freebsd-dev/sys/netinet/ip_fw.c

2260 lines
60 KiB
C
Raw Normal View History

/*
* Copyright (c) 1993 Daniel Boulet
* Copyright (c) 1994 Ugen J.S.Antsilevich
* Copyright (c) 1996 Alex Nash
* Copyright (c) 2000-2002 Luigi Rizzo
*
* Redistribution and use in source forms, with and without modification,
* are permitted provided that this entire comment appears intact.
*
* Redistribution in binary form may occur without any restrictions.
* Obviously, it would be nice if you gave credit where credit is due
* but requiring it would be too onerous.
*
* This software is provided ``AS IS'' without any warranties of any kind.
1995-07-23 05:36:31 +00:00
*
1999-08-28 01:08:13 +00:00
* $FreeBSD$
*/
#define DEB(x)
#define DDB(x) x
/*
* Implement IP packet firewall
*/
1996-06-12 19:34:33 +00:00
1999-04-20 14:29:59 +00:00
#if !defined(KLD_MODULE)
1996-06-12 19:34:33 +00:00
#include "opt_ipfw.h"
#include "opt_ipdn.h"
#include "opt_ipdivert.h"
#include "opt_inet.h"
#ifndef INET
#error IPFIREWALL requires INET.
#endif /* INET */
#endif
1996-06-12 19:34:33 +00:00
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/kernel.h>
#include <sys/proc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/ucred.h>
#include <net/if.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/in_pcb.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#include <netinet/ip_icmp.h>
#include <netinet/ip_fw.h>
#include <netinet/ip_dummynet.h>
#include <netinet/tcp.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcpip.h>
#include <netinet/udp.h>
#include <netinet/udp_var.h>
#include <netinet/if_ether.h> /* XXX ethertype_ip */
static int fw_debug = 1;
#ifdef IPFIREWALL_VERBOSE
static int fw_verbose = 1;
#else
static int fw_verbose = 0;
#endif
int fw_one_pass = 1 ;
#ifdef IPFIREWALL_VERBOSE_LIMIT
static int fw_verbose_limit = IPFIREWALL_VERBOSE_LIMIT;
#else
static int fw_verbose_limit = 0;
#endif
static int fw_permanent_rules = 0;
/*
* Right now, two fields in the IP header are changed to host format
* by the IP layer before calling the firewall. Ideally, we would like
* to have them in network format so that the packet can be
* used as it comes from the device driver (and is thus readonly).
*/
static u_int64_t counter; /* counter for ipfw_report(NULL...) */
#define IPFW_DEFAULT_RULE ((u_int)(u_short)~0)
LIST_HEAD (ip_fw_head, ip_fw) ip_fw_chain_head;
1995-07-04 05:29:30 +00:00
MALLOC_DEFINE(M_IPFW, "IpFw/IpAcct", "IpFw/IpAcct chain's");
#ifdef SYSCTL_NODE
SYSCTL_NODE(_net_inet_ip, OID_AUTO, fw, CTLFLAG_RW, 0, "Firewall");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, enable, CTLFLAG_RW,
&fw_enable, 0, "Enable ipfw");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO,one_pass,CTLFLAG_RW,
&fw_one_pass, 0,
"Only do a single pass through ipfw when using dummynet(4)");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, debug, CTLFLAG_RW,
&fw_debug, 0, "Enable printing of debug ip_fw statements");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, verbose, CTLFLAG_RW,
&fw_verbose, 0, "Log matches to ipfw rules");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, verbose_limit, CTLFLAG_RW,
&fw_verbose_limit, 0, "Set upper limit of matches of ipfw rules logged");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, permanent_rules, CTLFLAG_RW,
&fw_permanent_rules, 0, "Set rule number, below which rules are permanent");
/*
* Extension for stateful ipfw.
*
* Dynamic rules are stored in lists accessed through a hash table
* (ipfw_dyn_v) whose size is curr_dyn_buckets. This value can
* be modified through the sysctl variable dyn_buckets which is
* updated when the table becomes empty.
*
* XXX currently there is only one list, ipfw_dyn.
*
* When a packet is received, it is first hashed, then matched
* against the entries in the corresponding list.
* Matching occurs according to the rule type. The default is to
* match the four fields and the protocol, and rules are bidirectional.
*
* For a busy proxy/web server we will have lots of connections to
* the server. We could decide for a rule type where we ignore
* ports (different hashing) and avoid special SYN/RST/FIN handling.
