BSD 4_4_Lite2 release

[unix-history] / usr / src / sys / kern / uipc_socket.c

diff --git a/usr/src/sys/kern/uipc_socket.c b/usr/src/sys/kern/uipc_socket.c
index 8c0e7cc..a9c5453 100644 (file)
--- a/usr/src/sys/kern/uipc_socket.c
+++ b/usr/src/sys/kern/uipc_socket.c
@@ -1,21 +1,50 @@
 /*
 /*

- * Copyright (c) 1982, 1986 Regents of the University of California.
- * All rights reserved.  The Berkeley software License Agreement
- * specifies the terms and conditions for redistribution.
+ * Copyright (c) 1982, 1986, 1988, 1990, 1993
+ *     The Regents of the University of California.  All rights reserved.

  *
  *

- *     @(#)uipc_socket.c       7.6 (Berkeley) %G%
+ * 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.
+ *
+ *     @(#)uipc_socket.c       8.6 (Berkeley) 5/2/95

  */
 
  */
 

-#include "param.h"
-#include "dir.h"
-#include "user.h"
-#include "proc.h"
-#include "file.h"
-#include "mbuf.h"
-#include "domain.h"
-#include "protosw.h"
-#include "socket.h"
-#include "socketvar.h"
+#include <sys/param.h>
+#include <sys/systm.h>
+#include <sys/proc.h>
+#include <sys/file.h>
+#include <sys/malloc.h>
+#include <sys/mbuf.h>
+#include <sys/domain.h>
+#include <sys/kernel.h>
+#include <sys/protosw.h>
+#include <sys/socket.h>
+#include <sys/socketvar.h>
+#include <sys/resourcevar.h>

 
 /*
  * Socket operation routines.
 
 /*
  * Socket operation routines.


@@ -23,42 +52,36 @@
  * sys_socket.c or from a system process, and
  * implement the semantics of socket operations by
  * switching out to the protocol specific routines.
  * sys_socket.c or from a system process, and
  * implement the semantics of socket operations by
  * switching out to the protocol specific routines.

- *
- * TODO:
- *     test socketpair
- *     clean up async
- *     out-of-band is a kludge

  */
 /*ARGSUSED*/
  */
 /*ARGSUSED*/

+int

 socreate(dom, aso, type, proto)
 socreate(dom, aso, type, proto)

+       int dom;

        struct socket **aso;
        register int type;
        int proto;
 {
        struct socket **aso;
        register int type;
        int proto;
 {

+       struct proc *p = curproc;               /* XXX */

        register struct protosw *prp;
        register struct socket *so;
        register struct protosw *prp;
        register struct socket *so;

-       register struct mbuf *m;

        register int error;
 
        if (proto)
                prp = pffindproto(dom, proto, type);
        else
                prp = pffindtype(dom, type);
        register int error;
 
        if (proto)
                prp = pffindproto(dom, proto, type);
        else
                prp = pffindtype(dom, type);

-       if (prp == 0)
+       if (prp == 0 || prp->pr_usrreq == 0)

                return (EPROTONOSUPPORT);
        if (prp->pr_type != type)
                return (EPROTOTYPE);
                return (EPROTONOSUPPORT);
        if (prp->pr_type != type)
                return (EPROTOTYPE);

-       m = m_getclr(M_WAIT, MT_SOCKET);
-       so = mtod(m, struct socket *);
-       so->so_options = 0;
-       so->so_state = 0;
+       MALLOC(so, struct socket *, sizeof(*so), M_SOCKET, M_WAIT);
+       bzero((caddr_t)so, sizeof(*so));

        so->so_type = type;
        so->so_type = type;

-       if (u.u_uid == 0)
+       if (p->p_ucred->cr_uid == 0)

                so->so_state = SS_PRIV;
        so->so_proto = prp;
                so->so_state = SS_PRIV;
        so->so_proto = prp;

-       error =
-           (*prp->pr_usrreq)(so, PRU_ATTACH,
-               (struct mbuf *)0, (struct mbuf *)proto, (struct mbuf *)0);
+       error = (*prp->pr_usrreq)(so, PRU_ATTACH, (struct mbuf *)0,
+           (struct mbuf *)(long)proto, (struct mbuf *)0);

        if (error) {
                so->so_state |= SS_NOFDREF;
                sofree(so);
        if (error) {
                so->so_state |= SS_NOFDREF;
                sofree(so);


@@ -68,6 +91,7 @@ socreate(dom, aso, type, proto)
        return (0);
 }
 
        return (0);
 }
 

+int

 sobind(so, nam)
        struct socket *so;
        struct mbuf *nam;
 sobind(so, nam)
        struct socket *so;
        struct mbuf *nam;


@@ -82,6 +106,7 @@ sobind(so, nam)
        return (error);
 }
 
        return (error);
 }
 

+int

 solisten(so, backlog)
        register struct socket *so;
        int backlog;
 solisten(so, backlog)
        register struct socket *so;
        int backlog;


@@ -95,18 +120,16 @@ solisten(so, backlog)
                splx(s);
                return (error);
        }
                splx(s);
                return (error);
        }

-       if (so->so_q == 0) {
-               so->so_q = so;
-               so->so_q0 = so;
+       if (so->so_q == 0)

                so->so_options |= SO_ACCEPTCONN;
                so->so_options |= SO_ACCEPTCONN;

-       }

        if (backlog < 0)
                backlog = 0;
        if (backlog < 0)
                backlog = 0;

-       so->so_qlimit = MIN(backlog, SOMAXCONN);
+       so->so_qlimit = min(backlog, SOMAXCONN);

        splx(s);
        return (0);
 }
 
        splx(s);
        return (0);
 }
 

+int

 sofree(so)
        register struct socket *so;
 {
 sofree(so)
        register struct socket *so;
 {


