[PATCH net-next v12 5/5] openvswitch: Add support for unique flow IDs.

From: Joe Stringer
Date: Thu Jan 15 2015 - 16:56:36 EST

Previously, flows were manipulated by userspace specifying a full,
unmasked flow key. This adds significant burden onto flow
serialization/deserialization, particularly when dumping flows.

This patch adds an alternative way to refer to flows using a
variable-length "unique flow identifier" (UFID). At flow setup time,
userspace may specify a UFID for a flow, which is stored with the flow
and inserted into a separate table for lookup, in addition to the
standard flow table. Flows created using a UFID must be fetched or
deleted using the UFID.

All flow dump operations may now be made more terse with OVS_UFID_F_*
flags. For example, the OVS_UFID_F_OMIT_KEY flag allows responses to
omit the flow key from a datapath operation if the flow has a
corresponding UFID. This significantly reduces the time spent assembling
and transacting netlink messages. With all OVS_UFID_F_OMIT_* flags
enabled, the datapath only returns the UFID and statistics for each flow
during flow dump, increasing ovs-vswitchd revalidator performance by 40%
or more.

Signed-off-by: Joe Stringer <joestringer@xxxxxxxxxx>
---
Documentation/networking/openvswitch.txt | 13 ++
include/uapi/linux/openvswitch.h | 20 +++
net/openvswitch/datapath.c | 221 +++++++++++++++++++++++-------
net/openvswitch/flow.h | 28 +++-
net/openvswitch/flow_netlink.c | 67 ++++++++-
net/openvswitch/flow_netlink.h | 9 +-
net/openvswitch/flow_table.c | 187 ++++++++++++++++++++-----
net/openvswitch/flow_table.h | 8 +-
8 files changed, 461 insertions(+), 92 deletions(-)

diff --git a/Documentation/networking/openvswitch.txt b/Documentation/networking/openvswitch.txt
index 37c20ee..b3b9ac6 100644
--- a/Documentation/networking/openvswitch.txt
+++ b/Documentation/networking/openvswitch.txt
@@ -131,6 +131,19 @@ performs best-effort detection of overlapping wildcarded flows and may reject
some but not all of them. However, this behavior may change in future versions.


+Unique flow identifiers
+-----------------------
+
+An alternative to using the original match portion of a key as the handle for
+flow identification is a unique flow identifier, or "UFID". UFIDs are optional
+for both the kernel and user space program.
+
+User space programs that support UFID are expected to provide it during flow
+setup in addition to the flow, then refer to the flow using the UFID for all
+future operations. The kernel is not required to index flows by the original
+flow key if a UFID is specified.
+
+
Basic rule for evolving flow keys
---------------------------------

diff --git a/include/uapi/linux/openvswitch.h b/include/uapi/linux/openvswitch.h
index cd8d933..7a8785a 100644
--- a/include/uapi/linux/openvswitch.h
+++ b/include/uapi/linux/openvswitch.h
@@ -459,6 +459,14 @@ struct ovs_key_nd {
* a wildcarded match. Omitting attribute is treated as wildcarding all
* corresponding fields. Optional for all requests. If not present,
* all flow key bits are exact match bits.
+ * @OVS_FLOW_ATTR_UFID: A value between 1-16 octets specifying a unique
+ * identifier for the flow. Causes the flow to be indexed by this value rather
+ * than the value of the %OVS_FLOW_ATTR_KEY attribute. Optional for all
+ * requests. Present in notifications if the flow was created with this
+ * attribute.
+ * @OVS_FLOW_ATTR_UFID_FLAGS: A 32-bit value of OR'd %OVS_UFID_F_*
+ * flags that provide alternative semantics for flow installation and
+ * retrieval. Optional for all requests.
*
* These attributes follow the &struct ovs_header within the Generic Netlink
* payload for %OVS_FLOW_* commands.
@@ -474,12 +482,24 @@ enum ovs_flow_attr {
OVS_FLOW_ATTR_MASK, /* Sequence of OVS_KEY_ATTR_* attributes. */
OVS_FLOW_ATTR_PROBE, /* Flow operation is a feature probe, error
* logging should be suppressed. */
+ OVS_FLOW_ATTR_UFID, /* Variable length unique flow identifier. */
+ OVS_FLOW_ATTR_UFID_FLAGS,/* u32 of OVS_UFID_F_*. */
__OVS_FLOW_ATTR_MAX
};

#define OVS_FLOW_ATTR_MAX (__OVS_FLOW_ATTR_MAX - 1)

/**
+ * Omit attributes for notifications.
+ *
+ * If a datapath request contains an %OVS_UFID_F_OMIT_* flag, then the datapath
+ * may omit the corresponding %OVS_FLOW_ATTR_* from the response.
+ */
+#define OVS_UFID_F_OMIT_KEY (1 << 0)
+#define OVS_UFID_F_OMIT_MASK (1 << 1)
+#define OVS_UFID_F_OMIT_ACTIONS (1 << 2)
+
+/**
* enum ovs_sample_attr - Attributes for %OVS_ACTION_ATTR_SAMPLE action.
* @OVS_SAMPLE_ATTR_PROBABILITY: 32-bit fraction of packets to sample with
* @OVS_ACTION_ATTR_SAMPLE. A value of 0 samples no packets, a value of
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 3ded349..564e163 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -65,6 +65,8 @@ static struct genl_family dp_packet_genl_family;
static struct genl_family dp_flow_genl_family;
static struct genl_family dp_datapath_genl_family;

+static const struct nla_policy flow_policy[];
+
static const struct genl_multicast_group ovs_dp_flow_multicast_group = {
.name = OVS_FLOW_MCGROUP,
};
@@ -662,15 +664,48 @@ static void get_dp_stats(const struct datapath *dp, struct ovs_dp_stats *stats,
}
}

