freebsd-dev/sys/net/if.c
Garrett Wollman 2c37256e5a Modify the kernel to use the new pr_usrreqs interface rather than the old
pr_usrreq mechanism which was poorly designed and error-prone.  This
commit renames pr_usrreq to pr_ousrreq so that old code which depended on it
would break in an obvious manner.  This commit also implements the new
interface for TCP, although the old function is left as an example
(#ifdef'ed out).  This commit ALSO fixes a longstanding bug in the
TCP timer processing (introduced by davidg on 1995/04/12) which caused
timer processing on a TCB to always stop after a single timer had
expired (because it misinterpreted the return value from tcp_usrreq()
to indicate that the TCB had been deleted).  Finally, some code
related to polling has been deleted from if.c because it is not
relevant t -current and doesn't look at all like my current code.
1996-07-11 16:32:50 +00:00

747 lines
18 KiB
C

/*
* Copyright (c) 1980, 1986, 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.
*
* @(#)if.c 8.3 (Berkeley) 1/4/94
* $Id: if.c,v 1.32 1996/06/12 19:23:59 gpalmer Exp $
*/
#include <sys/param.h>
#include <sys/queue.h>
#include <sys/mbuf.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>
#include <sys/kernel.h>
#include <sys/ioctl.h>
#include <sys/errno.h>
#include <sys/syslog.h>
#include <sys/sysctl.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/if_types.h>
#include <net/radix.h>
/*
* System initialization
*/
static int ifconf __P((int, caddr_t));
static void ifinit __P((void *));
static void if_qflush __P((struct ifqueue *));
static void if_slowtimo __P((void *));
static void link_rtrequest __P((int, struct rtentry *, struct sockaddr *));
SYSINIT(interfaces, SI_SUB_PROTO_IF, SI_ORDER_FIRST, ifinit, NULL)
int ifqmaxlen = IFQ_MAXLEN;
struct ifnet *ifnet;
/*
* Network interface utility routines.
*
* Routines with ifa_ifwith* names take sockaddr *'s as
* parameters.
*
* This routine assumes that it will be called at splimp() or higher.
*/
/* ARGSUSED*/
void
ifinit(dummy)
void *dummy;
{
register struct ifnet *ifp;
for (ifp = ifnet; ifp; ifp = ifp->if_next)
if (ifp->if_snd.ifq_maxlen == 0)
ifp->if_snd.ifq_maxlen = ifqmaxlen;
if_slowtimo(0);
}
static int if_index = 0;
static struct ifaddr **ifnet_addrs;
/*
* Attach an interface to the
* list of "active" interfaces.
*/
void
if_attach(ifp)
struct ifnet *ifp;
{
unsigned socksize, ifasize;
int namelen, masklen;
char workbuf[64];
register struct ifnet **p = &ifnet;
register struct sockaddr_dl *sdl;
register struct ifaddr *ifa;
static int if_indexlim = 8;
while (*p)
p = &((*p)->if_next);
*p = ifp;
ifp->if_index = ++if_index;
microtime(&ifp->if_lastchange);
if (ifnet_addrs == 0 || if_index >= if_indexlim) {
unsigned n = (if_indexlim <<= 1) * sizeof(ifa);
struct ifaddr **q = (struct ifaddr **)
malloc(n, M_IFADDR, M_WAITOK);
bzero((caddr_t)q, n);
if (ifnet_addrs) {
bcopy((caddr_t)ifnet_addrs, (caddr_t)q, n/2);
free((caddr_t)ifnet_addrs, M_IFADDR);
}
ifnet_addrs = q;
}
/*
* create a Link Level name for this device
*/
namelen = sprintf(workbuf, "%s%d", ifp->if_name, ifp->if_unit);
#define _offsetof(t, m) ((int)((caddr_t)&((t *)0)->m))
masklen = _offsetof(struct sockaddr_dl, sdl_data[0]) + namelen;
socksize = masklen + ifp->if_addrlen;
#define ROUNDUP(a) (1 + (((a) - 1) | (sizeof(long) - 1)))
socksize = ROUNDUP(socksize);
if (socksize < sizeof(*sdl))
socksize = sizeof(*sdl);
ifasize = sizeof(*ifa) + 2 * socksize;
ifa = (struct ifaddr *)malloc(ifasize, M_IFADDR, M_WAITOK);
if (ifa) {
bzero((caddr_t)ifa, ifasize);
sdl = (struct sockaddr_dl *)(ifa + 1);
sdl->sdl_len = socksize;
sdl->sdl_family = AF_LINK;
bcopy(workbuf, sdl->sdl_data, namelen);
sdl->sdl_nlen = namelen;
sdl->sdl_index = ifp->if_index;
sdl->sdl_type = ifp->if_type;
ifnet_addrs[if_index - 1] = ifa;
ifa->ifa_ifp = ifp;
ifa->ifa_next = ifp->if_addrlist;
ifa->ifa_rtrequest = link_rtrequest;
ifp->if_addrlist = ifa;
ifa->ifa_addr = (struct sockaddr *)sdl;
sdl = (struct sockaddr_dl *)(socksize + (caddr_t)sdl);
ifa->ifa_netmask = (struct sockaddr *)sdl;
sdl->sdl_len = masklen;
while (namelen != 0)
sdl->sdl_data[--namelen] = 0xff;
}
}
/*
* Locate an interface based on a complete address.
