diff options
Diffstat (limited to 'net/openvswitch')
-rw-r--r-- | net/openvswitch/Kconfig | 14 | ||||
-rw-r--r-- | net/openvswitch/Makefile | 3 | ||||
-rw-r--r-- | net/openvswitch/actions.c | 8 | ||||
-rw-r--r-- | net/openvswitch/datapath.c | 371 | ||||
-rw-r--r-- | net/openvswitch/datapath.h | 4 | ||||
-rw-r--r-- | net/openvswitch/dp_notify.c | 2 | ||||
-rw-r--r-- | net/openvswitch/flow.c | 211 | ||||
-rw-r--r-- | net/openvswitch/flow.h | 47 | ||||
-rw-r--r-- | net/openvswitch/vport-gre.c | 275 | ||||
-rw-r--r-- | net/openvswitch/vport-internal_dev.c | 3 | ||||
-rw-r--r-- | net/openvswitch/vport-netdev.c | 9 | ||||
-rw-r--r-- | net/openvswitch/vport-netdev.h | 1 | ||||
-rw-r--r-- | net/openvswitch/vport.c | 34 | ||||
-rw-r--r-- | net/openvswitch/vport.h | 23 |
14 files changed, 884 insertions, 121 deletions
diff --git a/net/openvswitch/Kconfig b/net/openvswitch/Kconfig index d9ea33c361b..27ee56b688a 100644 --- a/net/openvswitch/Kconfig +++ b/net/openvswitch/Kconfig @@ -26,3 +26,17 @@ config OPENVSWITCH called openvswitch. If unsure, say N. + +config OPENVSWITCH_GRE + bool "Open vSwitch GRE tunneling support" + depends on INET + depends on OPENVSWITCH + depends on NET_IPGRE_DEMUX && !(OPENVSWITCH=y && NET_IPGRE_DEMUX=m) + default y + ---help--- + If you say Y here, then the Open vSwitch will be able create GRE + vport. + + Say N to exclude this support and reduce the binary size. + + If unsure, say Y. diff --git a/net/openvswitch/Makefile b/net/openvswitch/Makefile index 15e7384745c..01bddb2991e 100644 --- a/net/openvswitch/Makefile +++ b/net/openvswitch/Makefile @@ -10,5 +10,6 @@ openvswitch-y := \ dp_notify.o \ flow.o \ vport.o \ + vport-gre.o \ vport-internal_dev.o \ - vport-netdev.o \ + vport-netdev.o diff --git a/net/openvswitch/actions.c b/net/openvswitch/actions.c index 894b6cbdd92..22c5f399f1c 100644 --- a/net/openvswitch/actions.c +++ b/net/openvswitch/actions.c @@ -130,9 +130,13 @@ static int set_eth_addr(struct sk_buff *skb, if (unlikely(err)) return err; + skb_postpull_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2); + memcpy(eth_hdr(skb)->h_source, eth_key->eth_src, ETH_ALEN); memcpy(eth_hdr(skb)->h_dest, eth_key->eth_dst, ETH_ALEN); + ovs_skb_postpush_rcsum(skb, eth_hdr(skb), ETH_ALEN * 2); + return 0; } @@ -432,6 +436,10 @@ static int execute_set_action(struct sk_buff *skb, skb->mark = nla_get_u32(nested_attr); break; + case OVS_KEY_ATTR_IPV4_TUNNEL: + OVS_CB(skb)->tun_key = nla_data(nested_attr); + break; + case OVS_KEY_ATTR_ETHERNET: err = set_eth_addr(skb, nla_data(nested_attr)); break; diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c index d12d6b8b5e8..f7e3a0d84c4 100644 --- a/net/openvswitch/datapath.c +++ b/net/openvswitch/datapath.c @@ -362,6 +362,14 @@ static int queue_gso_packets(struct net *net, int dp_ifindex, static size_t key_attr_size(void) { 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(4) /* OVS_TUNNEL_KEY_ATTR_IPV4_SRC */ + + nla_total_size(4) /* OVS_TUNNEL_KEY_ATTR_IPV4_DST */ + + nla_total_size(1) /* OVS_TUNNEL_KEY_ATTR_TOS */ + + 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(4) /* OVS_KEY_ATTR_IN_PORT */ + nla_total_size(4) /* OVS_KEY_ATTR_SKB_MARK */ + nla_total_size(12) /* OVS_KEY_ATTR_ETHERNET */ @@ -464,16 +472,89 @@ static int flush_flows(struct datapath *dp) return 0; } -static int validate_actions(const struct nlattr *attr, - const struct sw_flow_key *key, int depth); +static struct nlattr *reserve_sfa_size(struct sw_flow_actions **sfa, int attr_len) +{ + + struct sw_flow_actions *acts; + int new_acts_size; + int req_size = NLA_ALIGN(attr_len); + int next_offset = offsetof(struct