-static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts)
+static bool should_fill_key(const struct sw_flow_id *sfid, uint32_t ufid_flags)
+{
+ return ovs_identifier_is_key(sfid) ||
+ !(ufid_flags & OVS_UFID_F_OMIT_KEY);
+}
+
+static bool should_fill_mask(uint32_t ufid_flags)
{
- return NLMSG_ALIGN(sizeof(struct ovs_header))
- + nla_total_size(ovs_key_attr_size()) /* OVS_FLOW_ATTR_KEY */
- + nla_total_size(ovs_key_attr_size()) /* OVS_FLOW_ATTR_MASK */
+ return !(ufid_flags & OVS_UFID_F_OMIT_MASK);
+}
+
+static bool should_fill_actions(uint32_t ufid_flags)
+{
+ return !(ufid_flags & OVS_UFID_F_OMIT_ACTIONS);
+}
+
+static size_t ovs_flow_cmd_msg_size(const struct sw_flow_actions *acts,
+ const struct sw_flow_id *sfid,
+ uint32_t ufid_flags)
+{
+ size_t len = NLMSG_ALIGN(sizeof(struct ovs_header));
+
+ /* OVS_FLOW_ATTR_UFID */
+ if (sfid && ovs_identifier_is_ufid(sfid))
+ len += nla_total_size(sfid->ufid_len);
+
+ /* OVS_FLOW_ATTR_KEY */
+ if (!sfid || should_fill_key(sfid, ufid_flags))
+ len += nla_total_size(ovs_key_attr_size());
+
+ /* OVS_FLOW_ATTR_MASK */
+ if (should_fill_mask(ufid_flags))
+ len += nla_total_size(ovs_key_attr_size());
+
+ /* OVS_FLOW_ATTR_ACTIONS */
+ if (should_fill_actions(ufid_flags))
+ len += nla_total_size(acts->actions_len);
+
+ return len
+ nla_total_size(sizeof(struct ovs_flow_stats)) /* OVS_FLOW_ATTR_STATS */
+ nla_total_size(1) /* OVS_FLOW_ATTR_TCP_FLAGS */
- + nla_total_size(8) /* OVS_FLOW_ATTR_USED */
- + nla_total_size(acts->actions_len); /* OVS_FLOW_ATTR_ACTIONS */
+ + nla_total_size(8); /* OVS_FLOW_ATTR_USED */
}

/* Called with ovs_mutex or RCU read lock. */
@@ -741,7 +776,7 @@ static int ovs_flow_cmd_fill_actions(const struct sw_flow *flow,
/* 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)
+ u32 seq, u32 flags, u8 cmd, u32 ufid_flags)
{
const int skb_orig_len = skb->len;
struct ovs_header *ovs_header;
@@ -754,21 +789,31 @@ static int ovs_flow_cmd_fill_info(const struct sw_flow *flow, int dp_ifindex,

ovs_header->dp_ifindex = dp_ifindex;

- err = ovs_nla_put_unmasked_key(flow, skb);
+ err = ovs_nla_put_identifier(flow, skb);
if (err)
goto error;

- err = ovs_nla_put_mask(flow, skb);
- if (err)
- goto error;
+ if (should_fill_key(flow->id, ufid_flags)) {
+ err = ovs_nla_put_masked_key(flow, skb);
+ if (err)
+ goto error;
+ }
+
+ if (should_fill_mask(ufid_flags)) {
+ err = ovs_nla_put_mask(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;
+ if (should_fill_actions(ufid_flags)) {
+ err = ovs_flow_cmd_fill_actions(flow, skb, skb_orig_len);
+ if (err)
+ goto error;
+ }

return genlmsg_end(skb, ovs_header);

@@ -779,15 +824,19 @@ error:

/* May not be called with RCU read lock. */
static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *acts,
+ const struct sw_flow_id *sfid,
struct genl_info *info,
- bool always)
+ bool always,
+ uint32_t ufid_flags)
{
struct sk_buff *skb;
+ size_t len;

if (!always && !ovs_must_notify(&dp_flow_genl_family, info, 0))
return NULL;

- skb = genlmsg_new_unicast(ovs_flow_cmd_msg_size(acts), info, GFP_KERNEL);
+ len = ovs_flow_cmd_msg_size(acts, sfid, ufid_flags);
+ skb = genlmsg_new_unicast(len, info, GFP_KERNEL);
if (!skb)
return ERR_PTR(-ENOMEM);

@@ -798,19 +847,19 @@ static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *act
static struct sk_buff *ovs_flow_cmd_build_info(const struct sw_flow *flow,
int dp_ifindex,
struct genl_info *info, u8 cmd,
- bool always)
+ bool always, u32 ufid_flags)
{
struct sk_buff *skb;
int retval;

- skb = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts), info,
- always);
+ skb = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts),
+ flow->id, info, always, ufid_flags);
if (IS_ERR_OR_NULL(skb))
return skb;

retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb,
info->snd_portid, info->snd_seq, 0,
- cmd);
+ cmd, ufid_flags);
BUG_ON(retval < 0);
return skb;
}
@@ -819,12 +868,15 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
{
struct nlattr **a = info->attrs;
struct ovs_header *ovs_header = info->userhdr;
- struct sw_flow *flow, *new_flow;
+ struct sw_flow *flow = NULL, *new_flow;
struct sw_flow_mask mask;
struct sk_buff *reply;
struct datapath *dp;
+ struct sw_flow_key key;
+ struct sw_flow_id *sfid;
struct sw_flow_actions *acts;
struct sw_flow_match match;
+ u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);
int error;
bool log = !a[OVS_FLOW_ATTR_PROBE];