*/
/*ARGSUSED*/
struct ifaddr *
ifa_ifwithaddr(addr)
register struct sockaddr *addr;
{
register struct ifnet *ifp;
register struct ifaddr *ifa;
#define equal(a1, a2) \
(bcmp((caddr_t)(a1), (caddr_t)(a2), ((struct sockaddr *)(a1))->sa_len) == 0)
for (ifp = ifnet; ifp; ifp = ifp->if_next)
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next) {
if (ifa->ifa_addr->sa_family != addr->sa_family)
continue;
if (equal(addr, ifa->ifa_addr))
return (ifa);
if ((ifp->if_flags & IFF_BROADCAST) && ifa->ifa_broadaddr &&
equal(ifa->ifa_broadaddr, addr))
return (ifa);
}
return ((struct ifaddr *)0);
}
/*
* Locate the point to point interface with a given destination address.
*/
/*ARGSUSED*/
struct ifaddr *
ifa_ifwithdstaddr(addr)
register struct sockaddr *addr;
{
register struct ifnet *ifp;
register struct ifaddr *ifa;
for (ifp = ifnet; ifp; ifp = ifp->if_next)
if (ifp->if_flags & IFF_POINTOPOINT)
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next) {
if (ifa->ifa_addr->sa_family != addr->sa_family)
continue;
if (ifa->ifa_dstaddr && equal(addr, ifa->ifa_dstaddr))
return (ifa);
}
return ((struct ifaddr *)0);
}
/*
* Find an interface on a specific network. If many, choice
* is most specific found.
*/
struct ifaddr *
ifa_ifwithnet(addr)
struct sockaddr *addr;
{
register struct ifnet *ifp;
register struct ifaddr *ifa;
struct ifaddr *ifa_maybe = (struct ifaddr *) 0;
u_int af = addr->sa_family;
char *addr_data = addr->sa_data, *cplim;
if (af == AF_LINK) {
register struct sockaddr_dl *sdl = (struct sockaddr_dl *)addr;
if (sdl->sdl_index && sdl->sdl_index <= if_index)
return (ifnet_addrs[sdl->sdl_index - 1]);
}
for (ifp = ifnet; ifp; ifp = ifp->if_next) {
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next) {
register char *cp, *cp2, *cp3;
if (ifa->ifa_addr->sa_family != af)
next: continue;
if (ifp->if_flags & IFF_POINTOPOINT) {
if (equal(addr, ifa->ifa_dstaddr))
return (ifa);
} else {
if (ifa->ifa_netmask == 0)
continue;
cp = addr_data;
cp2 = ifa->ifa_addr->sa_data;
cp3 = ifa->ifa_netmask->sa_data;
cplim = ifa->ifa_netmask->sa_len + (char *)ifa->ifa_netmask;
while (cp3 < cplim)
if ((*cp++ ^ *cp2++) & *cp3++)
goto next;
if (ifa_maybe == 0 ||
rn_refines((caddr_t)ifa->ifa_netmask,
(caddr_t)ifa_maybe->ifa_netmask))
ifa_maybe = ifa;
}
}
}
return (ifa_maybe);
}
/*
* Find an interface address specific to an interface best matching
* a given address.