sw_flow_actions, actions) + + (*sfa)->actions_len; + + if (req_size <= (ksize(*sfa) - next_offset)) + goto out; + + new_acts_size = ksize(*sfa) * 2; + + if (new_acts_size > MAX_ACTIONS_BUFSIZE) { + if ((MAX_ACTIONS_BUFSIZE - next_offset) < req_size) + return ERR_PTR(-EMSGSIZE); + new_acts_size = MAX_ACTIONS_BUFSIZE; + } + + acts = ovs_flow_actions_alloc(new_acts_size); + if (IS_ERR(acts)) + return (void *)acts; + + memcpy(acts->actions, (*sfa)->actions, (*sfa)->actions_len); + acts->actions_len = (*sfa)->actions_len; + kfree(*sfa); + *sfa = acts; + +out: + (*sfa)->actions_len += req_size; + return (struct nlattr *) ((unsigned char *)(*sfa) + next_offset); +} + +static int add_action(struct sw_flow_actions **sfa, int attrtype, void *data, int len) +{ + struct nlattr *a; + + a = reserve_sfa_size(sfa, nla_attr_size(len)); + if (IS_ERR(a)) + return PTR_ERR(a); + + a->nla_type = attrtype; + a->nla_len = nla_attr_size(len); + + if (data) + memcpy(nla_data(a), data, len); + memset((unsigned char *) a + a->nla_len, 0, nla_padlen(len)); + + return 0; +} + +static inline int add_nested_action_start(struct sw_flow_actions **sfa, int attrtype) +{ + int used = (*sfa)->actions_len; + int err; + + err = add_action(sfa, attrtype, NULL, 0); + if (err) + return err; + + return used; +} -static int validate_sample(const struct nlattr *attr, - const struct sw_flow_key *key, int depth) +static inline void add_nested_action_end(struct sw_flow_actions *sfa, int st_offset) +{ + struct nlattr *a = (struct nlattr *) ((unsigned char *)sfa->actions + st_offset); + + a->nla_len = sfa->actions_len - st_offset; +} + +static int validate_and_copy_actions(const struct nlattr *attr, + const struct sw_flow_key *key, int depth, + struct sw_flow_actions **sfa); + +static int validate_and_copy_sample(const struct nlattr *attr, + const struct sw_flow_key *key, int depth, + struct sw_flow_actions **sfa) { const struct nlattr *attrs[OVS_SAMPLE_ATTR_MAX + 1]; const struct nlattr *probability, *actions; const struct nlattr *a; - int rem; + int rem, start, err, st_acts; memset(attrs, 0, sizeof(attrs)); nla_for_each_nested(a, attr, rem) { @@ -492,7 +573,26 @@ static int validate_sample(const struct nlattr *attr, actions = attrs[OVS_SAMPLE_ATTR_ACTIONS]; if (!actions || (nla_len(actions) && nla_len(actions) < NLA_HDRLEN)) return -EINVAL; - return validate_actions(actions, key, depth + 1); + + /* validation done, copy sample action. */ + start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SAMPLE); + if (start < 0) + return start; + err = add_action(sfa, OVS_SAMPLE_ATTR_PROBABILITY, nla_data(probability), sizeof(u32)); + if (err) + return err; + st_acts = add_nested_action_start(sfa, OVS_SAMPLE_ATTR_ACTIONS); + if (st_acts < 0) + return st_acts; + + err = validate_and_copy_actions(actions, key, depth + 1, sfa); + if (err) + return err; + + add_nested_action_end(*sfa, st_acts); + add_nested_action_end(*sfa, start); + + return 0; } static int validate_tp_port(const struct sw_flow_key *flow_key) @@ -508,8 +608,30 @@ static int validate_tp_port(const struct sw_flow_key *flow_key) return -EINVAL; } +static int validate_and_copy_set_tun(const struct nlattr *attr, + struct sw_flow_actions **sfa) +{ + struct ovs_key_ipv4_tunnel tun_key; + int err, start; + + err = ovs_ipv4_tun_from_nlattr(nla_data(attr), &tun_key); + if (err) + return err; + + start = add_nested_action_start(sfa, OVS_ACTION_ATTR_SET); + if (start < 0) + return start; + + err = add_action(sfa, OVS_KEY_ATTR_IPV4_TUNNEL, &tun_key, sizeof(tun_key)); + add_nested_action_end(*sfa, start); + + return err; +} + static int validate_set(const struct nlattr *a, - const struct sw_flow_key *flow_key) + const