@@ -120,7 +143,7 @@ sofree(so)
        }
        sbrelease(&so->so_snd);
        sorflush(so);
        }
        sbrelease(&so->so_snd);
        sorflush(so);

-       (void) m_free(dtom(so));
+       FREE(so, M_SOCKET);

 }
 
 /*
 }
 
 /*


@@ -128,16 +151,17 @@ sofree(so)
  * Initiate disconnect if connected.
  * Free socket when disconnect complete.
  */
  * Initiate disconnect if connected.
  * Free socket when disconnect complete.
  */

+int

 soclose(so)
        register struct socket *so;
 {
        int s = splnet();               /* conservative */
 soclose(so)
        register struct socket *so;
 {
        int s = splnet();               /* conservative */

-       int error;
+       int error = 0;

 
        if (so->so_options & SO_ACCEPTCONN) {
 
        if (so->so_options & SO_ACCEPTCONN) {

-               while (so->so_q0 != so)
+               while (so->so_q0)

                        (void) soabort(so->so_q0);
                        (void) soabort(so->so_q0);

-               while (so->so_q != so)
+               while (so->so_q)

                        (void) soabort(so->so_q);
        }
        if (so->so_pcb == 0)
                        (void) soabort(so->so_q);
        }
        if (so->so_pcb == 0)


@@ -153,7 +177,9 @@ soclose(so)
                            (so->so_state & SS_NBIO))
                                goto drop;
                        while (so->so_state & SS_ISCONNECTED)
                            (so->so_state & SS_NBIO))
                                goto drop;
                        while (so->so_state & SS_ISCONNECTED)

-                               sleep((caddr_t)&so->so_timeo, PZERO+1);
+                               if (error = tsleep((caddr_t)&so->so_timeo,
+                                   PSOCK | PCATCH, netcls, so->so_linger * hz))
+                                       break;

                }
        }
 drop:
                }
        }
 drop:


@@ -176,6 +202,7 @@ discard:
 /*
  * Must be called at splnet...
  */
 /*
  * Must be called at splnet...
  */

+int

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


@@ -185,6 +212,7 @@ soabort(so)
                (struct mbuf *)0, (struct mbuf *)0, (struct mbuf *)0));
 }
 
                (struct mbuf *)0, (struct mbuf *)0, (struct mbuf *)0));
 }
 

+int

 soaccept(so, nam)
        register struct socket *so;
        struct mbuf *nam;
 soaccept(so, nam)
        register struct socket *so;
        struct mbuf *nam;


@@ -201,6 +229,7 @@ soaccept(so, nam)
        return (error);
 }
 
        return (error);
 }
 

+int

 soconnect(so, nam)
        register struct socket *so;
        struct mbuf *nam;
 soconnect(so, nam)
        register struct socket *so;
        struct mbuf *nam;


@@ -228,6 +257,7 @@ soconnect(so, nam)
        return (error);
 }
 
        return (error);
 }
 

+int

 soconnect2(so1, so2)
        register struct socket *so1;
        struct socket *so2;
 soconnect2(so1, so2)
        register struct socket *so1;
        struct socket *so2;


@@ -241,6 +271,7 @@ soconnect2(so1, so2)
        return (error);
 }
 
        return (error);
 }
 

+int

 sodisconnect(so)
        register struct socket *so;
 {
 sodisconnect(so)
        register struct socket *so;
 {


@@ -262,6 +293,7 @@ bad:
        return (error);
 }
 
        return (error);
 }
 

+#define        SBLOCKWAIT(f)   (((f) & MSG_DONTWAIT) ? M_NOWAIT : M_WAITOK)

 /*
  * Send on a socket.
  * If send must go all at once and message is larger than
 /*
  * Send on a socket.
  * If send must go all at once and message is larger than


@@ -270,117 +302,172 @@ bad:
  * If must go all at once and not enough room now, then
  * inform user that this would block and do nothing.
  * Otherwise, if nonblocking, send as much as possible.
  * If must go all at once and not enough room now, then
  * inform user that this would block and do nothing.
  * Otherwise, if nonblocking, send as much as possible.

+ * The data to be sent is described by "uio" if nonzero,
+ * otherwise by the mbuf chain "top" (which must be null
+ * if uio is not).  Data provided in mbuf chain must be small
+ * enough to send all at once.
+ *
+ * Returns nonzero on error, timeout or signal; callers
+ * must check for short counts if EINTR/ERESTART are returned.
+ * Data and control buffers are freed on return.

  */
  */

-sosend(so, nam, uio, flags, rights)
+int
+sosend(so, addr, uio, top, control, flags)

        register struct socket *so;
        register struct socket *so;

-       struct mbuf *nam;
-       register struct uio *uio;
+       struct mbuf *addr;
+       struct uio *uio;
+       struct mbuf *top;
+       struct mbuf *control;

        int flags;
        int flags;

-       struct mbuf *rights;

 {
 {

-       struct mbuf *top = 0;
-       register struct mbuf *m, **mp;
-       register int space;
-       int len, rlen = 0, error = 0, s, dontroute, first = 1;
+       struct proc *p = curproc;               /* XXX */
+       struct mbuf **mp;
+       register struct mbuf *m;
+       register long space, len, resid;
+       int clen = 0, error, s, dontroute, mlen;
+       int atomic = sosendallatonce(so) || top;

 
 

-       if (sosendallatonce(so) && uio->uio_resid > so->so_snd.sb_hiwat)
-               return (EMSGSIZE);
+       if (uio)
+               resid = uio->uio_resid;
+       else
+               resid = top->m_pkthdr.len;
+       /*
+        * In theory resid should be unsigned.
+        * However, space must be signed, as it might be less than 0
+        * if we over-committed, and we must use a signed comparison
+        * of space and resid.  On the other hand, a negative resid
+        * causes us to loop sending 0-length segments to the protocol.
+        */
+       if (resid < 0)
+               return (EINVAL);

        dontroute =
            (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 &&
            (so->so_proto->pr_flags & PR_ATOMIC);
        dontroute =
            (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 &&
            (so->so_proto->pr_flags & PR_ATOMIC);

