Commit a7f5e7f1 authored by Arnaldo Carvalho de Melo's avatar Arnaldo Carvalho de Melo Committed by David S. Miller

[INET]: Generalise tcp_v4_hash_connect

Renaming it to inet_hash_connect, making it possible to ditch
dccp_v4_hash_connect and share the same code with TCP instead.
Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 6d6ee43e
...@@ -1554,10 +1554,8 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, ...@@ -1554,10 +1554,8 @@ __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr,
EXPORT_SYMBOL(secure_tcp_sequence_number); EXPORT_SYMBOL(secure_tcp_sequence_number);
/* Generate secure starting point for ephemeral IPV4 transport port search */
u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport)
/* Generate secure starting point for ephemeral TCP port search */
u32 secure_tcp_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport)
{ {
struct keydata *keyptr = get_keyptr(); struct keydata *keyptr = get_keyptr();
u32 hash[4]; u32 hash[4];
......
...@@ -52,7 +52,7 @@ extern void get_random_bytes(void *buf, int nbytes); ...@@ -52,7 +52,7 @@ extern void get_random_bytes(void *buf, int nbytes);
void generate_random_uuid(unsigned char uuid_out[16]); void generate_random_uuid(unsigned char uuid_out[16]);
extern __u32 secure_ip_id(__u32 daddr); extern __u32 secure_ip_id(__u32 daddr);
extern u32 secure_tcp_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport); extern u32 secure_ipv4_port_ephemeral(__u32 saddr, __u32 daddr, __u16 dport);
extern u32 secure_tcpv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr, extern u32 secure_tcpv6_port_ephemeral(const __u32 *saddr, const __u32 *daddr,
__u16 dport); __u16 dport);
extern __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr, extern __u32 secure_tcp_sequence_number(__u32 saddr, __u32 daddr,
......
...@@ -434,4 +434,7 @@ static inline struct sock *inet_lookup(struct inet_hashinfo *hashinfo, ...@@ -434,4 +434,7 @@ static inline struct sock *inet_lookup(struct inet_hashinfo *hashinfo,
return sk; return sk;
} }
extern int inet_hash_connect(struct inet_timewait_death_row *death_row,
struct sock *sk);
#endif /* _INET_HASHTABLES_H */ #endif /* _INET_HASHTABLES_H */
...@@ -54,164 +54,6 @@ void dccp_unhash(struct sock *sk) ...@@ -54,164 +54,6 @@ void dccp_unhash(struct sock *sk)
EXPORT_SYMBOL_GPL(dccp_unhash); EXPORT_SYMBOL_GPL(dccp_unhash);
/* called with local bh disabled */
static int __dccp_v4_check_established(struct sock *sk, const __u16 lport,
struct inet_timewait_sock **twp)
{
struct inet_sock *inet = inet_sk(sk);
const u32 daddr = inet->rcv_saddr;
const u32 saddr = inet->daddr;
const int dif = sk->sk_bound_dev_if;
INET_ADDR_COOKIE(acookie, saddr, daddr)
const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
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, 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, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
/* Must record num and sport now. Otherwise we will see
* in hash table socket with a funny identity. */
inet->num = lport;
inet->sport = htons(lport);
sk->sk_hash = hash;
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
sock_prot_inc_use(sk->sk_prot);
write_unlock(&head->lock);
if (twp != NULL) {
*twp = tw;
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
} else if (tw != NULL) {
/* Silly. Should hash-dance instead... */
inet_twsk_deschedule(tw, &dccp_death_row);
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
inet_twsk_put(tw);
}
return 0;
not_unique:
write_unlock(&head->lock);
return -EADDRNOTAVAIL;
}
/*
* Bind a port for a connect operation and hash it.
*/
static int dccp_v4_hash_connect(struct sock *sk)
{
const unsigned short snum = inet_sk(sk)->num;
struct inet_bind_hashbucket *head;
struct inet_bind_bucket *tb;
int ret;
if (snum == 0) {
int low = sysctl_local_port_range[0];
int high = sysctl_local_port_range[1];
int remaining = (high - low) + 1;
int rover = net_random() % (high - low) + low;
struct hlist_node *node;
struct inet_timewait_sock *tw = NULL;
local_bh_disable();
do {
head = &dccp_hashinfo.bhash[inet_bhashfn(rover,
dccp_hashinfo.bhash_size)];
spin_lock(&head->lock);
/* Does not bother with rcv_saddr checks,
* because the established check is already
* unique enough.
