duplicate TIOCGPGRP for controller - minus the check for controlling terminal

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

diff --git a/usr/src/sys/netinet/ip_output.c b/usr/src/sys/netinet/ip_output.c
index b51bc7b..cd844b8 100644 (file)
--- a/usr/src/sys/netinet/ip_output.c
+++ b/usr/src/sys/netinet/ip_output.c
@@ -1,200 +1,496 @@
-/* ip_output.c 1.1 81/10/14 */
-#include "../h/param.h"
-#include "../bbnnet/net.h"
-#include "../bbnnet/tcp.h"
-#include "../bbnnet/ip.h"
-#include "../bbnnet/imp.h"
-#include "../bbnnet/ucb.h"
-
-/*****************************************************************************
-*                                                                            *
-*         internet level output:  called from higher level protocol          *
-*         or "raw internet driver."  passed a pointer to an mbuf             *
-*         chain containing the message to be sent, a partially filled        *
-*         in ip leader, and room for an 1822 leader and 2 pointers.          *
-*         this routine does fragmentation and mapping of ip parameters       *
-*         to 1822 ones.                                                      *
-*                                                                            *
-*****************************************************************************/
-
-ip_output(mp)
-struct mbuf *mp;
-{
-       register i, rnd;
-       register struct mbuf *m, *n;
-       register struct ip *p;
-       struct mbuf *mm;
-       int hlen, adj, max, len, off;
-
-COUNT(IP_OUTPUT);
-       p = (struct ip *)((int)mp + mp->m_off); /* -> ip header */
-       hlen = sizeof(struct ip);               /* header length */
-
-       /* fill in unspecified fields and byte swap others */
-
-       p->ip_v = IPVERSION;    
-       p->ip_hl = hlen >> 2;
-       p->ip_off = 0 | (p->ip_off & ip_df);
-       p->ip_ttl = MAXTTL;
-       p->ip_id = netcb.n_ip_cnt++;
-
-       if (p->ip_len > MTU) {          /* must fragment */
-               if (p->ip_off & ip_df)
-                       return(FALSE);
-               max = MTU - hlen;       /* maximum data length in fragment */
-               len = p->ip_len - hlen; /* data length */
-               off = 0;                /* fragment offset */
-               m = mp;
-
-               while (len > 0) {
-
-                       /* correct the header */
-
-                       p->ip_off |= off >> 3;
-
-                       /* find the end of the fragment */
-
-                       i = -hlen;
-                       while (m != NULL) {
-                               i += m->m_len;
-                               if (i > max)
-                                       break;
-                               n = m;
-                               m = m->m_next;
-                       }
-
-                       if (i < max || m == NULL) {     /* last fragment */
-                               p->ip_off = p->ip_off & ~ip_mf;
-                               p->ip_len = i + hlen;
-                               break;
-
-                       } else {                        /* more fragments */
-
-                               /* allocate header mbuf for next fragment */
-
-                               if ((mm = m_get(1)) == NULL)    /* no more bufs */
-                                       return(FALSE);
-
-                               p->ip_off |= ip_mf;
-
-                               /* terminate fragment at 8 byte boundary (round down) */
-
-                               i -= m->m_len;
-                               rnd = i & ~7;           /* fragment length */
-                               adj = i - rnd;          /* leftover in mbuf */
-                               p->ip_len = rnd + hlen; 
+/*
+ * Copyright (c) 1982, 1986, 1988 Regents of the University of California.
+ * 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.
+ *
+ *     @(#)ip_output.c 7.18 (Berkeley) %G%
+ */

 
 

-                               /* setup header for next fragment and 
-                                  append remaining fragment data */
+#include "param.h"
+#include "malloc.h"
+#include "mbuf.h"
+#include "errno.h"
+#include "protosw.h"
+#include "socket.h"
+#include "socketvar.h"

 
 

-                               n->m_next = NULL;                   
-                               mm->m_next = m;        
-                               m = mm;
-                               m->m_off = MSIZE - hlen - adj;
-                               m->m_len = hlen + adj;
+#include "../net/if.h"
+#include "../net/route.h"