*/
struct ifaddr *
ifaof_ifpforaddr(addr, ifp)
struct sockaddr *addr;
register struct ifnet *ifp;
{
register struct ifaddr *ifa;
register char *cp, *cp2, *cp3;
register char *cplim;
struct ifaddr *ifa_maybe = 0;
u_int af = addr->sa_family;
if (af >= AF_MAX)
return (0);
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next) {
if (ifa->ifa_addr->sa_family != af)
continue;
ifa_maybe = ifa;
if (ifa->ifa_netmask == 0) {
if (equal(addr, ifa->ifa_addr) ||
(ifa->ifa_dstaddr && equal(addr, ifa->ifa_dstaddr)))
return (ifa);
continue;
}
if (ifp->if_flags & IFF_POINTOPOINT) {
if (equal(addr, ifa->ifa_dstaddr))
return (ifa);
} else {
cp = addr->sa_data;
cp2 = ifa->ifa_addr->sa_data;
cp3 = ifa->ifa_netmask->sa_data;
cplim = ifa->ifa_netmask->sa_len + (char *)ifa->ifa_netmask;
for (; cp3 < cplim; cp3++)
if ((*cp++ ^ *cp2++) & *cp3)
break;
if (cp3 == cplim)
return (ifa);
}
}
return (ifa_maybe);
}
#include <net/route.h>
/*
* Default action when installing a route with a Link Level gateway.
* Lookup an appropriate real ifa to point to.
* This should be moved to /sys/net/link.c eventually.
*/
static void
link_rtrequest(cmd, rt, sa)
int cmd;
register struct rtentry *rt;
struct sockaddr *sa;
{
register struct ifaddr *ifa;
struct sockaddr *dst;
struct ifnet *ifp;
if (cmd != RTM_ADD || ((ifa = rt->rt_ifa) == 0) ||
((ifp = ifa->ifa_ifp) == 0) || ((dst = rt_key(rt)) == 0))
return;
ifa = ifaof_ifpforaddr(dst, ifp);
if (ifa) {
IFAFREE(rt->rt_ifa);
rt->rt_ifa = ifa;
ifa->ifa_refcnt++;
if (ifa->ifa_rtrequest && ifa->ifa_rtrequest != link_rtrequest)
ifa->ifa_rtrequest(cmd, rt, sa);
}
}
/*
* Mark an interface down and notify protocols of
* the transition.
* NOTE: must be called at splnet or eqivalent.
*/
void
if_down(ifp)
register struct ifnet *ifp;
{
register struct ifaddr *ifa;
ifp->if_flags &= ~IFF_UP;
microtime(&ifp->if_lastchange);
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next)
pfctlinput(PRC_IFDOWN, ifa->ifa_addr);
if_qflush(&ifp->if_snd);
rt_ifmsg(ifp);
}
/*
* Mark an interface up and notify protocols of
* the transition.
* NOTE: must be called at splnet or eqivalent.
*/
void
if_up(ifp)
register struct ifnet *ifp;
{
ifp->if_flags |= IFF_UP;
microtime(&ifp->if_lastchange);
#ifdef notyet
register struct ifaddr *ifa;
/* this has no effect on IP, and will kill all iso connections XXX */
for (ifa = ifp->if_addrlist; ifa; ifa = ifa->ifa_next)
pfctlinput(PRC_IFUP, ifa->ifa_addr);
#endif
rt_ifmsg(ifp);
}
/*
* Flush an interface queue.
*/
static void
if_qflush(ifq)
register struct ifqueue *ifq;
{
register struct mbuf *m, *n;
n = ifq->ifq_head;
while ((m = n) != 0) {
n = m->m_act;
m_freem(m);
}
ifq->ifq_head = 0;
ifq->ifq_tail = 0;
ifq->ifq_len = 0;
}
/*
* Handle interface watchdog timer routines. Called
* from softclock, we decrement timers (if set) and
* call the appropriate interface routine on expiration.
*/
static void
if_slowtimo(arg)
void *arg;
{
register struct ifnet *ifp;
int s = splimp();
for (ifp = ifnet; ifp; ifp = ifp->if_next) {
if (ifp->if_timer == 0 || --ifp->if_timer)
continue;
if (ifp->if_watchdog)
(*ifp->if_watchdog)(ifp);
}
splx(s);
timeout(if_slowtimo, (void *)0, hz / IFNET_SLOWHZ);
}
/*
* Map interface name to
* interface structure pointer.
*/
struct ifnet *
ifunit(name)
register char *name;
{
register char *cp;
register struct ifnet *ifp;
int unit;
unsigned len;
char *ep, c;
for (cp = name; cp < name + IFNAMSIZ && *cp; cp++)
if (*cp >= '0' && *cp <= '9')
break;
if (*cp == '\0' || cp == name + IFNAMSIZ)
return ((struct ifnet *)0);
/*
* Save first char of unit, and pointer to it,
* so we can put a null there to avoid matching
* initial substrings of interface names.
