sock.h

GNU Hurd 源代码

/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the AF_INET socket handler.
 *
 * Version:	@(#)sock.h	1.0.4	05/13/93
 *
 * Authors:	Ross Biro, <[e-mail]>
 *		Fred N. van Kempen, <[e-mail]>
 *		Corey Minyard <wf-rch![e-mail]>
 *		Florian La Roche <[e-mail]>
 *
 * Fixes:
 *		Alan Cox	:	Volatiles in skbuff pointers. See
 *					skbuff comments. May be overdone,
 *					better to prove they can be removed
 *					than the reverse.
 *		Alan Cox	:	Added a zapped field for tcp to note
 *					a socket is reset and must stay shut up
 *		Alan Cox	:	New fields for options
 *	Pauline Middelink	:	identd support
 *		Alan Cox	:	Eliminate low level recv/recvfrom
 *		David S. Miller	:	New socket lookup architecture.
 *              Steve Whitehouse:       Default routines for sock_ops
 *
 *		This program is free software; you can redistribute it and/or
 *		modify it under the terms of the GNU General Public License
 *		as published by the Free Software Foundation; either version
 *		2 of the License, or (at your option) any later version.
 */
#ifndef _SOCK_H
#define _SOCK_H

#include <linux/config.h>
#include <linux/timer.h>
#include <linux/in.h>		/* struct sockaddr_in */

#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
#include <linux/in6.h>		/* struct sockaddr_in6 */
#include <linux/ipv6.h>		/* dest_cache, inet6_options */
#include <linux/icmpv6.h>
#include <net/if_inet6.h>	/* struct ipv6_mc_socklist */
#endif

#if defined(CONFIG_INET) || defined (CONFIG_INET_MODULE)
#include <linux/icmp.h>
#endif
#include <linux/tcp.h>		/* struct tcphdr */

#include <linux/netdevice.h>
#include <linux/skbuff.h>	/* struct sk_buff */
#include <net/protocol.h>		/* struct inet_protocol */
#if defined(CONFIG_X25) || defined(CONFIG_X25_MODULE)
#include <net/x25.h>
#endif
#if defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE)
#include <net/ax25.h>
#if defined(CONFIG_NETROM) || defined(CONFIG_NETROM_MODULE)
#include <net/netrom.h>
#endif
#if defined(CONFIG_ROSE) || defined(CONFIG_ROSE_MODULE)
#include <net/rose.h>
#endif
#endif

#if defined(CONFIG_IPX) || defined(CONFIG_IPX_MODULE)
#if defined(CONFIG_SPX) || defined(CONFIG_SPX_MODULE)
#include <net/spx.h>
#else
#include <net/ipx.h>
#endif /* CONFIG_SPX */
#endif /* CONFIG_IPX */

#if defined(CONFIG_ATALK) || defined(CONFIG_ATALK_MODULE)
#include <linux/atalk.h>
#endif

#if defined(CONFIG_DECNET) || defined(CONFIG_DECNET_MODULE)
#include <net/dn.h>
#endif

#if defined(CONFIG_IRDA) || defined(CONFIG_IRDA_MODULE)
#include <net/irda/irda.h>
#endif

#ifdef CONFIG_FILTER
#include <linux/filter.h>
#endif

#include <asm/atomic.h>

#define MIN_WRITE_SPACE	2048

/* The AF_UNIX specific socket options */
struct unix_opt {
	int 			family;
	char *			name;
	int  			locks;
	struct unix_address	*addr;
	struct dentry *		dentry;
	struct semaphore	readsem;
	struct sock *		other;
	struct sock **		list;
	struct sock *		gc_tree;
	int			inflight;
};

#ifdef CONFIG_NETLINK
struct netlink_callback;

struct netlink_opt {
	pid_t			pid;
	unsigned		groups;
	pid_t			dst_pid;
	unsigned		dst_groups;
	int			(*handler)(int unit, struct sk_buff *skb);
	atomic_t		locks;
	struct netlink_callback	*cb;
};
#endif