 
 

-                               /* copy old header to new */
+#include "in.h"
+#include "in_systm.h"
+#include "ip.h"
+#include "in_pcb.h"
+#include "in_var.h"
+#include "ip_var.h"

 
 

-                               bcopy(p, (caddr_t)((int)m + m->m_off), hlen);
+#ifdef vax
+#include "machine/mtpr.h"
+#endif

 
 

-                               /* copy leftover data from previous frag */
+struct mbuf *ip_insertoptions();

 
 

-                               if (adj) {
-                                       n->m_len -= adj;
-                                       bcopy((caddr_t)((int)n + n->m_len + n->m_off),
-                                             (caddr_t)((int)m + m->m_off + hlen), adj);
-                               }
-                       }
+/*
+ * IP output.  The packet in mbuf chain m contains a skeletal IP
+ * header (with len, off, ttl, proto, tos, src, dst).
+ * The mbuf chain containing the packet will be freed.
+ * The mbuf opt, if present, will not be freed.
+ */
+ip_output(m0, opt, ro, flags)
+       struct mbuf *m0;
+       struct mbuf *opt;
+       struct route *ro;
+       int flags;
+{
+       register struct ip *ip, *mhip;
+       register struct ifnet *ifp;
+       register struct mbuf *m = m0;
+       register int hlen = sizeof (struct ip);
+       int len, off, error = 0;
+       struct route iproute;
+       struct sockaddr_in *dst;
+       struct in_ifaddr *ia;
+
+if ((m->m_flags & M_PKTHDR) == 0)
+panic("ip_output no HDR");
+       if (opt) {
+               m = ip_insertoptions(m, opt, &len);
+               hlen = len;
+       }
+       ip = mtod(m, struct ip *);
+       /*
+        * Fill in IP header.
+        */
+       if ((flags & IP_FORWARDING) == 0) {
+               ip->ip_v = IPVERSION;
+               ip->ip_off &= IP_DF;
+               ip->ip_id = htons(ip_id++);
+               ip->ip_hl = hlen >> 2;
+       } else {
+               hlen = ip->ip_hl << 2;
+               ipstat.ips_localout++;
+       }
+       /*
+        * Route packet.
+        */
+       if (ro == 0) {
+               ro = &iproute;
+               bzero((caddr_t)ro, sizeof (*ro));
+       }
+       dst = (struct sockaddr_in *)&ro->ro_dst;
+       /*
+        * If there is a cached route,
+        * check that it is to the same destination
+        * and is still up.  If not, free it and try again.
+        */
+       if (ro->ro_rt && ((ro->ro_rt->rt_flags & RTF_UP) == 0 ||
+          dst->sin_addr.s_addr != ip->ip_dst.s_addr)) {
+               RTFREE(ro->ro_rt);
+               ro->ro_rt = (struct rtentry *)0;
+       }
+       if (ro->ro_rt == 0) {
+               dst->sin_family = AF_INET;
+               dst->sin_len = sizeof(*dst);
+               dst->sin_addr = ip->ip_dst;
+       }
+       /*
+        * If routing to interface only,
+        * short circuit routing lookup.
+        */
+       if (flags & IP_ROUTETOIF) {
+
+               ia = (struct in_ifaddr *)ifa_ifwithdstaddr((struct sockaddr *)dst);
+               if (ia == 0)
+                       ia = in_iaonnetof(in_netof(ip->ip_dst));
+               if (ia == 0) {
+                       error = ENETUNREACH;
+                       goto bad;
+               }
+               ifp = ia->ia_ifp;
+       } else {
+               if (ro->ro_rt == 0)
+                       rtalloc(ro);
+               if (ro->ro_rt == 0 || (ifp = ro->ro_rt->rt_ifp) == 0) {
+                       if (in_localaddr(ip->ip_dst))
+                               error = EHOSTUNREACH;
+                       else
+                               error = ENETUNREACH;
+                       goto bad;
+               }
+               ia = (struct in_ifaddr *)ro->ro_rt->rt_ifa;
+               ro->ro_rt->rt_use++;
+               if (ro->ro_rt->rt_flags & RTF_GATEWAY)
+                       dst = (struct sockaddr_in *)ro->ro_rt->rt_gateway;
+       }
+#ifndef notdef
+       /*
+        * If source address not specified yet, use address
+        * of outgoing interface.
+        */
+       if (ip->ip_src.s_addr == INADDR_ANY)
+               ip->ip_src = IA_SIN(ia)->sin_addr;
+#endif
+       /*
+        * Look for broadcast address and
+        * and verify user is allowed to send
+        * such a packet.
+        */
+       if (in_broadcast(dst->sin_addr)) {
+               if ((ifp->if_flags & IFF_BROADCAST) == 0) {
+                       error = EADDRNOTAVAIL;
+                       goto bad;
+               }
+               if ((flags & IP_ALLOWBROADCAST) == 0) {
+                       error = EACCES;
+                       goto bad;
+               }
+               /* don't allow broadcast messages to be fragmented */
+               if (ip->ip_len > ifp->if_mtu) {
+                       error = EMSGSIZE;
+                       goto bad;
+               }
+               m->m_flags |= M_BCAST;
+       }

