BSD 4_3_Reno release

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

diff --git a/usr/src/sys/netinet/tcp_output.c b/usr/src/sys/netinet/tcp_output.c
index 38f3d17..aef4809 100644 (file)
--- a/usr/src/sys/netinet/tcp_output.c
+++ b/usr/src/sys/netinet/tcp_output.c
@@ -1,24 +1,28 @@
 /*
 /*

- * Copyright (c) 1982, 1986 Regents of the University of California.
+ * Copyright (c) 1982, 1986, 1988, 1990 Regents of the University of California.

  * All rights reserved.
  *
  * All rights reserved.
  *

- * Redistribution and use in source and binary forms are permitted
- * provided that the above copyright notice and this paragraph are
- * duplicated in all such forms and that any documentation,
- * advertising materials, and other materials related to such
- * distribution and use acknowledge that the software was developed
- * by the University of California, Berkeley.  The name of the
- * University may not be used to endorse or promote products derived
- * from this software without specific prior written permission.
- * THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR
- * IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED
- * WARRANTIES OF MERCHANTIBILITY AND FITNESS FOR A PARTICULAR PURPOSE.
+ * Redistribution is only permitted until one year after the first shipment
+ * of 4.4BSD by the Regents.  Otherwise, redistribution and use in source and
+ * binary forms are permitted provided that: (1) source distributions retain
+ * this entire copyright notice and comment, and (2) distributions including
+ * binaries display the following acknowledgement:  This product includes
+ * software developed by the University of California, Berkeley and its
+ * contributors'' in the documentation or other materials provided with the
+ * distribution and in all advertising materials mentioning features or use
+ * of this software.  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 AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
+ * WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

  *
  *

- *     @(#)tcp_output.c        7.17 (Berkeley) 6/29/88
+ *     @(#)tcp_output.c        7.21 (Berkeley) 6/28/90

  */
 
 #include "param.h"
 #include "systm.h"
  */
 
 #include "param.h"
 #include "systm.h"

+#include "malloc.h"

 #include "mbuf.h"
 #include "protosw.h"
 #include "socket.h"
 #include "mbuf.h"
 #include "protosw.h"
 #include "socket.h"


@@ -28,9 +32,9 @@
 #include "../net/route.h"
 
 #include "in.h"
 #include "../net/route.h"
 
 #include "in.h"

-#include "in_pcb.h"

 #include "in_systm.h"
 #include "ip.h"
 #include "in_systm.h"
 #include "ip.h"

+#include "in_pcb.h"

 #include "ip_var.h"
 #include "tcp.h"
 #define        TCPOUTFLAGS
 #include "ip_var.h"
 #include "tcp.h"
 #define        TCPOUTFLAGS


@@ -41,6 +45,10 @@
 #include "tcpip.h"
 #include "tcp_debug.h"
 
 #include "tcpip.h"
 #include "tcp_debug.h"
 

+#ifdef notyet
+extern struct mbuf *m_copypack();
+#endif
+

 /*
  * Initial options.
  */
 /*
  * Initial options.
  */


@@ -54,12 +62,11 @@ tcp_output(tp)
 {
        register struct socket *so = tp->t_inpcb->inp_socket;
        register long len, win;
 {
        register struct socket *so = tp->t_inpcb->inp_socket;
        register long len, win;

-       struct mbuf *m0;

        int off, flags, error;
        register struct mbuf *m;
        register struct tcpiphdr *ti;
        u_char *opt;
        int off, flags, error;
        register struct mbuf *m;
        register struct tcpiphdr *ti;
        u_char *opt;

-       unsigned optlen = 0;
+       unsigned optlen, hdrlen;

        int idle, sendalot;
 
        /*
        int idle, sendalot;
 
        /*


@@ -69,10 +76,17 @@ tcp_output(tp)
         * to send, then transmit; otherwise, investigate further.
         */
        idle = (tp->snd_max == tp->snd_una);
         * to send, then transmit; otherwise, investigate further.
         */
        idle = (tp->snd_max == tp->snd_una);

+       if (idle && tp->t_idle >= tp->t_rxtcur)
+               /*
+                * We have been idle for "a while" and no acks are
+                * expected to clock out any data we send --
+                * slow start to get ack "clock" running again.
+                */
+               tp->snd_cwnd = tp->t_maxseg;

 again:
        sendalot = 0;
        off = tp->snd_nxt - tp->snd_una;
 again:
        sendalot = 0;
        off = tp->snd_nxt - tp->snd_una;

-       win = MIN(tp->snd_wnd, tp->snd_cwnd);
+       win = min(tp->snd_wnd, tp->snd_cwnd);

 
        /*
         * If in persist timeout with window of 0, send 1 byte.
 
