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);
}
void ovs_dp_process_packet_with_key(struct sk_buff *skb,
- struct sw_flow_key *pkt_key)
+ struct sw_flow_key *pkt_key,
+ bool recirc)
{
const struct vport *p = OVS_CB(skb)->input_vport;
struct datapath *dp = p->dp;
u32 n_mask_hit;
stats = this_cpu_ptr(dp->stats_percpu);
+ OVS_CB(skb)->pkt_key = pkt_key;
/* Look up flow. */
- flow = ovs_flow_tbl_lookup_stats(&dp->table, pkt_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 = pkt_key;
upcall.userdata = NULL;
upcall.portid = ovs_vport_find_upcall_portid(p, skb);
ovs_dp_upcall(dp, skb, &upcall);
goto out;
}
- OVS_CB(skb)->pkt_key = pkt_key;
OVS_CB(skb)->flow = flow;
ovs_flow_stats_update(OVS_CB(skb)->flow, pkt_key->tp.flags, skb);
- ovs_execute_actions(dp, skb);
+ ovs_execute_actions(dp, skb, recirc);
stats_counter = &stats->n_hit;
out:
}
/* Must be called with rcu_read_lock. */
-void ovs_dp_process_received_packet(struct vport *p, struct sk_buff *skb)
+void ovs_dp_process_received_packet(struct sk_buff *skb)
{
int error;
struct sw_flow_key key;
- OVS_CB(skb)->input_vport = p;
-
/* Extract flow from 'skb' into 'key'. */
- error = ovs_flow_extract(skb, p->port_no, &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);
+ ovs_dp_process_packet_with_key(skb, &key, false);
}
int ovs_dp_upcall(struct datapath *dp, struct sk_buff *skb,
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)
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;
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();
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;
}
/* 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);
/* 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])
+ }
+ if (!a[OVS_FLOW_ATTR_ACTIONS]) {
+ OVS_NLERR("Flow actions attribute not present in new flow.\n");
goto error;
+ }
/* Most of the time we need to allocate a new flow, do it before
* locking. */
goto err_kfree_flow;
error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &new_flow->key,
- 0, &acts);
+ &acts);
if (error) {
OVS_NLERR("Flow actions may not be safe on all matching packets.\n");
goto err_kfree_acts;
}
if (unlikely(reply)) {
- error = ovs_flow_cmd_fill_info(new_flow,
+ error = ovs_flow_cmd_fill_info(dp, new_flow,
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
}
/* The unmasked key has to be the same for flow updates. */
if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) {
- error = -EEXIST;
- goto err_unlock_ovs;
+ /* 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);
if (unlikely(reply)) {
- error = ovs_flow_cmd_fill_info(flow,
+ error = ovs_flow_cmd_fill_info(dp, flow,
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
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;
/* 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_acts;
}
}
goto err_unlock_ovs;
}
/* Check that the flow exists. */
- flow = ovs_flow_tbl_lookup(&dp->table, &key);
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
if (unlikely(!flow)) {
error = -ENOENT;
goto err_unlock_ovs;
}
- /* The unmasked key has to be the same for flow updates. */
- if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) {
- error = -EEXIST;
- goto err_unlock_ovs;
- }
+
/* Update actions, if present. */
if (likely(acts)) {
old_acts = ovsl_dereference(flow->sf_acts);
rcu_assign_pointer(flow->sf_acts, acts);
if (unlikely(reply)) {
- error = ovs_flow_cmd_fill_info(flow,
+ error = ovs_flow_cmd_fill_info(dp, flow,
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
}
} else {
/* Could not alloc without acts before locking. */
- reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
+ 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 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,
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)
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);
}
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;
projects / cascardo / ovs.git / blobdiff