*/
len = cp - name + 1;
c = *cp;
ep = cp;
for (unit = 0; *cp >= '0' && *cp <= '9'; )
unit = unit * 10 + *cp++ - '0';
if (*cp != '\0')
return 0; /* no trailing garbage allowed */
*ep = 0;
for (ifp = ifnet; ifp; ifp = ifp->if_next) {
if (bcmp(ifp->if_name, name, len))
continue;
if (unit == ifp->if_unit)
break;
}
*ep = c;
return (ifp);
}
/*
* Interface ioctls.
*/
int
ifioctl(so, cmd, data, p)
struct socket *so;
int cmd;
caddr_t data;
struct proc *p;
{
register struct ifnet *ifp;
register struct ifreq *ifr;
int error;
switch (cmd) {
case SIOCGIFCONF:
case OSIOCGIFCONF:
return (ifconf(cmd, data));
}
ifr = (struct ifreq *)data;
ifp = ifunit(ifr->ifr_name);
if (ifp == 0)
return (ENXIO);
switch (cmd) {
case SIOCGIFFLAGS:
ifr->ifr_flags = ifp->if_flags;
break;
case SIOCGIFMETRIC:
ifr->ifr_metric = ifp->if_metric;
break;
case SIOCGIFMTU:
ifr->ifr_mtu = ifp->if_mtu;
break;
case SIOCGIFPHYS:
ifr->ifr_phys = ifp->if_physical;
break;
case SIOCSIFFLAGS:
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return (error);
if (ifp->if_flags & IFF_UP && (ifr->ifr_flags & IFF_UP) == 0) {
int s = splimp();
if_down(ifp);
splx(s);
}
if (ifr->ifr_flags & IFF_UP && (ifp->if_flags & IFF_UP) == 0) {
int s = splimp();
if_up(ifp);
splx(s);
}
ifp->if_flags = (ifp->if_flags & IFF_CANTCHANGE) |
(ifr->ifr_flags &~ IFF_CANTCHANGE);
if (ifp->if_ioctl)
(void) (*ifp->if_ioctl)(ifp, cmd, data);
microtime(&ifp->if_lastchange);
break;
case SIOCSIFMETRIC:
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return (error);
ifp->if_metric = ifr->ifr_metric;
microtime(&ifp->if_lastchange);
break;
case SIOCSIFPHYS:
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return error;
if (!ifp->if_ioctl)
return EOPNOTSUPP;
error = (*ifp->if_ioctl)(ifp, cmd, data);
if (error == 0)
microtime(&ifp->if_lastchange);
return(error);
case SIOCSIFMTU:
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return (error);
if (ifp->if_ioctl == NULL)
return (EOPNOTSUPP);
/*
* 72 was chosen below because it is the size of a TCP/IP
* header (40) + the minimum mss (32).
*/
if (ifr->ifr_mtu < 72 || ifr->ifr_mtu > 65535)
return (EINVAL);
error = (*ifp->if_ioctl)(ifp, cmd, data);
if (error == 0)
microtime(&ifp->if_lastchange);
return(error);
case SIOCADDMULTI:
case SIOCDELMULTI:
error = suser(p->p_ucred, &p->p_acflag);
if (error)
return (error);
if (ifp->if_ioctl == NULL)
return (EOPNOTSUPP);
error = (*ifp->if_ioctl)(ifp, cmd, data);
if (error == 0 )
microtime(&ifp->if_lastchange);
return(error);
default:
if (so->so_proto == 0)
return (EOPNOTSUPP);
#ifndef COMPAT_43
return ((*so->so_proto->pr_usrreqs->pru_control)(so, cmd,
data,
ifp));
#else
{
int ocmd = cmd;
switch (cmd) {
case SIOCSIFDSTADDR:
case SIOCSIFADDR:
case SIOCSIFBRDADDR:
case SIOCSIFNETMASK:
#if BYTE_ORDER != BIG_ENDIAN
if (ifr->ifr_addr.sa_family == 0 &&
ifr->ifr_addr.sa_len < 16) {
ifr->ifr_addr.sa_family = ifr->ifr_addr.sa_len;
ifr->ifr_addr.sa_len = 16;
}
#else
if (ifr->ifr_addr.sa_len == 0)
ifr->ifr_addr.sa_len = 16;
#endif
break;
case OSIOCGIFADDR:
cmd = SIOCGIFADDR;
break;
case OSIOCGIFDSTADDR:
cmd = SIOCGIFDSTADDR;
break;
case OSIOCGIFBRDADDR:
cmd = SIOCGIFBRDADDR;
break;
case OSIOCGIFNETMASK:
cmd = SIOCGIFNETMASK;
}
error = ((*so->so_proto->pr_usrreqs->pru_control)(so,
cmd,
data,
ifp));
switch (ocmd) {
case OSIOCGIFADDR:
case OSIOCGIFDSTADDR:
case OSIOCGIFBRDADDR:
case OSIOCGIFNETMASK:
*(u_short *)&ifr->ifr_addr = ifr->ifr_addr.sa_family;
}
return (error);
}
#endif
}
return (0);
}
/*
* Set/clear promiscuous mode on interface ifp based on the truth value
* of pswitch. The calls are reference counted so that only the first
* "on" request actually has an effect, as does the final "off" request.
