From 77d8bf9c6208eb535f05718168ffcc476be0ca8c Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 9 Aug 2005 20:00:51 -0700 Subject: [INET]: Move the TCP hashtable functions/structs to inet_hashtables.[ch] Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 122 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 122 insertions(+) (limited to 'include/net/inet_hashtables.h') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index c4c9e39f450..3a6c11ca421 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -14,8 +14,107 @@ #ifndef _INET_HASHTABLES_H #define _INET_HASHTABLES_H +#include +#include +#include +#include #include +/* This is for all connections with a full identity, no wildcards. + * New scheme, half the table is for TIME_WAIT, the other half is + * for the rest. I'll experiment with dynamic table growth later. + */ +struct inet_ehash_bucket { + rwlock_t lock; + struct hlist_head chain; +} __attribute__((__aligned__(8))); + +/* There are a few simple rules, which allow for local port reuse by + * an application. In essence: + * + * 1) Sockets bound to different interfaces may share a local port. + * Failing that, goto test 2. + * 2) If all sockets have sk->sk_reuse set, and none of them are in + * TCP_LISTEN state, the port may be shared. + * Failing that, goto test 3. + * 3) If all sockets are bound to a specific inet_sk(sk)->rcv_saddr local + * address, and none of them are the same, the port may be + * shared. + * Failing this, the port cannot be shared. + * + * The interesting point, is test #2. This is what an FTP server does + * all day. To optimize this case we use a specific flag bit defined + * below. As we add sockets to a bind bucket list, we perform a + * check of: (newsk->sk_reuse && (newsk->sk_state != TCP_LISTEN)) + * As long as all sockets added to a bind bucket pass this test, + * the flag bit will be set. + * The resulting situation is that tcp_v[46]_verify_bind() can just check + * for this flag bit, if it is set and the socket trying to bind has + * sk->sk_reuse set, we don't even have to walk the owners list at all, + * we return that it is ok to bind this socket to the requested local port. + * + * Sounds like a lot of work, but it is worth it. In a more naive + * implementation (ie. current FreeBSD etc.) the entire list of ports + * must be walked for each data port opened by an ftp server. Needless + * to say, this does not scale at all. With a couple thousand FTP + * users logged onto your box, isn't it nice to know that new data + * ports are created in O(1) time? I thought so. ;-) -DaveM + */ +struct inet_bind_bucket { + unsigned short port; + signed short fastreuse; + struct hlist_node node; + struct hlist_head owners; +}; + +#define inet_bind_bucket_for_each(tb, node, head) \ + hlist_for_each_entry(tb, node, head, node) + +struct inet_bind_hashbucket { + spinlock_t lock; + struct hlist_head chain; +}; + +/* This is for listening sockets, thus all sockets which possess wildcards. */ +#define INET_LHTABLE_SIZE 32 /* Yes, really, this is all you need. */ + +struct inet_hashinfo { + /* This is for sockets with full identity only. Sockets here will + * always be without wildcards and will have the following invariant: + * + * TCP_ESTABLISHED <= sk->sk_state < TCP_CLOSE + * + * First half of the table is for sockets not in TIME_WAIT, second half + * is for TIME_WAIT sockets only. + */ + struct inet_ehash_bucket *ehash; + + /* Ok, let's try this, I give up, we do need a local binding + * TCP hash as well as the others for fast bind/connect. + */ + struct inet_bind_hashbucket *bhash; + + int bhash_size; + int ehash_size; + + /* All sockets in TCP_LISTEN state will be in here. This is the only + * table where wildcard'd TCP sockets can exist. Hash function here + * is just local port number. + */ + struct hlist_head listening_hash[INET_LHTABLE_SIZE]; + + /* All the above members are written once at bootup and + * never written again _or_ are predominantly read-access. + * + * Now align to a new cache line as all the following members + * are often dirty. + */ + rwlock_t lhash_lock ____cacheline_aligned; + atomic_t lhash_users; + wait_queue_head_t lhash_wait; + spinlock_t portalloc_lock; +}; + static inline int inet_ehashfn(const __u32 laddr, const __u16 lport, const __u32 faddr, const __u16 fport, const int ehash_size) @@ -37,4 +136,27 @@ static inline int inet_sk_ehashfn(const struct sock *sk, const int ehash_size) return inet_ehashfn(laddr, lport, faddr, fport, ehash_size); } +extern struct inet_bind_bucket * + inet_bind_bucket_create(kmem_cache_t *cachep, + struct inet_bind_hashbucket *head, + const unsigned short snum); +extern void inet_bind_bucket_destroy(kmem_cache_t *cachep, + struct inet_bind_bucket *tb); + +static inline int inet_bhashfn(const __u16 lport, const int bhash_size) +{ + return lport & (bhash_size - 1); +} + +/* These can have wildcards, don't try too hard. */ +static inline int inet_lhashfn(const unsigned short num) +{ + return num & (INET_LHTABLE_SIZE - 1); +} + +static inline int inet_sk_listen_hashfn(const struct sock *sk) +{ + return inet_lhashfn(inet_sk(sk)->num); +} + #endif /* _INET_HASHTABLES_H */ -- cgit v1.2.3-18-g5258