[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/net/atm/common.c b/net/atm/common.c
index 801a581..63feea4 100644
--- a/net/atm/common.c
+++ b/net/atm/common.c
@@ -46,7 +46,7 @@
struct atm_vcc *vcc = atm_sk(sk);
struct hlist_head *head = &vcc_hash[vcc->vci &
(VCC_HTABLE_SIZE - 1)];
- sk->sk_hashent = vcc->vci & (VCC_HTABLE_SIZE - 1);
+ sk->sk_hash = vcc->vci & (VCC_HTABLE_SIZE - 1);
sk_add_node(sk, head);
}
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 40fe6af..ae088d1 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -62,27 +62,27 @@
const int dif = sk->sk_bound_dev_if;
INET_ADDR_COOKIE(acookie, saddr, daddr)
const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
- const int hash = inet_ehashfn(daddr, lport, saddr, inet->dport,
- dccp_hashinfo.ehash_size);
- struct inet_ehash_bucket *head = &dccp_hashinfo.ehash[hash];
+ unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport);
+ struct inet_ehash_bucket *head = inet_ehash_bucket(&dccp_hashinfo, hash);
const struct sock *sk2;
const struct hlist_node *node;
struct inet_timewait_sock *tw;
+ prefetch(head->chain.first);
write_lock(&head->lock);
/* Check TIME-WAIT sockets first. */
sk_for_each(sk2, node, &(head + dccp_hashinfo.ehash_size)->chain) {
tw = inet_twsk(sk2);
- if (INET_TW_MATCH(sk2, acookie, saddr, daddr, ports, dif))
+ if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
tw = NULL;
/* And established part... */
sk_for_each(sk2, node, &head->chain) {
- if (INET_MATCH(sk2, acookie, saddr, daddr, ports, dif))
+ if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
@@ -90,7 +90,7 @@
* in hash table socket with a funny identity. */
inet->num = lport;
inet->sport = htons(lport);
- sk->sk_hashent = hash;
+ sk->sk_hash = hash;
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
sock_prot_inc_use(sk->sk_prot);
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 4d1502a..f9076ef 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -20,7 +20,7 @@
struct inet_bind_hashbucket *bhead;
struct inet_bind_bucket *tb;
/* Unlink from established hashes. */
- struct inet_ehash_bucket *ehead = &hashinfo->ehash[tw->tw_hashent];
+ struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, tw->tw_hash);
write_lock(&ehead->lock);
if (hlist_unhashed(&tw->tw_node)) {
@@ -60,7 +60,7 @@
{
const struct inet_sock *inet = inet_sk(sk);
const struct inet_connection_sock *icsk = inet_csk(sk);
- struct inet_ehash_bucket *ehead = &hashinfo->ehash[sk->sk_hashent];
+ struct inet_ehash_bucket *ehead = inet_ehash_bucket(hashinfo, sk->sk_hash);
struct inet_bind_hashbucket *bhead;
/* Step 1: Put TW into bind hash. Original socket stays there too.