-       u.u_ru.ru_msgsnd++;
-       if (rights)
-               rlen = rights->m_len;
+       p->p_stats->p_ru.ru_msgsnd++;
+       if (control)
+               clen = control->m_len;

 #define        snderr(errno)   { error = errno; splx(s); goto release; }
 
 restart:
 #define        snderr(errno)   { error = errno; splx(s); goto release; }
 
 restart:

-       sblock(&so->so_snd);
+       if (error = sblock(&so->so_snd, SBLOCKWAIT(flags)))
+               goto out;

        do {
                s = splnet();
                if (so->so_state & SS_CANTSENDMORE)
                        snderr(EPIPE);
        do {
                s = splnet();
                if (so->so_state & SS_CANTSENDMORE)
                        snderr(EPIPE);

-               if (so->so_error) {
-                       error = so->so_error;
-                       so->so_error = 0;                       /* ??? */
-                       splx(s);
-                       goto release;
-               }
+               if (so->so_error)
+                       snderr(so->so_error);

                if ((so->so_state & SS_ISCONNECTED) == 0) {
                if ((so->so_state & SS_ISCONNECTED) == 0) {

-                       if (so->so_proto->pr_flags & PR_CONNREQUIRED)
-                               snderr(ENOTCONN);
-                       if (nam == 0)
+                       if (so->so_proto->pr_flags & PR_CONNREQUIRED) {
+                               if ((so->so_state & SS_ISCONFIRMING) == 0 &&
+                                   !(resid == 0 && clen != 0))
+                                       snderr(ENOTCONN);
+                       } else if (addr == 0)

                                snderr(EDESTADDRREQ);
                }
                                snderr(EDESTADDRREQ);
                }

+               space = sbspace(&so->so_snd);

                if (flags & MSG_OOB)
                if (flags & MSG_OOB)

-                       space = 1024;
-               else {
-                       space = sbspace(&so->so_snd);
-                       if (space <= rlen ||
-                          (sosendallatonce(so) &&
-                               space < uio->uio_resid + rlen) ||
-                          (uio->uio_resid >= CLBYTES && space < CLBYTES &&
-                          so->so_snd.sb_cc >= CLBYTES &&
-                          (so->so_state & SS_NBIO) == 0)) {
-                               if (so->so_state & SS_NBIO) {
-                                       if (first)
-                                               error = EWOULDBLOCK;
-                                       splx(s);
-                                       goto release;
-                               }
-                               sbunlock(&so->so_snd);
-                               sbwait(&so->so_snd);
-                               splx(s);
-                               goto restart;
-                       }
+                       space += 1024;
+               if (atomic && resid > so->so_snd.sb_hiwat ||
+                   clen > so->so_snd.sb_hiwat)
+                       snderr(EMSGSIZE);
+               if (space < resid + clen && uio &&
+                   (atomic || space < so->so_snd.sb_lowat || space < clen)) {
+                       if (so->so_state & SS_NBIO)
+                               snderr(EWOULDBLOCK);
+                       sbunlock(&so->so_snd);
+                       error = sbwait(&so->so_snd);
+                       splx(s);
+                       if (error)
+                               goto out;
+                       goto restart;

                }
                splx(s);
                mp = &top;
                }
                splx(s);
                mp = &top;

-               space -= rlen;
-               while (space > 0) {
-                       MGET(m, M_WAIT, MT_DATA);
-                       if (uio->uio_resid >= CLBYTES / 2 && space >= CLBYTES) {
-                               MCLGET(m);
-                               if (m->m_len != CLBYTES)
+               space -= clen;
+               do {
+                   if (uio == NULL) {
+                       /*
+                        * Data is prepackaged in "top".
+                        */
+                       resid = 0;
+                       if (flags & MSG_EOR)
+                               top->m_flags |= M_EOR;
+                   } else do {
+                       if (top == 0) {
+                               MGETHDR(m, M_WAIT, MT_DATA);
+                               mlen = MHLEN;
+                               m->m_pkthdr.len = 0;
+                               m->m_pkthdr.rcvif = (struct ifnet *)0;
+                       } else {
+                               MGET(m, M_WAIT, MT_DATA);
+                               mlen = MLEN;
+                       }
+                       if (resid >= MINCLSIZE && space >= MCLBYTES) {
+                               MCLGET(m, M_WAIT);
+                               if ((m->m_flags & M_EXT) == 0)

                                        goto nopages;
                                        goto nopages;

-                               len = MIN(CLBYTES, uio->uio_resid);
-                               space -= CLBYTES;
+                               mlen = MCLBYTES;
+#ifdef MAPPED_MBUFS
+                               len = min(MCLBYTES, resid);
+#else
+                               if (atomic && top == 0) {
+                                       len = min(MCLBYTES - max_hdr, resid);
+                                       m->m_data += max_hdr;
+                               } else
+                                       len = min(MCLBYTES, resid);
+#endif
+                               space -= MCLBYTES;

                        } else {
 nopages:
                        } else {
 nopages:

-                               len = MIN(MIN(MLEN, uio->uio_resid), space);
+                               len = min(min(mlen, resid), space);

                                space -= len;
                                space -= len;

+                               /*
+                                * For datagram protocols, leave room
+                                * for protocol headers in first mbuf.
+                                */
+                               if (atomic && top == 0 && len < mlen)
+                                       MH_ALIGN(m, len);

                        }
                        }

-                       error = uiomove(mtod(m, caddr_t), len, UIO_WRITE, uio);
+                       error = uiomove(mtod(m, caddr_t), (int)len, uio);
+                       resid = uio->uio_resid;