        /*
         * If in persist timeout with window of 0, send 1 byte.


@@ -89,8 +103,8 @@ again:
                }
        }
 
                }
        }
 

-       len = MIN(so->so_snd.sb_cc, win) - off;

        flags = tcp_outflags[tp->t_state];
        flags = tcp_outflags[tp->t_state];

+       len = min(so->so_snd.sb_cc, win) - off;

 
        if (len < 0) {
                /*
 
        if (len < 0) {
                /*


@@ -115,26 +129,8 @@ again:
        }
        if (SEQ_LT(tp->snd_nxt + len, tp->snd_una + so->so_snd.sb_cc))
                flags &= ~TH_FIN;
        }
        if (SEQ_LT(tp->snd_nxt + len, tp->snd_una + so->so_snd.sb_cc))
                flags &= ~TH_FIN;

-       win = sbspace(&so->so_rcv);
-

 
 

-       /*
-        * If our state indicates that FIN should be sent
-        * and we have not yet done so, or we're retransmitting the FIN,
-        * then we need to send.
-        */
-       if (flags & TH_FIN &&
-           ((tp->t_flags & TF_SENTFIN) == 0 || tp->snd_nxt == tp->snd_una))
-               goto send;
-       /*
-        * Send if we owe peer an ACK.
-        */
-       if (tp->t_flags & TF_ACKNOW)
-               goto send;
-       if (flags & (TH_SYN|TH_RST))
-               goto send;
-       if (SEQ_GT(tp->snd_up, tp->snd_una))
-               goto send;
+       win = sbspace(&so->so_rcv);

 
        /*
         * Sender silly window avoidance.  If connection is idle
 
        /*
         * Sender silly window avoidance.  If connection is idle


@@ -164,18 +160,39 @@ again:
         * Compare available window to amount of window
         * known to peer (as advertised window less
         * next expected input).  If the difference is at least two
         * Compare available window to amount of window
         * known to peer (as advertised window less
         * next expected input).  If the difference is at least two

-        * max size segments or at least 35% of the maximum possible
+        * max size segments, or at least 50% of the maximum possible

         * window, then want to send a window update to peer.
         */
        if (win > 0) {
                int adv = win - (tp->rcv_adv - tp->rcv_nxt);
 
         * window, then want to send a window update to peer.
         */
        if (win > 0) {
                int adv = win - (tp->rcv_adv - tp->rcv_nxt);
 

-               if (so->so_rcv.sb_cc == 0 && adv >= 2 * tp->t_maxseg)
+               /* this was:                                    XXX
+                * if (so->so_rcv.sb_cc == 0 && adv >= 2 * tp->t_maxseg)
+                */
+               if (adv >= 2 * tp->t_maxseg)

                        goto send;
                        goto send;

-               if (100 * adv / so->so_rcv.sb_hiwat >= 35)
+               if (2 * adv >= so->so_rcv.sb_hiwat)

                        goto send;
        }
 
                        goto send;
        }
 

+       /*
+        * Send if we owe peer an ACK.
+        */
+       if (tp->t_flags & TF_ACKNOW)
+               goto send;
+       if (flags & (TH_SYN|TH_RST))
+               goto send;
+       if (SEQ_GT(tp->snd_up, tp->snd_una))
+               goto send;
+       /*
+        * If our state indicates that FIN should be sent
+        * and we have not yet done so, or we're retransmitting the FIN,
+        * then we need to send.
+        */
+       if (flags & TH_FIN &&
+           ((tp->t_flags & TF_SENTFIN) == 0 || tp->snd_nxt == tp->snd_una))
+               goto send;
+

        /*
         * TCP window updates are not reliable, rather a polling protocol
         * using ``persist'' packets is used to insure receipt of window
        /*
         * TCP window updates are not reliable, rather a polling protocol
         * using ``persist'' packets is used to insure receipt of window


@@ -210,19 +227,32 @@ again:
        return (0);
 
 send:
        return (0);
 
 send:

