BSD 4_4_Lite1 release

[unix-history] / usr / src / sys / netinet / tcp_usrreq.c

diff --git a/usr/src/sys/netinet/tcp_usrreq.c b/usr/src/sys/netinet/tcp_usrreq.c
index f82be14..38a08d6 100644 (file)
--- a/usr/src/sys/netinet/tcp_usrreq.c
+++ b/usr/src/sys/netinet/tcp_usrreq.c
@@ -1,33 +1,68 @@
-/*     tcp_usrreq.c    1.69    82/10/30        */
-
-#include "../h/param.h"
-#include "../h/systm.h"
-#include "../h/mbuf.h"
-#include "../h/socket.h"
-#include "../h/socketvar.h"
-#include "../h/protosw.h"
-
-#include "../net/if.h"
-#include "../net/route.h"
-#include "../netinet/in.h"
-#include "../netinet/in_pcb.h"
-#include "../netinet/in_systm.h"
-#include "../netinet/ip.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"
-#include "../netinet/tcp_debug.h"
-#include <errno.h>
+/*
+ * 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.
+ *
+ *     @(#)tcp_usrreq.c        8.2 (Berkeley) 1/3/94
+ */
+
+#include <sys/param.h>
+#include <sys/systm.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 <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/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>
+#include <netinet/tcp_debug.h>

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

-struct tcpcb *tcp_newtcpcb();

 
 /*
  * Process a TCP user request for TCP tb.  If this is a send request
 
 /*
  * Process a TCP user request for TCP tb.  If this is a send request


@@ -35,18 +70,30 @@ struct      tcpcb *tcp_newtcpcb();
  * (called from the software clock routine), then timertype tells which timer.
  */
 /*ARGSUSED*/
  * (called from the software clock routine), then timertype tells which timer.
  */
 /*ARGSUSED*/

-tcp_usrreq(so, req, m, nam, opt)
+int
+tcp_usrreq(so, req, m, nam, control)

        struct socket *so;
        int req;
        struct socket *so;
        int req;

-       struct mbuf *m, *nam;
-       struct socketopt *opt;
+       struct mbuf *m, *nam, *control;

 {
 {

-       register struct inpcb *inp = sotoinpcb(so);
+       register struct inpcb *inp;

        register struct tcpcb *tp;
        register struct tcpcb *tp;

-       int s = splnet();
+       int s;

        int error = 0;
        int ostate;
 
        int error = 0;
        int ostate;
 

+       if (req == PRU_CONTROL)
+               return (in_control(so, (int)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
        /*
         * When a TCP is attached to a socket, then there will be
         * a (struct inpcb) pointed at by the socket, and this


@@ -79,7 +126,7 @@ tcp_usrreq(so, req, m, nam, opt)
                error = tcp_attach(so);
                if (error)
                        break;
                error = tcp_attach(so);
                if (error)
                        break;

-               if ((so->so_options & SO_DONTLINGER) == 0)
+               if ((so->so_options & SO_LINGER) && so->so_linger == 0)

                        so->so_linger = TCP_LINGERTIME;
                tp = sototcpcb(so);
                break;
                        so->so_linger = TCP_LINGERTIME;
                tp = sototcpcb(so);
                break;


@@ -93,11 +140,9 @@ tcp_usrreq(so, req, m, nam, opt)
         */
        case PRU_DETACH:
                if (tp->t_state > TCPS_LISTEN)
         */
        case PRU_DETACH:
                if (tp->t_state > TCPS_LISTEN)

-                       tcp_disconnect(tp);
-               else {
-                       tcp_close(tp);
-                       tp = 0;
-               }
+                       tp = tcp_disconnect(tp);
+               else
+                       tp = tcp_close(tp);

                break;
 
        /*
                break;
 
        /*


@@ -141,14 +186,26 @@ tcp_usrreq(so, req, m, nam, opt)
                        error = ENOBUFS;
                        break;
                }
                        error = ENOBUFS;
                        break;
                }

+               /* Compute window scaling to request.  */
+               while (tp->request_r_scale < TCP_MAX_WINSHIFT &&
+                   (TCP_MAXWIN << tp->request_r_scale) < so->so_rcv.sb_hiwat)
+                       tp->request_r_scale++;

                soisconnecting(so);
                soisconnecting(so);

+               tcpstat.tcps_connattempt++;

                tp->t_state = TCPS_SYN_SENT;
                tp->t_state = TCPS_SYN_SENT;