                        m->m_len = len;
                        *mp = m;
                        m->m_len = len;
                        *mp = m;

+                       top->m_pkthdr.len += len;

                        if (error)
                                goto release;
                        mp = &m->m_next;
                        if (error)
                                goto release;
                        mp = &m->m_next;

-                       if (uio->uio_resid <= 0)
+                       if (resid <= 0) {
+                               if (flags & MSG_EOR)
+                                       top->m_flags |= M_EOR;

                                break;
                                break;

-               }
-               if (dontroute)
-                       so->so_options |= SO_DONTROUTE;
-               s = splnet();                                   /* XXX */
-               error = (*so->so_proto->pr_usrreq)(so,
-                   (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND,
-                   top, (caddr_t)nam, rights);
-               splx(s);
-               if (dontroute)
-                       so->so_options &= ~SO_DONTROUTE;
-               rights = 0;
-               rlen = 0;
-               top = 0;
-               first = 0;
-               if (error)
-                       break;
-       } while (uio->uio_resid);
+                       }
+                   } while (space > 0 && atomic);
+                   if (dontroute)
+                           so->so_options |= SO_DONTROUTE;
+                   s = splnet();                               /* XXX */
+                   error = (*so->so_proto->pr_usrreq)(so,
+                       (flags & MSG_OOB) ? PRU_SENDOOB : PRU_SEND,
+                       top, addr, control);
+                   splx(s);
+                   if (dontroute)
+                           so->so_options &= ~SO_DONTROUTE;
+                   clen = 0;
+                   control = 0;
+                   top = 0;
+                   mp = &top;
+                   if (error)
+                       goto release;
+               } while (resid && space > 0);
+       } while (resid);

 
 release:
        sbunlock(&so->so_snd);
 
 release:
        sbunlock(&so->so_snd);

+out:

        if (top)
                m_freem(top);
        if (top)
                m_freem(top);

-       if (error == EPIPE)
-               psignal(u.u_procp, SIGPIPE);
+       if (control)
+               m_freem(control);

        return (error);
 }
 
        return (error);
 }
 


@@ -389,42 +476,51 @@ release:
  * We depend on the way that records are added to the sockbuf
  * by sbappend*.  In particular, each record (mbufs linked through m_next)
  * must begin with an address if the protocol so specifies,
  * We depend on the way that records are added to the sockbuf
  * by sbappend*.  In particular, each record (mbufs linked through m_next)
  * must begin with an address if the protocol so specifies,

- * followed by an optional mbuf containing access rights if supported
- * by the protocol, and then zero or more mbufs of data.
+ * followed by an optional mbuf or mbufs containing ancillary data,
+ * and then zero or more mbufs of data.

  * In order to avoid blocking network interrupts for the entire time here,
  * we splx() while doing the actual copy to user space.
  * Although the sockbuf is locked, new data may still be appended,
  * and thus we must maintain consistency of the sockbuf during that time.
  * In order to avoid blocking network interrupts for the entire time here,
  * we splx() while doing the actual copy to user space.
  * Although the sockbuf is locked, new data may still be appended,
  * and thus we must maintain consistency of the sockbuf during that time.

+ *
+ * The caller may receive the data as a single mbuf chain by supplying
+ * an mbuf **mp0 for use in returning the chain.  The uio is then used
+ * only for the count in uio_resid.

  */
  */

-soreceive(so, aname, uio, flags, rightsp)
+int
+soreceive(so, paddr, uio, mp0, controlp, flagsp)

        register struct socket *so;
        register struct socket *so;

-       struct mbuf **aname;
-       register struct uio *uio;
-       int flags;
-       struct mbuf **rightsp;
+       struct mbuf **paddr;
+       struct uio *uio;
+       struct mbuf **mp0;
+       struct mbuf **controlp;
+       int *flagsp;

 {
 {

-       register struct mbuf *m;
-       register int len, error = 0, s, offset;
+       register struct mbuf *m, **mp;
+       register int flags, len, error, s, offset;

        struct protosw *pr = so->so_proto;
        struct mbuf *nextrecord;
        struct protosw *pr = so->so_proto;
        struct mbuf *nextrecord;

-       int moff;
+       int moff, type;
+       int orig_resid = uio->uio_resid;

 
 

-       if (rightsp)
-               *rightsp = 0;
-       if (aname)
-               *aname = 0;
+       mp = mp0;
+       if (paddr)
+               *paddr = 0;
+       if (controlp)
+               *controlp = 0;
+       if (flagsp)
+               flags = *flagsp &~ MSG_EOR;
+       else
+               flags = 0;

        if (flags & MSG_OOB) {
                m = m_get(M_WAIT, MT_DATA);
        if (flags & MSG_OOB) {
                m = m_get(M_WAIT, MT_DATA);

-               error = (*pr->pr_usrreq)(so, PRU_RCVOOB,
-                   m, (struct mbuf *)(flags & MSG_PEEK), (struct mbuf *)0);
+               error = (*pr->pr_usrreq)(so, PRU_RCVOOB, m,
+                   (struct mbuf *)(long)(flags & MSG_PEEK), (struct mbuf *)0);

                if (error)
                        goto bad;
                do {
                if (error)
                        goto bad;
                do {

-                       len = uio->uio_resid;
-                       if (len > m->m_len)
-                               len = m->m_len;
-                       error =
-                           uiomove(mtod(m, caddr_t), (int)len, UIO_READ, uio);
+                       error = uiomove(mtod(m, caddr_t),
+                           (int) min(uio->uio_resid, m->m_len), uio);

                        m = m_free(m);
                } while (uio->uio_resid && error == 0 && m);
 bad:
                        m = m_free(m);
                } while (uio->uio_resid && error == 0 && m);
 bad:


