Merge branch 'net-sched-conditional-notification-of-events-for-cls-and-act'

Pedro Tammela says:

====================
net/sched: conditional notification of events for cls and act

This is an optimization we have been leveraging on P4TC but we believe
it will benefit rtnl users in general.

It's common to allocate an skb, build a notification message and then
broadcast an event. In the absence of any user space listeners, these
resources (cpu and memory operations) are wasted. In cases where the subsystem
is lockless (such as in tc-flower) this waste is more prominent. For the
scenarios where the rtnl_lock is held it is not as prominent.

The idea is simple. Build and send the notification iif:
   - The user requests via NLM_F_ECHO or
   - Someone is listening to the rtnl group (tc mon)

On a simple test with tc-flower adding 1M entries, using just a single core,
there's already a noticeable difference in the cycles spent in tc_new_tfilter
with this patchset.

before:
   - 43.68% tc_new_tfilter
      + 31.73% fl_change
      + 6.35% tfilter_notify
      + 1.62% nlmsg_notify
        0.66% __tcf_qdisc_find.part.0
        0.64% __tcf_chain_get
        0.54% fl_get
      + 0.53% tcf_proto_lookup_ops

after:
   - 39.20% tc_new_tfilter
      + 34.58% fl_change
        0.69% __tcf_qdisc_find.part.0
        0.67% __tcf_chain_get
      + 0.61% tcf_proto_lookup_ops

Note, the above test is using iproute2:tc which execs a shell.
We expect people using netlink directly to observe even greater
reductions.

The qdisc side needs some refactoring of the notification routines to fit in
this new model, so they will be sent in a later patchset.
====================

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

1 files changed, 29 insertions, 0 deletions

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 3d6cf306cd55..6a8543b34e2c 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -10,6 +10,13 @@
 #include <uapi/linux/rtnetlink.h>
 
 extern int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, u32 group, int echo);
+
+static inline int rtnetlink_maybe_send(struct sk_buff *skb, struct net *net,
+				       u32 pid, u32 group, int echo)
+{
+	return !skb ? 0 : rtnetlink_send(skb, net, pid, group, echo);
+}
+
 extern int rtnl_unicast(struct sk_buff *skb, struct net *net, u32 pid);
 extern void rtnl_notify(struct sk_buff *skb, struct net *net, u32 pid,
 			u32 group, const struct nlmsghdr *nlh, gfp_t flags);
@@ -130,4 +137,26 @@ extern int ndo_dflt_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq,
 
 extern void rtnl_offload_xstats_notify(struct net_device *dev);
 
+static inline int rtnl_has_listeners(const struct net *net, u32 group)
+{
+	struct sock *rtnl = net->rtnl;
+
+	return netlink_has_listeners(rtnl, group);
+}
+
+/**
+ * rtnl_notify_needed - check if notification is needed
+ * @net: Pointer to the net namespace
+ * @nlflags: netlink ingress message flags
+ * @group: rtnl group
+ *
+ * Based on the ingress message flags and rtnl group, returns true
+ * if a notification is needed, false otherwise.
+ */
+static inline bool
+rtnl_notify_needed(const struct net *net, u16 nlflags, u32 group)
+{
+	return (nlflags & NLM_F_ECHO) || rtnl_has_listeners(net, group);
+}
+
 #endif	/* __LINUX_RTNETLINK_H */