@@ -849,13 +901,21 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
}

/* Extract key. */
- ovs_match_init(&match, &new_flow->unmasked_key, &mask);
+ ovs_match_init(&match, &key, &mask);
error = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY],
a[OVS_FLOW_ATTR_MASK], log);
if (error)
goto err_kfree_flow;

- ovs_flow_mask_key(&new_flow->key, &new_flow->unmasked_key, &mask);
+ ovs_flow_mask_key(&new_flow->key, &key, &mask);
+
+ /* Extract flow identifier. */
+ sfid = ovs_nla_copy_identifier(a[OVS_FLOW_ATTR_UFID], &key, log);
+ if (IS_ERR(sfid)) {
+ error = PTR_ERR(sfid);
+ goto err_kfree_flow;
+ }
+ new_flow->id = sfid;

/* Validate actions. */
error = ovs_nla_copy_actions(a[OVS_FLOW_ATTR_ACTIONS], &new_flow->key,
@@ -865,7 +925,8 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
goto err_kfree_flow;
}

- reply = ovs_flow_cmd_alloc_info(acts, info, false);
+ reply = ovs_flow_cmd_alloc_info(acts, new_flow->id, info, false,
+ ufid_flags);
if (IS_ERR(reply)) {
error = PTR_ERR(reply);
goto err_kfree_acts;
@@ -877,8 +938,12 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
error = -ENODEV;
goto err_unlock_ovs;
}
+
/* Check if this is a duplicate flow */
- flow = ovs_flow_tbl_lookup(&dp->table, &new_flow->unmasked_key);
+ if (ovs_identifier_is_ufid(new_flow->id))
+ flow = ovs_flow_tbl_lookup_ufid(&dp->table, new_flow->id);
+ if (!flow)
+ flow = ovs_flow_tbl_lookup(&dp->table, &new_flow->key);
if (likely(!flow)) {
rcu_assign_pointer(new_flow->sf_acts, acts);

@@ -894,7 +959,8 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
- OVS_FLOW_CMD_NEW);
+ OVS_FLOW_CMD_NEW,
+ ufid_flags);
BUG_ON(error < 0);
}
ovs_unlock();
@@ -912,10 +978,15 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
error = -EEXIST;
goto err_unlock_ovs;
}
- /* The unmasked key has to be the same for flow updates. */
- if (unlikely(!ovs_flow_cmp_unmasked_key(flow, &match))) {
- /* Look for any overlapping flow. */
- flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ /* The flow identifier has to be the same for flow updates.
+ * Look for any overlapping flow.
+ */
+ if (unlikely(!ovs_flow_cmp(flow, &match))) {
+ if (ovs_identifier_is_key(flow->id))
+ flow = ovs_flow_tbl_lookup_exact(&dp->table,
+ &match);
+ else /* UFID matches but key is different */
+ flow = NULL;
if (!flow) {
error = -ENOENT;
goto err_unlock_ovs;
@@ -930,7 +1001,8 @@ static int ovs_flow_cmd_new(struct sk_buff *skb, struct genl_info *info)
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
- OVS_FLOW_CMD_NEW);
+ OVS_FLOW_CMD_NEW,
+ ufid_flags);
BUG_ON(error < 0);
}
ovs_unlock();
@@ -986,6 +1058,8 @@ static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
struct datapath *dp;
struct sw_flow_actions *old_acts = NULL, *acts = NULL;
struct sw_flow_match match;
+ struct sw_flow_id *sfid = NULL;
+ u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);
int error;
bool log = !a[OVS_FLOW_ATTR_PROBE];

@@ -1002,17 +1076,27 @@ static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
if (error)
goto error;

+ if (a[OVS_FLOW_ATTR_UFID]) {
+ sfid = ovs_nla_copy_identifier(a[OVS_FLOW_ATTR_UFID], &key,
+ log);
+ if (IS_ERR(sfid)) {
+ error = PTR_ERR(sfid);
+ goto error;
+ }
+ }
+
/* Validate actions. */
if (a[OVS_FLOW_ATTR_ACTIONS]) {
acts = get_flow_actions(a[OVS_FLOW_ATTR_ACTIONS], &key, &mask,
log);
if (IS_ERR(acts)) {
error = PTR_ERR(acts);
- goto error;
+ goto err_kfree_id;
}

/* Can allocate before locking if have acts. */
- reply = ovs_flow_cmd_alloc_info(acts, info, false);
+ reply = ovs_flow_cmd_alloc_info(acts, sfid, info, false,
+ ufid_flags);
if (IS_ERR(reply)) {
error = PTR_ERR(reply);
goto err_kfree_acts;
@@ -1026,7 +1110,10 @@ static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
goto err_unlock_ovs;
}
/* Check that the flow exists. */
- flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (sfid)
+ flow = ovs_flow_tbl_lookup_ufid(&dp->table, sfid);
+ else
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
if (unlikely(!flow)) {
error = -ENOENT;
goto err_unlock_ovs;
@@ -1042,13 +1129,16 @@ static int ovs_flow_cmd_set(struct sk_buff *skb, struct genl_info *info)
ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
- OVS_FLOW_CMD_NEW);
+ OVS_FLOW_CMD_NEW,
+ ufid_flags);
BUG_ON(error < 0);
}
} else {
/* Could not alloc without acts before locking. */
reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
- info, OVS_FLOW_CMD_NEW, false);
+ info, OVS_FLOW_CMD_NEW, false,
+ ufid_flags);
+
if (unlikely(IS_ERR(reply))) {
error = PTR_ERR(reply);
goto err_unlock_ovs;
@@ -1072,6 +1162,8 @@ err_unlock_ovs:
kfree_skb(reply);
err_kfree_acts:
kfree(acts);
+err_kfree_id:
+ kfree(sfid);
error:
return error;
}
@@ -1085,17 +1177,22 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info)
struct sw_flow *flow;
struct datapath *dp;
struct sw_flow_match match;
- int err;
+ struct sw_flow_id ufid;
+ u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);
+ int err = 0;
bool log = !a[OVS_FLOW_ATTR_PROBE];
+ bool ufid_present;

