Commit 86c1aee1 authored by David S. Miller's avatar David S. Miller
Browse files

Merge branch 'listener_refactor_part_15' 



Eric Dumazet says:

====================
tcp listener refactoring part 15

I am trying to make the final patch pushing request socks into ehash
as small as possible. In this patch series, I made various adjustments
for the SYNACK generation, allowing me to reach 1 Mpps SYNACK in my
stress test (still hitting LISTENER spinlock of course, and the syn_wait
spinlock)

I also converted the ICMP handlers a bit ahead of time :

They no longer need to get the LISTENER socket, and can use
only a lookup in ehash table. No big deal if we ignore ICMP
for requests socks before the final steps.
====================

Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents c9231f82 52036a43

include/linux/dccp.h

+3 −1
Original line number Diff line number Diff line
@@ -43,6 +43,7 @@ enum dccp_state { 
	DCCP_CLOSING	     = TCP_CLOSING,

	DCCP_TIME_WAIT	     = TCP_TIME_WAIT,

	DCCP_CLOSED	     = TCP_CLOSE,

	DCCP_NEW_SYN_RECV    = TCP_NEW_SYN_RECV,

	DCCP_PARTOPEN	     = TCP_MAX_STATES,

	DCCP_PASSIVE_CLOSEREQ,			/* clients receiving CloseReq */

	DCCP_MAX_STATES
@@ -57,6 +58,7 @@ enum { 
	DCCPF_CLOSING	      = TCPF_CLOSING,

	DCCPF_TIME_WAIT	      = TCPF_TIME_WAIT,

	DCCPF_CLOSED	      = TCPF_CLOSE,

	DCCPF_NEW_SYN_RECV    = TCPF_NEW_SYN_RECV,

	DCCPF_PARTOPEN	      = (1 << DCCP_PARTOPEN),

};
@@ -317,6 +319,6 @@ static inline const char *dccp_role(const struct sock *sk) 
	return NULL;

}



extern void dccp_syn_ack_timeout(struct sock *sk, struct request_sock *req);

extern void dccp_syn_ack_timeout(const struct request_sock *req);



#endif /* _LINUX_DCCP_H */

include/net/request_sock.h

+4 −10
Original line number Diff line number Diff line
@@ -39,8 +39,7 @@ struct request_sock_ops { 
	void		(*send_reset)(struct sock *sk,

				      struct sk_buff *skb);

	void		(*destructor)(struct request_sock *req);

	void		(*syn_ack_timeout)(struct sock *sk,

					   struct request_sock *req);

	void		(*syn_ack_timeout)(const struct request_sock *req);

};



int inet_rtx_syn_ack(struct sock *parent, struct request_sock *req);
@@ -174,11 +173,6 @@ struct fastopen_queue { 
 * %syn_wait_lock is necessary only to avoid proc interface having to grab the main

 * lock sock while browsing the listening hash (otherwise it's deadlock prone).

 *

 * This lock is acquired in read mode only from listening_get_next() seq_file

 * op and it's acquired in write mode _only_ from code that is actively

 * changing rskq_accept_head. All readers that are holding the master sock lock

 * don't need to grab this lock in read mode too as rskq_accept_head. writes

 * are always protected from the main sock lock.

 */

struct request_sock_queue {

	struct request_sock	*rskq_accept_head;
@@ -193,7 +187,7 @@ struct request_sock_queue { 
					     */



	/* temporary alignment, our goal is to get rid of this lock */

	rwlock_t		syn_wait_lock ____cacheline_aligned_in_smp;

	spinlock_t		syn_wait_lock ____cacheline_aligned_in_smp;

};



int reqsk_queue_alloc(struct request_sock_queue *queue,
@@ -224,14 +218,14 @@ static inline void reqsk_queue_unlink(struct request_sock_queue *queue, 
	struct listen_sock *lopt = queue->listen_opt;

	struct request_sock **prev;



	write_lock(&queue->syn_wait_lock);

	spin_lock(&queue->syn_wait_lock);



	prev = &lopt->syn_table[req->rsk_hash];

	while (*prev != req)

		prev = &(*prev)->dl_next;