+       /*
+        * Before ESTABLISHED, force sending of initial options
+        * unless TCP set not to do any options.
+        * NOTE: we assume that the IP/TCP header plus TCP options
+        * always fit in a single mbuf, leaving room for a maximum
+        * link header, i.e.
+        *      max_linkhdr + sizeof (struct tcpiphdr) + optlen <= MHLEN
+        */
+       optlen = 0;
+       hdrlen = sizeof (struct tcpiphdr);
+       if (flags & TH_SYN && (tp->t_flags & TF_NOOPT) == 0) {
+               opt = tcp_initopt;
+               optlen = sizeof (tcp_initopt);
+               hdrlen += sizeof (tcp_initopt);
+               *(u_short *)(opt + 2) = htons((u_short) tcp_mss(tp, 0));
+#ifdef DIAGNOSTIC
+               if (max_linkhdr + hdrlen > MHLEN)
+                       panic("tcphdr too big");
+#endif
+       }
+

        /*
         * Grab a header mbuf, attaching a copy of data to
         * be transmitted, and initialize the header from
         * the template for sends on this connection.
         */
        /*
         * Grab a header mbuf, attaching a copy of data to
         * be transmitted, and initialize the header from
         * the template for sends on this connection.
         */

-       MGET(m, M_DONTWAIT, MT_HEADER);
-       if (m == NULL)
-               return (ENOBUFS);
-#define        MAXLINKHDR      32              /* belongs elsewhere */
-#define        DATASPACE  (MMAXOFF - (MMINOFF + MAXLINKHDR + sizeof (struct tcpiphdr)))
-       m->m_off = MMINOFF + MAXLINKHDR;
-       m->m_len = sizeof (struct tcpiphdr);
-       ti = mtod(m, struct tcpiphdr *);

        if (len) {
                if (tp->t_force && len == 1)
                        tcpstat.tcps_sndprobe++;
        if (len) {
                if (tp->t_force && len == 1)
                        tcpstat.tcps_sndprobe++;


@@ -233,24 +263,63 @@ send:
                        tcpstat.tcps_sndpack++;
                        tcpstat.tcps_sndbyte += len;
                }
                        tcpstat.tcps_sndpack++;
                        tcpstat.tcps_sndbyte += len;
                }

-               if (len <= DATASPACE) {
+#ifdef notyet
+               if ((m = m_copypack(so->so_snd.sb_mb, off,
+                   (int)len, max_linkhdr + hdrlen)) == 0) {
+                       error = ENOBUFS;
+                       goto out;
+               }
+               /*
+                * m_copypack left space for our hdr; use it.
+                */
+               m->m_len += hdrlen;
+               m->m_data -= hdrlen;
+#else
+               MGETHDR(m, M_DONTWAIT, MT_HEADER);
+               if (m == NULL) {
+                       error = ENOBUFS;
+                       goto out;
+               }
+               m->m_data += max_linkhdr;
+               m->m_len = hdrlen;
+               if (len <= MHLEN - hdrlen - max_linkhdr) {

                        m_copydata(so->so_snd.sb_mb, off, (int) len,
                        m_copydata(so->so_snd.sb_mb, off, (int) len,

-                           mtod(m, caddr_t) + sizeof(struct tcpiphdr));
+                           mtod(m, caddr_t) + hdrlen);

                        m->m_len += len;
                } else {
                        m->m_next = m_copy(so->so_snd.sb_mb, off, (int) len);
                        if (m->m_next == 0)
                                len = 0;
                }
                        m->m_len += len;
                } else {
                        m->m_next = m_copy(so->so_snd.sb_mb, off, (int) len);
                        if (m->m_next == 0)
                                len = 0;
                }

-       } else if (tp->t_flags & TF_ACKNOW)
-               tcpstat.tcps_sndacks++;
-       else if (flags & (TH_SYN|TH_FIN|TH_RST))
-               tcpstat.tcps_sndctrl++;
-       else if (SEQ_GT(tp->snd_up, tp->snd_una))
-               tcpstat.tcps_sndurg++;
-       else
-               tcpstat.tcps_sndwinup++;
+#endif
+               /*
+                * If we're sending everything we've got, set PUSH.
+                * (This will keep happy those implementations which only
+                * give data to the user when a buffer fills or
+                * a PUSH comes in.)
+                */
+               if (off + len == so->so_snd.sb_cc)
+                       flags |= TH_PUSH;
+       } else {
+               if (tp->t_flags & TF_ACKNOW)
+                       tcpstat.tcps_sndacks++;
+               else if (flags & (TH_SYN|TH_FIN|TH_RST))
+                       tcpstat.tcps_sndctrl++;
+               else if (SEQ_GT(tp->snd_up, tp->snd_una))
+                       tcpstat.tcps_sndurg++;
+               else
+                       tcpstat.tcps_sndwinup++;

