openvswitch: Reduce locking requirements.
Reduce and clarify locking requirements for ovs_flow_cmd_alloc_info(),
ovs_flow_cmd_fill_info() and ovs_flow_cmd_build_info().
A datapath pointer is available only when holding a lock. Change
ovs_flow_cmd_fill_info() and ovs_flow_cmd_build_info() to take a
dp_ifindex directly, rather than a datapath pointer that is then
(only) used to get the dp_ifindex. This is useful, since the
dp_ifindex is available even when the datapath pointer is not, both
before and after taking a lock, which makes further critical section
reduction possible.
Make ovs_flow_cmd_alloc_info() take an 'acts' argument instead a
'flow' pointer. This allows some future patches to do the allocation
before acquiring the flow pointer.
The locking requirements after this patch are:
ovs_flow_cmd_alloc_info(): May be called without locking, must not be
called while holding the RCU read lock (due to memory allocation).
If 'acts' belong to a flow in the flow table, however, then the
caller must hold ovs_mutex.
ovs_flow_cmd_fill_info(): Either ovs_mutex or RCU read lock must be held.
ovs_flow_cmd_build_info(): This calls both of the above, so the caller
must hold ovs_mutex.
Signed-off-by: Jarno Rajahalme <jrajahalme@nicira.com>
Signed-off-by: Pravin B Shelar <pshelar@nicira.com>
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index f3bcdac..949fc7f 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -663,7 +663,7 @@
}
/* Called with ovs_mutex or RCU read lock. */
-static int ovs_flow_cmd_fill_info(struct sw_flow *flow, struct datapath *dp,
+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)
{
@@ -680,7 +680,7 @@
if (!ovs_header)
return -EMSGSIZE;
- ovs_header->dp_ifindex = get_dpifindex(dp);
+ ovs_header->dp_ifindex = dp_ifindex;
/* Fill flow key. */
nla = nla_nest_start(skb, OVS_FLOW_ATTR_KEY);
@@ -703,6 +703,7 @@
nla_nest_end(skb, nla);
ovs_flow_stats_get(flow, &stats, &used, &tcp_flags);
+
if (used &&
nla_put_u64(skb, OVS_FLOW_ATTR_USED, ovs_flow_used_time(used)))
goto nla_put_failure;
@@ -730,9 +731,9 @@
const struct sw_flow_actions *sf_acts;
sf_acts = rcu_dereference_ovsl(flow->sf_acts);
-
err = ovs_nla_put_actions(sf_acts->actions,
sf_acts->actions_len, skb);
+
if (!err)
nla_nest_end(skb, start);
else {
@@ -753,41 +754,40 @@
return err;
}
-/* Must be called with ovs_mutex. */
-static struct sk_buff *ovs_flow_cmd_alloc_info(struct sw_flow *flow,
+/* May not be called with RCU read lock. */
+static struct sk_buff *ovs_flow_cmd_alloc_info(const struct sw_flow_actions *acts,
struct genl_info *info,
bool always)
{
struct sk_buff *skb;
- size_t len;
if (!always && !ovs_must_notify(info, &ovs_dp_flow_multicast_group))
return NULL;
- len = ovs_flow_cmd_msg_size(ovsl_dereference(flow->sf_acts));
-
- skb = genlmsg_new_unicast(len, info, GFP_KERNEL);
+ skb = genlmsg_new_unicast(ovs_flow_cmd_msg_size(acts), info, GFP_KERNEL);
if (!skb)
return ERR_PTR(-ENOMEM);
return skb;
}
-/* Must be called with ovs_mutex. */
-static struct sk_buff *ovs_flow_cmd_build_info(struct sw_flow *flow,
- struct datapath *dp,
- struct genl_info *info,
- u8 cmd, bool always)
+/* Called with ovs_mutex. */
+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)
{
struct sk_buff *skb;
int retval;
- skb = ovs_flow_cmd_alloc_info(flow, info, always);
+ skb = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts), info,
+ always);
if (!skb || IS_ERR(skb))
return skb;
- retval = ovs_flow_cmd_fill_info(flow, dp, skb, info->snd_portid,
- info->snd_seq, 0, cmd);
+ retval = ovs_flow_cmd_fill_info(flow, dp_ifindex, skb,
+ info->snd_portid, info->snd_seq, 0,
+ cmd);
BUG_ON(retval < 0);
return skb;
}
@@ -868,8 +868,8 @@
goto err_flow_free;
}
- reply = ovs_flow_cmd_build_info(flow, dp, info,
- OVS_FLOW_CMD_NEW, false);
+ reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
+ info, OVS_FLOW_CMD_NEW, false);
} else {
/* We found a matching flow. */
/* Bail out if we're not allowed to modify an existing flow.
@@ -895,8 +895,9 @@
rcu_assign_pointer(flow->sf_acts, acts);
ovs_nla_free_flow_actions(old_acts);
}
- reply = ovs_flow_cmd_build_info(flow, dp, info,
- OVS_FLOW_CMD_NEW, false);
+
+ reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex,
+ info, OVS_FLOW_CMD_NEW, false);
/* Clear stats. */
if (a[OVS_FLOW_ATTR_CLEAR])
@@ -958,7 +959,8 @@
goto unlock;
}
- reply = ovs_flow_cmd_build_info(flow, dp, info, OVS_FLOW_CMD_NEW, true);
+ reply = ovs_flow_cmd_build_info(flow, ovs_header->dp_ifindex, info,
+ OVS_FLOW_CMD_NEW, true);
if (IS_ERR(reply)) {
err = PTR_ERR(reply);
goto unlock;
@@ -1005,7 +1007,8 @@
goto unlock;
}
- reply = ovs_flow_cmd_alloc_info(flow, info, false);
+ reply = ovs_flow_cmd_alloc_info(ovsl_dereference(flow->sf_acts), info,
+ false);
if (IS_ERR(reply)) {
err = PTR_ERR(reply);
goto unlock;
@@ -1014,7 +1017,8 @@
ovs_flow_tbl_remove(&dp->table, flow);
if (reply) {
- err = ovs_flow_cmd_fill_info(flow, dp, reply, info->snd_portid,
+ err = ovs_flow_cmd_fill_info(flow, ovs_header->dp_ifindex,
+ reply, info->snd_portid,
info->snd_seq, 0,
OVS_FLOW_CMD_DEL);
BUG_ON(err < 0);
@@ -1054,7 +1058,7 @@
if (!flow)
break;
- if (ovs_flow_cmd_fill_info(flow, dp, skb,
+ 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)