@@ -432,73 +528,115 @@ bad:
                        m_freem(m);
                return (error);
        }
                        m_freem(m);
                return (error);
        }

+       if (mp)
+               *mp = (struct mbuf *)0;
+       if (so->so_state & SS_ISCONFIRMING && uio->uio_resid)
+               (*pr->pr_usrreq)(so, PRU_RCVD, (struct mbuf *)0,
+                   (struct mbuf *)0, (struct mbuf *)0);

 
 restart:
 
 restart:

-       sblock(&so->so_rcv);
+       if (error = sblock(&so->so_rcv, SBLOCKWAIT(flags)))
+               return (error);

        s = splnet();
 
        s = splnet();
 

-       if (so->so_rcv.sb_cc == 0) {
+       m = so->so_rcv.sb_mb;
+       /*
+        * If we have less data than requested, block awaiting more
+        * (subject to any timeout) if:
+        *   1. the current count is less than the low water mark, or
+        *   2. MSG_WAITALL is set, and it is possible to do the entire
+        *      receive operation at once if we block (resid <= hiwat), or
+        *   3. MSG_DONTWAIT is not set.
+        * If MSG_WAITALL is set but resid is larger than the receive buffer,
+        * we have to do the receive in sections, and thus risk returning
+        * a short count if a timeout or signal occurs after we start.
+        */
+       if (m == 0 || ((flags & MSG_DONTWAIT) == 0 &&
+           so->so_rcv.sb_cc < uio->uio_resid) &&
+           (so->so_rcv.sb_cc < so->so_rcv.sb_lowat ||
+           ((flags & MSG_WAITALL) && uio->uio_resid <= so->so_rcv.sb_hiwat)) &&
+           m->m_nextpkt == 0 && (pr->pr_flags & PR_ATOMIC) == 0) {
+#ifdef DIAGNOSTIC
+               if (m == 0 && so->so_rcv.sb_cc)
+                       panic("receive 1");
+#endif

                if (so->so_error) {
                if (so->so_error) {

+                       if (m)
+                               goto dontblock;

                        error = so->so_error;
                        error = so->so_error;

-                       so->so_error = 0;
+                       if ((flags & MSG_PEEK) == 0)
+                               so->so_error = 0;

                        goto release;
                }
                        goto release;
                }

-               if (so->so_state & SS_CANTRCVMORE)
-                       goto release;
-               if ((so->so_state & SS_ISCONNECTED) == 0 &&
+               if (so->so_state & SS_CANTRCVMORE) {
+                       if (m)
+                               goto dontblock;
+                       else
+                               goto release;
+               }
+               for (; m; m = m->m_next)
+                       if (m->m_type == MT_OOBDATA  || (m->m_flags & M_EOR)) {
+                               m = so->so_rcv.sb_mb;
+                               goto dontblock;
+                       }
+               if ((so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING)) == 0 &&

                    (so->so_proto->pr_flags & PR_CONNREQUIRED)) {
                        error = ENOTCONN;
                        goto release;
                }
                if (uio->uio_resid == 0)
                        goto release;
                    (so->so_proto->pr_flags & PR_CONNREQUIRED)) {
                        error = ENOTCONN;
                        goto release;
                }
                if (uio->uio_resid == 0)
                        goto release;

-               if (so->so_state & SS_NBIO) {
+               if ((so->so_state & SS_NBIO) || (flags & MSG_DONTWAIT)) {

                        error = EWOULDBLOCK;
                        goto release;
                }
                sbunlock(&so->so_rcv);
                        error = EWOULDBLOCK;
                        goto release;
                }
                sbunlock(&so->so_rcv);

-               sbwait(&so->so_rcv);
+               error = sbwait(&so->so_rcv);

                splx(s);
                splx(s);

+               if (error)
+                       return (error);

                goto restart;
        }
                goto restart;
        }

-       u.u_ru.ru_msgrcv++;
-       m = so->so_rcv.sb_mb;
-       if (m == 0)
-               panic("receive 1");
-       nextrecord = m->m_act;
+dontblock:
+       if (uio->uio_procp)
+               uio->uio_procp->p_stats->p_ru.ru_msgrcv++;
+       nextrecord = m->m_nextpkt;

        if (pr->pr_flags & PR_ADDR) {
        if (pr->pr_flags & PR_ADDR) {

+#ifdef DIAGNOSTIC

                if (m->m_type != MT_SONAME)
                        panic("receive 1a");
                if (m->m_type != MT_SONAME)
                        panic("receive 1a");

+#endif
+               orig_resid = 0;

                if (flags & MSG_PEEK) {
                if (flags & MSG_PEEK) {

-                       if (aname)
-                               *aname = m_copy(m, 0, m->m_len);
+                       if (paddr)
+                               *paddr = m_copy(m, 0, m->m_len);

                        m = m->m_next;
                } else {
                        sbfree(&so->so_rcv, m);
                        m = m->m_next;
                } else {
                        sbfree(&so->so_rcv, m);

-                       if (aname) {
-                               *aname = m;
-                               m = m->m_next;
-                               (*aname)->m_next = 0;
-                               so->so_rcv.sb_mb = m;
+                       if (paddr) {
+                               *paddr = m;
+                               so->so_rcv.sb_mb = m->m_next;
+                               m->m_next = 0;
+                               m = so->so_rcv.sb_mb;

                        } else {
                                MFREE(m, so->so_rcv.sb_mb);
                                m = so->so_rcv.sb_mb;
                        }
                        } else {
                                MFREE(m, so->so_rcv.sb_mb);
                                m = so->so_rcv.sb_mb;
                        }

-                       if (m)
-                               m->m_act = nextrecord;