 
 

+               MGETHDR(m, M_DONTWAIT, MT_HEADER);
+               if (m == NULL) {
+                       error = ENOBUFS;
+                       goto out;
+               }
+               m->m_data += max_linkhdr;
+               m->m_len = hdrlen;
+       }
+       m->m_pkthdr.rcvif = (struct ifnet *)0;
+       ti = mtod(m, struct tcpiphdr *);

        if (tp->t_template == 0)
                panic("tcp_output");
        bcopy((caddr_t)tp->t_template, (caddr_t)ti, sizeof (struct tcpiphdr));
        if (tp->t_template == 0)
                panic("tcp_output");
        bcopy((caddr_t)tp->t_template, (caddr_t)ti, sizeof (struct tcpiphdr));


@@ -265,39 +334,8 @@ send:
                tp->snd_nxt--;
        ti->ti_seq = htonl(tp->snd_nxt);
        ti->ti_ack = htonl(tp->rcv_nxt);
                tp->snd_nxt--;
        ti->ti_seq = htonl(tp->snd_nxt);
        ti->ti_ack = htonl(tp->rcv_nxt);

-       /*
-        * Before ESTABLISHED, force sending of initial options
-        * unless TCP set to not do any options.
-        */
-       opt = NULL;
-       if (flags & TH_SYN && (tp->t_flags & TF_NOOPT) == 0) {
-               u_short mss;
-
-               mss = MIN(so->so_rcv.sb_hiwat / 2, tcp_mss(tp));
-               if (mss > IP_MSS - sizeof(struct tcpiphdr)) {
-                       opt = tcp_initopt;
-                       optlen = sizeof (tcp_initopt);
-                       *(u_short *)(opt + 2) = htons(mss);
-               }
-       }
-       if (opt) {
-               m0 = m->m_next;
-               m->m_next = m_get(M_DONTWAIT, MT_DATA);
-               if (m->m_next == 0) {
-                       (void) m_free(m);
-                       m_freem(m0);
-                       return (ENOBUFS);
-               }
-               m->m_next->m_next = m0;
-               m0 = m->m_next;
-               m0->m_len = optlen;
-               bcopy((caddr_t)opt, mtod(m0, caddr_t), optlen);
-               opt = (u_char *)(mtod(m0, caddr_t) + optlen);
-               while (m0->m_len & 0x3) {
-                       *opt++ = TCPOPT_EOL;
-                       m0->m_len++;
-               }
-               optlen = m0->m_len;
+       if (optlen) {
+               bcopy((caddr_t)opt, (caddr_t)(ti + 1), optlen);

                ti->ti_off = (sizeof (struct tcphdr) + optlen) >> 2;
        }
        ti->ti_flags = flags;
                ti->ti_off = (sizeof (struct tcphdr) + optlen) >> 2;
        }
        ti->ti_flags = flags;


@@ -307,8 +345,8 @@ send:
         */
        if (win < (long)(so->so_rcv.sb_hiwat / 4) && win < (long)tp->t_maxseg)
                win = 0;
         */
        if (win < (long)(so->so_rcv.sb_hiwat / 4) && win < (long)tp->t_maxseg)
                win = 0;

-       if (win > IP_MAXPACKET)
-               win = IP_MAXPACKET;
+       if (win > TCP_MAXWIN)
+               win = TCP_MAXWIN;

        if (win < (long)(tp->rcv_adv - tp->rcv_nxt))
                win = (long)(tp->rcv_adv - tp->rcv_nxt);
        ti->ti_win = htons((u_short)win);
        if (win < (long)(tp->rcv_adv - tp->rcv_nxt))
                win = (long)(tp->rcv_adv - tp->rcv_nxt);
        ti->ti_win = htons((u_short)win);