*/
inet_bind_bucket_for_each(tb, node, &head->chain) {
if (tb->port == rover) {
BUG_TRAP(!hlist_empty(&tb->owners));
if (tb->fastreuse >= 0)
goto next_port;
if (!__dccp_v4_check_established(sk,
rover,
&tw))
goto ok;
goto next_port;
}
}
tb = inet_bind_bucket_create(dccp_hashinfo.bind_bucket_cachep,
head, rover);
if (tb == NULL) {
spin_unlock(&head->lock);
break;
}
tb->fastreuse = -1;
goto ok;
next_port:
spin_unlock(&head->lock);
if (++rover > high)
rover = low;
} while (--remaining > 0);
local_bh_enable();
return -EADDRNOTAVAIL;
ok:
/* All locks still held and bhs disabled */
inet_bind_hash(sk, tb, rover);
if (sk_unhashed(sk)) {
inet_sk(sk)->sport = htons(rover);
__inet_hash(&dccp_hashinfo, sk, 0);
}
spin_unlock(&head->lock);
if (tw != NULL) {
inet_twsk_deschedule(tw, &dccp_death_row);
inet_twsk_put(tw);
}
ret = 0;
goto out;
}
head = &dccp_hashinfo.bhash[inet_bhashfn(snum,
dccp_hashinfo.bhash_size)];
tb = inet_csk(sk)->icsk_bind_hash;
spin_lock_bh(&head->lock);
if (sk_head(&tb->owners) == sk && sk->sk_bind_node.next == NULL) {
__inet_hash(&dccp_hashinfo, sk, 0);
spin_unlock_bh(&head->lock);
return 0;
} else {
spin_unlock(&head->lock);
/* No definite answer... Walk to established hash table */
ret = __dccp_v4_check_established(sk, snum, NULL);
out:
local_bh_enable();
return ret;
}
}
int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{ {
struct inet_sock *inet = inet_sk(sk); struct inet_sock *inet = inet_sk(sk);
...@@ -272,7 +114,7 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) ...@@ -272,7 +114,7 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
* complete initialization after this. * complete initialization after this.
*/ */
dccp_set_state(sk, DCCP_REQUESTING); dccp_set_state(sk, DCCP_REQUESTING);
err = dccp_v4_hash_connect(sk); err = inet_hash_connect(&dccp_death_row, sk);
if (err != 0) if (err != 0)
goto failure; goto failure;
......
...@@ -15,12 +15,14 @@ ...@@ -15,12 +15,14 @@
#include <linux/config.h> #include <linux/config.h>
#include <linux/module.h> #include <linux/module.h>
#include <linux/random.h>
#include <linux/sched.h> #include <linux/sched.h>
#include <linux/slab.h> #include <linux/slab.h>
#include <linux/wait.h> #include <linux/wait.h>
#include <net/inet_connection_sock.h> #include <net/inet_connection_sock.h>
#include <net/inet_hashtables.h> #include <net/inet_hashtables.h>
#include <net/ip.h>
/* /*
* Allocate and initialize a new local port bind bucket. * Allocate and initialize a new local port bind bucket.
...@@ -163,3 +165,179 @@ struct sock *__inet_lookup_listener(const struct hlist_head *head, const u32 dad ...@@ -163,3 +165,179 @@ struct sock *__inet_lookup_listener(const struct hlist_head *head, const u32 dad
} }
EXPORT_SYMBOL_GPL(__inet_lookup_listener); EXPORT_SYMBOL_GPL(__inet_lookup_listener);
/* called with local bh disabled */
static int __inet_check_established(struct inet_timewait_death_row *death_row,
struct sock *sk, __u16 lport,
struct inet_timewait_sock **twp)
{
struct inet_hashinfo *hinfo = death_row->hashinfo;
struct inet_sock *inet = inet_sk(sk);
u32 daddr = inet->rcv_saddr;
u32 saddr = inet->daddr;
int dif = sk->sk_bound_dev_if;
INET_ADDR_COOKIE(acookie, saddr, daddr)
const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport);
struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, 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 + hinfo->ehash_size)->chain) {
tw = inet_twsk(sk2);
if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) {
if (twsk_unique(sk, sk2, twp))
goto unique;
else
goto not_unique;
}
}
tw = NULL;
/* And established part... */
sk_for_each(sk2, node, &head->chain) {
if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
unique:
/* Must record num and sport now. Otherwise we will see
* in hash table socket with a funny identity. */
inet->num = lport;
inet->sport = htons(lport);
sk->sk_hash = hash;
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
sock_prot_inc_use(sk->sk_prot);
write_unlock(&head->lock);
if (twp) {
*twp = tw;
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
} else if (tw) {
/* Silly. Should hash-dance instead... */
inet_twsk_deschedule(tw, death_row);
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
inet_twsk_put(tw);
}
return 0;
not_unique:
write_unlock(&head->lock);
return -EADDRNOTAVAIL;
}
static inline u32 inet_sk_port_offset(const struct sock *sk)
{
const struct inet_sock *inet = inet_sk(sk);
return secure_ipv4_port_ephemeral(inet->rcv_saddr, inet->daddr,
inet->dport);
}
/*
* Bind a port for a connect operation and hash it.