struct sw_flow_key *flow_key, + struct sw_flow_actions **sfa, + bool *set_tun) { const struct nlattr *ovs_key = nla_data(a); int key_type = nla_type(ovs_key); @@ -519,18 +641,27 @@ static int validate_set(const struct nlattr *a, return -EINVAL; if (key_type > OVS_KEY_ATTR_MAX || - nla_len(ovs_key) != ovs_key_lens[key_type]) + (ovs_key_lens[key_type] != nla_len(ovs_key) && + ovs_key_lens[key_type] != -1)) return -EINVAL; switch (key_type) { const struct ovs_key_ipv4 *ipv4_key; const struct ovs_key_ipv6 *ipv6_key; + int err; case OVS_KEY_ATTR_PRIORITY: case OVS_KEY_ATTR_SKB_MARK: case OVS_KEY_ATTR_ETHERNET: break; + case OVS_KEY_ATTR_TUNNEL: + *set_tun = true; + err = validate_and_copy_set_tun(a, sfa); + if (err) + return err; + break; + case OVS_KEY_ATTR_IPV4: if (flow_key->eth.type != htons(ETH_P_IP)) return -EINVAL; @@ -606,8 +737,24 @@ static int validate_userspace(const struct nlattr *attr) return 0; } -static int validate_actions(const struct nlattr *attr, - const struct sw_flow_key *key, int depth) +static int copy_action(const struct nlattr *from, + struct sw_flow_actions **sfa) +{ + int totlen = NLA_ALIGN(from->nla_len); + struct nlattr *to; + + to = reserve_sfa_size(sfa, from->nla_len); + if (IS_ERR(to)) + return PTR_ERR(to); + + memcpy(to, from, totlen); + return 0; +} + +static int validate_and_copy_actions(const struct nlattr *attr, + const struct sw_flow_key *key, + int depth, + struct sw_flow_actions **sfa) { const struct nlattr *a; int rem, err; @@ -627,12 +774,14 @@ static int validate_actions(const struct nlattr *attr, }; const struct ovs_action_push_vlan *vlan; int type = nla_type(a); + bool skip_copy; if (type > OVS_ACTION_ATTR_MAX || (action_lens[type] != nla_len(a) && action_lens[type] != (u32)-1)) return -EINVAL; + skip_copy = false; switch (type) { case OVS_ACTION_ATTR_UNSPEC: return -EINVAL; @@ -661,20 +810,26 @@ static int validate_actions(const struct nlattr *attr, break; case OVS_ACTION_ATTR_SET: - err = validate_set(a, key); + err = validate_set(a, key, sfa, &skip_copy); if (err) return err; break; case OVS_ACTION_ATTR_SAMPLE: - err = validate_sample(a, key, depth); + err = validate_and_copy_sample(a, key, depth, sfa); if (err) return err; + skip_copy = true; break; default: return -EINVAL; } + if (!skip_copy) { + err = copy_action(a, sfa); + if (err) + return err; + } } if (rem > 0) @@ -739,24 +894,18 @@ static int ovs_packet_cmd_execute(struct sk_buff *skb, struct genl_info *info) if (err) goto err_flow_free; - err = ovs_flow_metadata_from_nlattrs(&flow->key.phy.priority, - &flow->key.phy.skb_mark, - &flow->key.phy.in_port, - a[OVS_PACKET_ATTR_KEY]); - if (err) - goto err_flow_free; - - err = validate_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0); + err = ovs_flow_metadata_from_nlattrs(flow, key_len, a[OVS_PACKET_ATTR_KEY]); if (err) goto err_flow_free; - - flow->hash = ovs_flow_hash(&flow->key, key_len); - - acts = ovs_flow_actions_alloc(a[OVS_PACKET_ATTR_ACTIONS]); + acts = ovs_flow_actions_alloc(nla_len(a[OVS_PACKET_ATTR_ACTIONS])); err = PTR_ERR(acts); if (IS_ERR(acts)) goto err_flow_free; + + err = validate_and_copy_actions(a[OVS_PACKET_ATTR_ACTIONS], &flow->key, 0, &acts); rcu_assign_pointer(flow->sf_acts, acts); + if (err) + goto err_flow_free; OVS_CB(packet)->flow = flow; packet->priority = flow->key.phy.priority; @@ -846,6 +995,99 @@ static struct genl_multicast_group ovs_dp_flow_multicast_group = { .name = OVS_FLOW_MCGROUP }; +static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb); +static int sample_action_to_attr(const struct nlattr *attr, struct sk_buff *skb) +{ + const struct nlattr *a; + struct nlattr *start; + int