* Results are undefined if the "off" and "on" requests are not matched.
*/
int
ifpromisc(ifp, pswitch)
struct ifnet *ifp;
int pswitch;
{
struct ifreq ifr;
if (pswitch) {
/*
* If the device is not configured up, we cannot put it in
* promiscuous mode.
*/
if ((ifp->if_flags & IFF_UP) == 0)
return (ENETDOWN);
if (ifp->if_pcount++ != 0)
return (0);
ifp->if_flags |= IFF_PROMISC;
log(LOG_INFO, "%s%d: promiscuous mode enabled\n",
ifp->if_name, ifp->if_unit);
} else {
if (--ifp->if_pcount > 0)
return (0);
ifp->if_flags &= ~IFF_PROMISC;
}
ifr.ifr_flags = ifp->if_flags;
return ((*ifp->if_ioctl)(ifp, SIOCSIFFLAGS, (caddr_t)&ifr));
}
/*
* Return interface configuration
* of system. List may be used
* in later ioctl's (above) to get
* other information.
*/
/*ARGSUSED*/
static int
ifconf(cmd, data)
int cmd;
caddr_t data;
{
register struct ifconf *ifc = (struct ifconf *)data;
register struct ifnet *ifp = ifnet;
register struct ifaddr *ifa;
struct ifreq ifr, *ifrp;
int space = ifc->ifc_len, error = 0;
ifrp = ifc->ifc_req;
for (; space > sizeof (ifr) && ifp; ifp = ifp->if_next) {
char workbuf[64];
int ifnlen;
ifnlen = sprintf(workbuf, "%s%d", ifp->if_name, ifp->if_unit);
if(ifnlen + 1 > sizeof ifr.ifr_name) {
error = ENAMETOOLONG;
} else {
strcpy(ifr.ifr_name, workbuf);
}
if ((ifa = ifp->if_addrlist) == 0) {
bzero((caddr_t)&ifr.ifr_addr, sizeof(ifr.ifr_addr));
error = copyout((caddr_t)&ifr, (caddr_t)ifrp,
sizeof (ifr));
if (error)
break;
space -= sizeof (ifr), ifrp++;
} else
for ( ; space > sizeof (ifr) && ifa; ifa = ifa->ifa_next) {
register struct sockaddr *sa = ifa->ifa_addr;
#ifdef COMPAT_43
if (cmd == OSIOCGIFCONF) {
struct osockaddr *osa =
(struct osockaddr *)&ifr.ifr_addr;
ifr.ifr_addr = *sa;
osa->sa_family = sa->sa_family;
error = copyout((caddr_t)&ifr, (caddr_t)ifrp,
sizeof (ifr));
ifrp++;
} else
#endif
if (sa->sa_len <= sizeof(*sa)) {
ifr.ifr_addr = *sa;
error = copyout((caddr_t)&ifr, (caddr_t)ifrp,
sizeof (ifr));
ifrp++;
} else {
space -= sa->sa_len - sizeof(*sa);
if (space < sizeof (ifr))
break;
error = copyout((caddr_t)&ifr, (caddr_t)ifrp,
sizeof (ifr.ifr_name));
if (error == 0)
error = copyout((caddr_t)sa,
(caddr_t)&ifrp->ifr_addr, sa->sa_len);
ifrp = (struct ifreq *)
(sa->sa_len + (caddr_t)&ifrp->ifr_addr);
}
if (error)
break;
space -= sizeof (ifr);
}
}
ifc->ifc_len -= space;
return (error);
}
SYSCTL_NODE(_net, PF_LINK, link, CTLFLAG_RW, 0, "Link layers");
SYSCTL_NODE(_net_link, 0, generic, CTLFLAG_RW, 0, "Generic link-management");