udp: introduce struct udp_table and multiple spinlocks
UDP sockets are hashed in a 128 slots hash table.
This hash table is protected by *one* rwlock.
This rwlock is readlocked each time an incoming UDP message is handled.
This rwlock is writelocked each time a socket must be inserted in
hash table (bind time), or deleted from this table (close time)
This is not scalable on SMP machines :
1) Even in read mode, lock() and unlock() are atomic operations and
must dirty a contended cache line, shared by all cpus.
2) A writer might be starved if many readers are 'in flight'. This can
happen on a machine with some NIC receiving many UDP messages. User
process can be delayed a long time at socket creation/dismantle time.
This patch prepares RCU migration, by introducing 'struct udp_table
and struct udp_hslot', and using one spinlock per chain, to reduce
contention on central rwlock.
Introducing one spinlock per chain reduces latencies, for port
randomization on heavily loaded UDP servers. This also speedup
bindings to specific ports.
udp_lib_unhash() was uninlined, becoming to big.
Some cleanups were done to ease review of following patch
(RCUification of UDP Unicast lookups)
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h
index 92dd7da..2377920 100644
--- a/net/ipv6/udp_impl.h
+++ b/net/ipv6/udp_impl.h
@@ -7,9 +7,9 @@
#include <net/inet_common.h>
#include <net/transp_v6.h>
-extern int __udp6_lib_rcv(struct sk_buff *, struct hlist_head [], int );
+extern int __udp6_lib_rcv(struct sk_buff *, struct udp_table *, int );
extern void __udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *,
- int , int , int , __be32 , struct hlist_head []);
+ int , int , int , __be32 , struct udp_table *);
extern int udp_v6_get_port(struct sock *sk, unsigned short snum);