- if (!a[OVS_FLOW_ATTR_KEY]) {
+ ufid_present = ovs_nla_get_ufid(a[OVS_FLOW_ATTR_UFID], &ufid, log);
+ if (a[OVS_FLOW_ATTR_KEY]) {
+ ovs_match_init(&match, &key, NULL);
+ err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL,
+ log);
+ } else if (!ufid_present) {
OVS_NLERR(log,
"Flow get message rejected, Key attribute missing.");
- return -EINVAL;
+ err = -EINVAL;
}
-
- ovs_match_init(&match, &key, NULL);
- err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL, log);
if (err)
return err;

@@ -1106,14 +1203,17 @@ static int ovs_flow_cmd_get(struct sk_buff *skb, struct genl_info *info)
goto unlock;
}

- flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (ufid_present)
+ flow = ovs_flow_tbl_lookup_ufid(&dp->table, &ufid);
+ else
+ 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,
- OVS_FLOW_CMD_NEW, true);
+ OVS_FLOW_CMD_NEW, true, ufid_flags);
if (IS_ERR(reply)) {
err = PTR_ERR(reply);
goto unlock;
@@ -1132,13 +1232,17 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
struct ovs_header *ovs_header = info->userhdr;
struct sw_flow_key key;
struct sk_buff *reply;
- struct sw_flow *flow;
+ struct sw_flow *flow = NULL;
struct datapath *dp;
struct sw_flow_match match;
+ struct sw_flow_id ufid;
+ u32 ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);
int err;
bool log = !a[OVS_FLOW_ATTR_PROBE];
+ bool ufid_present;

- if (likely(a[OVS_FLOW_ATTR_KEY])) {
+ ufid_present = ovs_nla_get_ufid(a[OVS_FLOW_ATTR_UFID], &ufid, log);
+ if (a[OVS_FLOW_ATTR_KEY]) {
ovs_match_init(&match, &key, NULL);
err = ovs_nla_get_match(&match, a[OVS_FLOW_ATTR_KEY], NULL,
log);
@@ -1153,12 +1257,15 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
goto unlock;
}

- if (unlikely(!a[OVS_FLOW_ATTR_KEY])) {
+ if (unlikely(!a[OVS_FLOW_ATTR_KEY] && !ufid_present)) {
err = ovs_flow_tbl_flush(&dp->table);
goto unlock;
}

- flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
+ if (ufid_present)
+ flow = ovs_flow_tbl_lookup_ufid(&dp->table, &ufid);
+ else
+ flow = ovs_flow_tbl_lookup_exact(&dp->table, &match);
if (unlikely(!flow)) {
err = -ENOENT;
goto unlock;
@@ -1168,14 +1275,15 @@ static int ovs_flow_cmd_del(struct sk_buff *skb, struct genl_info *info)
ovs_unlock();

reply = ovs_flow_cmd_alloc_info((const struct sw_flow_actions __force *) flow->sf_acts,
- info, false);
+ flow->id, info, false, ufid_flags);
if (likely(reply)) {
if (likely(!IS_ERR(reply))) {
rcu_read_lock(); /*To keep RCU checker happy. */
err = ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex,
reply, info->snd_portid,
info->snd_seq, 0,
- OVS_FLOW_CMD_DEL);
+ OVS_FLOW_CMD_DEL,
+ ufid_flags);
rcu_read_unlock();
BUG_ON(err < 0);

@@ -1194,9 +1302,18 @@ unlock:

static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
+ struct nlattr *a[__OVS_FLOW_ATTR_MAX];
struct ovs_header *ovs_header = genlmsg_data(nlmsg_data(cb->nlh));
struct table_instance *ti;
struct datapath *dp;
+ u32 ufid_flags;
+ int err;
+
+ err = genlmsg_parse(cb->nlh, &dp_flow_genl_family, a,
+ OVS_FLOW_ATTR_MAX, flow_policy);
+ if (err)
+ return err;
+ ufid_flags = ovs_nla_get_ufid_flags(a[OVS_FLOW_ATTR_UFID_FLAGS]);

rcu_read_lock();
dp = get_dp_rcu(sock_net(skb->sk), ovs_header->dp_ifindex);
@@ -1219,7 +1336,7 @@ static int ovs_flow_cmd_dump(struct sk_buff *skb, struct netlink_callback *cb)
if (ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex, skb,
NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, NLM_F_MULTI,
- OVS_FLOW_CMD_NEW) < 0)
+ OVS_FLOW_CMD_NEW, ufid_flags) < 0)
break;

cb->args[0] = bucket;
@@ -1235,6 +1352,8 @@ static const struct nla_policy flow_policy[OVS_FLOW_ATTR_MAX + 1] = {
[OVS_FLOW_ATTR_ACTIONS] = { .type = NLA_NESTED },
[OVS_FLOW_ATTR_CLEAR] = { .type = NLA_FLAG },
[OVS_FLOW_ATTR_PROBE] = { .type = NLA_FLAG },
+ [OVS_FLOW_ATTR_UFID] = { .type = NLA_UNSPEC, .len = 1 },
+ [OVS_FLOW_ATTR_UFID_FLAGS] = { .type = NLA_U32 },
};

static const struct genl_ops dp_flow_genl_ops[] = {
diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
index d3d0a40..84f259d 100644
--- a/net/openvswitch/flow.h
+++ b/net/openvswitch/flow.h
@@ -197,6 +197,16 @@ struct sw_flow_match {
struct sw_flow_mask *mask;
};

