uipc_socket.c

RTEMS (Real-Time Executive for Multiprocessor Systems) is a free open source real-time operating sys

/*
 * 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
 * $Id: uipc_socket.c,v 1.7 2003/01/03 18:09:25 joel Exp $
 */

#include <sys/param.h>
#include <sys/queue.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>
#include <sys/signalvar.h>
#include <sys/sysctl.h>
#include <limits.h>

static int somaxconn = SOMAXCONN;
SYSCTL_INT(_kern, KIPC_SOMAXCONN, somaxconn, CTLFLAG_RW, &somaxconn, 0, "");

/*
 * 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.
 */
/*ARGSUSED*/
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 == 0)
		return (EPROTONOSUPPORT);
	if (prp->pr_type != type)
		return (EPROTOTYPE);
	MALLOC(so, struct socket *, sizeof(*so), M_SOCKET, M_WAIT);
	bzero((caddr_t)so, sizeof(*so));
	TAILQ_INIT(&so->so_incomp);
	TAILQ_INIT(&so->so_comp);
	so->so_type = type;
	so->so_state = SS_PRIV;
	so->so_uid = 0;
	so->so_proto = prp;
	error = (*prp->pr_usrreqs->pru_attach)(so, proto);
	if (error) {
		so->so_state |= SS_NOFDREF;
		sofree(so);
		return (error);
	}
	*aso = so;
	return (0);
}

int
sobind(so, nam)
	struct socket *so;
	struct mbuf *nam;
{
	int s = splnet();
	int error;

	error = (*so->so_proto->pr_usrreqs->pru_bind)(so, nam);
	splx(s);
	return (error);
}

int
solisten(so, backlog)
	register struct socket *so;
	int backlog;
{
	int s = splnet(), error;

	error = (*so->so_proto->pr_usrreqs->pru_listen)(so);
	if (error) {
		splx(s);
		return (error);
	}
	if (so->so_comp.tqh_first == NULL)
		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) {
			TAILQ_REMOVE(&head->so_comp, so, so_list);
		} else {
			panic("sofree: not queued");
		}
		head->so_qlen--;
		so->so_state &= ~(SS_INCOMP|SS_COMP);
		so->so_head = NULL;
	}
	sbrelease(&so->so_snd);
	sorflush(so);
	FREE(so, M_SOCKET);
}

/*
 * 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;

	if (so->so_options & SO_ACCEPTCONN) {
		struct socket *sp, *sonext;

		for (sp = so->so_incomp.tqh_first; sp != NULL; sp = sonext) {
			sonext = sp->so_list.tqe_next;
			(void) soabort(sp);
		}
		for (sp = so->so_comp.tqh_first; sp != NULL; sp = sonext) {
			sonext = sp->so_list.tqe_next;
			(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) {
				soconnsleep (so);
			}
		}
	}
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;
{

	return (*so->so_proto->pr_usrreqs->pru_abort)(so);
}

int
soaccept(so, nam)
	register struct socket *so;
	struct mbuf *nam;
{
	int s = splnet();
	int error;

	if ((so->so_state & SS_NOFDREF) == 0)
		panic("soaccept: !NOFDREF");
	so->so_state &= ~SS_NOFDREF;
	error = (*so->so_proto->pr_usrreqs->pru_accept)(so, nam);
	splx(s);
	return (error);
}

int
soconnect(so, nam)
	register struct socket *so;
	struct mbuf *nam;
{
	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);
	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)
	register struct socket *so;
	struct mbuf *addr;
	struct uio *uio;
	struct mbuf *top;
	struct mbuf *control;
	int flags;
{
	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 (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 = &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);
				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) {
				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 */
		    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 : 0,
			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);
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, paddr, uio, mp0, controlp, flagsp)
	register struct socket *so;
	struct mbuf **paddr;
	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 (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);
		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