uipc_socket.c

来自「eCos操作系统源码」· C语言 代码 · 共 1,522 行 · 第 1/3 页

					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 = &top;
		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		     * done could be out of date.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 = &top;
		    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)) {
		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:
	nextrecord = m->m_nextpkt;
	if (pr->pr_flags & PR_ADDR) {
		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;
		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;

	if (!(how == SHUT_RD || how == SHUT_WR || how == SHUT_RDWR))
		return (EINVAL);

	if (how != SHUT_WR)
		sorflush(so);
	if (how != SHUT_RD)
		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);
}

#ifdef INET
static int
do_setopt_accept_filter(so, sopt)
	struct	socket *so;
	struct	sockopt *sopt;
{
	struct accept_filter_arg	*afap = NULL;
	struct accept_filter	*afp;
	struct so_accf	*af = so->so_accf;
	int	error = 0;

	/* do not set/remove accept filters on non listen sockets */
	if ((so->so_options & SO_ACCEPTCONN) == 0) {
		error = EINVAL;
		goto out;
	}

	/* removing the filter */
	if (sopt == NULL) {
		if (af != NULL) {
			if (af->so_accept_filter != NULL && 
				af->so_accept_filter->accf_destroy != NULL) {
				af->so_accept_filter->accf_destroy(so);
			}
			if (af->so_accept_filter_str != NULL) {
				FREE(af->so_accept_filter_str, M_ACCF);
			}
			FREE(af, M_ACCF);
			so->so_accf = NULL;
		}
		so->so_options &= ~SO_ACCEPTFILTER;
		return (0);
	}
	/* adding a filter */
	/* must remove previous filter first */
	if (af != NULL) {
		error = EINVAL;
		goto out;
	}
	/* don't put large objects on the kernel stack */
	MALLOC(afap, struct accept_filter_arg *, sizeof(*afap), M_TEMP, M_WAITOK);