 
 

-                       ip_send(p);             /* pass frag to local net level */
+       /*
+        * If small enough for interface, can just send directly.
+        */
+       if (ip->ip_len <= ifp->if_mtu) {
+               ip->ip_len = htons((u_short)ip->ip_len);
+               ip->ip_off = htons((u_short)ip->ip_off);
+               ip->ip_sum = 0;
+               ip->ip_sum = in_cksum(m, hlen);
+               error = (*ifp->if_output)(ifp, m,
+                               (struct sockaddr *)dst, ro->ro_rt);
+               goto done;
+       }
+       ipstat.ips_fragmented++;
+       /*
+        * Too large for interface; fragment if possible.
+        * Must be able to put at least 8 bytes per fragment.
+        */
+       if (ip->ip_off & IP_DF) {
+               error = EMSGSIZE;
+               goto bad;
+       }
+       len = (ifp->if_mtu - hlen) &~ 7;
+       if (len < 8) {
+               error = EMSGSIZE;
+               goto bad;
+       }

 
 

-                       p = (struct ip *)((int)m + m->m_off);   /* -> new hdr */
-                       len -= rnd;
-                       off += rnd;
+    {
+       int mhlen, firstlen = len;
+       struct mbuf **mnext = &m->m_nextpkt;
+
+       /*
+        * Loop through length of segment after first fragment,
+        * make new header and copy data of each part and link onto chain.
+        */
+       m0 = m;
+       mhlen = sizeof (struct ip);
+       for (off = hlen + len; off < ip->ip_len; off += len) {
+               MGETHDR(m, M_DONTWAIT, MT_HEADER);
+               if (m == 0) {
+                       error = ENOBUFS;
+                       goto sendorfree;