                }
        }
                }
        }

-       if (m && m->m_type == MT_RIGHTS) {
-               if ((pr->pr_flags & PR_RIGHTS) == 0)
-                       panic("receive 2");
+       while (m && m->m_type == MT_CONTROL && error == 0) {

                if (flags & MSG_PEEK) {
                if (flags & MSG_PEEK) {

-                       if (rightsp)
-                               *rightsp = m_copy(m, 0, m->m_len);
+                       if (controlp)
+                               *controlp = m_copy(m, 0, m->m_len);

                        m = m->m_next;
                } else {
                        sbfree(&so->so_rcv, m);
                        m = m->m_next;
                } else {
                        sbfree(&so->so_rcv, m);

-                       if (rightsp) {
-                               *rightsp = m;
+                       if (controlp) {
+                               if (pr->pr_domain->dom_externalize &&
+                                   mtod(m, struct cmsghdr *)->cmsg_type ==
+                                   SCM_RIGHTS)
+                                  error = (*pr->pr_domain->dom_externalize)(m);
+                               *controlp = m;

                                so->so_rcv.sb_mb = m->m_next;
                                m->m_next = 0;
                                m = so->so_rcv.sb_mb;
                                so->so_rcv.sb_mb = m->m_next;
                                m->m_next = 0;
                                m = so->so_rcv.sb_mb;


@@ -506,42 +644,79 @@ restart:
                                MFREE(m, so->so_rcv.sb_mb);
                                m = so->so_rcv.sb_mb;
                        }
                                MFREE(m, so->so_rcv.sb_mb);
                                m = so->so_rcv.sb_mb;
                        }

-                       if (m)
-                               m->m_act = nextrecord;
+               }
+               if (controlp) {
+                       orig_resid = 0;
+                       controlp = &(*controlp)->m_next;

                }
        }
                }
        }

+       if (m) {
+               if ((flags & MSG_PEEK) == 0)
+                       m->m_nextpkt = nextrecord;
+               type = m->m_type;
+               if (type == MT_OOBDATA)
+                       flags |= MSG_OOB;
+       }

        moff = 0;
        offset = 0;
        while (m && uio->uio_resid > 0 && error == 0) {
        moff = 0;
        offset = 0;
        while (m && uio->uio_resid > 0 && error == 0) {

-               if (m->m_type != MT_DATA && m->m_type != MT_HEADER)
+               if (m->m_type == MT_OOBDATA) {
+                       if (type != MT_OOBDATA)
+                               break;
+               } else if (type == MT_OOBDATA)
+                       break;
+#ifdef DIAGNOSTIC
+               else if (m->m_type != MT_DATA && m->m_type != MT_HEADER)

                        panic("receive 3");
                        panic("receive 3");

-               len = uio->uio_resid;
+#endif

                so->so_state &= ~SS_RCVATMARK;
                so->so_state &= ~SS_RCVATMARK;

+               len = uio->uio_resid;

                if (so->so_oobmark && len > so->so_oobmark - offset)
                        len = so->so_oobmark - offset;
                if (len > m->m_len - moff)
                        len = m->m_len - moff;
                if (so->so_oobmark && len > so->so_oobmark - offset)
                        len = so->so_oobmark - offset;
                if (len > m->m_len - moff)
                        len = m->m_len - moff;

-               splx(s);
-               error =
-                   uiomove(mtod(m, caddr_t) + moff, (int)len, UIO_READ, uio);
-               s = splnet();
+               /*
+                * If mp is set, just pass back the mbufs.
+                * Otherwise copy them out via the uio, then free.
+                * Sockbuf must be consistent here (points to current mbuf,
+                * it points to next record) when we drop priority;
+                * we must note any additions to the sockbuf when we
+                * block interrupts again.
+                */
+               if (mp == 0) {
+                       splx(s);
+                       error = uiomove(mtod(m, caddr_t) + moff, (int)len, uio);
+                       s = splnet();
+               } else
+                       uio->uio_resid -= len;

                if (len == m->m_len - moff) {
                if (len == m->m_len - moff) {

+                       if (m->m_flags & M_EOR)
+                               flags |= MSG_EOR;

                        if (flags & MSG_PEEK) {
                                m = m->m_next;
                                moff = 0;
                        } else {
                        if (flags & MSG_PEEK) {
                                m = m->m_next;
                                moff = 0;
                        } else {

-                               nextrecord = m->m_act;
+                               nextrecord = m->m_nextpkt;

                                sbfree(&so->so_rcv, m);
                                sbfree(&so->so_rcv, m);

-                               MFREE(m, so->so_rcv.sb_mb);
-                               m = so->so_rcv.sb_mb;
+                               if (mp) {
+                                       *mp = m;
+                                       mp = &m->m_next;
+                                       so->so_rcv.sb_mb = m = m->m_next;
+                                       *mp = (struct mbuf *)0;
+                               } else {
+                                       MFREE(m, so->so_rcv.sb_mb);
+                                       m = so->so_rcv.sb_mb;
+                               }

                                if (m)
                                if (m)

-                                       m->m_act = nextrecord;
+                                       m->m_nextpkt = nextrecord;

                        }
                } else {
                        if (flags & MSG_PEEK)
                                moff += len;
                        else {
                        }
                } else {
                        if (flags & MSG_PEEK)
                                moff += len;
                        else {

-                               m->m_off += len;
+                               if (mp)
+                                       *mp = m_copym(m, 0, len, M_WAIT);
+                               m->m_data += len;

                                m->m_len -= len;
                                so->so_rcv.sb_cc -= len;
                        }
                                m->m_len -= len;
                                so->so_rcv.sb_cc -= len;
                        }


@@ -553,28 +728,65 @@ restart:
                                        so->so_state |= SS_RCVATMARK;
                                        break;
                                }
                                        so->so_state |= SS_RCVATMARK;
                                        break;
                                }

-                       } else
+                       } else {

                                offset += len;
                                offset += len;

