aboutsummaryrefslogtreecommitdiff
path: root/net/openvswitch
diff options
context:
space:
mode:
Diffstat (limited to 'net/openvswitch')
-rw-r--r--net/openvswitch/Kconfig14
-rw-r--r--net/openvswitch/Makefile3
-rw-r--r--net/openvswitch/actions.c8
-rw-r--r--net/openvswitch/datapath.c371
-rw-r--r--net/openvswitch/datapath.h4
-rw-r--r--net/openvswitch/dp_notify.c2
-rw-r--r--net/openvswitch/flow.c211
-rw-r--r--net/openvswitch/flow.h47
-rw-r--r--net/openvswitch/vport-gre.c275
-rw-r--r--net/openvswitch/vport-internal_dev.c3
-rw-r--r--net/openvswitch/vport-netdev.c9
-rw-r--r--net/openvswitch/vport-netdev.h1
-rw-r--r--net/openvswitch/vport.c34
-rw-r--r--net/openvswitch/vport.h23
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:
- *priority = nla_get_u32(nla);
+ flow->key.phy.priority = nla_get_u32(nla);
+ break;
+
+ case OVS_KEY_ATTR_TUNNEL:
+ err = ovs_ipv4_tun_from_nlattr(nla, tun_key);
+ if (err)
+ return err;
break;
case OVS_KEY_ATTR_IN_PORT:
if (nla_get_u32(nla) >= DP_MAX_PORTS)
return -EINVAL;
- *in_port = nla_get_u32(nla);
+ flow->key.phy.in_port = nla_get_u32(nla);
break;
case OVS_KEY_ATTR_SKB_MARK:
- *mark = nla_get_u32(nla);
+ flow->key.phy.skb_mark = nla_get_u32(nla);
break;
}
}
}
if (rem)
return -EINVAL;
+
+ flow->hash = ovs_flow_hash(&flow->key,
+ flow_key_start(&flow->key), key_len);
+
return 0;
}
@@ -1181,6 +1326,10 @@ int ovs_flow_to_nlattrs(const struct sw_flow_key *swkey, struct sk_buff *skb)
nla_put_u32(skb, OVS_KEY_ATTR_PRIORITY, swkey->phy.priority))
goto nla_put_failure;
+ if (swkey->tun_key.ipv4_dst &&
+ ovs_ipv4_tun_to_nlattr(skb, &swkey->tun_key))
+ goto nla_put_failure;
+
if (swkey->phy.in_port != DP_MAX_PORTS &&
nla_put_u32(skb, OVS_KEY_ATTR_IN_PORT, swkey->phy.in_port))
goto nla_put_failure;
diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
index 0875fde65b9..66ef7220293 100644
--- a/net/openvswitch/flow.h
+++ b/net/openvswitch/flow.h
@@ -40,7 +40,38 @@ struct sw_flow_actions {
struct nlattr actions[];
};
+/* Used to memset ovs_key_ipv4_tunnel padding. */
+#define OVS_TUNNEL_KEY_SIZE \
+ (offsetof(struct ovs_key_ipv4_tunnel, ipv4_ttl) + \
+ FIELD_SIZEOF(struct ovs_key_ipv4_tunnel, ipv4_ttl))
+
+struct ovs_key_ipv4_tunnel {
+ __be64 tun_id;
+ __be32 ipv4_src;
+ __be32 ipv4_dst;
+ __be16 tun_flags;
+ u8 ipv4_tos;
+ u8 ipv4_ttl;
+};
+
+static inline void ovs_flow_tun_key_init(struct ovs_key_ipv4_tunnel *tun_key,
+ const struct iphdr *iph, __be64 tun_id,
+ __be16 tun_flags)
+{
+ tun_key->tun_id = tun_id;
+ tun_key->ipv4_src = iph->saddr;
+ tun_key->ipv4_dst = iph->daddr;
+ tun_key->ipv4_tos = iph->tos;
+ tun_key->ipv4_ttl = iph->ttl;
+ tun_key->tun_flags = tun_flags;
+
+ /* clear struct padding. */
+ memset((unsigned char *) tun_key + OVS_TUNNEL_KEY_SIZE, 0,
+ sizeof(*tun_key) - OVS_TUNNEL_KEY_SIZE);
+}
+
struct sw_flow_key {
+ struct ovs_key_ipv4_tunnel tun_key; /* Encapsulating tunnel key. */
struct {
u32 priority; /* Packet QoS priority. */
u32 skb_mark; /* SKB mark. */