err = 0, rem; + + start = nla_nest_start(skb, OVS_ACTION_ATTR_SAMPLE); + if (!start) + return -EMSGSIZE; + + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + struct nlattr *st_sample; + + switch (type) { + case OVS_SAMPLE_ATTR_PROBABILITY: + if (nla_put(skb, OVS_SAMPLE_ATTR_PROBABILITY, sizeof(u32), nla_data(a))) + return -EMSGSIZE; + break; + case OVS_SAMPLE_ATTR_ACTIONS: + st_sample = nla_nest_start(skb, OVS_SAMPLE_ATTR_ACTIONS); + if (!st_sample) + return -EMSGSIZE; + err = actions_to_attr(nla_data(a), nla_len(a), skb); + if (err) + return err; + nla_nest_end(skb, st_sample); + break; + } + } + + nla_nest_end(skb, start); + return err; +} + +static int set_action_to_attr(const struct nlattr *a, struct sk_buff *skb) +{ + const struct nlattr *ovs_key = nla_data(a); + int key_type = nla_type(ovs_key); + struct nlattr *start; + int err; + + switch (key_type) { + case OVS_KEY_ATTR_IPV4_TUNNEL: + start = nla_nest_start(skb, OVS_ACTION_ATTR_SET); + if (!start) + return -EMSGSIZE; + + err = ovs_ipv4_tun_to_nlattr(skb, nla_data(ovs_key)); + if (err) + return err; + nla_nest_end(skb, start); + break; + default: + if (nla_put(skb, OVS_ACTION_ATTR_SET, nla_len(a), ovs_key)) + return -EMSGSIZE; + break; + } + + return 0; +} + +static int actions_to_attr(const struct nlattr *attr, int len, struct sk_buff *skb) +{ + const struct nlattr *a; + int rem, err; + + nla_for_each_attr(a, attr, len, rem) { + int type = nla_type(a); + + switch (type) { + case OVS_ACTION_ATTR_SET: + err = set_action_to_attr(a, skb); + if (err) + return err; + break; + + case OVS_ACTION_ATTR_SAMPLE: + err = sample_action_to_attr(a, skb); + if (err) + return err; + break; + default: + if (nla_put(skb, type, nla_len(a), nla_data(a))) + return -EMSGSIZE; + break; + } + } + + return 0; +} + static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts) { return NLMSG_ALIGN(sizeof(struct ovs_header)) @@ -863,6 +1105,7 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, { const int skb_orig_len = skb->len; const struct sw_flow_actions *sf_acts; + struct nlattr *start; struct ovs_flow_stats stats; struct ovs_header *ovs_header; struct nlattr *nla; @@ -916,10 +1159,19 @@ static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp, * This can only fail for dump operations because the skb is always * properly sized for single flows. */ - err = nla_put(skb, OVS_FLOW_ATTR_ACTIONS, sf_acts->actions_len, - sf_acts->actions); - if (err < 0 && skb_orig_len) - goto error; + start = nla_nest_start(skb, OVS_FLOW_ATTR_ACTIONS); + if (start) { + err = actions_to_attr(sf_acts->actions, sf_acts->actions_len, skb); + if (!err) + nla_nest_end(skb, start); + else { + if (skb_orig_len) + goto error; + + nla_nest_cancel(skb, start); + } + } else if (skb_orig_len) + goto nla_put_failure; return genlmsg_end(skb, ovs_header); @@ -964,6 +1216,7 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) struct sk_buff *reply; struct datapath *dp; struct flow_table *table; + struct sw_flow_actions *acts = NULL; int error; int key_len; @@ -977,9 +1230,14 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) /* Validate actions. */ if (a[OVS_FLOW_ATTR_ACTIONS]) { - error = validate_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, 0); - if (error) + acts = ovs_flow_actions_alloc(nla_len(a[OVS_FLOW_ATTR_ACTIONS])); + error = PTR_ERR(acts); + if (IS_ERR(acts)) goto error; + + error = validate_and_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, 0, &acts); + if (error) + goto err_kfree; } else if (info->genlhdr->cmd == OVS_FLOW_CMD_NEW) { error = -EINVAL; goto error; @@ -994,8 +1252,6 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) table = ovsl_dereference(dp->table); flow = ovs_flow_tbl_lookup(table, &key, key_len); if (!flow) { - struct sw_flow_actions *acts; - /* Bail out if we're not allowed to create a new flow. */ error = -ENOENT; if (info->genlhdr->cmd == OVS_FLOW_CMD_SET) @@ -1019,19 +1275,12 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) error = PTR_ERR(flow); goto err_unlock_ovs; } - flow->key = key; clear_stats(flow); - /* Obtain actions. */ - acts = ovs_flow_actions_alloc(a[OVS_FLOW_ATTR_ACTIONS]); - error = PTR_ERR(acts); - if (IS_ERR(acts)) - goto error_free_flow; rcu_assign_pointer(flow->sf_acts, acts); /* Put flow in bucket. */ - flow->hash = ovs_flow_hash(&key, key_len); - ovs_flow_tbl_insert(table, flow); + ovs_flow_tbl_insert(table, flow, &key, key_len); reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, info->snd_seq, @@ -1039,7 +1288,6 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) } else { /* We found a matching flow. */ struct sw_flow_actions *old_acts; - struct nlattr *acts_attrs; /* Bail out if we're not allowed to modify an existing flow. * We accept NLM_F_CREATE in place of the intended NLM_F_EXCL @@ -1054,21 +1302,8 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) /* Update actions. */ old_acts = ovsl_dereference(flow->sf_acts); - acts_attrs = a[OVS_FLOW_ATTR_ACTIONS]; - if (acts_attrs && - (old_acts->actions_len != nla_len(acts_attrs) || - memcmp(old_acts->actions, nla_data(acts_attrs), - old_acts->actions_len))) { - struct sw_flow_actions *new_acts; - - new_acts = ovs_flow_actions_alloc(acts_attrs); - error = PTR_ERR(new_acts); - if (IS_ERR(new_acts)) - goto err_unlock_ovs; - - rcu_assign_pointer(flow->sf_acts, new_acts); - ovs_flow_deferred_free_acts(old_acts); - } + rcu_assign_pointer(flow->sf_acts, acts); + ovs_flow_deferred_free_acts(old_acts); reply = ovs_flow_cmd_build_info(flow, dp, info->snd_portid, info->snd_seq, OVS_FLOW_CMD_NEW); @@ -1089,10 +1324,10 @@ static int ovs_flow_cmd_new_or_set(struct sk_buff *skb, struct genl_info *info) ovs_dp_flow_multicast_group.id, PTR_ERR(reply)); return 0; -error_free_flow: - ovs_flow_free(flow); err_unlock_ovs: ovs_unlock(); +err_kfree: + kfree(acts); error: return error; } @@ -1812,10 +2047,11 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(vport)) goto exit_unlock; - err = 0; if (a[OVS_VPORT_ATTR_TYPE] && - nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) + nla_get_u32(a[OVS_VPORT_ATTR_TYPE]) != vport->ops->type) { err = -EINVAL; + goto exit_unlock; + } reply = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); if (!reply) { @@ -1823,10 +2059,11 @@ static int ovs_vport_cmd_set(struct sk_buff *skb, struct genl_info *info) goto exit_unlock; } - if (!err && a[OVS_VPORT_ATTR_OPTIONS]) + if (a[OVS_VPORT_ATTR_OPTIONS]) { err = ovs_vport_set_options(vport, a[OVS_VPORT_ATTR_OPTIONS]); - if (err) - goto exit_free; + if (err) + goto exit_free; + } if (a[OVS_VPORT_ATTR_UPCALL_PID]) vport->upcall_portid = nla_get_u32(a[OVS_VPORT_ATTR_UPCALL_PID]); @@ -1867,8 +2104,8 @@ static int ovs_vport_cmd_del(struct sk_buff *skb, struct genl_info *info) goto exit_unlock; } - reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq, - OVS_VPORT_CMD_DEL); + reply = ovs_vport_cmd_build_info(vport, info->snd_portid, + info->snd_seq, OVS_VPORT_CMD_DEL); err = PTR_ERR(reply); if (IS_ERR(reply)) goto exit_unlock; @@ -1897,8 +2134,8 @@ static int ovs_vport_cmd_get(struct sk_buff *skb, struct genl_info *info) if (IS_ERR(vport)) goto exit_unlock; - reply = ovs_vport_cmd_build_info(vport, info->snd_portid, info->snd_seq, - OVS_VPORT_CMD_NEW); + reply = ovs_vport_cmd_build_info(vport, info->snd_portid, + info->snd_seq, OVS_VPORT_CMD_NEW); err = PTR_ERR(reply); if (IS_ERR(reply)) goto exit_unlock; diff --git a/net/openvswitch/datapath.h b/net/openvswitch/datapath.h index 16b84069521..a9148648491 100644 --- a/net/openvswitch/datapath.h +++ b/net/openvswitch/datapath.h @@ -88,9 +88,12 @@ struct datapath { /** * struct ovs_skb_cb - OVS data in skb CB * @flow: The flow associated with this packet. May be %NULL if no flow. + * @tun_key: Key for the tunnel that encapsulated this packet. NULL if the + * packet is not being tunneled. */ struct ovs_skb_cb { struct sw_flow *flow; + struct ovs_key_ipv4_tunnel *tun_key; }; #define OVS_CB(skb) ((struct ovs_skb_cb *)(skb)->cb) @@ -119,6 +122,7 @@ struct dp_upcall_info { struct ovs_net { struct list_head dps; struct work_struct dp_notify_work; + struct vport_net vport_net; }; extern int ovs_net_id; diff --git a/net/openvswitch/dp_notify.c b/net/openvswitch/dp_notify.c index ef4feec6cd8..c3235675f35 100644 --- a/net/openvswitch/dp_notify.c +++ b/net/openvswitch/dp_notify.c @@ -78,7 +78,7 @@ static int dp_device_event(struct notifier_block *unused, unsigned long event, void *ptr) { struct ovs_net *ovs_net; - struct net_device *dev = ptr; + struct net_device *dev = netdev_notifier_info_to_dev(ptr); struct vport *vport = NULL; if (!ovs_is_internal_dev(dev)) diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index b15321a2228..5c519b121e1 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -40,6 +40,7 @@ #include <linux/icmpv6.h> #include <linux/rculist.h> #include <net/ip.h> +#include <net/ip_tunnels.h> #include <net/ipv6.h> #include <net/ndisc.h> @@ -198,20 +199,18 @@ void ovs_flow_used(struct sw_flow *flow, struct sk_buff *skb) spin_unlock(&flow->lock); } -struct sw_flow_actions *ovs_flow_actions_alloc(const struct nlattr *actions) +struct sw_flow_actions *ovs_flow_actions_alloc(int size) { - int actions_len = nla_len(actions); struct sw_flow_actions *sfa; - if (actions_len > MAX_ACTIONS_BUFSIZE) + if (size > MAX_ACTIONS_BUFSIZE) return ERR_PTR(-EINVAL); - sfa = kmalloc(sizeof(*sfa) + actions_len, GFP_KERNEL); + sfa = kmalloc(sizeof(*sfa) + size, GFP_KERNEL); if (!sfa) return ERR_PTR(-ENOMEM); - sfa->actions_len = actions_len; - nla_memcpy(sfa->actions, actions, actions_len); + sfa->actions_len = 0; return sfa; } @@ -354,6 +353,14 @@ struct sw_flow *ovs_flow_tbl_next(struct flow_table *table, u32 *bucket, u32 *la return NULL; } +static void __flow_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; @@ -370,7 +377,7 @@ static void flow_table_copy_flows(struct flow_table *old, struct flow_table *new head = flex_array_get(old->buckets, i); hlist_for_each_entry(flow, head, hash_node[old_ver]) - ovs_flow_tbl_insert(new, flow); + __flow_tbl_insert(new, flow); } old->keep_flows = true; } @@ -590,10 +597,10 @@ out: * - skb->network_header: just past the Ethernet header, or just past the * VLAN header, to the first byte of the Ethernet payload. * - * - skb->transport_header: If key->dl_type is ETH_P_IP or ETH_P_IPV6 + * - skb->transport_header: If key->eth.type is ETH_P_IP or ETH_P_IPV6 * on output, then just past the IP header, if one is present and * of a correct length, otherwise the same as skb->network_header. - * For other key->dl_type values it is left untouched. + * For other key->eth.type values it is left untouched. */ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key, int *key_lenp) @@ -605,6 +612,8 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key, memset(key, 0, sizeof(*key)); key->phy.priority = skb->priority; + if (OVS_CB(skb)->tun_key) + memcpy(&key->tun_key, OVS_CB(skb)->tun_key, sizeof(key->tun_key)); key->phy.in_port = in_port; key->phy.skb_mark = skb->mark; @@ -618,6 +627,9 @@ int ovs_flow_extract(struct sk_buff *skb, u16 in_port, struct sw_flow_key *key, memcpy(key->eth.dst, eth->h_dest, ETH_ALEN); __skb_pull(skb, 2 * ETH_ALEN); + /* We are going to push all headers that we pull, so no need to + * update skb->csum here. + */ if (vlan_tx_tag_present(skb)) key->eth.tci = htons(skb->vlan_tci); @@ -759,9 +771,18 @@ out: return error; } -u32 ovs_flow_hash(const struct sw_flow_key *key, int key_len) +static u32 ovs_flow_hash(const struct sw_flow_key *key, int key_start, int key_len) +{ + return jhash2((u32 *)((u8 *)key + key_start), + DIV_ROUND_UP(key_len - key_start, sizeof(u32)), 0); +} + +static int flow_key_start(struct sw_flow_key *key) { - return jhash2((u32 *)key, DIV_ROUND_UP(key_len, sizeof(u32)), 0); + if (key->tun_key.ipv4_dst) + return 0; + else + return offsetof(struct sw_flow_key, phy); } struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *table, @@ -769,28 +790,31 @@ struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *table, { struct sw_flow *flow; struct hlist_head *head; + u8 *_key; + int key_start; u32 hash; - hash = ovs_flow_hash(key, key_len); + key_start = flow_key_start(key); + hash = ovs_flow_hash(key, key_start, key_len); + _key = (u8 *) key + key_start; head = find_bucket(table, hash); hlist_for_each_entry_rcu(flow, head, hash_node[table->node_ver]) { if (flow->hash == hash && - !memcmp(&flow->key, key, key_len)) { + !memcmp((u8 *)&flow->key + key_start, _key, key_len - key_start)) { return flow; } } return NULL; } -void ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow) +void ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow, + struct sw_flow_key *key, int key_len) { - struct hlist_head *head; - - head = find_bucket(table, flow->hash); - hlist_add_head_rcu(&flow->hash_node[table->node_ver], head); - table->count++; + flow->hash = ovs_flow_hash(key, flow_key_start(key), key_len); + memcpy(&flow->key, key, sizeof(flow->key)); + __flow_tbl_insert(table, flow); } void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow) @@ -817,6 +841,7 @@ const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1] = { [OVS_KEY_ATTR_ICMPV6] = sizeof(struct ovs_key_icmpv6), [OVS_KEY_ATTR_ARP] = sizeof(struct ovs_key_arp), [OVS_KEY_ATTR_ND] = sizeof(struct ovs_key_nd), + [OVS_KEY_ATTR_TUNNEL] = -1, }; static int ipv4_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_len, @@ -954,6 +979,105 @@ static int parse_flow_nlattrs(const struct nlattr *attr, return 0; } +int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr, + struct ovs_key_ipv4_tunnel *tun_key) +{ + struct nlattr *a; + int rem; + bool ttl = false; + + memset(tun_key, 0, sizeof(*tun_key)); + + nla_for_each_nested(a, attr, rem) { + int type = nla_type(a); + static const u32 ovs_tunnel_key_lens[OVS_TUNNEL_KEY_ATTR_MAX + 1] = { + [OVS_TUNNEL_KEY_ATTR_ID] = sizeof(u64), + [OVS_TUNNEL_KEY_ATTR_IPV4_SRC] = sizeof(u32), + [OVS_TUNNEL_KEY_ATTR_IPV4_DST] = sizeof(u32), + [OVS_TUNNEL_KEY_ATTR_TOS] = 1, + [OVS_TUNNEL_KEY_ATTR_TTL] = 1, + [OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT] = 0, + [OVS_TUNNEL_KEY_ATTR_CSUM] = 0, + }; + + if (type > OVS_TUNNEL_KEY_ATTR_MAX || + ovs_tunnel_key_lens[type] != nla_len(a)) + return -EINVAL; + + switch (type) { + case OVS_TUNNEL_KEY_ATTR_ID: + tun_key->tun_id = nla_get_be64(a); + tun_key->tun_flags |= TUNNEL_KEY; + break; + case OVS_TUNNEL_KEY_ATTR_IPV4_SRC: + tun_key->ipv4_src = nla_get_be32(a); + break; + case OVS_TUNNEL_KEY_ATTR_IPV4_DST: + tun_key->ipv4_dst = nla_get_be32(a); + break; + case OVS_TUNNEL_KEY_ATTR_TOS: + tun_key->ipv4_tos = nla_get_u8(a); + break; + case