	*prev = req->dl_next;



	write_unlock(&queue->syn_wait_lock);

	spin_unlock(&queue->syn_wait_lock);

	if (del_timer(&req->rsk_timer))

		reqsk_put(req);

}

include/net/tcp.h

+2 −1
Original line number Diff line number Diff line
@@ -433,7 +433,7 @@ int compat_tcp_getsockopt(struct sock *sk, int level, int optname, 
int compat_tcp_setsockopt(struct sock *sk, int level, int optname,

			  char __user *optval, unsigned int optlen);

void tcp_set_keepalive(struct sock *sk, int val);

void tcp_syn_ack_timeout(struct sock *sk, struct request_sock *req);

void tcp_syn_ack_timeout(const struct request_sock *req);

int tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int nonblock,

		int flags, int *addr_len);

void tcp_parse_options(const struct sk_buff *skb,
@@ -447,6 +447,7 @@ const u8 *tcp_parse_md5sig_option(const struct tcphdr *th); 


void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb);

void tcp_v4_mtu_reduced(struct sock *sk);

void tcp_req_err(struct sock *sk, u32 seq);

int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb);

struct sock *tcp_create_openreq_child(struct sock *sk,

				      struct request_sock *req,

net/core/request_sock.c

+7 −7
Original line number Diff line number Diff line
@@ -58,14 +58,14 @@ int reqsk_queue_alloc(struct request_sock_queue *queue, 
		return -ENOMEM;



	get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd));

	rwlock_init(&queue->syn_wait_lock);

	spin_lock_init(&queue->syn_wait_lock);

	queue->rskq_accept_head = NULL;

	lopt->nr_table_entries = nr_table_entries;

	lopt->max_qlen_log = ilog2(nr_table_entries);



	write_lock_bh(&queue->syn_wait_lock);

	spin_lock_bh(&queue->syn_wait_lock);

	queue->listen_opt = lopt;

	write_unlock_bh(&queue->syn_wait_lock);

	spin_unlock_bh(&queue->syn_wait_lock);



	return 0;

}
@@ -81,10 +81,10 @@ static inline struct listen_sock *reqsk_queue_yank_listen_sk( 
{

	struct listen_sock *lopt;



	write_lock_bh(&queue->syn_wait_lock);

	spin_lock_bh(&queue->syn_wait_lock);

	lopt = queue->listen_opt;

	queue->listen_opt = NULL;

	write_unlock_bh(&queue->syn_wait_lock);

	spin_unlock_bh(&queue->syn_wait_lock);



	return lopt;

}
@@ -100,7 +100,7 @@ void reqsk_queue_destroy(struct request_sock_queue *queue) 
		for (i = 0; i < lopt->nr_table_entries; i++) {

			struct request_sock *req;



			write_lock_bh(&queue->syn_wait_lock);

			spin_lock_bh(&queue->syn_wait_lock);

			while ((req = lopt->syn_table[i]) != NULL) {

				lopt->syn_table[i] = req->dl_next;

				atomic_inc(&lopt->qlen_dec);
@@ -108,7 +108,7 @@ void reqsk_queue_destroy(struct request_sock_queue *queue) 
					reqsk_put(req);

				reqsk_put(req);

			}

			write_unlock_bh(&queue->syn_wait_lock);

			spin_unlock_bh(&queue->syn_wait_lock);

		}

	}

net/dccp/dccp.h

+1 −0
Original line number Diff line number Diff line
@@ -317,6 +317,7 @@ int inet_dccp_listen(struct socket *sock, int backlog); 
unsigned int dccp_poll(struct file *file, struct socket *sock,

		       poll_table *wait);

int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len);

void dccp_req_err(struct sock *sk, u64 seq);



struct sk_buff *dccp_ctl_make_reset(struct sock *sk, struct sk_buff *skb);

int dccp_send_reset(struct sock *sk, enum dccp_reset_codes code);

CRA Git | Maintained and supported by SUSTech CRA and CCSE