/* * Copyright (c) 1982, 1986, 1988, 1990, 1993 * The Regents of the University of California. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by the University of * California, Berkeley and its contributors. * 4. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * @(#)uipc_socket.c 8.3 (Berkeley) 4/15/94 * $FreeBSD$ */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include struct vm_zone *socket_zone; so_gen_t so_gencnt; /* generation count for sockets */ MALLOC_DEFINE(M_SONAME, "soname", "socket name"); MALLOC_DEFINE(M_PCB, "pcb", "protocol control block"); SYSCTL_DECL(_kern_ipc); static int somaxconn = SOMAXCONN; SYSCTL_INT(_kern_ipc, KIPC_SOMAXCONN, somaxconn, CTLFLAG_RW, &somaxconn, 0, "Maximum pending socket connection queue size"); /* * Socket operation routines. * These routines are called by the routines in * sys_socket.c or from a system process, and * implement the semantics of socket operations by * switching out to the protocol specific routines. */ /* * Get a socket structure from our zone, and initialize it. * We don't implement `waitok' yet (see comments in uipc_domain.c). * Note that it would probably be better to allocate socket * and PCB at the same time, but I'm not convinced that all * the protocols can be easily modified to do this. */ struct socket * soalloc(waitok) int waitok; { struct socket *so; so = zalloci(socket_zone); if (so) { /* XXX race condition for reentrant kernel */ bzero(so, sizeof *so); so->so_gencnt = ++so_gencnt; so->so_zone = socket_zone; TAILQ_INIT(&so->so_aiojobq); } return so; } int socreate(dom, aso, type, proto, p) int dom; struct socket **aso; register int type; int proto; struct proc *p; { register struct protosw *prp; register struct socket *so; register int error; if (proto) prp = pffindproto(dom, proto, type); else prp = pffindtype(dom, type); if (prp == 0 || prp->pr_usrreqs->pru_attach == 0) return (EPROTONOSUPPORT); if (prp->pr_type != type) return (EPROTOTYPE); so = soalloc(p != 0); if (so == 0) return (ENOBUFS); TAILQ_INIT(&so->so_incomp); TAILQ_INIT(&so->so_comp); so->so_type = type; so->so_cred = p->p_ucred; crhold(so->so_cred); so->so_proto = prp; error = (*prp->pr_usrreqs->pru_attach)(so, proto, p); if (error) { so->so_state |= SS_NOFDREF; sofree(so); return (error); } *aso = so; return (0); } int sobind(so, nam, p) struct socket *so; struct sockaddr *nam; struct proc *p; { int s = splnet(); int error; error = (*so->so_proto->pr_usrreqs->pru_bind)(so, nam, p); splx(s); return (error); } void sodealloc(so) struct socket *so; { so->so_gencnt = ++so_gencnt; if (so->so_rcv.sb_hiwat) (void)chgsbsize(so->so_cred->cr_uid, -(rlim_t)so->so_rcv.sb_hiwat); if (so->so_snd.sb_hiwat) (void)chgsbsize(so->so_cred->cr_uid, -(rlim_t)so->so_snd.sb_hiwat); crfree(so->so_cred); zfreei(so->so_zone, so); } int solisten(so, backlog, p) register struct socket *so; int backlog; struct proc *p; { int s, error; s = splnet(); error = (*so->so_proto->pr_usrreqs->pru_listen)(so, p); if (error) { splx(s); return (error); } if (TAILQ_EMPTY(&so->so_comp)) so->so_options |= SO_ACCEPTCONN; if (backlog < 0 || backlog > somaxconn) backlog = somaxconn; so->so_qlimit = backlog; splx(s); return (0); } void sofree(so) register struct socket *so; { struct socket *head = so->so_head; if (so->so_pcb || (so->so_state & SS_NOFDREF) == 0) return; if (head != NULL) { if (so->so_state & SS_INCOMP) { TAILQ_REMOVE(&head->so_incomp, so, so_list); head->so_incqlen--; } else if (so->so_state & SS_COMP) { /* * We must not decommission a socket that's * on the accept(2) queue. If we do, then * accept(2) may hang after select(2) indicated * that the listening socket was ready. */ return; } else { panic("sofree: not queued"); } head->so_qlen--; so->so_state &= ~SS_INCOMP; so->so_head = NULL; } sbrelease(&so->so_snd, so); sorflush(so); sodealloc(so); } /* * Close a socket on last file table reference removal. * Initiate disconnect if connected. * Free socket when disconnect complete. */ int soclose(so) register struct socket *so; { int s = splnet(); /* conservative */ int error = 0; funsetown(so->so_sigio); if (so->so_options & SO_ACCEPTCONN) { struct socket *sp, *sonext; sp = TAILQ_FIRST(&so->so_incomp); for (; sp != NULL; sp = sonext) { sonext = TAILQ_NEXT(sp, so_list); (void) soabort(sp); } for (sp = TAILQ_FIRST(&so->so_comp); sp != NULL; sp = sonext) { sonext = TAILQ_NEXT(sp, so_list); /* Dequeue from so_comp since sofree() won't do it */ TAILQ_REMOVE(&so->so_comp, sp, so_list); so->so_qlen--; sp->so_state &= ~SS_COMP; sp->so_head = NULL; (void) soabort(sp); } } if (so->so_pcb == 0) goto discard; if (so->so_state & SS_ISCONNECTED) { if ((so->so_state & SS_ISDISCONNECTING) == 0) { error = sodisconnect(so); if (error) goto drop; } if (so->so_options & SO_LINGER) { if ((so->so_state & SS_ISDISCONNECTING) && (so->so_state & SS_NBIO)) goto drop; while (so->so_state & SS_ISCONNECTED) { error = tsleep((caddr_t)&so->so_timeo, PSOCK | PCATCH, "soclos", so->so_linger * hz); if (error) break; } } } drop: if (so->so_pcb) { int error2 = (*so->so_proto->pr_usrreqs->pru_detach)(so); if (error == 0) error = error2; } discard: if (so->so_state & SS_NOFDREF) panic("soclose: NOFDREF"); so->so_state |= SS_NOFDREF; sofree(so); splx(s); return (error); } /* * Must be called at splnet... */ int soabort(so) struct socket *so; { int error; error = (*so->so_proto->pr_usrreqs->pru_abort)(so); if (error) { sofree(so); return error; } return (0); } int soaccept(so, nam) register struct socket *so; struct sockaddr **nam; { int s = splnet(); int error; if ((so->so_state & SS_NOFDREF) == 0) panic("soaccept: !NOFDREF"); so->so_state &= ~SS_NOFDREF; if ((so->so_state & SS_ISDISCONNECTED) == 0) error = (*so->so_proto->pr_usrreqs->pru_accept)(so, nam); else { if (nam) *nam = 0; error = 0; } splx(s); return (error); } int soconnect(so, nam, p) register struct socket *so; struct sockaddr *nam; struct proc *p; { int s; int error; if (so->so_options & SO_ACCEPTCONN) return (EOPNOTSUPP); s = splnet(); /* * If protocol is connection-based, can only connect once. * Otherwise, if connected, try to disconnect first. * This allows user to disconnect by connecting to, e.g., * a null address. */ if (so->so_state & (SS_ISCONNECTED|SS_ISCONNECTING) && ((so->so_proto->pr_flags & PR_CONNREQUIRED) || (error = sodisconnect(so)))) error = EISCONN; else error = (*so->so_proto->pr_usrreqs->pru_connect)(so, nam, p); splx(s); return (error); } int soconnect2(so1, so2) register struct socket *so1; struct socket *so2; { int s = splnet(); int error; error = (*so1->so_proto->pr_usrreqs->pru_connect2)(so1, so2); splx(s); return (error); } int sodisconnect(so) register struct socket *so; { int s = splnet(); int error; if ((so->so_state & SS_ISCONNECTED) == 0) { error = ENOTCONN; goto bad; } if (so->so_state & SS_ISDISCONNECTING) { error = EALREADY; goto bad; } error = (*so->so_proto->pr_usrreqs->pru_disconnect)(so); bad: splx(s); 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 buffering, then hard error. * Lock against other senders. * 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. */ int sosend(so, addr, uio, top, control, flags, p) register struct socket *so; struct sockaddr *addr; struct uio *uio; struct mbuf *top; struct mbuf *control; int flags; struct proc *p; { 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 (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. * * Also check to make sure that MSG_EOR isn't used on SOCK_STREAM * type sockets since that's an error. */ if (resid < 0 || (so->so_type == SOCK_STREAM && (flags & MSG_EOR))) { error = EINVAL; goto out; } dontroute = (flags & MSG_DONTROUTE) && (so->so_options & SO_DONTROUTE) == 0 && (so->so_proto->pr_flags & PR_ATOMIC); if (p) p->p_stats->p_ru.ru_msgsnd++; if (control) clen = control->m_len; #define snderr(errno) { error = errno; splx(s); goto release; } restart: error = sblock(&so->so_snd, SBLOCKWAIT(flags)); if (error) goto out; 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_state & SS_ISCONNECTED) == 0) { /* * `sendto' and `sendmsg' is allowed on a connection- * based socket if it supports implied connect. * Return ENOTCONN if not connected and no address is * supplied. */ if ((so->so_proto->pr_flags & PR_CONNREQUIRED) && (so->so_proto->pr_flags & PR_IMPLOPCL) == 0) { if ((so->so_state & SS_ISCONFIRMING) == 0 && !(resid == 0 && clen != 0)) snderr(ENOTCONN); } else if (addr == 0) snderr(so->so_proto->pr_flags & PR_CONNREQUIRED ? ENOTCONN : EDESTADDRREQ); } space = sbspace(&so->so_snd); if (flags & MSG_OOB) 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 = ⊤ 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); if (m == NULL) { error = ENOBUFS; goto release; } mlen = MHLEN; m->m_pkthdr.len = 0; m->m_pkthdr.rcvif = (struct ifnet *)0; } else { MGET(m, M_WAIT, MT_DATA); if (m == NULL) { error = ENOBUFS; goto release; } mlen = MLEN; } if (resid >= MINCLSIZE) { MCLGET(m, M_WAIT); if ((m->m_flags & M_EXT) == 0) goto nopages; mlen = MCLBYTES; len = min(min(mlen, resid), space); } else { nopages: len = min(min(mlen, resid), space); /* * For datagram protocols, leave room * for protocol headers in first mbuf. */ if (atomic && top == 0 && len < mlen) MH_ALIGN(m, len); } space -= len; error = uiomove(mtod(m, caddr_t), (int)len, uio); resid = uio->uio_resid; m->m_len = len; *mp = m; top->m_pkthdr.len += len; if (error) goto release; mp = &m->m_next; if (resid <= 0) { if (flags & MSG_EOR) top->m_flags |= M_EOR; break; } } while (space > 0 && atomic); if (dontroute) so->so_options |= SO_DONTROUTE; s = splnet(); /* XXX */ /* * XXX all the SS_CANTSENDMORE checks previously * done could be out of date. We could have recieved * a reset packet in an interrupt or maybe we slept * while doing page faults in uiomove() etc. We could * probably recheck again inside the splnet() protection * here, but there are probably other places that this * also happens. We must rethink this. */ error = (*so->so_proto->pr_usrreqs->pru_send)(so, (flags & MSG_OOB) ? PRUS_OOB : /* * If the user set MSG_EOF, the protocol * understands this flag and