DEFINE_SNMP_STAT(struct tcp_mib, tcp_statistics);
-kmem_cache_t *tcp_bucket_cachep;
kmem_cache_t *tcp_timewait_cachep;
atomic_t tcp_orphan_count = ATOMIC_INIT(0);
}
sk->sk_state = TCP_CLOSE;
- reqsk_queue_destroy(&tp->accept_queue);
+ __reqsk_queue_destroy(&tp->accept_queue);
return -EADDRINUSE;
}
static void tcp_listen_stop (struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct listen_sock *lopt;
struct request_sock *acc_req;
struct request_sock *req;
- int i;
tcp_delete_keepalive_timer(sk);
/* make all the listen_opt local to us */
- lopt = reqsk_queue_yank_listen_sk(&tp->accept_queue);
acc_req = reqsk_queue_yank_acceptq(&tp->accept_queue);
- if (lopt->qlen) {
- for (i = 0; i < TCP_SYNQ_HSIZE; i++) {
- while ((req = lopt->syn_table[i]) != NULL) {
- lopt->syn_table[i] = req->dl_next;
- lopt->qlen--;
- reqsk_free(req);
-
- /* Following specs, it would be better either to send FIN
- * (and enter FIN-WAIT-1, it is normal close)
- * or to send active reset (abort).
- * Certainly, it is pretty dangerous while synflood, but it is
- * bad justification for our negligence 8)
- * To be honest, we are not able to make either
- * of the variants now. --ANK
- */
- }
- }
- }
- BUG_TRAP(!lopt->qlen);
-
- kfree(lopt);
+ /* Following specs, it would be better either to send FIN
+ * (and enter FIN-WAIT-1, it is normal close)
+ * or to send active reset (abort).
+ * Certainly, it is pretty dangerous while synflood, but it is
+ * bad justification for our negligence 8)
+ * To be honest, we are not able to make either
+ * of the variants now. --ANK
+ */
+ reqsk_queue_destroy(&tp->accept_queue);
while ((req = acc_req) != NULL) {
struct sock *child = req->sk;
sk_charge_skb(sk, skb);
if (!sk->sk_send_head)
sk->sk_send_head = skb;
- else if (tp->nonagle&TCP_NAGLE_PUSH)
+ if (tp->nonagle & TCP_NAGLE_PUSH)
tp->nonagle &= ~TCP_NAGLE_PUSH;
}
size_t psize, int flags)
{
struct tcp_sock *tp = tcp_sk(sk);
- int mss_now;
+ int mss_now, size_goal;
int err;
ssize_t copied;
long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+ size_goal = tp->xmit_size_goal;
copied = 0;
err = -EPIPE;
int offset = poffset % PAGE_SIZE;
int size = min_t(size_t, psize, PAGE_SIZE - offset);
- if (!sk->sk_send_head || (copy = mss_now - skb->len) <= 0) {
+ if (!sk->sk_send_head || (copy = size_goal - skb->len) <= 0) {
new_segment:
if (!sk_stream_memory_free(sk))
goto wait_for_sndbuf;
goto wait_for_memory;
skb_entail(sk, tp, skb);
- copy = mss_now;
+ copy = size_goal;
}
if (copy > size)
if (!(psize -= copy))
goto out;
- if (skb->len != mss_now || (flags & MSG_OOB))
+ if (skb->len < mss_now || (flags & MSG_OOB))
continue;
if (forced_push(tp)) {
goto do_error;
mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+ size_goal = tp->xmit_size_goal;
}
out:
static inline int select_size(struct sock *sk, struct tcp_sock *tp)
{
- int tmp = tp->mss_cache_std;
+ int tmp = tp->mss_cache;
if (sk->sk_route_caps & NETIF_F_SG) {
- int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER);
+ if (sk->sk_route_caps & NETIF_F_TSO)
+ tmp = 0;
+ else {
+ int pgbreak = SKB_MAX_HEAD(MAX_TCP_HEADER);
- if (tmp >= pgbreak &&
- tmp <= pgbreak + (MAX_SKB_FRAGS - 1) * PAGE_SIZE)
- tmp = pgbreak;
+ if (tmp >= pgbreak &&
+ tmp <= pgbreak + (MAX_SKB_FRAGS - 1) * PAGE_SIZE)
+ tmp = pgbreak;
+ }
}
+
return tmp;
}
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *skb;
int iovlen, flags;
- int mss_now;
+ int mss_now, size_goal;
int err, copied;
long timeo;
clear_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+ size_goal = tp->xmit_size_goal;
/* Ok commence sending. */
iovlen = msg->msg_iovlen;
skb = sk->sk_write_queue.prev;
if (!sk->sk_send_head ||
- (copy = mss_now - skb->len) <= 0) {
+ (copy = size_goal - skb->len) <= 0) {
new_segment:
/* Allocate new segment. If the interface is SG,
skb->ip_summed = CHECKSUM_HW;
skb_entail(sk, tp, skb);
- copy = mss_now;
+ copy = size_goal;
}
/* Try to append data to the end of skb. */
tcp_mark_push(tp, skb);
goto new_segment;
} else if (page) {
- /* If page is cached, align
- * offset to L1 cache boundary
- */
- off = (off + L1_CACHE_BYTES - 1) &