*/
int inet_hash_connect(struct inet_timewait_death_row *death_row,
struct sock *sk)
{
struct inet_hashinfo *hinfo = death_row->hashinfo;
const unsigned short snum = inet_sk(sk)->num;
struct inet_bind_hashbucket *head;
struct inet_bind_bucket *tb;
int ret;
if (!snum) {
int low = sysctl_local_port_range[0];
int high = sysctl_local_port_range[1];
int range = high - low;
int i;
int port;
static u32 hint;
u32 offset = hint + inet_sk_port_offset(sk);
struct hlist_node *node;
struct inet_timewait_sock *tw = NULL;
local_bh_disable();
for (i = 1; i <= range; i++) {
port = low + (i + offset) % range;
head = &hinfo->bhash[inet_bhashfn(port, hinfo->bhash_size)];
spin_lock(&head->lock);
/* Does not bother with rcv_saddr checks,
* because the established check is already
* unique enough.
*/
inet_bind_bucket_for_each(tb, node, &head->chain) {
if (tb->port == port) {
BUG_TRAP(!hlist_empty(&tb->owners));
if (tb->fastreuse >= 0)
goto next_port;
if (!__inet_check_established(death_row,
sk, port,
&tw))
goto ok;
goto next_port;
}
}
tb = inet_bind_bucket_create(hinfo->bind_bucket_cachep, head, port);
if (!tb) {
spin_unlock(&head->lock);
break;
}
tb->fastreuse = -1;
goto ok;
next_port:
spin_unlock(&head->lock);
}
local_bh_enable();
return -EADDRNOTAVAIL;
ok:
hint += i;
/* Head lock still held and bh's disabled */
inet_bind_hash(sk, tb, port);
if (sk_unhashed(sk)) {
inet_sk(sk)->sport = htons(port);
__inet_hash(hinfo, sk, 0);
}
spin_unlock(&head->lock);
if (tw) {
inet_twsk_deschedule(tw, death_row);;
inet_twsk_put(tw);
}
ret = 0;
goto out;
}
head = &hinfo->bhash[inet_bhashfn(snum, hinfo->bhash_size)];
tb = inet_csk(sk)->icsk_bind_hash;
spin_lock_bh(&head->lock);
if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
__inet_hash(hinfo, sk, 0);
spin_unlock_bh(&head->lock);
return 0;
} else {
spin_unlock(&head->lock);
/* No definite answer... Walk to established hash table */
ret = __inet_check_established(death_row, sk, snum, NULL);
out:
local_bh_enable();
return ret;
}
}
EXPORT_SYMBOL_GPL(inet_hash_connect);
...@@ -152,177 +152,6 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp) ...@@ -152,177 +152,6 @@ int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp)
EXPORT_SYMBOL_GPL(tcp_twsk_unique); EXPORT_SYMBOL_GPL(tcp_twsk_unique);
/* called with local bh disabled */
static int __tcp_v4_check_established(struct sock *sk, __u16 lport,
struct inet_timewait_sock **twp)
{
struct inet_sock *inet = inet_sk(sk);
u32 daddr = inet->rcv_saddr;
u32 saddr = inet->daddr;
int dif = sk->sk_bound_dev_if;
INET_ADDR_COOKIE(acookie, saddr, daddr)
const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport);
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, hash, acookie, saddr, daddr, ports, dif)) {
if (twsk_unique(sk, sk2, twp))
goto unique;
else
goto not_unique;
}
}
tw = NULL;
/* And established part... */
sk_for_each(sk2, node, &head->chain) {
if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif))
goto not_unique;
}
unique:
/* Must record num and sport now. Otherwise we will see
* in hash table socket with a funny identity. */
inet->num = lport;
inet->sport = htons(lport);
sk->sk_hash = hash;
BUG_TRAP(sk_unhashed(sk));
__sk_add_node(sk, &head->chain);
sock_prot_inc_use(sk->sk_prot);
write_unlock(&head->lock);
if (twp) {
*twp = tw;
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
} else if (tw) {
/* Silly. Should hash-dance instead... */
inet_twsk_deschedule(tw, &tcp_death_row);
NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED);
inet_twsk_put(tw);
}
return 0;
not_unique:
write_unlock(&head->lock);
return -EADDRNOTAVAIL;
}
static inline u32 connect_port_offset(const struct sock *sk)
{
const struct inet_sock *inet = inet_sk(sk);
return secure_tcp_port_ephemeral(inet->rcv_saddr, inet->daddr,
inet->dport);
}
/*
* Bind a port for a connect operation and hash it.