@@ -130,7 +161,7 @@ struct sw_flow *ovs_flow_alloc(void);
void ovs_flow_deferred_free(struct sw_flow *);
void ovs_flow_free(struct sw_flow *flow);
-struct sw_flow_actions *ovs_flow_actions_alloc(const struct nlattr *);
+struct sw_flow_actions *ovs_flow_actions_alloc(int actions_len);
void ovs_flow_deferred_free_acts(struct sw_flow_actions *);
int ovs_flow_extract(struct sk_buff *, u16 in_port, struct sw_flow_key *,
@@ -141,10 +172,10 @@ u64 ovs_flow_used_time(unsigned long flow_jiffies);
int ovs_flow_to_nlattrs(const struct sw_flow_key *, struct sk_buff *);
int ovs_flow_from_nlattrs(struct sw_flow_key *swkey, int *key_lenp,
const struct nlattr *);
-int ovs_flow_metadata_from_nlattrs(u32 *priority, u32 *mark, u16 *in_port,
- const struct nlattr *);
+int ovs_flow_metadata_from_nlattrs(struct sw_flow *flow, int key_len,
+ const struct nlattr *attr);
-#define MAX_ACTIONS_BUFSIZE (16 * 1024)
+#define MAX_ACTIONS_BUFSIZE (32 * 1024)
#define TBL_MIN_BUCKETS 1024
struct flow_table {
@@ -173,11 +204,15 @@ void ovs_flow_tbl_deferred_destroy(struct flow_table *table);
struct flow_table *ovs_flow_tbl_alloc(int new_size);
struct flow_table *ovs_flow_tbl_expand(struct flow_table *table);
struct flow_table *ovs_flow_tbl_rehash(struct flow_table *table);
-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);
void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow);
-u32 ovs_flow_hash(const struct sw_flow_key *key, int key_len);
struct sw_flow *ovs_flow_tbl_next(struct flow_table *table, u32 *bucket, u32 *idx);
extern const int ovs_key_lens[OVS_KEY_ATTR_MAX + 1];
+int ovs_ipv4_tun_from_nlattr(const struct nlattr *attr,
+ struct ovs_key_ipv4_tunnel *tun_key);
+int ovs_ipv4_tun_to_nlattr(struct sk_buff *skb,
+ const struct ovs_key_ipv4_tunnel *tun_key);
#endif /* flow.h */
diff --git a/net/openvswitch/vport-gre.c b/net/openvswitch/vport-gre.c
new file mode 100644
index 00000000000..493e9775dcd
--- /dev/null
+++ b/net/openvswitch/vport-gre.c
@@ -0,0 +1,275 @@
+/*
+ * 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
+ */
+
+#ifdef CONFIG_OPENVSWITCH_GRE
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/if.h>
+#include <linux/skbuff.h>
+#include <linux/ip.h>
+#include <linux/if_tunnel.h>
+#include <linux/if_vlan.h>
+#include <linux/in.h>
+#include <linux/if_vlan.h>
+#include <linux/in.h>
+#include <linux/in_route.h>
+#include <linux/inetdevice.h>
+#include <linux/jhash.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/workqueue.h>
+#include <linux/rculist.h>
+#include <net/route.h>
+#include <net/xfrm.h>
+
+#include <net/icmp.h>
+#include <net/ip.h>
+#include <net/ip_tunnels.h>
+#include <net/gre.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/protocol.h>
+
+#include "datapath.h"
+#include "vport.h"
+
+/* Returns the least-significant 32 bits of a __be64. */
+static __be32 be64_get_low32(__be64 x)
+{
+#ifdef __BIG_ENDIAN
+ return (__force __be32)x;
+#else
+ return (__force __be32)((__force u64)x >> 32);