+#define MAX_UFID_LENGTH 16 /* 128 bits */
+
+struct sw_flow_id {
+ u32 ufid_len;
+ union {
+ u32 ufid[MAX_UFID_LENGTH / 4];
+ struct sw_flow_key flow_key;
+ };
+};
+
struct sw_flow_actions {
struct rcu_head rcu;
u32 actions_len;
@@ -213,13 +223,15 @@ struct flow_stats {

struct sw_flow {
struct rcu_head rcu;
- struct hlist_node hash_node[2];
- u32 hash;
+ struct {
+ struct hlist_node node[2];
+ u32 hash;
+ } flow_table, ufid_table;
int stats_last_writer; /* NUMA-node id of the last writer on
* 'stats[0]'.
*/
struct sw_flow_key key;
- struct sw_flow_key unmasked_key;
+ struct sw_flow_id *id;
struct sw_flow_mask *mask;
struct sw_flow_actions __rcu *sf_acts;
struct flow_stats __rcu *stats[]; /* One for each NUMA node. First one
@@ -243,6 +255,16 @@ struct arp_eth_header {
unsigned char ar_tip[4]; /* target IP address */
} __packed;

+static inline bool ovs_identifier_is_ufid(const struct sw_flow_id *sfid)
+{
+ return sfid->ufid_len;
+}
+
+static inline bool ovs_identifier_is_key(const struct sw_flow_id *sfid)
+{
+ return !ovs_identifier_is_ufid(sfid);
+}
+
void ovs_flow_stats_update(struct sw_flow *, __be16 tcp_flags,
const struct sk_buff *);
void ovs_flow_stats_get(const struct sw_flow *, struct ovs_flow_stats *,
diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c
index 398f110..cdba8eb 100644
--- a/net/openvswitch/flow_netlink.c
+++ b/net/openvswitch/flow_netlink.c
@@ -1180,6 +1180,58 @@ free_newmask:
return err;
}

+static size_t get_ufid_len(const struct nlattr *attr, bool log)
+{
+ size_t len;
+
+ if (!attr)
+ return 0;
+
+ len = nla_len(attr);
+ if (len < 1 || len > MAX_UFID_LENGTH) {
+ OVS_NLERR(log, "Flow ufid size %u bytes is outside the range "
+ "(1, %d)", nla_len(attr), MAX_UFID_LENGTH);
+ return 0;
+ }
+
+ return len;
+}
+
+/* Initializes 'flow->ufid', returning true if 'attr' contains a valid UFID,
+ * or false otherwise. */
+bool ovs_nla_get_ufid(const struct nlattr *attr, struct sw_flow_id *sfid,
+ bool log)
+{
+ sfid->ufid_len = get_ufid_len(attr, log);
+ if (sfid->ufid_len)
+ memcpy(sfid->ufid, nla_data(attr), sfid->ufid_len);
+
+ return sfid->ufid_len;
+}
+
+struct sw_flow_id *ovs_nla_copy_identifier(const struct nlattr *ufid,
+ const struct sw_flow_key *key,
+ bool log)
+{
+ struct sw_flow_id *sfid;
+
+ sfid = kmalloc(sizeof(*sfid), GFP_KERNEL);
+ if (!sfid)
+ return ERR_PTR(-ENOMEM);
+
+ /* If UFID was not provided, use unmasked key. */
+ if (!ovs_nla_get_ufid(ufid, sfid, log))
+ memcpy(&sfid->flow_key, key, sizeof(*key));
+
+ return sfid;
+}
+
+
+u32 ovs_nla_get_ufid_flags(const struct nlattr *attr)
+{
+ return attr ? nla_get_u32(attr) : 0;
+}
+
/**
* ovs_nla_get_flow_metadata - parses Netlink attributes into a flow key.
* @key: Receives extracted in_port, priority, tun_key and skb_mark.
@@ -1450,9 +1502,20 @@ int ovs_nla_put_key(const struct sw_flow_key *swkey,
}

/* Called with ovs_mutex or RCU read lock. */
-int ovs_nla_put_unmasked_key(const struct sw_flow *flow, struct sk_buff *skb)
+int ovs_nla_put_identifier(const struct sw_flow *flow, struct sk_buff *skb)
+{
+ if (ovs_identifier_is_ufid(flow->id))
+ return nla_put(skb, OVS_FLOW_ATTR_UFID, flow->id->ufid_len,
+ flow->id->ufid);
+
+ return ovs_nla_put_key(&flow->id->flow_key, &flow->id->flow_key,
+ OVS_FLOW_ATTR_KEY, false, skb);
+}
+
+/* Called with ovs_mutex or RCU read lock. */
+int ovs_nla_put_masked_key(const struct sw_flow *flow, struct sk_buff *skb)
{
- return ovs_nla_put_key(&flow->unmasked_key, &flow->unmasked_key,
+ return ovs_nla_put_key(&flow->mask->key, &flow->key,
OVS_FLOW_ATTR_KEY, false, skb);
}

diff --git a/net/openvswitch/flow_netlink.h b/net/openvswitch/flow_netlink.h
index 9ed09e6..07fc9b6 100644
--- a/net/openvswitch/flow_netlink.h
+++ b/net/openvswitch/flow_netlink.h
@@ -48,7 +48,8 @@ int ovs_nla_put_key(const struct sw_flow_key *, const struct sw_flow_key *,
int ovs_nla_get_flow_metadata(const struct nlattr *, struct sw_flow_key *,
bool log);