                }
                }

-       } 
-
-       return(ip_send(p));     /* pass datagram to local net level */
+               m->m_data += max_linkhdr;
+               mhip = mtod(m, struct ip *);
+               *mhip = *ip;
+               if (hlen > sizeof (struct ip)) {
+                       mhlen = ip_optcopy(ip, mhip) + sizeof (struct ip);
+                       mhip->ip_hl = mhlen >> 2;
+               }
+               m->m_len = mhlen;
+               mhip->ip_off = ((off - hlen) >> 3) + (ip->ip_off & ~IP_MF);
+               if (ip->ip_off & IP_MF)
+                       mhip->ip_off |= IP_MF;
+               if (off + len >= ip->ip_len)
+                       len = ip->ip_len - off;
+               else
+                       mhip->ip_off |= IP_MF;
+               mhip->ip_len = htons((u_short)(len + mhlen));
+               m->m_next = m_copy(m0, off, len);
+               if (m->m_next == 0) {
+                       error = ENOBUFS;        /* ??? */
+                       goto sendorfree;
+               }
+               m->m_pkthdr.len = mhlen + len;
+               m->m_pkthdr.rcvif = (struct ifnet *)0;
+               mhip->ip_off = htons((u_short)mhip->ip_off);
+               mhip->ip_sum = 0;
+               mhip->ip_sum = in_cksum(m, mhlen);
+               *mnext = m;
+               mnext = &m->m_nextpkt;
+               ipstat.ips_ofragments++;
+       }
+       /*
+        * Update first fragment by trimming what's been copied out
+        * and updating header, then send each fragment (in order).
+        */
+       m = m0;
+       m_adj(m, hlen + firstlen - ip->ip_len);
+       m->m_pkthdr.len = hlen + firstlen;
+       ip->ip_len = htons((u_short)m->m_pkthdr.len);
+       ip->ip_off = htons((u_short)(ip->ip_off | IP_MF));
+       ip->ip_sum = 0;
+       ip->ip_sum = in_cksum(m, hlen);
+sendorfree:
+       for (m = m0; m; m = m0) {
+               m0 = m->m_nextpkt;
+               m->m_nextpkt = 0;
+               if (error == 0)
+                       error = (*ifp->if_output)(ifp, m,
+                           (struct sockaddr *)dst, ro->ro_rt);
+               else
+                       m_freem(m);
+       }
+    }
+done:
+       if (ro == &iproute && (flags & IP_ROUTETOIF) == 0 && ro->ro_rt)
+               RTFREE(ro->ro_rt);
+       return (error);
+bad:
+       m_freem(m0);
+       goto done;

 }
 
 }
 

-ip_send(p)      /* format header and send message to 1822 level */
-struct ip *p;
-{
+/*
+ * Insert IP options into preformed packet.
+ * Adjust IP destination as required for IP source routing,
+ * as indicated by a non-zero in_addr at the start of the options.
+ */
+struct mbuf *
+ip_insertoptions(m, opt, phlen)

        register struct mbuf *m;
        register struct mbuf *m;

-       register struct imp *l;
-COUNT(IP_SEND);
-
-       m = dtom(p);                    /* ->header mbuf */
-
-       /* set up 1822 leader fields for transmit */
+       struct mbuf *opt;
+       int *phlen;
+{
+       register struct ipoption *p = mtod(opt, struct ipoption *);
+       struct mbuf *n;
+       register struct ip *ip = mtod(m, struct ip *);
+       unsigned optlen;
+
+       optlen = opt->m_len - sizeof(p->ipopt_dst);
+       if (p->ipopt_dst.s_addr)
+               ip->ip_dst = p->ipopt_dst;
+       if (m->m_flags & M_EXT || m->m_data - optlen < m->m_pktdat) {
+               MGETHDR(n, M_DONTWAIT, MT_HEADER);
+               if (n == 0)
+                       return (m);
+               n->m_pkthdr.len = m->m_pkthdr.len + optlen;
+               m->m_len -= sizeof(struct ip);
+               m->m_data += sizeof(struct ip);
+               n->m_next = m;
+               m = n;
+               m->m_len = optlen + sizeof(struct ip);
+               m->m_data += max_linkhdr;
+               bcopy((caddr_t)ip, mtod(m, caddr_t), sizeof(struct ip));
+       } else {
+               m->m_data -= optlen;
+               m->m_len += optlen;
+               m->m_pkthdr.len += optlen;
+               ovbcopy((caddr_t)ip, mtod(m, caddr_t), sizeof(struct ip));
+       }
+       ip = mtod(m, struct ip *);
+       bcopy((caddr_t)p->ipopt_list, (caddr_t)(ip + 1), (unsigned)optlen);
+       *phlen = sizeof(struct ip) + optlen;
+       ip->ip_len += optlen;
+       return (m);
+}

 
 

-       l = (struct imp *)((int)m + m->m_off - L1822);