nothing left to * send then use PRU_SEND_EOF instead of PRU_SEND. */ ((flags & MSG_EOF) && (so->so_proto->pr_flags & PR_IMPLOPCL) && (resid <= 0)) ? PRUS_EOF : /* If there is more to send set PRUS_MORETOCOME */ (resid > 0 && space > 0) ? PRUS_MORETOCOME : 0, top, addr, control, p); splx(s); if (dontroute) so->so_options &= ~SO_DONTROUTE; clen = 0; control = 0; top = 0; mp = ⊤ if (error) goto release; } while (resid && space > 0); } while (resid); release: sbunlock(&so->so_snd); out: if (top) m_freem(top); if (control) m_freem(control); return (error); } /* * Implement receive operations on a socket. * 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 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. * * 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. */ int soreceive(so, psa, uio, mp0, controlp, flagsp) register struct socket *so; struct sockaddr **psa; struct uio *uio; struct mbuf **mp0; struct mbuf **controlp; int *flagsp; { register struct mbuf *m, **mp; register int flags, len, error, s, offset; struct protosw *pr = so->so_proto; struct mbuf *nextrecord; int moff, type = 0; int orig_resid = uio->uio_resid; mp = mp0; if (psa) *psa = 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 (m == NULL) return (ENOBUFS); error = (*pr->pr_usrreqs->pru_rcvoob)(so, m, flags & MSG_PEEK); if (error) goto bad; do { 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: if (m) m_freem(m); return (error); } if (mp) *mp = (struct mbuf *)0; if (so->so_state & SS_ISCONFIRMING && uio->uio_resid) (*pr->pr_usrreqs->pru_rcvd)(so, 0); restart: error = sblock(&so->so_rcv, SBLOCKWAIT(flags)); if (error) return (error); s = splnet(); 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). * 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)) { KASSERT(m != 0 || !so->so_rcv.sb_cc, ("receive 1")); if (so->so_error) { if (m) goto dontblock; error = so->so_error; if ((flags & MSG_PEEK) == 0) so->so_error = 0; goto release; } 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; if ((so->so_state & SS_NBIO) || (flags & MSG_DONTWAIT)) { error = EWOULDBLOCK; goto release; } sbunlock(&so->so_rcv); error = sbwait(&so->so_rcv); splx(s); if (error) return (error); goto restart; } dontblock: if (uio->uio_procp) uio->uio_procp->p_stats->p_ru.ru_msgrcv++; nextrecord = m->m_nextpkt; if (pr->pr_flags & PR_ADDR) { KASSERT(m->m_type == MT_SONAME, ("receive 1a")); orig_resid = 0; if (psa) *psa = dup_sockaddr(mtod(m, struct sockaddr *), mp0 == 0); if (flags & MSG_PEEK) { m = m->m_next; } else { sbfree(&so->so_rcv, m); MFREE(m, so->so_rcv.sb_mb); m = so->so_rcv.sb_mb; } } while (m && m->m_type == MT_CONTROL && error == 0) { if (flags & MSG_PEEK) { if (controlp) *controlp = m_copy(m, 0, m->m_len); m = m->m_next; } else { sbfree(&so->so_rcv, 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; } else { MFREE(m, so->so_rcv.sb_mb); m = so->so_rcv.sb_mb; } } 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) { if (m->m_type == MT_OOBDATA) { if (type != MT_OOBDATA) break; } else if (type == MT_OOBDATA) break; else KASSERT(m->m_type == MT_DATA || m->m_type == MT_HEADER, ("receive 3")); 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 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(); if (error) goto release; } else uio->uio_resid -= len; 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 { nextrecord = m->m_nextpkt; sbfree(&so->so_rcv, m); 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) m->m_nextpkt = nextrecord; } } else { if (flags & MSG_PEEK) moff += len; else { if (mp) *mp = m_copym(m, 0, len, M_WAIT); m->m_data += len; m->m_len -= len; so->so_rcv.sb_cc -= len; } } if (so->so_oobmark) { if ((flags & MSG_PEEK) == 0) { so->so_oobmark -= len; if (so->so_oobmark == 0) { so->so_state |= SS_RCVATMARK; break; } } else { 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); } m = so->so_rcv.sb_mb; if (m) 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 (pr->pr_flags & PR_WANTRCVD && so->so_pcb) (*pr->pr_usrreqs->pru_rcvd)(so, flags); } 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); } int soshutdown(so, how) register struct socket *so; register int how; { register struct protosw *pr = so->so_proto; how++; if (how & FREAD) sorflush(so); if (how & FWRITE) return ((*pr->pr_usrreqs->pru_shutdown)(so)); return (0); } void sorflush(so) register struct socket *so; { register struct sockbuf *sb = &so->so_rcv; register struct protosw *pr = so->so_proto; register int s; struct sockbuf asb; sb->sb_flags |= SB_NOINTR; (void) sblock(sb, M_WAITOK); s = splimp(); socantrcvmore(so); sbunlock(sb); asb = *sb; bzero((caddr_t)sb, sizeof (*sb)); splx(s); if (pr->pr_flags & PR_RIGHTS && pr->pr_domain->dom_dispose) (*pr->pr_domain->dom_dispose)(asb.sb_mb); sbrelease(&asb, so); } /* * Perhaps this routine, and sooptcopyout(), below, ought to come in * an additional variant to handle the case where the option value needs * to be some kind of integer, but not a specific size. * In addition to their use here, these functions are also called by the * protocol-level pr_ctloutput() routines. */ int sooptcopyin(sopt, buf, len, minlen) struct sockopt *sopt; void *buf; size_t len; size_t minlen; { size_t valsize; /* * If the user gives us more than we wanted, we ignore it, * but if we don't get the minimum length the caller * wants, we return EINVAL. On success, sopt->sopt_valsize * is set to however much we actually retrieved. */ if ((valsize = sopt->sopt_valsize) < minlen) return EINVAL; if (valsize > len) sopt->sopt_valsize = valsize = len; if (sopt->sopt_p != 0) return (copyin(sopt->sopt_val, buf, valsize)); bcopy(sopt->sopt_val, buf, valsize); return 0; } int sosetopt(so, sopt) struct socket *so; struct sockopt *sopt; { int error, optval; struct linger l; struct timeval tv; u_long val; error = 0; if (sopt->sopt_level != SOL_SOCKET) { if (so->so_proto && so->so_proto->pr_ctloutput) return ((*so->so_proto->pr_ctloutput) (so, sopt)); error = ENOPROTOOPT; } else { switch (sopt->sopt_name) { case SO_LINGER: error = sooptcopyin(sopt, &l, sizeof l, sizeof l); if (error) goto bad; so->so_linger = l.l_linger; if (l.l_onoff) so->so_options |= SO_LINGER; else so->so_options &= ~SO_LINGER; break; case SO_DEBUG: case SO_KEEPALIVE: case SO_DONTROUTE: case SO_USELOOPBACK: case SO_BROADCAST: case SO_REUSEADDR: case SO_REUSEPORT: case SO_OOBINLINE: case SO_TIMESTAMP: error = sooptcopyin(sopt, &optval, sizeof optval, sizeof optval); if (error) goto bad; if (optval) so->so_options |= sopt->sopt_name; else so->so_options &= ~sopt->sopt_name; break; case SO_SNDBUF: case SO_RCVBUF: case SO_SNDLOWAT: case SO_RCVLOWAT: error = sooptcopyin(sopt, &optval, sizeof optval, sizeof optval); if (error) goto bad; /* * Values < 1 make no sense for any of these * options, so disallow them. */ if (optval < 1) { error = EINVAL; goto bad; } switch (sopt->sopt_name) { case SO_SNDBUF: case SO_RCVBUF: if (sbreserve(sopt->sopt_name == SO_SNDBUF ? &so->so_snd : &so->so_rcv, (u_long)optval, so, curproc) == 0) { error = ENOBUFS; goto bad; } break; /* * Make sure the low-water is never greater than * the high-water. */ case SO_SNDLOWAT: so->so_snd.sb_lowat = (optval > so->so_snd.sb_hiwat) ? so->so_snd.sb_hiwat : optval; break; case SO_RCVLOWAT: so->so_rcv.sb_lowat = (optval > so->so_rcv.sb_hiwat) ? so->so_rcv.sb_hiwat : optval; break; } break; case SO_SNDTIMEO: case SO_RCVTIMEO: error = sooptcopyin(sopt, &tv, sizeof tv, sizeof tv); if (error) goto bad; /* assert(hz > 0); */ if (tv.tv_sec < 0 || tv.tv_sec > SHRT_MAX / hz || tv.tv_usec < 0 || tv.tv_usec >= 1000000) { error = EDOM; goto bad; } /* assert(tick > 0); */ /* assert(ULONG_MAX - SHRT_MAX >= 1000000); */ val = (u_long)(tv.tv_sec * hz) + tv.tv_usec / tick; if (val > SHRT_MAX) { error = EDOM; goto bad; } switch (sopt->sopt_name) { case SO_SNDTIMEO: so->so_snd.sb_timeo = val; break; case SO_RCVTIMEO: so->so_rcv.sb_timeo = val; break; } break; default: error = ENOPROTOOPT; break; } if (error == 0 && so->so_proto && so->so_proto->pr_ctloutput) { (void) ((*so->so_proto->pr_ctloutput) (so, sopt)); } } bad: return (error); } /* Helper routine for getsockopt */ int sooptcopyout(sopt, buf, len) struct sockopt *sopt; void *buf; size_t len; { int error; size_t valsize; error = 0; /* * Documented get behavior is that we always return a value, * possibly truncated to fit in the user's buffer. * Traditional behavior is that we always tell the user * precisely how much we copied, rather than something useful * like the total amount we had available for her. * Note that this interface is not idempotent; the entire answer must * generated ahead of time. */ valsize = min(len, sopt->sopt_valsize); sopt->sopt_valsize = valsize; if (sopt->sopt_val != 0) { if (sopt->sopt_p != 0) error = copyout(buf, sopt->sopt_val, valsize); else bcopy(buf, sopt->sopt_val, valsize); } return error; } int sogetopt(so, sopt) struct socket *so; struct sockopt *sopt; { int error, optval; struct linger l; struct timeval tv; error = 0; if (sopt->sopt_level != SOL_SOCKET) { if (so->so_proto && so->so_proto->pr_ctloutput) { return ((*so->so_proto->pr_ctloutput) (so, sopt)); } else return (ENOPROTOOPT); } else { switch (sopt->sopt_name) { case SO_LINGER: l.l_onoff = so->so_options & SO_LINGER; l.l_linger = so->so_linger; error = sooptcopyout(sopt, &l, sizeof l); break; case SO_USELOOPBACK: case SO_DONTROUTE: case SO_DEBUG: case SO_KEEPALIVE: case SO_REUSEADDR: case SO_REUSEPORT: case SO_BROADCAST: case SO_OOBINLINE: case SO_TIMESTAMP: optval = so->so_options & sopt->sopt_name; integer: error = sooptcopyout(sopt, &optval, sizeof optval); break; case SO_TYPE: optval = so->so_type; goto integer; case SO_ERROR: optval = so->so_error; so->so_error = 0; goto integer; case SO_SNDBUF: optval = so->so_snd.sb_hiwat; goto integer; case SO_RCVBUF: optval = so->so_rcv.sb_hiwat; goto integer; case SO_SNDLOWAT: optval = so->so_snd.sb_lowat; goto integer; case SO_RCVLOWAT: optval = so->so_rcv.sb_lowat; goto integer; case SO_SNDTIMEO: case SO_RCVTIMEO: optval = (sopt->sopt_name == SO_SNDTIMEO ? so->so_snd.sb_timeo : so->so_rcv.sb_timeo); tv.tv_sec = optval / hz; tv.tv_usec = (optval % hz) * tick; error = sooptcopyout(sopt, &tv, sizeof tv); break; default: error = ENOPROTOOPT; break; } return (error); } } /* XXX; prepare mbuf for (__FreeBSD__ < 3) routines. */ int soopt_getm(struct sockopt *sopt, struct mbuf **mp) { struct mbuf *m, *m_prev; int sopt_size = sopt->sopt_valsize; MGET(m, sopt->sopt_p ? M_WAIT : M_DONTWAIT, MT_DATA); if (m == 0) return ENOBUFS; if (sopt_size > MLEN) { MCLGET(m, sopt->sopt_p ? M_WAIT : M_DONTWAIT); if ((m->m_flags & M_EXT) == 0) { m_free(m); return ENOBUFS; } m->m_len = min(MCLBYTES, sopt_size); } else { m->m_len = min(MLEN, sopt_size); } sopt_size -= m->m_len; *mp = m; m_prev = m; while (sopt_size) { MGET(m, sopt->sopt_p ? M_WAIT : M_DONTWAIT, MT_DATA); if (m == 0) { m_freem(*mp); return ENOBUFS; } if (sopt_size > MLEN) { MCLGET(m, sopt->sopt_p ? M_WAIT : M_DONTWAIT); if ((m->m_flags & M_EXT) == 0) { m_freem(*mp); return ENOBUFS; } m->m_len = min(MCLBYTES, sopt_size); } else { m->m_len = min(MLEN, sopt_size); } sopt_size -= m->m_len; m_prev->m_next = m; m_prev = m; } return 0; } /* XXX; copyin sopt data into mbuf chain for (__FreeBSD__ < 3) routines. */ int soopt_mcopyin(struct sockopt *sopt, struct mbuf *m) { struct mbuf *m0 = m; if (sopt->sopt_val == NULL) return 0; while (m != NULL && sopt->sopt_valsize >= m->m_len) { if (sopt->sopt_p != NULL) { int error; error = copyin(sopt->sopt_val, mtod(m, char *), m->m_len); if (error != 0) { m_freem(m0); return(error); } } else bcopy(sopt->sopt_val, mtod(m, char *), m->m_len); sopt->sopt_valsize -= m->m_len; (caddr_t)sopt->sopt_val += m->m_len; m = m->m_next; } if (m != NULL) /* should be allocated enoughly at ip6_sooptmcopyin() */ panic("ip6_sooptmcopyin"); return 0; } /* XXX; copyout mbuf chain data into soopt for (__FreeBSD__ < 3) routines. */ int soopt_mcopyout(struct sockopt *sopt, struct mbuf *m) { struct mbuf *m0 = m; size_t valsize = 0; if (sopt->sopt_val == NULL) return 0; while (m != NULL && sopt->sopt_valsize >= m->m_len) { if (sopt->sopt_p != NULL) { int error; error = copyout(mtod(m, char *), sopt->sopt_val, m->m_len); if (error != 0) { m_freem(m0); return(error); } } else bcopy(mtod(m, char *), sopt->sopt_val, m->m_len); sopt->sopt_valsize -= m->m_len; (caddr_t)sopt->sopt_val += m->m_len; valsize += m->m_len; m = m->m_next; } if (m != NULL) { /* enough soopt buffer should be given from user-land */ m_freem(m0); return(EINVAL); } sopt->sopt_valsize = valsize; return 0; } void sohasoutofband(so) register struct socket *so; { if (so->so_sigio != NULL) pgsigio(so->so_sigio, SIGURG, 0); selwakeup(&so->so_rcv.sb_sel); } int sopoll(struct socket *so, int events, struct ucred *cred, struct proc *p) { int revents = 0; int s = splnet(); if (events & (POLLIN | POLLRDNORM)) if (soreadable(so)) revents |= events & (POLLIN | POLLRDNORM); if (events & (POLLOUT | POLLWRNORM)) if (sowriteable(so)) revents |= events & (POLLOUT | POLLWRNORM); if (events & (POLLPRI | POLLRDBAND)) if (so->so_oobmark || (so->so_state & SS_RCVATMARK)) revents |= events & (POLLPRI | POLLRDBAND); if (revents == 0) { if (events & (POLLIN | POLLPRI | POLLRDNORM | POLLRDBAND)) { selrecord(p, &so->so_rcv.sb_sel); so->so_rcv.sb_flags |= SB_SEL; } if (events & (POLLOUT | POLLWRNORM)) { selrecord(p, &so->so_snd.sb_sel); so->so_snd.sb_flags |= SB_SEL; } } splx(s); return (revents); }