-int ovs_nla_put_unmasked_key(const struct sw_flow *flow, struct sk_buff *skb);
+int ovs_nla_put_identifier(const struct sw_flow *flow, struct sk_buff *skb);
+int ovs_nla_put_masked_key(const struct sw_flow *flow, struct sk_buff *skb);
int ovs_nla_put_mask(const struct sw_flow *flow, struct sk_buff *skb);

int ovs_nla_get_match(struct sw_flow_match *, const struct nlattr *key,
@@ -56,6 +57,12 @@ int ovs_nla_get_match(struct sw_flow_match *, const struct nlattr *key,
int ovs_nla_put_egress_tunnel_key(struct sk_buff *,
const struct ovs_tunnel_info *);

+bool ovs_nla_get_ufid(const struct nlattr *, struct sw_flow_id *, bool log);
+struct sw_flow_id *ovs_nla_copy_identifier(const struct nlattr *ufid,
+ const struct sw_flow_key *key,
+ bool log);
+u32 ovs_nla_get_ufid_flags(const struct nlattr *attr);
+
int ovs_nla_copy_actions(const struct nlattr *attr,
const struct sw_flow_key *key,
struct sw_flow_actions **sfa, bool log);
diff --git a/net/openvswitch/flow_table.c b/net/openvswitch/flow_table.c
index 9a3f41f..623bbe0 100644
--- a/net/openvswitch/flow_table.c
+++ b/net/openvswitch/flow_table.c
@@ -85,6 +85,7 @@ struct sw_flow *ovs_flow_alloc(void)

flow->sf_acts = NULL;
flow->mask = NULL;
+ flow->id = NULL;
flow->stats_last_writer = NUMA_NO_NODE;

/* Initialize the default stat node. */
@@ -139,6 +140,7 @@ static void flow_free(struct sw_flow *flow)
{
int node;

+ kfree(flow->id);
kfree((struct sw_flow_actions __force *)flow->sf_acts);
for_each_node(node)
if (flow->stats[node])
@@ -200,18 +202,28 @@ static struct table_instance *table_instance_alloc(int new_size)

int ovs_flow_tbl_init(struct flow_table *table)
{
- struct table_instance *ti;
+ struct table_instance *ti, *ufid_ti;

ti = table_instance_alloc(TBL_MIN_BUCKETS);

if (!ti)
return -ENOMEM;

+ ufid_ti = table_instance_alloc(TBL_MIN_BUCKETS);
+ if (!ufid_ti)
+ goto free_ti;
+
rcu_assign_pointer(table->ti, ti);
+ rcu_assign_pointer(table->ufid_ti, ufid_ti);
INIT_LIST_HEAD(&table->mask_list);
table->last_rehash = jiffies;
table->count = 0;
+ table->ufid_count = 0;
return 0;
+
+free_ti:
+ __table_instance_destroy(ti);
+ return -ENOMEM;
}

static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu)
@@ -221,13 +233,16 @@ static void flow_tbl_destroy_rcu_cb(struct rcu_head *rcu)
__table_instance_destroy(ti);
}

-static void table_instance_destroy(struct table_instance *ti, bool deferred)
+static void table_instance_destroy(struct table_instance *ti,
+ struct table_instance *ufid_ti,
+ bool deferred)
{
int i;

if (!ti)
return;

+ BUG_ON(!ufid_ti);
if (ti->keep_flows)
goto skip_flows;

@@ -236,18 +251,24 @@ static void table_instance_destroy(struct table_instance *ti, bool deferred)
struct hlist_head *head = flex_array_get(ti->buckets, i);
struct hlist_node *n;
int ver = ti->node_ver;
+ int ufid_ver = ufid_ti->node_ver;

- hlist_for_each_entry_safe(flow, n, head, hash_node[ver]) {
- hlist_del_rcu(&flow->hash_node[ver]);
+ hlist_for_each_entry_safe(flow, n, head, flow_table.node[ver]) {
+ hlist_del_rcu(&flow->flow_table.node[ver]);
+ if (ovs_identifier_is_ufid(flow->id))
+ hlist_del_rcu(&flow->ufid_table.node[ufid_ver]);
ovs_flow_free(flow, deferred);
}
}

skip_flows:
- if (deferred)
+ if (deferred) {
call_rcu(&ti->rcu, flow_tbl_destroy_rcu_cb);
- else
+ call_rcu(&ufid_ti->rcu, flow_tbl_destroy_rcu_cb);
+ } else {
__table_instance_destroy(ti);
+ __table_instance_destroy(ufid_ti);
+ }
}

/* No need for locking this function is called from RCU callback or
@@ -256,8 +277,9 @@ skip_flows:
void ovs_flow_tbl_destroy(struct flow_table *table)
{
struct table_instance *ti = rcu_dereference_raw(table->ti);
+ struct table_instance *ufid_ti = rcu_dereference_raw(table->ufid_ti);

- table_instance_destroy(ti, false);
+ table_instance_destroy(ti, ufid_ti, false);
}

struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti,
@@ -272,7 +294,7 @@ struct sw_flow *ovs_flow_tbl_dump_next(struct table_instance *ti,
while (*bucket < ti->n_buckets) {
i = 0;
head = flex_array_get(ti->buckets, *bucket);
- hlist_for_each_entry_rcu(flow, head, hash_node[ver]) {
+ hlist_for_each_entry_rcu(flow, head, flow_table.node[ver]) {
if (i < *last) {
i++;
continue;
@@ -294,16 +316,26 @@ static struct hlist_head *find_bucket(struct table_instance *ti, u32 hash)
(hash & (ti->n_buckets - 1)));
}

-static void table_instance_insert(struct table_instance *ti, struct sw_flow *flow)
+static void table_instance_insert(struct table_instance *ti,
+ struct sw_flow *flow)
{
struct hlist_head *head;