- ~(L1_CACHE_BYTES - 1);
if (off == PAGE_SIZE) {
put_page(page);
TCP_PAGE(sk) = page = NULL;
if ((seglen -= copy) == 0 && iovlen == 0)
goto out;
- if (skb->len != mss_now || (flags & MSG_OOB))
+ if (skb->len < mss_now || (flags & MSG_OOB))
continue;
if (forced_push(tp)) {
goto do_error;
mss_now = tcp_current_mss(sk, !(flags&MSG_OOB));
+ size_goal = tp->xmit_size_goal;
}
}
if (!skb->len) {
if (sk->sk_send_head == skb)
sk->sk_send_head = NULL;
- __skb_unlink(skb, skb->list);
+ __skb_unlink(skb, &sk->sk_write_queue);
sk_stream_free_skb(sk, skb);
}
struct sk_buff *skb;
struct tcp_sock *tp = tcp_sk(sk);
- NET_ADD_STATS_USER(LINUX_MIB_TCPPREQUEUED, skb_queue_len(&tp->ucopy.prequeue));
+ NET_INC_STATS_USER(LINUX_MIB_TCPPREQUEUED);
/* RX process wants to run with disabled BHs, though it is not
* necessary */
* is not empty. It is more elegant, but eats cycles,
* unfortunately.
*/
- if (skb_queue_len(&tp->ucopy.prequeue))
+ if (!skb_queue_empty(&tp->ucopy.prequeue))
goto do_prequeue;
/* __ Set realtime policy in scheduler __ */
}
if (tp->rcv_nxt == tp->copied_seq &&
- skb_queue_len(&tp->ucopy.prequeue)) {
+ !skb_queue_empty(&tp->ucopy.prequeue)) {
do_prequeue:
tcp_prequeue_process(sk);
} while (len > 0);
if (user_recv) {
- if (skb_queue_len(&tp->ucopy.prequeue)) {
+ if (!skb_queue_empty(&tp->ucopy.prequeue)) {
int chunk;
tp->ucopy.len = copied > 0 ? len : 0;
BUG_TRAP(sk_unhashed(sk));
/* If it has not 0 inet_sk(sk)->num, it must be bound */
- BUG_TRAP(!inet_sk(sk)->num || tcp_sk(sk)->bind_hash);
+ BUG_TRAP(!inet_sk(sk)->num || inet_sk(sk)->bind_hash);
sk->sk_prot->destroy(sk);
xfrm_sk_free_policy(sk);
-#ifdef INET_REFCNT_DEBUG
- if (atomic_read(&sk->sk_refcnt) != 1) {
- printk(KERN_DEBUG "Destruction TCP %p delayed, c=%d\n",
- sk, atomic_read(&sk->sk_refcnt));
- }
-#endif
+ sk_refcnt_debug_release(sk);
atomic_dec(&tcp_orphan_count);
sock_put(sk);
tcp_sack_reset(&tp->rx_opt);
__sk_dst_reset(sk);
- BUG_TRAP(!inet->num || tp->bind_hash);
+ BUG_TRAP(!inet->num || inet->bind_hash);
sk->sk_error_report(sk);
return err;
return tp->af_specific->setsockopt(sk, level, optname,
optval, optlen);
+ /* This is a string value all the others are int's */
+ if (optname == TCP_CONGESTION) {
+ char name[TCP_CA_NAME_MAX];
+
+ if (optlen < 1)
+ return -EINVAL;
+
+ val = strncpy_from_user(name, optval,
+ min(TCP_CA_NAME_MAX-1, optlen));
+ if (val < 0)
+ return -EFAULT;
+ name[val] = 0;
+
+ lock_sock(sk);
+ err = tcp_set_congestion_control(tp, name);
+ release_sock(sk);
+ return err;
+ }
+
if (optlen < sizeof(int))
return -EINVAL;
info->tcpi_rto = jiffies_to_usecs(tp->rto);
info->tcpi_ato = jiffies_to_usecs(tp->ack.ato);
- info->tcpi_snd_mss = tp->mss_cache_std;
+ info->tcpi_snd_mss = tp->mss_cache;
info->tcpi_rcv_mss = tp->ack.rcv_mss;
info->tcpi_unacked = tp->packets_out;
switch (optname) {
case TCP_MAXSEG:
- val = tp->mss_cache_std;
+ val = tp->mss_cache;
if (!val && ((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
val = tp->rx_opt.user_mss;
break;
case TCP_QUICKACK:
val = !tp->ack.pingpong;
break;
+
+ case TCP_CONGESTION:
+ if (get_user(len, optlen))
+ return -EFAULT;
+ len = min_t(unsigned int, len, TCP_CA_NAME_MAX);
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, tp->ca_ops->name, len))
+ return -EFAULT;
+ return 0;
default:
return -ENOPROTOOPT;
};
extern void __skb_cb_too_small_for_tcp(int, int);
-extern void tcpdiag_init(void);
+extern struct tcp_congestion_ops tcp_reno;
static __initdata unsigned long thash_entries;
static int __init set_thash_entries(char *str)
__skb_cb_too_small_for_tcp(sizeof(struct tcp_skb_cb),
sizeof(skb->cb));
- tcp_bucket_cachep = kmem_cache_create("tcp_bind_bucket",
- sizeof(struct tcp_bind_bucket),
- 0, SLAB_HWCACHE_ALIGN,
- NULL, NULL);
- if (!tcp_bucket_cachep)
+ tcp_hashinfo.bind_bucket_cachep =
+ kmem_cache_create("tcp_bind_bucket",
+ sizeof(struct inet_bind_bucket), 0,
+ SLAB_HWCACHE_ALIGN, NULL, NULL);
+ if (!tcp_hashinfo.bind_bucket_cachep)
panic("tcp_init: Cannot alloc tcp_bind_bucket cache.");
tcp_timewait_cachep = kmem_cache_create("tcp_tw_bucket",
*
* The methodology is similar to that of the buffer cache.