/* Once the IPX ncpd patches are in these are going into protinfo. */
#if defined(CONFIG_IPX) || defined(CONFIG_IPX_MODULE)
struct ipx_opt {
	ipx_address		dest_addr;
	ipx_interface		*intrfc;
	unsigned short		port;
#ifdef CONFIG_IPX_INTERN
	unsigned char           node[IPX_NODE_LEN];
#endif
	unsigned short		type;
/* 
 * To handle special ncp connection-handling sockets for mars_nwe,
 * the connection number must be stored in the socket.
 */
	unsigned short		ipx_ncp_conn;
};
#endif

#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
struct ipv6_pinfo {
	struct in6_addr 	saddr;
	struct in6_addr 	rcv_saddr;
	struct in6_addr		daddr;
	struct in6_addr		*daddr_cache;

	__u32			flow_label;
	__u32			frag_size;
	int			hop_limit;
	int			mcast_hops;
	int			mcast_oif;

	/* pktoption flags */
	union {
		struct {
			__u8	srcrt:2,
			        rxinfo:1,
				rxhlim:1,
				hopopts:1,
				dstopts:1,
                                authhdr:1,
                                rxflow:1;
		} bits;
		__u8		all;
	} rxopt;

	/* sockopt flags */
	__u8			mc_loop:1,
	                        recverr:1,
	                        sndflow:1,
	                        pmtudisc:2;

	struct ipv6_mc_socklist	*ipv6_mc_list;
	struct ipv6_fl_socklist *ipv6_fl_list;
	__u32			dst_cookie;

	struct ipv6_txoptions	*opt;
	struct sk_buff		*pktoptions;
};

struct raw6_opt {
	__u32			checksum;	/* perform checksum */
	__u32			offset;		/* checksum offset  */

	struct icmp6_filter	filter;
};

#endif /* IPV6 */

#if defined(CONFIG_INET) || defined(CONFIG_INET_MODULE)
struct raw_opt {
	struct icmp_filter	filter;
};
#endif

/* This defines a selective acknowledgement block. */
struct tcp_sack_block {
	__u32	start_seq;
	__u32	end_seq;
};

struct tcp_opt {
	int	tcp_header_len;	/* Bytes of tcp header to send		*/

/*
 *	Header prediction flags
 *	0x5?10 << 16 + snd_wnd in net byte order
 */
	__u32	pred_flags;

/*
 *	RFC793 variables by their proper names. This means you can
 *	read the code and the spec side by side (and laugh ...)
 *	See RFC793 and RFC1122. The RFC writes these in capitals.
 */
 	__u32	rcv_nxt;	/* What we want to receive next 	*/
 	__u32	snd_nxt;	/* Next sequence we send		*/

 	__u32	snd_una;	/* First byte we want an ack for	*/
	__u32	rcv_tstamp;	/* timestamp of last received packet	*/
	__u32	lrcvtime;	/* timestamp of last received data packet*/
	__u32	srtt;		/* smothed round trip time << 3		*/

	__u32	ato;		/* delayed ack timeout			*/
	__u32	snd_wl1;	/* Sequence for window update		*/

	__u32	snd_wl2;	/* Ack sequence for update		*/
	__u32	snd_wnd;	/* The window we expect to receive	*/
	__u32	max_window;
	__u32	pmtu_cookie;	/* Last pmtu seen by socket		*/
	__u16	mss_cache;	/* Cached effective mss, not including SACKS */
	__u16	mss_clamp;	/* Maximal mss, negotiated at connection setup */
	__u16	ext_header_len;	/* Dave, do you allow mw to use this hole? 8) --ANK */
	__u8	pending;	/* pending events			*/
	__u8	retransmits;
	__u32	last_ack_sent;	/* last ack we sent			*/

	__u32	backoff;	/* backoff				*/
	__u32	mdev;		/* medium deviation			*/
 	__u32	snd_cwnd;	/* Sending congestion window		*/
	__u32	rto;		/* retransmit timeout			*/