+#endif
+}
+
+static __be16 filter_tnl_flags(__be16 flags)
+{
+ return flags & (TUNNEL_CSUM | TUNNEL_KEY);
+}
+
+static struct sk_buff *__build_header(struct sk_buff *skb,
+ int tunnel_hlen)
+{
+ const struct ovs_key_ipv4_tunnel *tun_key = OVS_CB(skb)->tun_key;
+ struct tnl_ptk_info tpi;
+
+ skb = gre_handle_offloads(skb, !!(tun_key->tun_flags & TUNNEL_CSUM));
+ if (IS_ERR(skb))
+ return NULL;
+
+ tpi.flags = filter_tnl_flags(tun_key->tun_flags);
+ tpi.proto = htons(ETH_P_TEB);
+ tpi.key = be64_get_low32(tun_key->tun_id);
+ tpi.seq = 0;
+ gre_build_header(skb, &tpi, tunnel_hlen);
+
+ return skb;
+}
+
+static __be64 key_to_tunnel_id(__be32 key, __be32 seq)
+{
+#ifdef __BIG_ENDIAN
+ return (__force __be64)((__force u64)seq << 32 | (__force u32)key);
+#else
+ return (__force __be64)((__force u64)key << 32 | (__force u32)seq);
+#endif
+}
+
+/* Called with rcu_read_lock and BH disabled. */
+static int gre_rcv(struct sk_buff *skb,
+ const struct tnl_ptk_info *tpi)
+{
+ struct ovs_key_ipv4_tunnel tun_key;
+ struct ovs_net *ovs_net;
+ struct vport *vport;
+ __be64 key;
+
+ ovs_net = net_generic(dev_net(skb->dev), ovs_net_id);
+ vport = rcu_dereference(ovs_net->vport_net.gre_vport);
+ if (unlikely(!vport))
+ return PACKET_REJECT;
+
+ key = key_to_tunnel_id(tpi->key, tpi->seq);
+ ovs_flow_tun_key_init(&tun_key, ip_hdr(skb), key,
+ filter_tnl_flags(tpi->flags));
+
+ ovs_vport_receive(vport, skb, &tun_key);
+ return PACKET_RCVD;
+}
+
+static int gre_tnl_send(struct vport *vport, struct sk_buff *skb)
+{
+ struct net *net = ovs_dp_get_net(vport->dp);
+ struct flowi4 fl;
+ struct rtable *rt;
+ int min_headroom;
+ int tunnel_hlen;
+ __be16 df;
+ int err;
+
+ if (unlikely(!OVS_CB(skb)->tun_key)) {
+ err = -EINVAL;
+ goto error;
+ }
+
+ /* Route lookup */
+ memset(&fl, 0, sizeof(fl));
+ fl.daddr = OVS_CB(skb)->tun_key->ipv4_dst;
+ fl.saddr = OVS_CB(skb)->tun_key->ipv4_src;
+ fl.flowi4_tos = RT_TOS(OVS_CB(skb)->tun_key->ipv4_tos);
+ fl.flowi4_mark = skb->mark;
+ fl.flowi4_proto = IPPROTO_GRE;
+
+ rt = ip_route_output_key(net, &fl);
+ if (IS_ERR(rt))
+ return PTR_ERR(rt);
+
+ tunnel_hlen = ip_gre_calc_hlen(OVS_CB(skb)->tun_key->tun_flags);
+
+ min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len
+ + tunnel_hlen + sizeof(struct iphdr)
+ + (vlan_tx_tag_present(skb) ? VLAN_HLEN : 0);
+ if (skb_headroom(skb) < min_headroom || skb_header_cloned(skb)) {
+ int head_delta = SKB_DATA_ALIGN(min_headroom -
+ skb_headroom(skb) +
+ 16);
+ err = pskb_expand_head(skb, max_t(int, head_delta, 0),
+ 0, GFP_ATOMIC);
+ if (unlikely(err))
+ goto err_free_rt;
+ }
+
+ if (vlan_tx_tag_present(skb)) {
+ if (unlikely(!__vlan_put_tag(skb,
+ skb->vlan_proto,
+ vlan_tx_tag_get(skb)))) {
+ err = -ENOMEM;
+ goto err_free_rt;
+ }
+ skb->vlan_tci = 0;
+ }
+
+ /* Push Tunnel header. */
+ skb = __build_header(skb, tunnel_hlen);
+ if (unlikely(!skb)) {
+ err = 0;
+ goto err_free_rt;
+ }
+
+ df = OVS_CB(skb)->tun_key->tun_flags & TUNNEL_DONT_FRAGMENT ?