- head = find_bucket(ti, flow->hash);
- hlist_add_head_rcu(&flow->hash_node[ti->node_ver], head);
+ head = find_bucket(ti, flow->flow_table.hash);
+ hlist_add_head_rcu(&flow->flow_table.node[ti->node_ver], head);
+}
+
+static void ufid_table_instance_insert(struct table_instance *ti,
+ struct sw_flow *flow)
+{
+ struct hlist_head *head;
+
+ head = find_bucket(ti, flow->ufid_table.hash);
+ hlist_add_head_rcu(&flow->ufid_table.node[ti->node_ver], head);
}

static void flow_table_copy_flows(struct table_instance *old,
- struct table_instance *new)
+ struct table_instance *new, bool ufid)
{
int old_ver;
int i;
@@ -318,15 +350,21 @@ static void flow_table_copy_flows(struct table_instance *old,

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

- hlist_for_each_entry(flow, head, hash_node[old_ver])
- table_instance_insert(new, flow);
+ if (ufid)
+ hlist_for_each_entry(flow, head,
+ ufid_table.node[old_ver])
+ ufid_table_instance_insert(new, flow);
+ else
+ hlist_for_each_entry(flow, head,
+ flow_table.node[old_ver])
+ table_instance_insert(new, flow);
}

old->keep_flows = true;
}

static struct table_instance *table_instance_rehash(struct table_instance *ti,
- int n_buckets)
+ int n_buckets, bool ufid)
{
struct table_instance *new_ti;

@@ -334,27 +372,38 @@ static struct table_instance *table_instance_rehash(struct table_instance *ti,
if (!new_ti)
return NULL;

- flow_table_copy_flows(ti, new_ti);
+ flow_table_copy_flows(ti, new_ti, ufid);

return new_ti;
}

int ovs_flow_tbl_flush(struct flow_table *flow_table)
{
- struct table_instance *old_ti;
- struct table_instance *new_ti;
+ struct table_instance *old_ti, *new_ti;
+ struct table_instance *old_ufid_ti, *new_ufid_ti;

- old_ti = ovsl_dereference(flow_table->ti);
new_ti = table_instance_alloc(TBL_MIN_BUCKETS);
if (!new_ti)
return -ENOMEM;
+ new_ufid_ti = table_instance_alloc(TBL_MIN_BUCKETS);
+ if (!new_ufid_ti)
+ goto err_free_ti;
+
+ old_ti = ovsl_dereference(flow_table->ti);
+ old_ufid_ti = ovsl_dereference(flow_table->ufid_ti);

rcu_assign_pointer(flow_table->ti, new_ti);
+ rcu_assign_pointer(flow_table->ufid_ti, new_ufid_ti);
flow_table->last_rehash = jiffies;
flow_table->count = 0;
+ flow_table->ufid_count = 0;

- table_instance_destroy(old_ti, true);
+ table_instance_destroy(old_ti, old_ufid_ti, true);
return 0;
+
+err_free_ti:
+ __table_instance_destroy(new_ti);
+ return -ENOMEM;
}

static u32 flow_hash(const struct sw_flow_key *key,
@@ -402,14 +451,15 @@ static bool flow_cmp_masked_key(const struct sw_flow *flow,
return cmp_key(&flow->key, key, range->start, range->end);
}

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

- return cmp_key(&flow->unmasked_key, key, key_start, key_end);
+ BUG_ON(ovs_identifier_is_ufid(flow->id));
+ return cmp_key(&flow->id->flow_key, key, key_start, key_end);
}

static struct sw_flow *masked_flow_lookup(struct table_instance *ti,
@@ -424,8 +474,8 @@ static struct sw_flow *masked_flow_lookup(struct table_instance *ti,
ovs_flow_mask_key(&masked_key, unmasked, mask);
hash = flow_hash(&masked_key, &mask->range);
head = find_bucket(ti, hash);
- hlist_for_each_entry_rcu(flow, head, hash_node[ti->node_ver]) {
- if (flow->mask == mask && flow->hash == hash &&
+ hlist_for_each_entry_rcu(flow, head, flow_table.node[ti->node_ver]) {
+ if (flow->mask == mask && flow->flow_table.hash == hash &&
flow_cmp_masked_key(flow, &masked_key, &mask->range))
return flow;
}
@@ -468,7 +518,48 @@ struct sw_flow *ovs_flow_tbl_lookup_exact(struct flow_table *tbl,
/* Always called under ovs-mutex. */
list_for_each_entry(mask, &tbl->mask_list, list) {
flow = masked_flow_lookup(ti, match->key, mask);
- if (flow && ovs_flow_cmp_unmasked_key(flow, match)) /* Found */
+ if (flow && ovs_identifier_is_key(flow->id) &&
+ ovs_flow_cmp_unmasked_key(flow, match))
+ return flow;
+ }
+ return NULL;
+}
+
+static u32 ufid_hash(const struct sw_flow_id *sfid)
+{
+ return jhash(sfid->ufid, sfid->ufid_len, 0);
+}
+
+static bool ovs_flow_cmp_ufid(const struct sw_flow *flow,
+ const struct sw_flow_id *sfid)
+{
+ if (flow->id->ufid_len != sfid->ufid_len)
+ return false;
+
+ return !memcmp(flow->id->ufid, sfid->ufid, sfid->ufid_len);
+}
+
+bool ovs_flow_cmp(const struct sw_flow *flow, const struct sw_flow_match *match)
+{
+ if (ovs_identifier_is_ufid(flow->id))
+ return flow_cmp_masked_key(flow, match->key, &match->range);
+
+ return ovs_flow_cmp_unmasked_key(flow, match);
+}
+
+struct sw_flow *ovs_flow_tbl_lookup_ufid(struct flow_table *tbl,
+ const struct sw_flow_id *ufid)
+{
+ struct table_instance *ti = rcu_dereference_ovsl(tbl->ufid_ti);
+ struct sw_flow *flow;
+ struct hlist_head *head;
+ u32 hash;
+
+ hash = ufid_hash(ufid);
+ head = find_bucket(ti, hash);
+ hlist_for_each_entry_rcu(flow, head, ufid_table.node[ti->node_ver]) {
+ if (flow->ufid_table.hash == hash &&
+ ovs_flow_cmp_ufid(flow, ufid))
return flow;
}
return NULL;
@@ -485,9 +576,10 @@ int ovs_flow_tbl_num_masks(const struct flow_table *table)
return num;
}