*
* XXX when we decide to support more than one rule type, we should
* repeat the hashing multiple times uing only the useful fields.
* Or, we could run the various tests in parallel, because the
* 'move to front' technique should shorten the average search.
*
* The lifetime of dynamic rules is regulated by dyn_*_lifetime,
* measured in seconds and depending on the flags.
*
* The total number of dynamic rules is stored in dyn_count.
* The max number of dynamic rules is dyn_max. When we reach
* the maximum number of rules we do not create anymore. This is
* done to avoid consuming too much memory, but also too much
* time when searching on each packet (ideally, we should try instead
* to put a limit on the length of the list on each bucket...).
*
* Each dynamic rules holds a pointer to the parent ipfw rule so
* we know what action to perform. Dynamic rules are removed when
* the parent rule is deleted.
* There are some limitations with dynamic rules -- we do not
* obey the 'randomized match', and we do not do multiple
* passes through the firewall.
* XXX check the latter!!!
*/
static struct ipfw_dyn_rule **ipfw_dyn_v = NULL ;
static u_int32_t dyn_buckets = 256 ; /* must be power of 2 */
static u_int32_t curr_dyn_buckets = 256 ; /* must be power of 2 */
/*
* timeouts for various events in handing dynamic rules.
*/
static u_int32_t dyn_ack_lifetime = 300 ;
static u_int32_t dyn_syn_lifetime = 20 ;
static u_int32_t dyn_fin_lifetime = 1 ;
static u_int32_t dyn_rst_lifetime = 1 ;
static u_int32_t dyn_udp_lifetime = 10 ;
static u_int32_t dyn_short_lifetime = 5 ;
/*
* after reaching 0, dynamic rules are considered still valid for
* an additional grace time, unless there is lack of resources.
*/
static u_int32_t dyn_grace_time = 10 ;
static u_int32_t static_count = 0 ; /* # of static rules */
static u_int32_t dyn_count = 0 ; /* # of dynamic rules */
static u_int32_t dyn_max = 1000 ; /* max # of dynamic rules */
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_buckets, CTLFLAG_RW,
&dyn_buckets, 0, "Number of dyn. buckets");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, curr_dyn_buckets, CTLFLAG_RD,
&curr_dyn_buckets, 0, "Current Number of dyn. buckets");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_count, CTLFLAG_RD,
&dyn_count, 0, "Number of dyn. rules");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_max, CTLFLAG_RW,
&dyn_max, 0, "Max number of dyn. rules");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, static_count, CTLFLAG_RD,
&static_count, 0, "Number of static rules");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_ack_lifetime, CTLFLAG_RW,
&dyn_ack_lifetime, 0, "Lifetime of dyn. rules for acks");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_syn_lifetime, CTLFLAG_RW,
&dyn_syn_lifetime, 0, "Lifetime of dyn. rules for syn");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_fin_lifetime, CTLFLAG_RW,
&dyn_fin_lifetime, 0, "Lifetime of dyn. rules for fin");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_rst_lifetime, CTLFLAG_RW,
&dyn_rst_lifetime, 0, "Lifetime of dyn. rules for rst");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_udp_lifetime, CTLFLAG_RW,
&dyn_udp_lifetime, 0, "Lifetime of dyn. rules for UDP");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_short_lifetime, CTLFLAG_RW,
&dyn_short_lifetime, 0, "Lifetime of dyn. rules for other situations");
SYSCTL_INT(_net_inet_ip_fw, OID_AUTO, dyn_grace_time, CTLFLAG_RD,
&dyn_grace_time, 0, "Grace time for dyn. rules");
#endif /* SYSCTL_NODE */
#define dprintf(a) do { \
if (fw_debug) \
printf a; \
} while (0)