+ htons(IP_DF) : 0;
+
+ skb->local_df = 1;
+
+ return iptunnel_xmit(net, rt, skb, fl.saddr,
+ OVS_CB(skb)->tun_key->ipv4_dst, IPPROTO_GRE,
+ OVS_CB(skb)->tun_key->ipv4_tos,
+ OVS_CB(skb)->tun_key->ipv4_ttl, df);
+err_free_rt:
+ ip_rt_put(rt);
+error:
+ return err;
+}
+
+static struct gre_cisco_protocol gre_protocol = {
+ .handler = gre_rcv,
+ .priority = 1,
+};
+
+static int gre_ports;
+static int gre_init(void)
+{
+ int err;
+
+ gre_ports++;
+ if (gre_ports > 1)
+ return 0;
+
+ err = gre_cisco_register(&gre_protocol);
+ if (err)
+ pr_warn("cannot register gre protocol handler\n");
+
+ return err;
+}
+
+static void gre_exit(void)
+{
+ gre_ports--;
+ if (gre_ports > 0)
+ return;
+
+ gre_cisco_unregister(&gre_protocol);
+}
+
+static const char *gre_get_name(const struct vport *vport)
+{
+ return vport_priv(vport);
+}
+
+static struct vport *gre_create(const struct vport_parms *parms)
+{
+ struct net *net = ovs_dp_get_net(parms->dp);
+ struct ovs_net *ovs_net;
+ struct vport *vport;
+ int err;
+
+ err = gre_init();
+ if (err)
+ return ERR_PTR(err);
+
+ ovs_net = net_generic(net, ovs_net_id);
+ if (ovsl_dereference(ovs_net->vport_net.gre_vport)) {
+ vport = ERR_PTR(-EEXIST);
+ goto error;
+ }
+
+ vport = ovs_vport_alloc(IFNAMSIZ, &ovs_gre_vport_ops, parms);
+ if (IS_ERR(vport))
+ goto error;
+
+ strncpy(vport_priv(vport), parms->name, IFNAMSIZ);
+ rcu_assign_pointer(ovs_net->vport_net.gre_vport, vport);
+ return vport;
+
+error:
+ gre_exit();
+ return vport;
+}
+
+static void gre_tnl_destroy(struct vport *vport)
+{
+ struct net *net = ovs_dp_get_net(vport->dp);
+ struct ovs_net *ovs_net;
+
+ ovs_net = net_generic(net, ovs_net_id);
+
+ rcu_assign_pointer(ovs_net->vport_net.gre_vport, NULL);
+ ovs_vport_deferred_free(vport);
+ gre_exit();
+}
+
+const struct vport_ops ovs_gre_vport_ops = {
+ .type = OVS_VPORT_TYPE_GRE,
+ .create = gre_create,
+ .destroy = gre_tnl_destroy,
+ .get_name = gre_get_name,
+ .send = gre_tnl_send,
+};
+
+#endif /* OPENVSWITCH_GRE */
diff --git a/net/openvswitch/vport-internal_dev.c b/net/openvswitch/vport-internal_dev.c
index 84e0a037918..98d3edbbc23 100644
--- a/net/openvswitch/vport-internal_dev.c
+++ b/net/openvswitch/vport-internal_dev.c
@@ -67,7 +67,7 @@ static struct rtnl_link_stats64 *internal_dev_get_stats(struct net_device *netde
static int internal_dev_xmit(struct sk_buff *skb, struct net_device *netdev)
{
rcu_read_lock();
- ovs_vport_receive(internal_dev_priv(netdev)->vport, skb);
+ ovs_vport_receive(internal_dev_priv(netdev)->vport, skb, NULL);
rcu_read_unlock();
return 0;
}
@@ -221,6 +221,7 @@ static int internal_dev_recv(struct vport *vport, struct sk_buff *skb)
skb->dev = netdev;
skb->pkt_type = PACKET_HOST;
skb->protocol = eth_type_trans(skb, netdev);
+ skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
netif_rx(skb);
diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c
index 4f01c6d2ffa..5982f3f6283 100644
--- a/net/openvswitch/vport-netdev.c
+++ b/net/openvswitch/vport-netdev.c