@@ -323,23 +361,15 @@ send:
                 * number wraparound.
                 */
                tp->snd_up = tp->snd_una;               /* drag it along */
                 * number wraparound.
                 */
                tp->snd_up = tp->snd_una;               /* drag it along */

-       /*
-        * If anything to send and we can send it all, set PUSH.
-        * (This will keep happy those implementations which only
-        * give data to the user when a buffer fills or a PUSH comes in.)
-        */
-       if (len && off+len == so->so_snd.sb_cc)
-               ti->ti_flags |= TH_PUSH;

 
        /*
         * Put TCP length in extended header, and then
         * checksum extended header and data.
         */
        if (len + optlen)
 
        /*
         * Put TCP length in extended header, and then
         * checksum extended header and data.
         */
        if (len + optlen)

-               ti->ti_len = htons((u_short)(sizeof(struct tcphdr) +
+               ti->ti_len = htons((u_short)(sizeof (struct tcphdr) +

                    optlen + len));
                    optlen + len));

-       ti->ti_sum = in_cksum(m,
-           (int)(sizeof (struct tcpiphdr) + (int)optlen + len));
+       ti->ti_sum = in_cksum(m, (int)(hdrlen + len));

 
        /*
         * In transmit state, time the transmission and arrange for
 
        /*
         * In transmit state, time the transmission and arrange for


@@ -351,11 +381,13 @@ send:
                /*
                 * Advance snd_nxt over sequence space of this segment.
                 */
                /*
                 * Advance snd_nxt over sequence space of this segment.
                 */

-               if (flags & TH_SYN)
-                       tp->snd_nxt++;
-               if (flags & TH_FIN) {
-                       tp->snd_nxt++;
-                       tp->t_flags |= TF_SENTFIN;
+               if (flags & (TH_SYN|TH_FIN)) {
+                       if (flags & TH_SYN)
+                               tp->snd_nxt++;
+                       if (flags & TH_FIN) {
+                               tp->snd_nxt++;
+                               tp->t_flags |= TF_SENTFIN;
+                       }

                }
                tp->snd_nxt += len;
                if (SEQ_GT(tp->snd_nxt, tp->snd_max)) {
                }
                tp->snd_nxt += len;
                if (SEQ_GT(tp->snd_nxt, tp->snd_max)) {


@@ -399,17 +431,32 @@ send:
 
        /*
         * Fill in IP length and desired time to live and
 
        /*
         * Fill in IP length and desired time to live and

-        * send to IP level.
+        * send to IP level.  There should be a better way
+        * to handle ttl and tos; we could keep them in
+        * the template, but need a way to checksum without them.

         */
         */

-       ((struct ip *)ti)->ip_len = sizeof (struct tcpiphdr) + optlen + len;
-       ((struct ip *)ti)->ip_ttl = TCP_TTL;
+       m->m_pkthdr.len = hdrlen + len;
+       ((struct ip *)ti)->ip_len = m->m_pkthdr.len;
+       ((struct ip *)ti)->ip_ttl = tp->t_inpcb->inp_ip.ip_ttl; /* XXX */
+       ((struct ip *)ti)->ip_tos = tp->t_inpcb->inp_ip.ip_tos; /* XXX */
+#if BSD >= 43

        error = ip_output(m, tp->t_inpcb->inp_options, &tp->t_inpcb->inp_route,
            so->so_options & SO_DONTROUTE);
        error = ip_output(m, tp->t_inpcb->inp_options, &tp->t_inpcb->inp_route,
            so->so_options & SO_DONTROUTE);

+#else
+       error = ip_output(m, (struct mbuf *)0, &tp->t_inpcb->inp_route, 
+           so->so_options & SO_DONTROUTE);
+#endif

        if (error) {
        if (error) {

+out:

                if (error == ENOBUFS) {
                        tcp_quench(tp->t_inpcb);
                        return (0);
                }
                if (error == ENOBUFS) {
                        tcp_quench(tp->t_inpcb);
                        return (0);
                }

+               if ((error == EHOSTUNREACH || error == ENETDOWN)
+                   && TCPS_HAVERCVDSYN(tp->t_state)) {
+                       tp->t_softerror = error;
+                       return (0);
+               }

                return (error);
        }
        tcpstat.tcps_sndtotal++;
                return (error);
        }
        tcpstat.tcps_sndtotal++;