-               tp->t_timer[TCPT_KEEP] = TCPTV_KEEP;
+               tp->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT;

                tp->iss = tcp_iss; tcp_iss += TCP_ISSINCR/2;
                tcp_sendseqinit(tp);
                error = tcp_output(tp);
                break;
 
                tp->iss = tcp_iss; tcp_iss += TCP_ISSINCR/2;
                tcp_sendseqinit(tp);
                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;
        /*
         * Initiate disconnect from peer.
         * If connection never passed embryonic stage, just drop;


@@ -161,7 +218,7 @@ tcp_usrreq(so, req, m, nam, opt)
         * SHOULD IMPLEMENT LATER PRU_CONNECT VIA REALLOC TCPCB.
         */
        case PRU_DISCONNECT:
         * SHOULD IMPLEMENT LATER PRU_CONNECT VIA REALLOC TCPCB.
         */
        case PRU_DISCONNECT:

-               tcp_disconnect(tp);
+               tp = tcp_disconnect(tp);

                break;
 
        /*
                break;
 
        /*


@@ -169,23 +226,18 @@ tcp_usrreq(so, req, m, nam, opt)
         * done at higher levels; just return the address
         * of the peer, storing through addr.
         */
         * done at higher levels; just return the address
         * of the peer, storing through addr.
         */

-       case PRU_ACCEPT: {
-               struct sockaddr_in *sin = mtod(nam, struct sockaddr_in *);
-
-               nam->m_len = sizeof (struct sockaddr_in);
-               sin->sin_family = AF_INET;
-               sin->sin_port = inp->inp_fport;
-               sin->sin_addr = inp->inp_faddr;
+       case PRU_ACCEPT:
+               in_setpeeraddr(inp, nam);

                break;
                break;

-               }

 
        /*
         * Mark the connection as being incapable of further output.
         */
        case PRU_SHUTDOWN:
                socantsendmore(so);
 
        /*
         * Mark the connection as being incapable of further output.
         */
        case PRU_SHUTDOWN:
                socantsendmore(so);

-               tcp_usrclosed(tp);
-               error = tcp_output(tp);
+               tp = tcp_usrclosed(tp);
+               if (tp)
+                       error = tcp_output(tp);

                break;
 
        /*
                break;
 
        /*


@@ -201,10 +253,6 @@ tcp_usrreq(so, req, m, nam, opt)
         */
        case PRU_SEND:
                sbappend(&so->so_snd, m);
         */
        case PRU_SEND:
                sbappend(&so->so_snd, m);

-#ifdef notdef
-               if (tp->t_flags & TF_PUSH)
-                       tp->snd_end = tp->snd_una + so->so_snd.sb_cc;
-#endif

                error = tcp_output(tp);
                break;
 
                error = tcp_output(tp);
                break;
 


@@ -212,22 +260,19 @@ tcp_usrreq(so, req, m, nam, opt)
         * Abort the TCP.
         */
        case PRU_ABORT:
         * Abort the TCP.
         */
        case PRU_ABORT:

-               tcp_drop(tp, ECONNABORTED);
-               break;
-
-/* SOME AS YET UNIMPLEMENTED HOOKS */
-       case PRU_CONTROL:
-               error = EOPNOTSUPP;
+               tp = tcp_drop(tp, ECONNABORTED);

                break;
 
        case PRU_SENSE:
                break;
 
        case PRU_SENSE:

-               error = EOPNOTSUPP;
-               break;
-/* END UNIMPLEMENTED HOOKS */
+               ((struct stat *) m)->st_blksize = so->so_snd.sb_hiwat;
+               (void) splx(s);
+               return (0);

 
        case PRU_RCVOOB:
 
        case PRU_RCVOOB:

-               if (so->so_oobmark == 0 &&
-                   (so->so_state & SS_RCVATMARK) == 0) {
+               if ((so->so_oobmark == 0 &&
+                   (so->so_state & SS_RCVATMARK) == 0) ||
+                   so->so_options & SO_OOBINLINE ||
+                   tp->t_oobflags & TCPOOB_HADDATA) {

                        error = EINVAL;
                        break;
                }
                        error = EINVAL;
                        break;
                }


@@ -237,15 +282,26 @@ tcp_usrreq(so, req, m, nam, opt)
                }
                m->m_len = 1;
                *mtod(m, caddr_t) = tp->t_iobc;
                }
                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) {
                break;
 
        case PRU_SENDOOB:
                if (sbspace(&so->so_snd) < -512) {

+                       m_freem(m);