@@ -49,7 +49,9 @@ static void netdev_port_receive(struct vport *vport, struct sk_buff *skb)
return;
skb_push(skb, ETH_HLEN);
- ovs_vport_receive(vport, skb);
+ ovs_skb_postpush_rcsum(skb, skb->data, ETH_HLEN);
+
+ ovs_vport_receive(vport, skb, NULL);
return;
error:
@@ -170,7 +172,7 @@ static int netdev_send(struct vport *vport, struct sk_buff *skb)
net_warn_ratelimited("%s: dropped over-mtu packet: %d > %d\n",
netdev_vport->dev->name,
packet_length(skb), mtu);
- goto error;
+ goto drop;
}
skb->dev = netdev_vport->dev;
@@ -179,9 +181,8 @@ static int netdev_send(struct vport *vport, struct sk_buff *skb)
return len;
-error:
+drop:
kfree_skb(skb);
- ovs_vport_record_error(vport, VPORT_E_TX_DROPPED);
return 0;
}
diff --git a/net/openvswitch/vport-netdev.h b/net/openvswitch/vport-netdev.h
index a3cb3a32cd7..dd298b5c5cd 100644
--- a/net/openvswitch/vport-netdev.h
+++ b/net/openvswitch/vport-netdev.h
@@ -39,6 +39,5 @@ netdev_vport_priv(const struct vport *vport)
}
const char *ovs_netdev_get_name(const struct vport *);
-const char *ovs_netdev_get_config(const struct vport *);
#endif /* vport_netdev.h */
diff --git a/net/openvswitch/vport.c b/net/openvswitch/vport.c
index 720623190ea..d4c7fa04ce0 100644
--- a/net/openvswitch/vport.c
+++ b/net/openvswitch/vport.c
@@ -38,6 +38,10 @@
static const struct vport_ops *vport_ops_list[] = {
&ovs_netdev_vport_ops,
&ovs_internal_vport_ops,
+
+#ifdef CONFIG_OPENVSWITCH_GRE
+ &ovs_gre_vport_ops,
+#endif
};
/* Protected by RCU read lock for reading, ovs_mutex for writing. */
@@ -325,7 +329,8 @@ int ovs_vport_get_options(const struct vport *vport, struct sk_buff *skb)
* Must be called with rcu_read_lock. The packet cannot be shared and
* skb->data should point to the Ethernet header.
*/
-void ovs_vport_receive(struct vport *vport, struct sk_buff *skb)
+void ovs_vport_receive(struct vport *vport, struct sk_buff *skb,
+ struct ovs_key_ipv4_tunnel *tun_key)
{
struct pcpu_tstats *stats;
@@ -335,6 +340,7 @@ void ovs_vport_receive(struct vport *vport, struct sk_buff *skb)
stats->rx_bytes += skb->len;
u64_stats_update_end(&stats->syncp);
+ OVS_CB(skb)->tun_key = tun_key;
ovs_dp_process_received_packet(vport, skb);
}
@@ -351,7 +357,7 @@ int ovs_vport_send(struct vport *vport, struct sk_buff *skb)
{
int sent = vport->ops->send(vport, skb);
- if (likely(sent)) {
+ if (likely(sent > 0)) {
struct pcpu_tstats *stats;
stats = this_cpu_ptr(vport->percpu_stats);
@@ -360,7 +366,12 @@ int ovs_vport_send(struct vport *vport, struct sk_buff *skb)
stats->tx_packets++;
stats->tx_bytes += sent;
u64_stats_update_end(&stats->syncp);
- }
+ } else if (sent < 0) {
+ ovs_vport_record_error(vport, VPORT_E_TX_ERROR);
+ kfree_skb(skb);
+ } else
+ ovs_vport_record_error(vport, VPORT_E_TX_DROPPED);
+
return sent;
}
@@ -371,7 +382,7 @@ int ovs_vport_send(struct vport *vport, struct sk_buff *skb)
* @err_type: one of enum vport_err_type types to indicate the error type
*
* If using the vport generic stats layer indicate that an error of the given
- * type has occured.