+                               if (offset == so->so_oobmark)
+                                       break;
+                       }

                }
                }

+               if (flags & MSG_EOR)
+                       break;
+               /*
+                * If the MSG_WAITALL flag is set (for non-atomic socket),
+                * we must not quit until "uio->uio_resid == 0" or an error
+                * termination.  If a signal/timeout occurs, return
+                * with a short count but without error.
+                * Keep sockbuf locked against other readers.
+                */
+               while (flags & MSG_WAITALL && m == 0 && uio->uio_resid > 0 &&
+                   !sosendallatonce(so) && !nextrecord) {
+                       if (so->so_error || so->so_state & SS_CANTRCVMORE)
+                               break;
+                       error = sbwait(&so->so_rcv);
+                       if (error) {
+                               sbunlock(&so->so_rcv);
+                               splx(s);
+                               return (0);
+                       }
+                       if (m = so->so_rcv.sb_mb)
+                               nextrecord = m->m_nextpkt;
+               }
+       }
+
+       if (m && pr->pr_flags & PR_ATOMIC) {
+               flags |= MSG_TRUNC;
+               if ((flags & MSG_PEEK) == 0)
+                       (void) sbdroprecord(&so->so_rcv);

        }
        if ((flags & MSG_PEEK) == 0) {
                if (m == 0)
                        so->so_rcv.sb_mb = nextrecord;
        }
        if ((flags & MSG_PEEK) == 0) {
                if (m == 0)
                        so->so_rcv.sb_mb = nextrecord;

-               else if (pr->pr_flags & PR_ATOMIC)
-                       (void) sbdroprecord(&so->so_rcv);

                if (pr->pr_flags & PR_WANTRCVD && so->so_pcb)
                        (*pr->pr_usrreq)(so, PRU_RCVD, (struct mbuf *)0,
                if (pr->pr_flags & PR_WANTRCVD && so->so_pcb)
                        (*pr->pr_usrreq)(so, PRU_RCVD, (struct mbuf *)0,

-                           (struct mbuf *)0, (struct mbuf *)0);
-               if (error == 0 && rightsp && *rightsp &&
-                   pr->pr_domain->dom_externalize)
-                       error = (*pr->pr_domain->dom_externalize)(*rightsp);
+                           (struct mbuf *)(long)flags, (struct mbuf *)0,
+                           (struct mbuf *)0);

        }
        }

+       if (orig_resid == uio->uio_resid && orig_resid &&
+           (flags & MSG_EOR) == 0 && (so->so_state & SS_CANTRCVMORE) == 0) {
+               sbunlock(&so->so_rcv);
+               splx(s);
+               goto restart;
+       }
+               
+       if (flagsp)
+               *flagsp |= flags;

 release:
        sbunlock(&so->so_rcv);
        splx(s);
        return (error);
 }
 
 release:
        sbunlock(&so->so_rcv);
        splx(s);
        return (error);
 }
 

+int

 soshutdown(so, how)
        register struct socket *so;
        register int how;
 soshutdown(so, how)
        register struct socket *so;
        register int how;


@@ -590,6 +802,7 @@ soshutdown(so, how)
        return (0);
 }
 
        return (0);
 }
 

+void

 sorflush(so)
        register struct socket *so;
 {
 sorflush(so)
        register struct socket *so;
 {


@@ -598,7 +811,8 @@ sorflush(so)
        register int s;
        struct sockbuf asb;
 
        register int s;
        struct sockbuf asb;
 

-       sblock(sb);
+       sb->sb_flags |= SB_NOINTR;
+       (void) sblock(sb, M_WAITOK);

        s = splimp();
        socantrcvmore(so);
        sbunlock(sb);
        s = splimp();
        socantrcvmore(so);
        sbunlock(sb);


@@ -610,6 +824,7 @@ sorflush(so)
        sbrelease(&asb);
 }
 
        sbrelease(&asb);
 }
 

+int

 sosetopt(so, level, optname, m0)
        register struct socket *so;
        int level, optname;
 sosetopt(so, level, optname, m0)
        register struct socket *so;
        int level, optname;


@@ -640,6 +855,7 @@ sosetopt(so, level, optname, m0)
                case SO_USELOOPBACK:
                case SO_BROADCAST:
                case SO_REUSEADDR:
                case SO_USELOOPBACK:
                case SO_BROADCAST:
                case SO_REUSEADDR:

+               case SO_REUSEPORT:

                case SO_OOBINLINE:
                        if (m == NULL || m->m_len < sizeof (int)) {
                                error = EINVAL;
                case SO_OOBINLINE:
                        if (m == NULL || m->m_len < sizeof (int)) {
                                error = EINVAL;


@@ -655,8 +871,6 @@ sosetopt(so, level, optname, m0)
                case SO_RCVBUF:
                case SO_SNDLOWAT:
                case SO_RCVLOWAT:
                case SO_RCVBUF:
                case SO_SNDLOWAT:
                case SO_RCVLOWAT:

-               case SO_SNDTIMEO:
-               case SO_RCVTIMEO:

                        if (m == NULL || m->m_len < sizeof (int)) {
                                error = EINVAL;
                                goto bad;
                        if (m == NULL || m->m_len < sizeof (int)) {
                                error = EINVAL;
                                goto bad;


@@ -665,8 +879,9 @@ sosetopt(so, level, optname, m0)
 
                        case SO_SNDBUF:
                        case SO_RCVBUF:
 
                        case SO_SNDBUF:
                        case SO_RCVBUF:

-                               if (sbreserve(optname == SO_SNDBUF ? &so->so_snd :
-                                   &so->so_rcv, *mtod(m, int *)) == 0) {
+                               if (sbreserve(optname == SO_SNDBUF ?
+                                   &so->so_snd : &so->so_rcv,
+                                   (u_long) *mtod(m, int *)) == 0) {