	__u32	packets_out;	/* Packets which are "in flight"	*/
	__u32	fackets_out;	/* Non-retrans SACK'd packets		*/
	__u32	retrans_out;	/* Fast-retransmitted packets out	*/
	__u32	high_seq;	/* snd_nxt at onset of congestion	*/
/*
 *	Slow start and congestion control (see also Nagle, and Karn & Partridge)
 */
 	__u32	snd_ssthresh;	/* Slow start size threshold		*/
 	__u16	snd_cwnd_cnt;	/* Linear increase counter		*/
	__u8	dup_acks;	/* Consequetive duplicate acks seen from other end */
	__u8	delayed_acks;
	__u16	user_mss;  	/* mss requested by user in ioctl */

	/* Two commonly used timers in both sender and receiver paths. */
 	struct timer_list	retransmit_timer;	/* Resend (no ack)	*/
 	struct timer_list	delack_timer;		/* Ack delay 		*/

	struct sk_buff_head	out_of_order_queue; /* Out of order segments go here */

	struct tcp_func		*af_specific;	/* Operations which are AF_INET{4,6} specific	*/
	struct sk_buff		*send_head;	/* Front of stuff to transmit			*/
	struct sk_buff		*retrans_head;	/* retrans head can be 
						 * different to the head of
						 * write queue if we are doing
						 * fast retransmit
						 */

 	__u32	rcv_wnd;	/* Current receiver window		*/
	__u32	rcv_wup;	/* rcv_nxt on last window update sent	*/
	__u32	write_seq;
	__u32	copied_seq;
/*
 *      Options received (usually on last packet, some only on SYN packets).
 */
	char	tstamp_ok,	/* TIMESTAMP seen on SYN packet		*/
		wscale_ok,	/* Wscale seen on SYN packet		*/
		sack_ok;	/* SACK seen on SYN packet		*/
	char	saw_tstamp;	/* Saw TIMESTAMP on last packet		*/
        __u8	snd_wscale;	/* Window scaling received from sender	*/
        __u8	rcv_wscale;	/* Window scaling to send to receiver	*/
	__u8	rexmt_done;	/* Retransmitted up to send head?	*/
        __u32	rcv_tsval;	/* Time stamp value             	*/
        __u32	rcv_tsecr;	/* Time stamp echo reply        	*/
        __u32	ts_recent;	/* Time stamp to echo next		*/
        __u32	ts_recent_stamp;/* Time we stored ts_recent (for aging) */
	int	num_sacks;	/* Number of SACK blocks		*/
	struct tcp_sack_block selective_acks[4]; /* The SACKS themselves*/

 	struct timer_list	probe_timer;		/* Probes	*/
	__u32	window_clamp;	/* XXX Document this... -DaveM		*/
	__u32	probes_out;	/* unanswered 0 window probes		*/
	__u32	syn_seq;
	__u32	fin_seq;
	__u32	urg_seq;
	__u32	urg_data;

	__u32	last_seg_size;	/* Size of last incoming segment */
	__u32	rcv_mss;	/* MSS used for delayed ACK decisions */ 
	__u32 	partial_writers; /* # of clients wanting at the head packet */

	struct open_request	*syn_wait_queue;
	struct open_request	**syn_wait_last;

	int syn_backlog;	/* Backlog of received SYNs */
};

 	
/*
 * This structure really needs to be cleaned up.
 * Most of it is for TCP, and not used by any of
 * the other protocols.
 */

/*
 * The idea is to start moving to a newer struct gradualy
 * 
 * IMHO the newer struct should have the following format:
 * 
 *	struct sock {
 *		sockmem [mem, proto, callbacks]
 *
 *		union or struct {
 *			ax25;
 *		} ll_pinfo;
 *	
 *		union {
 *			ipv4;
 *			ipv6;
 *			ipx;
 *			netrom;
 *			rose;
 * 			x25;
 *		} net_pinfo;
 *
 *		union {
 *			tcp;
 *			udp;
 *			spx;
 *			netrom;
 *		} tp_pinfo;
 *
 *	}
 */

/* Define this to get the sk->debug debugging facility. */
#define SOCK_DEBUGGING
#ifdef SOCK_DEBUGGING
#define SOCK_DEBUG(sk, msg...) do { if((sk) && ((sk)->debug)) printk(KERN_DEBUG msg); } while (0)
#else
#define SOCK_DEBUG(sk, msg...) do { } while (0)
#endif

struct sock {
	/* This must be first. */
	struct sock		*sklist_next;
	struct sock		*sklist_prev;