                        error = ENOBUFS;
                        break;
                }
                        error = ENOBUFS;
                        break;
                }

-               tp->snd_up = tp->snd_una + so->so_snd.sb_cc + 1;
+               /*
+                * 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);
                sbappend(&so->so_snd, m);

+               tp->snd_up = tp->snd_una + so->so_snd.sb_cc;

                tp->t_force = 1;
                error = tcp_output(tp);
                tp->t_force = 0;
                tp->t_force = 1;
                error = tcp_output(tp);
                tp->t_force = 0;


@@ -255,12 +311,16 @@ tcp_usrreq(so, req, m, nam, opt)
                in_setsockaddr(inp, nam);
                break;
 
                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:
        /*
         * TCP slow timer went off; going through this
         * routine for tracing's sake.
         */
        case PRU_SLOWTIMO:

-               tcp_timers(tp, (int)nam);
+               tp = tcp_timers(tp, (int)nam);

                req |= (int)nam << 8;           /* for debug's sake */
                break;
 
                req |= (int)nam << 8;           /* for debug's sake */
                break;
 


@@ -273,13 +333,94 @@ tcp_usrreq(so, req, m, nam, opt)
        return (error);
 }
 
        return (error);
 }
 

-int    tcp_sendspace = 1024*2;
-int    tcp_recvspace = 1024*2;
+int
+tcp_ctloutput(op, so, level, optname, mp)
+       int op;
+       struct socket *so;
+       int level, optname;
+       struct mbuf **mp;
+{
+       int error = 0, s;
+       struct inpcb *inp;
+       register struct tcpcb *tp;
+       register struct mbuf *m;
+       register int i;
+
+       s = splnet();
+       inp = sotoinpcb(so);
+       if (inp == NULL) {
+               splx(s);
+               if (op == PRCO_SETOPT && *mp)
+                       (void) m_free(*mp);
+               return (ECONNRESET);
+       }
+       if (level != IPPROTO_TCP) {
+               error = ip_ctloutput(op, so, level, optname, mp);
+               splx(s);
+               return (error);
+       }
+       tp = intotcpcb(inp);
+
+       switch (op) {
+
+       case PRCO_SETOPT:
+               m = *mp;
+               switch (optname) {
+
+               case TCP_NODELAY:
+                       if (m == NULL || m->m_len < sizeof (int))
+                               error = EINVAL;
+                       else if (*mtod(m, int *))
+                               tp->t_flags |= TF_NODELAY;
+                       else
+                               tp->t_flags &= ~TF_NODELAY;
+                       break;
+
+               case TCP_MAXSEG:
+                       if (m && (i = *mtod(m, int *)) > 0 && i <= tp->t_maxseg)
+                               tp->t_maxseg = i;
+                       else
+                               error = EINVAL;
+                       break;
+
+               default:
+                       error = ENOPROTOOPT;
+                       break;
+               }
+               if (m)
+                       (void) m_free(m);
+               break;
+
+       case PRCO_GETOPT:
+               *mp = m = m_get(M_WAIT, MT_SOOPTS);
+               m->m_len = sizeof(int);
+
+               switch (optname) {
+               case TCP_NODELAY:
+                       *mtod(m, int *) = tp->t_flags & TF_NODELAY;
+                       break;
+               case TCP_MAXSEG:
+                       *mtod(m, int *) = tp->t_maxseg;
+                       break;
+               default:
+                       error = ENOPROTOOPT;
+                       break;
+               }
+               break;
+       }
+       splx(s);
+       return (error);
+}
+
+u_long tcp_sendspace = 1024*8;
+u_long tcp_recvspace = 1024*8;
+

 /*
  * Attach TCP protocol to socket, allocating
  * internet protocol control block, tcp control block,
  * bufer space, and entering LISTEN state if to accept connections.
  */
 /*
  * Attach TCP protocol to socket, allocating
  * internet protocol control block, tcp control block,
  * bufer space, and entering LISTEN state if to accept connections.
  */

+int

 tcp_attach(so)
        struct socket *so;
 {
 tcp_attach(so)
        struct socket *so;
 {


@@ -287,49 +428,54 @@ tcp_attach(so)
        struct inpcb *inp;
        int error;
 
        struct inpcb *inp;
        int error;
 

