freebsd-dev/lib/libc/net/getifaddrs.c
Gleb Smirnoff 08b68b0e4c A major overhaul of the CARP implementation. The ip_carp.c was started
from scratch, copying needed functionality from the old implemenation
on demand, with a thorough review of all code. The main change is that
interface layer has been removed from the CARP. Now redundant addresses
are configured exactly on the interfaces, they run on.

The CARP configuration itself is, as before, configured and read via
SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
which makes the prefix redundant.

ifconfig(8) semantics has been changed too: now one doesn't need
to clone carpXX interface, he/she should directly configure a vhid
on a Ethernet interface.

To supply vhid data from the kernel to an application the getifaddrs(8)
function had been changed to pass ifam_data with each address. [1]

The new implementation definitely closes all PRs related to carp(4)
being an interface, and may close several others. It also allows
to run a single redundant IP per interface.

Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
idea on using ifam_data and for several rounds of reviewing!

PR:		kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
Reviewed by:	bz
Submitted by:	bz [1]
2011-12-16 12:16:56 +00:00

419 lines
10 KiB
C

/* $KAME: getifaddrs.c,v 1.9 2001/08/20 02:31:20 itojun Exp $ */
/*
* Copyright (c) 1995, 1999
* Berkeley Software Design, Inc. 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.
*
* THIS SOFTWARE IS PROVIDED BY Berkeley Software Design, Inc. ``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 Berkeley Software Design, Inc. 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.
*
* BSDI getifaddrs.c,v 2.12 2000/02/23 14:51:59 dab Exp
*/
/*
* NOTE: SIOCGIFCONF case is not LP64 friendly. it also does not perform
* try-and-error for region size.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "namespace.h"
#include <sys/types.h>
#include <sys/ioctl.h>
#include <sys/socket.h>
#include <net/if.h>
#ifdef NET_RT_IFLIST
#include <sys/param.h>
#include <net/route.h>
#include <sys/sysctl.h>
#include <net/if_dl.h>
#endif
#include <errno.h>
#include <ifaddrs.h>
#include <stdlib.h>
#include <string.h>
#include "un-namespace.h"
#if !defined(AF_LINK)
#define SA_LEN(sa) sizeof(struct sockaddr)
#endif
#if !defined(SA_LEN)
#define SA_LEN(sa) (sa)->sa_len
#endif
#define SALIGN (sizeof(long) - 1)
#define SA_RLEN(sa) ((sa)->sa_len ? (((sa)->sa_len + SALIGN) & ~SALIGN) : (SALIGN + 1))
#ifndef ALIGNBYTES
/*
* On systems with a routing socket, ALIGNBYTES should match the value
* that the kernel uses when building the messages.
*/
#define ALIGNBYTES XXX
#endif
#ifndef ALIGN
#define ALIGN(p) (((u_long)(p) + ALIGNBYTES) &~ ALIGNBYTES)
#endif
#if _BSDI_VERSION >= 199701
#define HAVE_IFM_DATA
#endif
#if (_BSDI_VERSION >= 199802) || (__FreeBSD_version >= 1000003)
/* ifam_data is very specific to recent versions of bsdi */
#define HAVE_IFAM_DATA
#endif
#if defined(__NetBSD__) || defined(__OpenBSD__) || defined(__FreeBSD__)
#define HAVE_IFM_DATA
#endif
#define MAX_SYSCTL_TRY 5
int
getifaddrs(struct ifaddrs **pif)
{
int icnt = 1;
int dcnt = 0;
int ncnt = 0;
#ifdef NET_RT_IFLIST
int ntry = 0;
int mib[6];
size_t needed;
char *buf;
char *next;
struct ifaddrs *cif = 0;
char *p, *p0;
struct rt_msghdr *rtm;
struct if_msghdr *ifm;