	/* Local port binding hash linkage. */
	struct sock		*bind_next;
	struct sock		**bind_pprev;

	/* Socket demultiplex comparisons on incoming packets. */
	__u32			daddr;		/* Foreign IPv4 addr			*/
	__u32			rcv_saddr;	/* Bound local IPv4 addr		*/
	__u16			dport;		/* Destination port			*/
	unsigned short		num;		/* Local port				*/
	int			bound_dev_if;	/* Bound device index if != 0		*/

	/* Main hash linkage for various protocol lookup tables. */
	struct sock		*next;
	struct sock		**pprev;

	volatile unsigned char	state,		/* Connection state			*/
				zapped;		/* In ax25 & ipx means not linked	*/
	__u16			sport;		/* Source port				*/

	unsigned short		family;		/* Address family			*/
	unsigned char		reuse,		/* SO_REUSEADDR setting			*/
				nonagle;	/* Disable Nagle algorithm?		*/

	atomic_t		sock_readers;	/* User count				*/
	int			rcvbuf;		/* Size of receive buffer in bytes	*/

	struct wait_queue	**sleep;	/* Sock wait queue			*/
	struct dst_entry	*dst_cache;	/* Destination cache			*/
	atomic_t		rmem_alloc;	/* Receive queue bytes committed	*/
	struct sk_buff_head	receive_queue;	/* Incoming packets			*/
	atomic_t		wmem_alloc;	/* Transmit queue bytes committed	*/
	struct sk_buff_head	write_queue;	/* Packet sending queue			*/
	atomic_t		omem_alloc;	/* "o" is "option" or "other" */
	__u32			saddr;		/* Sending source			*/
	unsigned int		allocation;	/* Allocation mode			*/
	int			sndbuf;		/* Size of send buffer in bytes		*/
	struct sock		*prev;

	/* Not all are volatile, but some are, so we might as well say they all are.
	 * XXX Make this a flag word -DaveM
	 */
	volatile char		dead,
				done,
				urginline,
				keepopen,
				linger,
				destroy,
				no_check,
				broadcast,
				bsdism;
	unsigned char		debug;
	int			proc;
	unsigned long	        lingertime;

	int			hashent;
	struct sock		*pair;

	/* Error and backlog packet queues, rarely used. */
	struct sk_buff_head	back_log,
	                        error_queue;

	struct proto		*prot;

	unsigned short		shutdown;

#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
	union {
		struct ipv6_pinfo	af_inet6;
	} net_pinfo;
#endif

	union {
		struct tcp_opt		af_tcp;
#if defined(CONFIG_INET) || defined (CONFIG_INET_MODULE)
		struct raw_opt		tp_raw4;
#endif
#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
		struct raw6_opt		tp_raw;
#endif /* CONFIG_IPV6 */
#if defined(CONFIG_SPX) || defined (CONFIG_SPX_MODULE)
		struct spx_opt		af_spx;
#endif /* CONFIG_SPX */

	} tp_pinfo;

	int			err, err_soft;	/* Soft holds errors that don't
						   cause failure but are the cause
						   of a persistent failure not just
						   'timed out' */
	unsigned short		ack_backlog;
	unsigned short		max_ack_backlog;
	__u32			priority;
	unsigned short		type;
	unsigned char		localroute;	/* Route locally only */
	unsigned char		protocol;
	struct ucred		peercred;

#ifdef CONFIG_FILTER
	/* Socket Filtering Instructions */
	struct sk_filter      	*filter;
#endif /* CONFIG_FILTER */

	/* This is where all the private (optional) areas that don't
	 * overlap will eventually live. 
	 */
	union {
		void *destruct_hook;
	  	struct unix_opt	af_unix;
#if defined(CONFIG_ATALK) || defined(CONFIG_ATALK_MODULE)
		struct atalk_sock	af_at;
#endif
#if defined(CONFIG_IPX) || defined(CONFIG_IPX_MODULE)
		struct ipx_opt		af_ipx;
#endif
#if defined (CONFIG_DECNET) || defined(CONFIG_DECNET_MODULE)