                                        error = ENOBUFS;
                                        goto bad;
                                }
                                        error = ENOBUFS;
                                        goto bad;
                                }


@@ -678,19 +893,47 @@ sosetopt(so, level, optname, m0)
                        case SO_RCVLOWAT:
                                so->so_rcv.sb_lowat = *mtod(m, int *);
                                break;
                        case SO_RCVLOWAT:
                                so->so_rcv.sb_lowat = *mtod(m, int *);
                                break;

+                       }
+                       break;
+
+               case SO_SNDTIMEO:
+               case SO_RCVTIMEO:
+                   {
+                       struct timeval *tv;
+                       short val;
+
+                       if (m == NULL || m->m_len < sizeof (*tv)) {
+                               error = EINVAL;
+                               goto bad;
+                       }
+                       tv = mtod(m, struct timeval *);
+                       if (tv->tv_sec * hz + tv->tv_usec / tick > SHRT_MAX) {
+                               error = EDOM;
+                               goto bad;
+                       }
+                       val = tv->tv_sec * hz + tv->tv_usec / tick;
+
+                       switch (optname) {
+

                        case SO_SNDTIMEO:
                        case SO_SNDTIMEO:

-                               so->so_snd.sb_timeo = *mtod(m, int *);
+                               so->so_snd.sb_timeo = val;

                                break;
                        case SO_RCVTIMEO:
                                break;
                        case SO_RCVTIMEO:

-                               so->so_rcv.sb_timeo = *mtod(m, int *);
+                               so->so_rcv.sb_timeo = val;

                                break;
                        }
                        break;
                                break;
                        }
                        break;

+                   }

 
                default:
                        error = ENOPROTOOPT;
                        break;
                }
 
                default:
                        error = ENOPROTOOPT;
                        break;
                }

+               if (error == 0 && so->so_proto && so->so_proto->pr_ctloutput) {
+                       (void) ((*so->so_proto->pr_ctloutput)
+                                 (PRCO_SETOPT, so, level, optname, &m0));
+                       m = NULL;       /* freed by protocol */
+               }

        }
 bad:
        if (m)
        }
 bad:
        if (m)


@@ -698,6 +941,7 @@ bad:
        return (error);
 }
 
        return (error);
 }
 

+int

 sogetopt(so, level, optname, mp)
        register struct socket *so;
        int level, optname;
 sogetopt(so, level, optname, mp)
        register struct socket *so;
        int level, optname;


@@ -709,7 +953,7 @@ sogetopt(so, level, optname, mp)
                if (so->so_proto && so->so_proto->pr_ctloutput) {
                        return ((*so->so_proto->pr_ctloutput)
                                  (PRCO_GETOPT, so, level, optname, mp));
                if (so->so_proto && so->so_proto->pr_ctloutput) {
                        return ((*so->so_proto->pr_ctloutput)
                                  (PRCO_GETOPT, so, level, optname, mp));

-               } else 
+               } else

                        return (ENOPROTOOPT);
        } else {
                m = m_get(M_WAIT, MT_SOOPTS);
                        return (ENOPROTOOPT);
        } else {
                m = m_get(M_WAIT, MT_SOOPTS);


@@ -729,6 +973,7 @@ sogetopt(so, level, optname, mp)
                case SO_DEBUG:
                case SO_KEEPALIVE:
                case SO_REUSEADDR:
                case SO_DEBUG:
                case SO_KEEPALIVE:
                case SO_REUSEADDR:

+               case SO_REUSEPORT:

                case SO_BROADCAST:
                case SO_OOBINLINE:
                        *mtod(m, int *) = so->so_options & optname;
                case SO_BROADCAST:
                case SO_OOBINLINE:
                        *mtod(m, int *) = so->so_options & optname;


@@ -760,12 +1005,17 @@ sogetopt(so, level, optname, mp)
                        break;
 
                case SO_SNDTIMEO:
                        break;
 
                case SO_SNDTIMEO:

-                       *mtod(m, int *) = so->so_snd.sb_timeo;
-                       break;
-

                case SO_RCVTIMEO:
                case SO_RCVTIMEO:

-                       *mtod(m, int *) = so->so_rcv.sb_timeo;
+                   {
+                       int val = (optname == SO_SNDTIMEO ?
+                            so->so_snd.sb_timeo : so->so_rcv.sb_timeo);
+
+                       m->m_len = sizeof(struct timeval);
+                       mtod(m, struct timeval *)->tv_sec = val / hz;
+                       mtod(m, struct timeval *)->tv_usec =
+                           (val % hz) * tick;

                        break;
                        break;

+                   }

 
                default:
                        (void)m_free(m);
 
                default:
                        (void)m_free(m);


@@ -776,18 +1026,15 @@ sogetopt(so, level, optname, mp)
        }
 }
 
        }
 }
 

+void

 sohasoutofband(so)
        register struct socket *so;
 {
        struct proc *p;
 
 sohasoutofband(so)
        register struct socket *so;
 {
        struct proc *p;
 

-       if (so->so_pgrp < 0)
-               gsignal(-so->so_pgrp, SIGURG);
-       else if (so->so_pgrp > 0 && (p = pfind(so->so_pgrp)) != 0)
+       if (so->so_pgid < 0)
+               gsignal(-so->so_pgid, SIGURG);
+       else if (so->so_pgid > 0 && (p = pfind(so->so_pgid)) != 0)

                psignal(p, SIGURG);
                psignal(p, SIGURG);

-       if (so->so_rcv.sb_sel) {
-               selwakeup(so->so_rcv.sb_sel, so->so_rcv.sb_flags & SB_COLL);
-               so->so_rcv.sb_sel = 0;
-               so->so_rcv.sb_flags &= ~SB_COLL;
-       }
+       selwakeup(&so->so_rcv.sb_sel);

 }
 }