[NET] Generalise TCP's struct open_request minisock infrastructure
Kept this first changeset minimal, without changing existing names to
ease peer review.
Basicaly tcp_openreq_alloc now receives the or_calltable, that in turn
has two new members:
->slab, that replaces tcp_openreq_cachep
->obj_size, to inform the size of the openreq descendant for
a specific protocol
The protocol specific fields in struct open_request were moved to a
class hierarchy, with the things that are common to all connection
oriented PF_INET protocols in struct inet_request_sock, the TCP ones
in tcp_request_sock, that is an inet_request_sock, that is an
open_request.
I.e. this uses the same approach used for the struct sock class
hierarchy, with sk_prot indicating if the protocol wants to use the
open_request infrastructure by filling in sk_prot->rsk_prot with an
or_calltable.
Results? Performance is improved and TCP v4 now uses only 64 bytes per
open request minisock, down from 96 without this patch :-)
Next changeset will rename some of the structs, fields and functions
mentioned above, struct or_calltable is way unclear, better name it
struct request_sock_ops, s/struct open_request/struct request_sock/g,
etc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index eea1a17..1037401 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -692,6 +692,8 @@
struct sock *newsk = sk_alloc(PF_INET, GFP_ATOMIC, sk->sk_prot, 0);
if(newsk != NULL) {
+ struct inet_request_sock *ireq = inet_rsk(req);
+ struct tcp_request_sock *treq = tcp_rsk(req);
struct tcp_sock *newtp;
struct sk_filter *filter;
@@ -703,7 +705,7 @@
tcp_sk(newsk)->bind_hash = NULL;
/* Clone the TCP header template */
- inet_sk(newsk)->dport = req->rmt_port;
+ inet_sk(newsk)->dport = ireq->rmt_port;
sock_lock_init(newsk);
bh_lock_sock(newsk);
@@ -739,14 +741,14 @@
/* Now setup tcp_sock */
newtp = tcp_sk(newsk);
newtp->pred_flags = 0;
- newtp->rcv_nxt = req->rcv_isn + 1;
- newtp->snd_nxt = req->snt_isn + 1;
- newtp->snd_una = req->snt_isn + 1;
- newtp->snd_sml = req->snt_isn + 1;
+ newtp->rcv_nxt = treq->rcv_isn + 1;
+ newtp->snd_nxt = treq->snt_isn + 1;
+ newtp->snd_una = treq->snt_isn + 1;
+ newtp->snd_sml = treq->snt_isn + 1;
tcp_prequeue_init(newtp);
- tcp_init_wl(newtp, req->snt_isn, req->rcv_isn);
+ tcp_init_wl(newtp, treq->snt_isn, treq->rcv_isn);
newtp->retransmits = 0;
newtp->backoff = 0;
@@ -775,10 +777,10 @@
tcp_set_ca_state(newtp, TCP_CA_Open);
tcp_init_xmit_timers(newsk);
skb_queue_head_init(&newtp->out_of_order_queue);
- newtp->rcv_wup = req->rcv_isn + 1;
- newtp->write_seq = req->snt_isn + 1;
+ newtp->rcv_wup = treq->rcv_isn + 1;
+ newtp->write_seq = treq->snt_isn + 1;
newtp->pushed_seq = newtp->write_seq;
- newtp->copied_seq = req->rcv_isn + 1;
+ newtp->copied_seq = treq->rcv_isn + 1;
newtp->rx_opt.saw_tstamp = 0;
@@ -808,18 +810,18 @@
newsk->sk_socket = NULL;
newsk->sk_sleep = NULL;
- newtp->rx_opt.tstamp_ok = req->tstamp_ok;
- if((newtp->rx_opt.sack_ok = req->sack_ok) != 0) {
+ newtp->rx_opt.tstamp_ok = ireq->tstamp_ok;
+ if((newtp->rx_opt.sack_ok = ireq->sack_ok) != 0) {
if (sysctl_tcp_fack)
newtp->rx_opt.sack_ok |= 2;
}
newtp->window_clamp = req->window_clamp;
newtp->rcv_ssthresh = req->rcv_wnd;
newtp->rcv_wnd = req->rcv_wnd;
- newtp->rx_opt.wscale_ok = req->wscale_ok;
+ newtp->rx_opt.wscale_ok = ireq->wscale_ok;
if (newtp->rx_opt.wscale_ok) {
- newtp->rx_opt.snd_wscale = req->snd_wscale;
- newtp->rx_opt.rcv_wscale = req->rcv_wscale;
+ newtp->rx_opt.snd_wscale = ireq->snd_wscale;
+ newtp->rx_opt.rcv_wscale = ireq->rcv_wscale;
} else {
newtp->rx_opt.snd_wscale = newtp->rx_opt.rcv_wscale = 0;
newtp->window_clamp = min(newtp->window_clamp, 65535U);
@@ -881,7 +883,7 @@
}
/* Check for pure retransmitted SYN. */
- if (TCP_SKB_CB(skb)->seq == req->rcv_isn &&
+ if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn &&
flg == TCP_FLAG_SYN &&
!paws_reject) {
/*
@@ -959,7 +961,7 @@
* Invalid ACK: reset will be sent by listening socket
*/
if ((flg & TCP_FLAG_ACK) &&
- (TCP_SKB_CB(skb)->ack_seq != req->snt_isn+1))
+ (TCP_SKB_CB(skb)->ack_seq != tcp_rsk(req)->snt_isn + 1))
return sk;
/* Also, it would be not so bad idea to check rcv_tsecr, which
@@ -970,7 +972,7 @@
/* RFC793: "first check sequence number". */
if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq,
- req->rcv_isn+1, req->rcv_isn+1+req->rcv_wnd)) {
+ tcp_rsk(req)->rcv_isn + 1, tcp_rsk(req)->rcv_isn + 1 + req->rcv_wnd)) {
/* Out of window: send ACK and drop. */
if (!(flg & TCP_FLAG_RST))
req->class->send_ack(skb, req);
@@ -981,12 +983,12 @@
/* In sequence, PAWS is OK. */
- if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, req->rcv_isn+1))
+ if (tmp_opt.saw_tstamp && !after(TCP_SKB_CB(skb)->seq, tcp_rsk(req)->rcv_isn + 1))
req->ts_recent = tmp_opt.rcv_tsval;
- if (TCP_SKB_CB(skb)->seq == req->rcv_isn) {
+ if (TCP_SKB_CB(skb)->seq == tcp_rsk(req)->rcv_isn) {
/* Truncate SYN, it is out of window starting
- at req->rcv_isn+1. */
+ at tcp_rsk(req)->rcv_isn + 1. */
flg &= ~TCP_FLAG_SYN;
}
@@ -1003,8 +1005,8 @@
return NULL;
/* If TCP_DEFER_ACCEPT is set, drop bare ACK. */
- if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == req->rcv_isn+1) {
- req->acked = 1;
+ if (tp->defer_accept && TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) {
+ inet_rsk(req)->acked = 1;
return NULL;
}
@@ -1026,7 +1028,7 @@
listen_overflow:
if (!sysctl_tcp_abort_on_overflow) {
- req->acked = 1;
+ inet_rsk(req)->acked = 1;
return NULL;
}