[INET]: speedup inet (tcp/dccp) lookups
Arnaldo and I agreed it could be applied now, because I have other
pending patches depending on this one (Thank you Arnaldo)
(The other important patch moves skc_refcnt in a separate cache line,
so that the SMP/NUMA performance doesnt suffer from cache line ping pongs)
1) First some performance data :
--------------------------------
tcp_v4_rcv() wastes a *lot* of time in __inet_lookup_established()
The most time critical code is :
sk_for_each(sk, node, &head->chain) {
if (INET_MATCH(sk, acookie, saddr, daddr, ports, dif))
goto hit; /* You sunk my battleship! */
}
The sk_for_each() does use prefetch() hints but only the begining of
"struct sock" is prefetched.
As INET_MATCH first comparison uses inet_sk(__sk)->daddr, wich is far
away from the begining of "struct sock", it has to bring into CPU
cache cold cache line. Each iteration has to use at least 2 cache
lines.
This can be problematic if some chains are very long.
2) The goal
-----------
The idea I had is to change things so that INET_MATCH() may return
FALSE in 99% of cases only using the data already in the CPU cache,
using one cache line per iteration.
3) Description of the patch
---------------------------
Adds a new 'unsigned int skc_hash' field in 'struct sock_common',
filling a 32 bits hole on 64 bits platform.
struct sock_common {
unsigned short skc_family;
volatile unsigned char skc_state;
unsigned char skc_reuse;
int skc_bound_dev_if;
struct hlist_node skc_node;
struct hlist_node skc_bind_node;
atomic_t skc_refcnt;
+ unsigned int skc_hash;
struct proto *skc_prot;
};
Store in this 32 bits field the full hash, not masked by (ehash_size -
1) Using this full hash as the first comparison done in INET_MATCH
permits us immediatly skip the element without touching a second cache
line in case of a miss.
Suppress the sk_hashent/tw_hashent fields since skc_hash (aliased to
sk_hash and tw_hash) already contains the slot number if we mask with
(ehash_size - 1)
File include/net/inet_hashtables.h
64 bits platforms :
#define INET_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash))
((*((__u64 *)&(inet_sk(__sk)->daddr)))== (__cookie)) && \
((*((__u32 *)&(inet_sk(__sk)->dport))) == (__ports)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
32bits platforms:
#define TCP_IPV4_MATCH(__sk, __hash, __cookie, __saddr, __daddr, __ports, __dif)\
(((__sk)->sk_hash == (__hash)) && \
(inet_sk(__sk)->daddr == (__saddr)) && \
(inet_sk(__sk)->rcv_saddr == (__daddr)) && \
(!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif))))
- Adds a prefetch(head->chain.first) in
__inet_lookup_established()/__tcp_v4_check_established() and
__inet6_lookup_established()/__tcp_v6_check_established() and
__dccp_v4_check_established() to bring into cache the first element of the
list, before the {read|write}_lock(&head->lock);
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/include/net/inet6_hashtables.h b/include/net/inet6_hashtables.h
index 03df3b1..5a2beed 100644
--- a/include/net/inet6_hashtables.h
+++ b/include/net/inet6_hashtables.h
@@ -26,19 +26,18 @@
struct inet_hashinfo;
/* I have no idea if this is a good hash for v6 or not. -DaveM */
-static inline int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport,
- const struct in6_addr *faddr, const u16 fport,
- const int ehash_size)
+static inline unsigned int inet6_ehashfn(const struct in6_addr *laddr, const u16 lport,
+ const struct in6_addr *faddr, const u16 fport)
{
- int hashent = (lport ^ fport);
+ unsigned int hashent = (lport ^ fport);
hashent ^= (laddr->s6_addr32[3] ^ faddr->s6_addr32[3]);
hashent ^= hashent >> 16;
hashent ^= hashent >> 8;
- return (hashent & (ehash_size - 1));
+ return hashent;
}
-static inline int inet6_sk_ehashfn(const struct sock *sk, const int ehash_size)
+static inline int inet6_sk_ehashfn(const struct sock *sk)
{
const struct inet_sock *inet = inet_sk(sk);
const struct ipv6_pinfo *np = inet6_sk(sk);
@@ -46,7 +45,7 @@
const struct in6_addr *faddr = &np->daddr;
const __u16 lport = inet->num;
const __u16 fport = inet->dport;
- return inet6_ehashfn(laddr, lport, faddr, fport, ehash_size);
+ return inet6_ehashfn(laddr, lport, faddr, fport);
}
/*
@@ -69,14 +68,14 @@
/* Optimize here for direct hit, only listening connections can
* have wildcards anyways.
*/
- const int hash = inet6_ehashfn(daddr, hnum, saddr, sport,
- hashinfo->ehash_size);
- struct inet_ehash_bucket *head = &hashinfo->ehash[hash];
+ unsigned int hash = inet6_ehashfn(daddr, hnum, saddr, sport);
+ struct inet_ehash_bucket *head = inet_ehash_bucket(hashinfo, hash);
+ prefetch(head->chain.first);
read_lock(&head->lock);
sk_for_each(sk, node, &head->chain) {
/* For IPV6 do the cheaper port and family tests first. */
- if (INET6_MATCH(sk, saddr, daddr, ports, dif))
+ if (INET6_MATCH(sk, hash, saddr, daddr, ports, dif))
goto hit; /* You sunk my battleship! */
}
/* Must check for a TIME_WAIT'er before going to listener hash. */