+ * type has occurred.
*/
void ovs_vport_record_error(struct vport *vport, enum vport_err_type err_type)
{
@@ -397,3 +408,18 @@ void ovs_vport_record_error(struct vport *vport, enum vport_err_type err_type)
spin_unlock(&vport->stats_lock);
}
+
+static void free_vport_rcu(struct rcu_head *rcu)
+{
+ struct vport *vport = container_of(rcu, struct vport, rcu);
+
+ ovs_vport_free(vport);
+}
+
+void ovs_vport_deferred_free(struct vport *vport)
+{
+ if (!vport)
+ return;
+
+ call_rcu(&vport->rcu, free_vport_rcu);
+}
diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h
index 68a377bc084..376045c42f8 100644
--- a/net/openvswitch/vport.h
+++ b/net/openvswitch/vport.h
@@ -34,6 +34,11 @@ struct vport_parms;
/* The following definitions are for users of the vport subsytem: */
+/* The following definitions are for users of the vport subsytem: */
+struct vport_net {
+ struct vport __rcu *gre_vport;
+};
+
int ovs_vport_init(void);
void ovs_vport_exit(void);
@@ -123,9 +128,8 @@ struct vport_parms {
* existing vport to a &struct sk_buff. May be %NULL for a vport that does not
* have any configuration.
* @get_name: Get the device's name.
- * @get_config: Get the device's configuration.
- * May be null if the device does not have an ifindex.
- * @send: Send a packet on the device. Returns the length of the packet sent.
+ * @send: Send a packet on the device. Returns the length of the packet sent,
+ * zero for dropped packets or negative for error.
*/
struct vport_ops {
enum ovs_vport_type type;
@@ -139,7 +143,6 @@ struct vport_ops {
/* Called with rcu_read_lock or ovs_mutex. */
const char *(*get_name)(const struct vport *);
- void (*get_config)(const struct vport *, void *);
int (*send)(struct vport *, struct sk_buff *);
};
@@ -154,6 +157,7 @@ enum vport_err_type {
struct vport *ovs_vport_alloc(int priv_size, const struct vport_ops *,
const struct vport_parms *);
void ovs_vport_free(struct vport *);
+void ovs_vport_deferred_free(struct vport *vport);
#define VPORT_ALIGN 8
@@ -186,12 +190,21 @@ static inline struct vport *vport_from_priv(const void *priv)
return (struct vport *)(priv - ALIGN(sizeof(struct vport), VPORT_ALIGN));
}
-void ovs_vport_receive(struct vport *, struct sk_buff *);
+void ovs_vport_receive(struct vport *, struct sk_buff *,
+ struct ovs_key_ipv4_tunnel *);
void ovs_vport_record_error(struct vport *, enum vport_err_type err_type);
/* List of statically compiled vport implementations. Don't forget to also
* add yours to the list at the top of vport.c. */
extern const struct vport_ops ovs_netdev_vport_ops;
extern const struct vport_ops ovs_internal_vport_ops;
+extern const struct vport_ops ovs_gre_vport_ops;
+
+static inline void ovs_skb_postpush_rcsum(struct sk_buff *skb,
+ const void *start, unsigned int len)
+{
+ if (skb->ip_summed == CHECKSUM_COMPLETE)
+ skb->csum = csum_add(skb->csum, csum_partial(start, len, 0));
+}
#endif /* vport.h */
generated by cgit v1.2.3 (git 2.25.1) at 2024-05-28 17:23:35 +0000