-       error = in_pcbreserve(so, tcp_sendspace, tcp_recvspace);
-       if (error)
-               goto bad;
+       if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) {
+               error = soreserve(so, tcp_sendspace, tcp_recvspace);
+               if (error)
+                       return (error);
+       }

        error = in_pcballoc(so, &tcb);
        if (error)
        error = in_pcballoc(so, &tcb);
        if (error)

-               goto bad;
+               return (error);

        inp = sotoinpcb(so);
        tp = tcp_newtcpcb(inp);
        if (tp == 0) {
        inp = sotoinpcb(so);
        tp = tcp_newtcpcb(inp);
        if (tp == 0) {

-               error = ENOBUFS;
-               goto bad2;
+               int nofd = so->so_state & SS_NOFDREF;   /* XXX */
+
+               so->so_state &= ~SS_NOFDREF;    /* don't free the socket yet */
+               in_pcbdetach(inp);
+               so->so_state |= nofd;
+               return (ENOBUFS);

        }
        tp->t_state = TCPS_CLOSED;
        return (0);
        }
        tp->t_state = TCPS_CLOSED;
        return (0);

-bad2:
-       in_pcbdetach(inp);
-bad:
-       return (error);

 }
 
 /*
  * Initiate (or continue) disconnect.
  * If embryonic state, just send reset (once).
 }
 
 /*
  * Initiate (or continue) disconnect.
  * If embryonic state, just send reset (once).

- * If not in ``let data drain'' option, just drop.
+ * If in ``let data drain'' option and linger null, just drop.

  * Otherwise (hard), mark socket disconnecting and drop
  * current input data; switch states based on user close, and
  * send segment to peer (with FIN).
  */
  * Otherwise (hard), mark socket disconnecting and drop
  * current input data; switch states based on user close, and
  * send segment to peer (with FIN).
  */

+struct tcpcb *

 tcp_disconnect(tp)
 tcp_disconnect(tp)

-       struct tcpcb *tp;
+       register struct tcpcb *tp;

 {
        struct socket *so = tp->t_inpcb->inp_socket;
 
        if (tp->t_state < TCPS_ESTABLISHED)
 {
        struct socket *so = tp->t_inpcb->inp_socket;
 
        if (tp->t_state < TCPS_ESTABLISHED)

-               tcp_close(tp);
-       else if (so->so_linger == 0)
-               tcp_drop(tp, 0);
+               tp = tcp_close(tp);
+       else if ((so->so_options & SO_LINGER) && so->so_linger == 0)
+               tp = tcp_drop(tp, 0);

        else {
                soisdisconnecting(so);
                sbflush(&so->so_rcv);
        else {
                soisdisconnecting(so);
                sbflush(&so->so_rcv);

-               tcp_usrclosed(tp);
-               (void) tcp_output(tp);
+               tp = tcp_usrclosed(tp);
+               if (tp)
+                       (void) tcp_output(tp);

        }
        }

+       return (tp);

 }
 
 /*
 }
 
 /*


@@ -342,16 +488,18 @@ tcp_disconnect(tp)
  * for peer to send FIN or not respond to keep-alives, etc.
  * We can let the user exit from the close as soon as the FIN is acked.
  */
  * for peer to send FIN or not respond to keep-alives, etc.
  * We can let the user exit from the close as soon as the FIN is acked.
  */

+struct tcpcb *

 tcp_usrclosed(tp)
 tcp_usrclosed(tp)

-       struct tcpcb *tp;
+       register struct tcpcb *tp;

 {
 
        switch (tp->t_state) {
 
 {
 
        switch (tp->t_state) {
 

+       case TCPS_CLOSED:

        case TCPS_LISTEN:
        case TCPS_SYN_SENT:
                tp->t_state = TCPS_CLOSED;
        case TCPS_LISTEN:
        case TCPS_SYN_SENT:
                tp->t_state = TCPS_CLOSED;

-               tcp_close(tp);
+               tp = tcp_close(tp);

                break;
 
        case TCPS_SYN_RECEIVED:
                break;
 
        case TCPS_SYN_RECEIVED:


@@ -363,6 +511,7 @@ tcp_usrclosed(tp)
                tp->t_state = TCPS_LAST_ACK;
                break;
        }
                tp->t_state = TCPS_LAST_ACK;
                break;
        }

-       if (tp->t_state >= TCPS_FIN_WAIT_2)
+       if (tp && tp->t_state >= TCPS_FIN_WAIT_2)

                soisdisconnected(tp->t_inpcb->inp_socket);
                soisdisconnected(tp->t_inpcb->inp_socket);

+       return (tp);

 }
 }