struct ifa_msghdr *ifam;
struct sockaddr_dl *dl;
struct sockaddr *sa;
struct ifaddrs *ifa, *ift;
u_short idx = 0;
#else /* NET_RT_IFLIST */
char buf[1024];
int m, sock;
struct ifconf ifc;
struct ifreq *ifr;
struct ifreq *lifr;
#endif /* NET_RT_IFLIST */
int i;
size_t len, alen;
char *data;
char *names;
#ifdef NET_RT_IFLIST
mib[0] = CTL_NET;
mib[1] = PF_ROUTE;
mib[2] = 0; /* protocol */
mib[3] = 0; /* wildcard address family */
mib[4] = NET_RT_IFLIST;
mib[5] = 0; /* no flags */
do {
/*
* We'll try to get addresses several times in case that
* the number of addresses is unexpectedly increased during
* the two sysctl calls. This should rarely happen, but we'll
* try to do our best for applications that assume success of
* this library (which should usually be the case).
* Portability note: since FreeBSD does not add margin of
* memory at the first sysctl, the possibility of failure on
* the second sysctl call is a bit higher.
*/
if (sysctl(mib, 6, NULL, &needed, NULL, 0) < 0)
return (-1);
if ((buf = malloc(needed)) == NULL)
return (-1);
if (sysctl(mib, 6, buf, &needed, NULL, 0) < 0) {
if (errno != ENOMEM || ++ntry >= MAX_SYSCTL_TRY) {
free(buf);
return (-1);
}
free(buf);
buf = NULL;
}
} while (buf == NULL);
for (next = buf; next < buf + needed; next += rtm->rtm_msglen) {
rtm = (struct rt_msghdr *)(void *)next;
if (rtm->rtm_version != RTM_VERSION)
continue;
switch (rtm->rtm_type) {
case RTM_IFINFO:
ifm = (struct if_msghdr *)(void *)rtm;
if (ifm->ifm_addrs & RTA_IFP) {
idx = ifm->ifm_index;
++icnt;
dl = (struct sockaddr_dl *)(void *)(ifm + 1);
dcnt += SA_RLEN((struct sockaddr *)(void*)dl) +
ALIGNBYTES;
#ifdef HAVE_IFM_DATA
dcnt += sizeof(ifm->ifm_data);
#endif /* HAVE_IFM_DATA */
ncnt += dl->sdl_nlen + 1;
} else
idx = 0;
break;
case RTM_NEWADDR:
ifam = (struct ifa_msghdr *)(void *)rtm;
if (idx && ifam->ifam_index != idx)
abort(); /* this cannot happen */
#define RTA_MASKS (RTA_NETMASK | RTA_IFA | RTA_BRD)
if (idx == 0 || (ifam->ifam_addrs & RTA_MASKS) == 0)
break;
p = (char *)(void *)(ifam + 1);
++icnt;
#ifdef HAVE_IFAM_DATA
dcnt += sizeof(ifam->ifam_data) + ALIGNBYTES;
#endif /* HAVE_IFAM_DATA */
/* Scan to look for length of address */
alen = 0;
for (p0 = p, i = 0; i < RTAX_MAX; i++) {
if ((RTA_MASKS & ifam->ifam_addrs & (1 << i))
== 0)
continue;
sa = (struct sockaddr *)(void *)p;
len = SA_RLEN(sa);
if (i == RTAX_IFA) {
alen = len;
break;
}
p += len;
}
for (p = p0, i = 0; i < RTAX_MAX; i++) {
if ((RTA_MASKS & ifam->ifam_addrs & (1 << i))
== 0)
continue;
sa = (struct sockaddr *)(void *)p;
len = SA_RLEN(sa);
if (i == RTAX_NETMASK && SA_LEN(sa) == 0)
dcnt += alen;
else
dcnt += len;
p += len;
}
break;
}
}
#else /* NET_RT_IFLIST */
ifc.ifc_buf = buf;
ifc.ifc_len = sizeof(buf);
if ((sock = _socket(AF_INET, SOCK_STREAM, 0)) < 0)
return (-1);
i = _ioctl(sock, SIOCGIFCONF, (char *)&ifc);
_close(sock);
if (i < 0)
return (-1);
ifr = ifc.ifc_req;
lifr = (struct ifreq *)&ifc.ifc_buf[ifc.ifc_len];
while (ifr < lifr) {
struct sockaddr *sa;
sa = &ifr->ifr_addr;
++icnt;
dcnt += SA_RLEN(sa);
ncnt += sizeof(ifr->ifr_name) + 1;
if (SA_LEN(sa) < sizeof(*sa))
ifr = (struct ifreq *)(((char *)sa) + sizeof(*sa));
else
ifr = (struct ifreq *)(((char *)sa) + SA_LEN(sa));
}
#endif /* NET_RT_IFLIST */
if (icnt + dcnt + ncnt == 1) {
*pif = NULL;
free(buf);
return (0);
}