-static struct table_instance *table_instance_expand(struct table_instance *ti)
+static struct table_instance *table_instance_expand(struct table_instance *ti,
+ bool ufid)
{
- return table_instance_rehash(ti, ti->n_buckets * 2);
+ return table_instance_rehash(ti, ti->n_buckets * 2, ufid);
}

/* Remove 'mask' from the mask list, if it is not needed any more. */
@@ -512,10 +604,15 @@ static void flow_mask_remove(struct flow_table *tbl, struct sw_flow_mask *mask)
void ovs_flow_tbl_remove(struct flow_table *table, struct sw_flow *flow)
{
struct table_instance *ti = ovsl_dereference(table->ti);
+ struct table_instance *ufid_ti = ovsl_dereference(table->ufid_ti);

BUG_ON(table->count == 0);
- hlist_del_rcu(&flow->hash_node[ti->node_ver]);
+ hlist_del_rcu(&flow->flow_table.node[ti->node_ver]);
table->count--;
+ if (ovs_identifier_is_ufid(flow->id)) {
+ hlist_del_rcu(&flow->ufid_table.node[ufid_ti->node_ver]);
+ table->ufid_count--;
+ }

/* RCU delete the mask. 'flow->mask' is not NULLed, as it should be
* accessible as long as the RCU read lock is held.
@@ -589,25 +686,47 @@ static void flow_key_insert(struct flow_table *table, struct sw_flow *flow)
struct table_instance *new_ti = NULL;
struct table_instance *ti;

- flow->hash = flow_hash(&flow->key, &flow->mask->range);
+ flow->flow_table.hash = flow_hash(&flow->key, &flow->mask->range);
ti = ovsl_dereference(table->ti);
table_instance_insert(ti, flow);
table->count++;

/* Expand table, if necessary, to make room. */
if (table->count > ti->n_buckets)
- new_ti = table_instance_expand(ti);
+ new_ti = table_instance_expand(ti, false);
else if (time_after(jiffies, table->last_rehash + REHASH_INTERVAL))
- new_ti = table_instance_rehash(ti, ti->n_buckets);
+ new_ti = table_instance_rehash(ti, ti->n_buckets, false);

if (new_ti) {
rcu_assign_pointer(table->ti, new_ti);
- table_instance_destroy(ti, true);
+ call_rcu(&ti->rcu, flow_tbl_destroy_rcu_cb);
table->last_rehash = jiffies;
}
}

/* Must be called with OVS mutex held. */
+static void flow_ufid_insert(struct flow_table *table, struct sw_flow *flow)
+{
+ struct table_instance *ti;
+
+ flow->ufid_table.hash = ufid_hash(flow->id);
+ ti = ovsl_dereference(table->ufid_ti);
+ ufid_table_instance_insert(ti, flow);
+ table->ufid_count++;
+
+ /* Expand table, if necessary, to make room. */
+ if (table->ufid_count > ti->n_buckets) {
+ struct table_instance *new_ti;
+
+ new_ti = table_instance_expand(ti, true);
+ if (new_ti) {
+ rcu_assign_pointer(table->ufid_ti, new_ti);
+ call_rcu(&ti->rcu, flow_tbl_destroy_rcu_cb);
+ }
+ }
+}
+
+/* Must be called with OVS mutex held. */
int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow,
const struct sw_flow_mask *mask)
{
@@ -617,6 +736,8 @@ int ovs_flow_tbl_insert(struct flow_table *table, struct sw_flow *flow,
if (err)
return err;
flow_key_insert(table, flow);
+ if (ovs_identifier_is_ufid(flow->id))
+ flow_ufid_insert(table, flow);

return 0;
}
diff --git a/net/openvswitch/flow_table.h b/net/openvswitch/flow_table.h
index 309fa64..616eda1 100644
--- a/net/openvswitch/flow_table.h
+++ b/net/openvswitch/flow_table.h
@@ -47,9 +47,11 @@ struct table_instance {

struct flow_table {
struct table_instance __rcu *ti;
+ struct table_instance __rcu *ufid_ti;
struct list_head mask_list;
unsigned long last_rehash;
unsigned int count;
+ unsigned int ufid_count;
};

extern struct kmem_cache *flow_stats_cache;
@@ -78,8 +80,10 @@ struct sw_flow *ovs_flow_tbl_lookup(struct flow_table *,
const struct sw_flow_key *);
struct sw_flow *ovs_flow_tbl_lookup_exact(struct flow_table *tbl,
const struct sw_flow_match *match);
-bool ovs_flow_cmp_unmasked_key(const struct sw_flow *flow,
- const struct sw_flow_match *match);
+struct sw_flow *ovs_flow_tbl_lookup_ufid(struct flow_table *,
+ const struct sw_flow_id *);
+
+bool ovs_flow_cmp(const struct sw_flow *, const struct sw_flow_match *);

void ovs_flow_mask_key(struct sw_flow_key *dst, const struct sw_flow_key *src,
const struct sw_flow_mask *mask);
--
1.7.10.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/