAP_NET_ADMIN privilege. */
},
};
-struct genl_family_and_ops {
- struct genl_family *family;
- struct genl_ops *ops;
- int n_ops;
- struct genl_multicast_group *group;
+struct genl_family dp_vport_genl_family = {
+ .id = GENL_ID_GENERATE,
+ .hdrsize = sizeof(struct ovs_header),
+ .name = OVS_VPORT_FAMILY,
+ .version = OVS_VPORT_VERSION,
+ .maxattr = OVS_VPORT_ATTR_MAX,
+ .netnsok = true,
+ .parallel_ops = true,
+ .ops = dp_vport_genl_ops,
+ .n_ops = ARRAY_SIZE(dp_vport_genl_ops),
+ .mcgrps = &ovs_dp_vport_multicast_group,
+ .n_mcgrps = 1,
};
-static const struct genl_family_and_ops dp_genl_families[] = {
- { &dp_datapath_genl_family,
- dp_datapath_genl_ops, ARRAY_SIZE(dp_datapath_genl_ops),
- &ovs_dp_datapath_multicast_group },
- { &dp_vport_genl_family,
- dp_vport_genl_ops, ARRAY_SIZE(dp_vport_genl_ops),
- &ovs_dp_vport_multicast_group },
- { &dp_flow_genl_family,
- dp_flow_genl_ops, ARRAY_SIZE(dp_flow_genl_ops),
- &ovs_dp_flow_multicast_group },
- { &dp_packet_genl_family,
- dp_packet_genl_ops, ARRAY_SIZE(dp_packet_genl_ops),
- NULL },
+static struct genl_family *dp_genl_families[] = {
+ &dp_datapath_genl_family,
+ &dp_vport_genl_family,
+ &dp_flow_genl_family,
+ &dp_packet_genl_family,
};
static void dp_unregister_genl(int n_families)
int i;
for (i = 0; i < n_families; i++)
- genl_unregister_family(dp_genl_families[i].family);
+ genl_unregister_family(dp_genl_families[i]);
}
static int dp_register_genl(void)
{
- int n_registered;
int err;
int i;
- n_registered = 0;
for (i = 0; i < ARRAY_SIZE(dp_genl_families); i++) {
- const struct genl_family_and_ops *f = &dp_genl_families[i];
- err = genl_register_family_with_ops(f->family, f->ops,
- f->n_ops);
+ err = genl_register_family(dp_genl_families[i]);
if (err)
goto error;
- n_registered++;
-
- if (f->group) {
- err = genl_register_mc_group(f->family, f->group);
- if (err)
- goto error;
- }
}
return 0;
error:
- dp_unregister_genl(n_registered);
+ dp_unregister_genl(i);
return err;
}
projects / cascardo / ovs.git / blobdiff