*/
- tcp_ehash = (struct tcp_ehash_bucket *)
+ tcp_hashinfo.ehash =
alloc_large_system_hash("TCP established",
- sizeof(struct tcp_ehash_bucket),
+ sizeof(struct inet_ehash_bucket),
thash_entries,
(num_physpages >= 128 * 1024) ?
(25 - PAGE_SHIFT) :
(27 - PAGE_SHIFT),
HASH_HIGHMEM,
- &tcp_ehash_size,
+ &tcp_hashinfo.ehash_size,
NULL,
0);
- tcp_ehash_size = (1 << tcp_ehash_size) >> 1;
- for (i = 0; i < (tcp_ehash_size << 1); i++) {
- rwlock_init(&tcp_ehash[i].lock);
- INIT_HLIST_HEAD(&tcp_ehash[i].chain);
+ tcp_hashinfo.ehash_size = (1 << tcp_hashinfo.ehash_size) >> 1;
+ for (i = 0; i < (tcp_hashinfo.ehash_size << 1); i++) {
+ rwlock_init(&tcp_hashinfo.ehash[i].lock);
+ INIT_HLIST_HEAD(&tcp_hashinfo.ehash[i].chain);
}
- tcp_bhash = (struct tcp_bind_hashbucket *)
+ tcp_hashinfo.bhash =
alloc_large_system_hash("TCP bind",
- sizeof(struct tcp_bind_hashbucket),
- tcp_ehash_size,
+ sizeof(struct inet_bind_hashbucket),
+ tcp_hashinfo.ehash_size,
(num_physpages >= 128 * 1024) ?
(25 - PAGE_SHIFT) :
(27 - PAGE_SHIFT),
HASH_HIGHMEM,
- &tcp_bhash_size,
+ &tcp_hashinfo.bhash_size,
NULL,
64 * 1024);
- tcp_bhash_size = 1 << tcp_bhash_size;
- for (i = 0; i < tcp_bhash_size; i++) {
- spin_lock_init(&tcp_bhash[i].lock);
- INIT_HLIST_HEAD(&tcp_bhash[i].chain);
+ tcp_hashinfo.bhash_size = 1 << tcp_hashinfo.bhash_size;
+ for (i = 0; i < tcp_hashinfo.bhash_size; i++) {
+ spin_lock_init(&tcp_hashinfo.bhash[i].lock);
+ INIT_HLIST_HEAD(&tcp_hashinfo.bhash[i].chain);
}
/* Try to be a bit smarter and adjust defaults depending
* on available memory.
*/
for (order = 0; ((1 << order) << PAGE_SHIFT) <
- (tcp_bhash_size * sizeof(struct tcp_bind_hashbucket));
+ (tcp_hashinfo.bhash_size * sizeof(struct inet_bind_hashbucket));
order++)
;
if (order >= 4) {
sysctl_tcp_max_orphans >>= (3 - order);
sysctl_max_syn_backlog = 128;
}
- tcp_port_rover = sysctl_local_port_range[0] - 1;
+ tcp_hashinfo.port_rover = sysctl_local_port_range[0] - 1;
sysctl_tcp_mem[0] = 768 << order;
sysctl_tcp_mem[1] = 1024 << order;
printk(KERN_INFO "TCP: Hash tables configured "
"(established %d bind %d)\n",
- tcp_ehash_size << 1, tcp_bhash_size);
+ tcp_hashinfo.ehash_size << 1, tcp_hashinfo.bhash_size);
+
+ tcp_register_congestion_control(&tcp_reno);
}
EXPORT_SYMBOL(tcp_accept);