Note, that any socket with inet->num != 0 MUST be bound in
@@ -106,7 +106,7 @@
tw->tw_dport = inet->dport;
tw->tw_family = sk->sk_family;
tw->tw_reuse = sk->sk_reuse;
- tw->tw_hashent = sk->sk_hashent;
+ tw->tw_hash = sk->sk_hash;
tw->tw_ipv6only = 0;
tw->tw_prot = sk->sk_prot_creator;
atomic_set(&tw->tw_refcnt, 1);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 13dfb39..c85819d 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -130,19 +130,20 @@
int dif = sk->sk_bound_dev_if;
INET_ADDR_COOKIE(acookie, saddr, daddr)
const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
- const int hash = inet_ehashfn(daddr, lport, saddr, inet->dport, tcp_hashinfo.ehash_size);
- struct inet_ehash_bucket *head = &tcp_hashinfo.ehash[hash];
+ unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport);
+ struct inet_ehash_bucket *head = inet_ehash_bucket(&tcp_hashinfo, hash);
struct sock *sk2;
const struct hlist_node *node;
struct inet_timewait_sock *tw;
+ prefetch(head->chain.first);
write_lock(&head->lock);
/* Check TIME-WAIT sockets first. */
sk_for_each(sk2, node, &(head + tcp_hashinfo.ehash_size)->chain) {
tw = inet_twsk(sk2);
- if (INET_TW_MATCH(sk2, acookie, saddr, daddr, ports, dif)) {
+ if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) {
const struct tcp_timewait_sock *tcptw = tcp_twsk(sk2);
struct tcp_sock *tp = tcp_sk(sk);
@@ -179,7 +180,7 @@
/* And established part... */
sk_for_each(sk2, node, &head->chain) {
- if (INET_MATCH(sk2, acookie, saddr, daddr, ports, dif))
+ if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
@@ -188,7 +189,7 @@
* in hash table socket with a funny identity. */
inet->num = lport;
inet->sport = htons(lport);
- sk->sk_hashent = hash;
+ sk->sk_hash = hash;
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
sock_prot_inc_use(sk->sk_prot);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 80643e6..d693cb9 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -209,9 +209,11 @@
lock = &tcp_hashinfo.lhash_lock;
inet_listen_wlock(&tcp_hashinfo);
} else {
- sk->sk_hashent = inet6_sk_ehashfn(sk, tcp_hashinfo.ehash_size);
- list = &tcp_hashinfo.ehash[sk->sk_hashent].chain;
- lock = &tcp_hashinfo.ehash[sk->sk_hashent].lock;
+ unsigned int hash;
+ sk->sk_hash = hash = inet6_sk_ehashfn(sk);
+ hash &= (tcp_hashinfo.ehash_size - 1);
+ list = &tcp_hashinfo.ehash[hash].chain;
+ lock = &tcp_hashinfo.ehash[hash].lock;
write_lock(lock);
}
@@ -322,13 +324,13 @@
const struct in6_addr *saddr = &np->daddr;
const int dif = sk->sk_bound_dev_if;
const u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
- const int hash = inet6_ehashfn(daddr, inet->num, saddr, inet->dport,
- tcp_hashinfo.ehash_size);
- struct inet_ehash_bucket *head = &tcp_hashinfo.ehash[hash];
+ unsigned int hash = inet6_ehashfn(daddr, inet->num, saddr, inet->dport);
+ struct inet_ehash_bucket *head = inet_ehash_bucket(&tcp_hashinfo, hash);
struct sock *sk2;
const struct hlist_node *node;
struct inet_timewait_sock *tw;
+ prefetch(head->chain.first);
write_lock(&head->lock);
/* Check TIME-WAIT sockets first. */
@@ -365,14 +367,14 @@
/* And established part... */
sk_for_each(sk2, node, &head->chain) {
- if (INET6_MATCH(sk2, saddr, daddr, ports, dif))
+ if (INET6_MATCH(sk2, hash, saddr, daddr, ports, dif))
goto not_unique;
}
unique:
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
- sk->sk_hashent = hash;
+ sk->sk_hash = hash;
sock_prot_inc_use(sk->sk_prot);
write_unlock(&head->lock);
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index 00eae5f..cf68a59 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -393,10 +393,10 @@
dst->value = skb->sk->sk_route_caps;
}
-META_COLLECTOR(int_sk_hashent)
+META_COLLECTOR(int_sk_hash)
{
SKIP_NONLOCAL(skb);
- dst->value = skb->sk->sk_hashent;
+ dst->value = skb->sk->sk_hash;
}
META_COLLECTOR(int_sk_lingertime)
@@ -515,7 +515,7 @@
[META_ID(SK_FORWARD_ALLOCS)] = META_FUNC(int_sk_fwd_alloc),
[META_ID(SK_ALLOCS)] = META_FUNC(int_sk_alloc),
[META_ID(SK_ROUTE_CAPS)] = META_FUNC(int_sk_route_caps),
- [META_ID(SK_HASHENT)] = META_FUNC(int_sk_hashent),
+ [META_ID(SK_HASH)] = META_FUNC(int_sk_hash),
[META_ID(SK_LINGERTIME)] = META_FUNC(int_sk_lingertime),
[META_ID(SK_ACK_BACKLOG)] = META_FUNC(int_sk_ack_bl),
[META_ID(SK_MAX_ACK_BACKLOG)] = META_FUNC(int_sk_max_ack_bl),