OVS_TUNNEL_KEY_ATTR_TTL: + tun_key->ipv4_ttl = nla_get_u8(a); + ttl = true; + break; + case OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT: + tun_key->tun_flags |= TUNNEL_DONT_FRAGMENT; + break; + case OVS_TUNNEL_KEY_ATTR_CSUM: + tun_key->tun_flags |= TUNNEL_CSUM; + break; + default: + return -EINVAL; + + } + } + if (rem > 0) + return -EINVAL; + + if (!tun_key->ipv4_dst) + return -EINVAL; + + if (!ttl) + return -EINVAL; + + return 0; +} + +int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb, + const struct ovs_key_ipv4_tunnel *tun_key) +{ + struct nlattr *nla; + + nla = nla_nest_start(skb, OVS_KEY_ATTR_TUNNEL); + if (!nla) + return -EMSGSIZE; + + if (tun_key->tun_flags & TUNNEL_KEY && + nla_put_be64(skb, OVS_TUNNEL_KEY_ATTR_ID, tun_key->tun_id)) + return -EMSGSIZE; + if (tun_key->ipv4_src && + nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_SRC, tun_key->ipv4_src)) + return -EMSGSIZE; + if (nla_put_be32(skb, OVS_TUNNEL_KEY_ATTR_IPV4_DST, tun_key->ipv4_dst)) + return -EMSGSIZE; + if (tun_key->ipv4_tos && + nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TOS, tun_key->ipv4_tos)) + return -EMSGSIZE; + if (nla_put_u8(skb, OVS_TUNNEL_KEY_ATTR_TTL, tun_key->ipv4_ttl)) + return -EMSGSIZE; + if ((tun_key->tun_flags & TUNNEL_DONT_FRAGMENT) && + nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_DONT_FRAGMENT)) + return -EMSGSIZE; + if ((tun_key->tun_flags & TUNNEL_CSUM) && + nla_put_flag(skb, OVS_TUNNEL_KEY_ATTR_CSUM)) + return -EMSGSIZE; + + nla_nest_end(skb, nla); + return 0; +} + /** * ovs_flow_from_nlattrs - parses Netlink attributes into a flow key. * @swkey: receives the extracted flow key. @@ -996,6 +1120,14 @@ int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp, attrs &= ~(1 << OVS_KEY_ATTR_SKB_MARK); } + if (attrs & (1 << OVS_KEY_ATTR_TUNNEL)) { + err = ovs_ipv4_tun_from_nlattr(a[OVS_KEY_ATTR_TUNNEL], &swkey->tun_key); + if (err) + return err; + + attrs &= ~(1 << OVS_KEY_ATTR_TUNNEL); + } + /* Data attributes. */ if (!(attrs & (1 << OVS_KEY_ATTR_ETHERNET))) return -EINVAL; @@ -1122,10 +1254,9 @@ int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp, /** * ovs_flow_metadata_from_nlattrs - parses Netlink attributes into a flow key. - * @priority: receives the skb priority - * @mark: receives the skb mark - * @in_port: receives the extracted input port. - * @key: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute + * @flow: Receives extracted in_port, priority, tun_key and skb_mark. + * @key_len: Length of key in @flow. Used for calculating flow hash. + * @attr: Netlink attribute holding nested %OVS_KEY_ATTR_* Netlink attribute * sequence. * * This parses a series of Netlink attributes that form a flow key, which must @@ -1133,42 +1264,56 @@ int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp, * get the metadata, that is, the parts of the flow key that cannot be * extracted from the packet itself. */ -int ovs_flow_metadata_from_nlattrs(u32 *priority, u32 *mark, u16 *in_port, - const struct nlattr *attr) +int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow, int key_len, + const struct nlattr *attr) { + struct ovs_key_ipv4_tunnel *tun_key = &flow->key.tun_key; const struct nlattr *nla; int rem; - *in_port = DP_MAX_PORTS; - *priority = 0; - *mark = 0; + flow->key.phy.in_port = DP_MAX_PORTS; + flow->key.phy.priority = 0; + flow->key.phy.skb_mark = 0; + memset(tun_key, 0, sizeof(flow->key.tun_key)); nla_for_each_nested(nla, attr, rem) { int type = nla_type(nla); if (type <= OVS_KEY_ATTR_MAX && ovs_key_lens[type] > 0) { + int err; + if (nla_len(nla) != ovs_key_lens[type]) return -EINVAL; switch (type) { case OVS_KEY_ATTR_PRIORITY: |