data = malloc(sizeof(struct ifaddrs) * icnt + dcnt + ncnt);
if (data == NULL) {
free(buf);
return(-1);
}
ifa = (struct ifaddrs *)(void *)data;
data += sizeof(struct ifaddrs) * icnt;
names = data + dcnt;
memset(ifa, 0, sizeof(struct ifaddrs) * icnt);
ift = ifa;
#ifdef NET_RT_IFLIST
idx = 0;
for (next = buf; next < buf + needed; next += rtm->rtm_msglen) {
rtm = (struct rt_msghdr *)(void *)next;
if (rtm->rtm_version != RTM_VERSION)
continue;
switch (rtm->rtm_type) {
case RTM_IFINFO:
ifm = (struct if_msghdr *)(void *)rtm;
if (ifm->ifm_addrs & RTA_IFP) {
idx = ifm->ifm_index;
dl = (struct sockaddr_dl *)(void *)(ifm + 1);
cif = ift;
ift->ifa_name = names;
ift->ifa_flags = (int)ifm->ifm_flags;
memcpy(names, dl->sdl_data,
(size_t)dl->sdl_nlen);
names[dl->sdl_nlen] = 0;
names += dl->sdl_nlen + 1;
ift->ifa_addr = (struct sockaddr *)(void *)data;
memcpy(data, dl,
(size_t)SA_LEN((struct sockaddr *)
(void *)dl));
data += SA_RLEN((struct sockaddr *)(void *)dl);
#ifdef HAVE_IFM_DATA
/* ifm_data needs to be aligned */
ift->ifa_data = data = (void *)ALIGN(data);
memcpy(data, &ifm->ifm_data, sizeof(ifm->ifm_data));
data += sizeof(ifm->ifm_data);
#else /* HAVE_IFM_DATA */
ift->ifa_data = NULL;
#endif /* HAVE_IFM_DATA */
ift = (ift->ifa_next = ift + 1);
} else
idx = 0;
break;
case RTM_NEWADDR:
ifam = (struct ifa_msghdr *)(void *)rtm;
if (idx && ifam->ifam_index != idx)
abort(); /* this cannot happen */
if (idx == 0 || (ifam->ifam_addrs & RTA_MASKS) == 0)
break;
ift->ifa_name = cif->ifa_name;
ift->ifa_flags = cif->ifa_flags;
ift->ifa_data = NULL;
p = (char *)(void *)(ifam + 1);
/* Scan to look for length of address */
alen = 0;
for (p0 = p, i = 0; i < RTAX_MAX; i++) {
if ((RTA_MASKS & ifam->ifam_addrs & (1 << i))
== 0)
continue;
sa = (struct sockaddr *)(void *)p;
len = SA_RLEN(sa);
if (i == RTAX_IFA) {
alen = len;
break;
}
p += len;
}
for (p = p0, i = 0; i < RTAX_MAX; i++) {
if ((RTA_MASKS & ifam->ifam_addrs & (1 << i))
== 0)
continue;
sa = (struct sockaddr *)(void *)p;
len = SA_RLEN(sa);
switch (i) {
case RTAX_IFA:
ift->ifa_addr =
(struct sockaddr *)(void *)data;
memcpy(data, p, len);
data += len;
break;
case RTAX_NETMASK:
ift->ifa_netmask =
(struct sockaddr *)(void *)data;
if (SA_LEN(sa) == 0) {
memset(data, 0, alen);
data += alen;
break;
}
memcpy(data, p, len);
data += len;
break;
case RTAX_BRD:
ift->ifa_broadaddr =
(struct sockaddr *)(void *)data;
memcpy(data, p, len);
data += len;
break;
}
p += len;
}
#ifdef HAVE_IFAM_DATA
/* ifam_data needs to be aligned */
ift->ifa_data = data = (void *)ALIGN(data);
memcpy(data, &ifam->ifam_data, sizeof(ifam->ifam_data));
data += sizeof(ifam->ifam_data);
#endif /* HAVE_IFAM_DATA */
ift = (ift->ifa_next = ift + 1);
break;
}
}
free(buf);
#else /* NET_RT_IFLIST */
ifr = ifc.ifc_req;
lifr = (struct ifreq *)&ifc.ifc_buf[ifc.ifc_len];
while (ifr < lifr) {
struct sockaddr *sa;
ift->ifa_name = names;
names[sizeof(ifr->ifr_name)] = 0;
strncpy(names, ifr->ifr_name, sizeof(ifr->ifr_name));
while (*names++)
;
ift->ifa_addr = (struct sockaddr *)data;
sa = &ifr->ifr_addr;
memcpy(data, sa, SA_LEN(sa));
data += SA_RLEN(sa);
ifr = (struct ifreq *)(((char *)sa) + SA_LEN(sa));
ift = (ift->ifa_next = ift + 1);
}
#endif /* NET_RT_IFLIST */
if (--ift >= ifa) {
ift->ifa_next = NULL;
*pif = ifa;
} else {
*pif = NULL;
free(ifa);
}
return (0);
}
void
freeifaddrs(struct ifaddrs *ifp)
{
free(ifp);
}