 /*
 /*

-       l->i_hst = p->ip_dst.s_host;
-       l->i_impno = p->ip_dst.s_imp;
-       l->i_mlen = p->ip_len + L1822;
-       l->i_link = IPLINK;
-       l->i_type = 0;
-       l->i_htype = 0;
-       l->i_stype = 0;
-*/
-       l->i_shost = arpa_ether(p->ip_src.s_host);
-       l->i_dhost = arpa_ether(p->ip_dst.s_host);
-       l->i_type = IPTYPE;
-
-       /* finish ip leader by calculating checksum and doing
-          necessary byte-swapping  */
-
-       p->ip_sum = 0;
-       ip_bswap(p);
-       p->ip_sum = cksum(m, sizeof(struct ip));     
-
-       m->m_off -= L1822;              /* -> 1822 leader */
-       m->m_len += L1822;
-
-       return(imp_snd(m));             /* pass frag to 1822 */
+ * Copy options from ip to jp,
+ * omitting those not copied during fragmentation.
+ */
+ip_optcopy(ip, jp)
+       struct ip *ip, *jp;
+{
+       register u_char *cp, *dp;
+       int opt, optlen, cnt;
+
+       cp = (u_char *)(ip + 1);
+       dp = (u_char *)(jp + 1);
+       cnt = (ip->ip_hl << 2) - sizeof (struct ip);
+       for (; cnt > 0; cnt -= optlen, cp += optlen) {
+               opt = cp[0];
+               if (opt == IPOPT_EOL)
+                       break;
+               if (opt == IPOPT_NOP)
+                       optlen = 1;
+               else
+                       optlen = cp[IPOPT_OLEN];
+               /* bogus lengths should have been caught by ip_dooptions */
+               if (optlen > cnt)
+                       optlen = cnt;
+               if (IPOPT_COPIED(opt)) {
+                       bcopy((caddr_t)cp, (caddr_t)dp, (unsigned)optlen);
+                       dp += optlen;
+               }
+       }
+       for (optlen = dp - (u_char *)(jp+1); optlen & 0x3; optlen++)
+               *dp++ = IPOPT_EOL;
+       return (optlen);

 }
 
 }
 

-ip_setup(up, m, len)            /* setup an ip header for raw write */
-register struct ucb *up;
-register struct mbuf *m;
-int len;
+/*
+ * IP socket option processing.
+ */
+ip_ctloutput(op, so, level, optname, m)
+       int op;
+       struct socket *so;
+       int level, optname;
+       struct mbuf **m;

 {
 {

-       register struct ip *ip;
-COUNT(IP_SETUP);
+       int error = 0;
+       struct inpcb *inp = sotoinpcb(so);

 
 

-       m->m_off = MSIZE - sizeof(struct ip);
-       m->m_len = sizeof(struct ip);
+       if (level != IPPROTO_IP)
+               error = EINVAL;
+       else switch (op) {

 
 

-       ip = (struct ip *)((int)m + m->m_off);
+       case PRCO_SETOPT:
+               switch (optname) {
+               case IP_OPTIONS:
+                       return (ip_pcbopts(&inp->inp_options, *m));

 
 

-       ip->ip_tos = 0;
-       ip->ip_id = 0;
-       ip->ip_off = 0;
-       ip->ip_p = up->uc_lolink;
-       ip->ip_len = len + sizeof(struct ip);
-
-       ip->ip_src.s_addr = netcb.n_lhost.s_addr;
-        ip->ip_dst.s_addr = up->uc_host->h_addr.s_addr;
+               default:
+                       error = EINVAL;
+                       break;
+               }
+               break;
+
+       case PRCO_GETOPT:
+               switch (optname) {
+               case IP_OPTIONS:
+                       *m = m_get(M_WAIT, MT_SOOPTS);
+                       if (inp->inp_options) {
+                               (*m)->m_len = inp->inp_options->m_len;
+                               bcopy(mtod(inp->inp_options, caddr_t),
+                                   mtod(*m, caddr_t), (unsigned)(*m)->m_len);
+                       } else
+                               (*m)->m_len = 0;
+                       break;
+               default:
+                       error = EINVAL;
+                       break;
+               }
+               break;
+       }
+       if (op == PRCO_SETOPT && *m)
+               (void)m_free(*m);
+       return (error);

 }
 
 }
 

-

 /*
 /*

- * Convert logical host on imp to ethernet address.
- * (Primitive; use gateway table in next version.)
- *     0/78    arpavax         253
- *     1/78    ucb-c70         -
- *     -       csvax           252
- *     -       ucb-comet       223
+ * Set up IP options in pcb for insertion in output packets.
+ * Store in mbuf with pointer in pcbopt, adding pseudo-option
+ * with destination address if source routed.