*/
static inline int tcp_v4_hash_connect(struct sock *sk)
{
const unsigned short snum = inet_sk(sk)->num;
struct inet_bind_hashbucket *head;
struct inet_bind_bucket *tb;
int ret;
if (!snum) {
int low = sysctl_local_port_range[0];
int high = sysctl_local_port_range[1];
int range = high - low;
int i;
int port;
static u32 hint;
u32 offset = hint + connect_port_offset(sk);
struct hlist_node *node;
struct inet_timewait_sock *tw = NULL;
local_bh_disable();
for (i = 1; i <= range; i++) {
port = low + (i + offset) % range;
head = &tcp_hashinfo.bhash[inet_bhashfn(port, tcp_hashinfo.bhash_size)];
spin_lock(&head->lock);
/* Does not bother with rcv_saddr checks,
* because the established check is already
* unique enough.
*/
inet_bind_bucket_for_each(tb, node, &head->chain) {
if (tb->port == port) {
BUG_TRAP(!hlist_empty(&tb->owners));
if (tb->fastreuse >= 0)
goto next_port;
if (!__tcp_v4_check_established(sk,
port,
&tw))
goto ok;
goto next_port;
}
}
tb = inet_bind_bucket_create(tcp_hashinfo.bind_bucket_cachep, head, port);
if (!tb) {
spin_unlock(&head->lock);
break;
}
tb->fastreuse = -1;
goto ok;
next_port:
spin_unlock(&head->lock);
}
local_bh_enable();
return -EADDRNOTAVAIL;
ok:
hint += i;
/* Head lock still held and bh's disabled */
inet_bind_hash(sk, tb, port);
if (sk_unhashed(sk)) {
inet_sk(sk)->sport = htons(port);
__inet_hash(&tcp_hashinfo, sk, 0);
}
spin_unlock(&head->lock);
if (tw) {
inet_twsk_deschedule(tw, &tcp_death_row);;
inet_twsk_put(tw);
}
ret = 0;
goto out;
}
head = &tcp_hashinfo.bhash[inet_bhashfn(snum, tcp_hashinfo.bhash_size)];
tb = inet_csk(sk)->icsk_bind_hash;
spin_lock_bh(&head->lock);
if (sk_head(&tb->owners) == sk && !sk->sk_bind_node.next) {
__inet_hash(&tcp_hashinfo, sk, 0);
spin_unlock_bh(&head->lock);
return 0;
} else {
spin_unlock(&head->lock);
/* No definite answer... Walk to established hash table */
ret = __tcp_v4_check_established(sk, snum, NULL);
out:
local_bh_enable();
return ret;
}
}
/* This will initiate an outgoing connection. */ /* This will initiate an outgoing connection. */
int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{ {
...@@ -403,7 +232,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) ...@@ -403,7 +232,7 @@ int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
* complete initialization after this. * complete initialization after this.
*/ */
tcp_set_state(sk, TCP_SYN_SENT); tcp_set_state(sk, TCP_SYN_SENT);
err = tcp_v4_hash_connect(sk); err = inet_hash_connect(&tcp_death_row, sk);
if (err) if (err)
goto failure; goto failure;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment