tcp_usrreq.c

ReactOS是一些高手根据Windows XP的内核编写出的类XP。内核实现机理和API函数调用几乎相同。甚至可以兼容XP的程序。喜欢研究系统内核的人可以看一看。

/*
 * Copyright (c) 1982, 1986, 1988, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	From: @(#)tcp_usrreq.c	8.2 (Berkeley) 1/3/94
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>
#include <sys/errno.h>
#include <sys/stat.h>
#include <vm/vm.h>
#include <sys/sysctl.h>

#include <net/if.h>
#include <net/route.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/in_pcb.h>
#include <netinet/in_var.h>
#include <netinet/ip_var.h>
#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_seq.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcpip.h>
#ifdef TCPDEBUG
#include <netinet/tcp_debug.h>
#endif

/*
 * TCP protocol interface to socket abstraction.
 */
extern	char *tcpstates[];

/*
 * Process a TCP user request for TCP tb.  If this is a send request
 * then m is the mbuf chain of send data.  If this is a timer expiration
 * (called from the software clock routine), then timertype tells which timer.
 */
/*ARGSUSED*/
int
tcp_usrreq(so, req, m, nam, control)
	struct socket *so;
	int req;
	struct mbuf *m, *nam, *control;
{
	register struct inpcb *inp;
	register struct tcpcb *tp = 0;
	struct sockaddr_in *sinp;
	int s;
	int error = 0;
#ifdef TCPDEBUG
	int ostate;
#endif

	if (req == PRU_CONTROL)
		return (in_control(so, (u_long)m, (caddr_t)nam,
			(struct ifnet *)control));
	if (control && control->m_len) {
		m_freem(control);
		if (m)
			m_freem(m);
		return (EINVAL);
	}

	s = splnet();
	inp = sotoinpcb(so);
	/*
	 * When a TCP is attached to a socket, then there will be
	 * a (struct inpcb) pointed at by the socket, and this
	 * structure will point at a subsidary (struct tcpcb).
	 */
	if (inp == 0 && req != PRU_ATTACH) {
		splx(s);
#if 0
		/*
		 * The following corrects an mbuf leak under rare
		 * circumstances, but has not been fully tested.
		 */
		if (m && req != PRU_SENSE)
			m_freem(m);
#else
		/* safer version of fix for mbuf leak */
		if (m && (req == PRU_SEND || req == PRU_SENDOOB))
			m_freem(m);
#endif
		return (EINVAL);		/* XXX */
	}
	if (inp) {
		tp = intotcpcb(inp);
		/* WHAT IF TP IS 0? */
#ifdef KPROF
		tcp_acounts[tp->t_state][req]++;
#endif
#ifdef TCPDEBUG
		ostate = tp->t_state;
	} else
		ostate = 0;
#else /* TCPDEBUG */
	}
#endif /* TCPDEBUG */

	switch (req) {

	/*
	 * TCP attaches to socket via PRU_ATTACH, reserving space,
	 * and an internet control block.
	 */
	case PRU_ATTACH:
		if (inp) {
			error = EISCONN;
			break;
		}
		error = tcp_attach(so);
		if (error)
			break;
		if ((so->so_options & SO_LINGER) && so->so_linger == 0)
			so->so_linger = TCP_LINGERTIME * hz;
		tp = sototcpcb(so);
		break;

	/*
	 * PRU_DETACH detaches the TCP protocol from the socket.
	 * If the protocol state is non-embryonic, then can't
	 * do this directly: have to initiate a PRU_DISCONNECT,
	 * which may finish later; embryonic TCB's can just
	 * be discarded here.
	 */
	case PRU_DETACH:
		if (tp->t_state > TCPS_LISTEN)
			tp = tcp_disconnect(tp);
		else
			tp = tcp_close(tp);
		break;

	/*
	 * Give the socket an address.
	 */
	case PRU_BIND:
		/*
		 * Must check for multicast addresses and disallow binding
		 * to them.
		 */
		sinp = mtod(nam, struct sockaddr_in *);
		if (sinp->sin_family == AF_INET &&
		    IN_MULTICAST(ntohl(sinp->sin_addr.s_addr))) {
			error = EAFNOSUPPORT;
			break;
		}
		error = in_pcbbind(inp, nam);
		if (error)
			break;
		break;

	/*
	 * Prepare to accept connections.
	 */
	case PRU_LISTEN:
		if (inp->inp_lport == 0)
			error = in_pcbbind(inp, NULL);
		if (error == 0)
			tp->t_state = TCPS_LISTEN;
		break;

	/*
	 * Initiate connection to peer.
	 * Create a template for use in transmissions on this connection.
	 * Enter SYN_SENT state, and mark socket as connecting.
	 * Start keep-alive timer, and seed output sequence space.
	 * Send initial segment on connection.
	 */
	case PRU_CONNECT:
		/*
		 * Must disallow TCP ``connections'' to multicast addresses.
		 */
		sinp = mtod(nam, struct sockaddr_in *);
		if (sinp->sin_family == AF_INET
		    && IN_MULTICAST(ntohl(sinp->sin_addr.s_addr))) {
			error = EAFNOSUPPORT;
			break;
		}

		if ((error = tcp_connect(tp, nam)) != 0)
			break;
		error = tcp_output(tp);
		break;

	/*
	 * Create a TCP connection between two sockets.
	 */
	case PRU_CONNECT2:
		error = EOPNOTSUPP;
		break;

	/*
	 * Initiate disconnect from peer.
	 * If connection never passed embryonic stage, just drop;
	 * else if don't need to let data drain, then can just drop anyways,
	 * else have to begin TCP shutdown process: mark socket disconnecting,
	 * drain unread data, state switch to reflect user close, and
	 * send segment (e.g. FIN) to peer.  Socket will be really disconnected
	 * when peer sends FIN and acks ours.
	 *
	 * SHOULD IMPLEMENT LATER PRU_CONNECT VIA REALLOC TCPCB.
	 */
	case PRU_DISCONNECT:
		tp = tcp_disconnect(tp);
		break;

	/*
	 * Accept a connection.  Essentially all the work is
	 * done at higher levels; just return the address
	 * of the peer, storing through addr.
	 */
	case PRU_ACCEPT:
		in_setpeeraddr(inp, nam);
		break;

	/*
	 * Mark the connection as being incapable of further output.
	 */
	case PRU_SHUTDOWN:
		socantsendmore(so);
		tp = tcp_usrclosed(tp);
		if (tp)
			error = tcp_output(tp);
		break;

	/*
	 * After a receive, possibly send window update to peer.
	 */
	case PRU_RCVD:
		(void) tcp_output(tp);
		break;

	/*
	 * Do a send by putting data in output queue and updating urgent
	 * marker if URG set.  Possibly send more data.
	 */
	case PRU_SEND_EOF:
	case PRU_SEND:
		sbappend(&so->so_snd, m);
		if (nam && tp->t_state < TCPS_SYN_SENT) {
			/*
			 * Do implied connect if not yet connected,
			 * initialize window to default value, and
			 * initialize maxseg/maxopd using peer's cached
			 * MSS.
			 */
			error = tcp_connect(tp, nam);
			if (error)
				break;
			tp->snd_wnd = TTCP_CLIENT_SND_WND;
			tcp_mss(tp, -1);
		}

		if (req == PRU_SEND_EOF) {
			/*
			 * Close the send side of the connection after
			 * the data is sent.
			 */
			socantsendmore(so);
			tp = tcp_usrclosed(tp);
		}
		if (tp != NULL)
			error = tcp_output(tp);
		break;

	/*
	 * Abort the TCP.
	 */
	case PRU_ABORT:
		tp = tcp_drop(tp, ECONNABORTED);
		break;

	case PRU_SENSE:
		((struct stat *) m)->st_blksize = so->so_snd.sb_hiwat;
		(void) splx(s);
		return (0);

	case PRU_RCVOOB:
		if ((so->so_oobmark == 0 &&
		    (so->so_state & SS_RCVATMARK) == 0) ||
		    so->so_options & SO_OOBINLINE ||
		    tp->t_oobflags & TCPOOB_HADDATA) {
			error = EINVAL;
			break;
		}
		if ((tp->t_oobflags & TCPOOB_HAVEDATA) == 0) {
			error = EWOULDBLOCK;
			break;
		}
		m->m_len = 1;
		*mtod(m, caddr_t) = tp->t_iobc;
		if (((int)nam & MSG_PEEK) == 0)
			tp->t_oobflags ^= (TCPOOB_HAVEDATA | TCPOOB_HADDATA);
		break;

	case PRU_SENDOOB:
		if (sbspace(&so->so_snd) < -512) {
			m_freem(m);
			error = ENOBUFS;
			break;
		}
		/*
		 * According to RFC961 (Assigned Protocols),
		 * the urgent pointer points to the last octet
		 * of urgent data.  We continue, however,
		 * to consider it to indicate the first octet
		 * of data past the urgent section.
		 * Otherwise, snd_up should be one lower.
		 */
		sbappend(&so->so_snd, m);
		if (nam && tp->t_state < TCPS_SYN_SENT) {
			/*
			 * Do implied connect if not yet connected,
			 * initialize window to default value, and
			 * initialize maxseg/maxopd using peer's cached
			 * MSS.
			 */
			error = tcp_connect(tp, nam);
			if (error)
				break;
			tp->snd_wnd = TTCP_CLIENT_SND_WND;
			tcp_mss(tp, -1);
		}
		tp->snd_up = tp->snd_una + so->so_snd.sb_cc;
		tp->t_force = 1;
		error = tcp_output(tp);
		tp->t_force = 0;
		break;

	case PRU_SOCKADDR:
		in_setsockaddr(inp, nam);
		break;

	case PRU_PEERADDR:
		in_setpeeraddr(inp, nam);
		break;

	/*
	 * TCP slow timer went off; going through this
	 * routine for tracing's sake.
	 */
	case PRU_SLOWTIMO:
		tp = tcp_timers(tp, (int)nam);
#ifdef TCPDEBUG
		req |= (int)nam << 8;		/* for debug's sake */
#endif
		break;