  */
  */

-arpa_ether(n)
-       int n;
+ip_pcbopts(pcbopt, m)
+       struct mbuf **pcbopt;
+       register struct mbuf *m;

 {
 {

-COUNT(ARPA_ETHER);
-       switch (n) {
-       case 0:                 /* arpavax */
-               return 253;
-       default:
-               return n;
+       register cnt, optlen;
+       register u_char *cp;
+       u_char opt;
+
+       /* turn off any old options */
+       if (*pcbopt)
+               (void)m_free(*pcbopt);
+       *pcbopt = 0;
+       if (m == (struct mbuf *)0 || m->m_len == 0) {
+               /*
+                * Only turning off any previous options.
+                */
+               if (m)
+                       (void)m_free(m);
+               return (0);
+       }
+
+#ifndef        vax
+       if (m->m_len % sizeof(long))
+               goto bad;
+#endif
+       /*
+        * IP first-hop destination address will be stored before
+        * actual options; move other options back
+        * and clear it when none present.
+        */
+       if (m->m_data + m->m_len + sizeof(struct in_addr) >= &m->m_dat[MLEN])
+               goto bad;
+       cnt = m->m_len;
+       m->m_len += sizeof(struct in_addr);
+       cp = mtod(m, u_char *) + sizeof(struct in_addr);
+       ovbcopy(mtod(m, caddr_t), (caddr_t)cp, (unsigned)cnt);
+       bzero(mtod(m, caddr_t), sizeof(struct in_addr));
+
+       for (; cnt > 0; cnt -= optlen, cp += optlen) {
+               opt = cp[IPOPT_OPTVAL];
+               if (opt == IPOPT_EOL)
+                       break;
+               if (opt == IPOPT_NOP)
+                       optlen = 1;
+               else {
+                       optlen = cp[IPOPT_OLEN];
+                       if (optlen <= IPOPT_OLEN || optlen > cnt)
+                               goto bad;
+               }
+               switch (opt) {
+
+               default:
+                       break;
+
+               case IPOPT_LSRR:
+               case IPOPT_SSRR:
+                       /*
+                        * user process specifies route as:
+                        *      ->A->B->C->D
+                        * D must be our final destination (but we can't
+                        * check that since we may not have connected yet).
+                        * A is first hop destination, which doesn't appear in
+                        * actual IP option, but is stored before the options.
+                        */
+                       if (optlen < IPOPT_MINOFF - 1 + sizeof(struct in_addr))
+                               goto bad;
+                       m->m_len -= sizeof(struct in_addr);
+                       cnt -= sizeof(struct in_addr);
+                       optlen -= sizeof(struct in_addr);
+                       cp[IPOPT_OLEN] = optlen;
+                       /*
+                        * Move first hop before start of options.
+                        */
+                       bcopy((caddr_t)&cp[IPOPT_OFFSET+1], mtod(m, caddr_t),
+                           sizeof(struct in_addr));
+                       /*
+                        * Then copy rest of options back
+                        * to close up the deleted entry.
+                        */
+                       ovbcopy((caddr_t)(&cp[IPOPT_OFFSET+1] +
+                           sizeof(struct in_addr)),
+                           (caddr_t)&cp[IPOPT_OFFSET+1],
+                           (unsigned)cnt + sizeof(struct in_addr));
+                       break;
+               }

        }
        }

+       if (m->m_len > MAX_IPOPTLEN + sizeof(struct in_addr))
+               goto bad;
+       *pcbopt = m;
+       return (0);
+
+bad:
+       (void)m_free(m);
+       return (EINVAL);

 }
 }