#define SNPARGS(buf, len) buf + len, sizeof(buf) > len ? sizeof(buf) - len : 0
static int add_entry (struct ip_fw_head *chainptr, struct ip_fw *frwl);
static int del_entry (struct ip_fw_head *chainptr, u_short number);
static int zero_entry (struct ip_fw *, int);
static int check_ipfw_struct (struct ip_fw *m);
static int iface_match (struct ifnet *ifp, union ip_fw_if *ifu,
int byname);
static int ipopts_match (struct ip *ip, struct ip_fw *f);
static int iptos_match (struct ip *ip, struct ip_fw *f);
1998-04-15 17:47:40 +00:00
static __inline int
port_match (u_short *portptr, int nports, u_short port,
int range_flag, int mask);
static int tcpflg_match (struct tcphdr *tcp, struct ip_fw *f);
static int icmptype_match (struct icmp * icmp, struct ip_fw * f);
static void ipfw_report (struct ip_fw *f, struct ip *ip, int ip_off,
int ip_len, struct ifnet *rif,
struct ifnet *oif);
static void flush_rule_ptrs(void);
static ip_fw_chk_t ip_fw_chk;
static int ip_fw_ctl (struct sockopt *sopt);
ip_dn_ruledel_t *ip_dn_ruledel_ptr = NULL;
static char err_prefix[] = "ip_fw_ctl:";
/*
* Returns 1 if the port is matched by the vector, 0 otherwise
*/
1998-04-15 17:47:40 +00:00
static __inline int
port_match(u_short *portptr, int nports, u_short port, int range_flag, int mask)
{
if (!nports)
return 1;
if (mask) {
if ( 0 == ((portptr[0] ^ port) & portptr[1]) )
return 1;
nports -= 2;
portptr += 2;
}
if (range_flag) {
if (portptr[0] <= port && port <= portptr[1])
return 1;
nports -= 2;
portptr += 2;
}
while (nports-- > 0)
if (*portptr++ == port)
return 1;
return 0;
}
static int
tcpflg_match(struct tcphdr *tcp, struct ip_fw *f)
{
u_char flg_set, flg_clr;
/*
* If an established connection is required, reject packets that
* have only SYN of RST|ACK|SYN set. Otherwise, fall through to
* other flag requirements.
*/
if ((f->fw_ipflg & IP_FW_IF_TCPEST) &&
((tcp->th_flags & (TH_RST | TH_ACK | TH_SYN)) == TH_SYN))
return 0;
flg_set = tcp->th_flags & f->fw_tcpf;
flg_clr = tcp->th_flags & f->fw_tcpnf;
if (flg_set != f->fw_tcpf)
return 0;
if (flg_clr)
return 0;
return 1;
}
static int
icmptype_match(struct icmp *icmp, struct ip_fw *f)
{
int type;
if (!(f->fw_flg & IP_FW_F_ICMPBIT))
return(1);
type = icmp->icmp_type;
/* check for matching type in the bitmap */
if (type < IP_FW_ICMPTYPES_MAX &&
1999-07-28 22:27:27 +00:00
(f->fw_uar.fw_icmptypes[type / (sizeof(unsigned) * NBBY)] &
(1U << (type % (sizeof(unsigned) * NBBY)))))
return(1);
return(0); /* no match */
}
static int
is_icmp_query(struct ip *ip)
{
const struct icmp *icmp;
int icmp_type;
icmp = (struct icmp *)((u_int32_t *)ip + ip->ip_hl);
icmp_type = icmp->icmp_type;
if (icmp_type == ICMP_ECHO || icmp_type == ICMP_ROUTERSOLICIT ||
icmp_type == ICMP_TSTAMP || icmp_type == ICMP_IREQ ||
icmp_type == ICMP_MASKREQ)
return(1);
return(0);
}
static int
ipopts_match(struct ip *ip, struct ip_fw *f)
{
register u_char *cp;
int opt, optlen, cnt;
u_char opts, nopts, nopts_sve;
cp = (u_char *)(ip + 1);
cnt = (ip->ip_hl << 2) - sizeof (struct ip);
opts = f->fw_ipopt;
nopts = nopts_sve = f->fw_ipnopt;
for (; cnt > 0; cnt -= optlen, cp += optlen) {
opt = cp[IPOPT_OPTVAL];
if (opt == IPOPT_EOL)
break;
if (opt == IPOPT_NOP)
optlen = 1;
else {
optlen = cp[IPOPT_OLEN];
if (optlen <= 0 || optlen > cnt) {
return 0; /*XXX*/
}
}
switch (opt) {
default:
break;
case IPOPT_LSRR:
opts &= ~IP_FW_IPOPT_LSRR;
nopts &= ~IP_FW_IPOPT_LSRR;
break;
case IPOPT_SSRR:
opts &= ~IP_FW_IPOPT_SSRR;
nopts &= ~IP_FW_IPOPT_SSRR;
break;
case IPOPT_RR:
opts &= ~IP_FW_IPOPT_RR;
nopts &= ~IP_FW_IPOPT_RR;
break;
case IPOPT_TS:
opts &= ~IP_FW_IPOPT_TS;
nopts &= ~IP_FW_IPOPT_TS;
break;
}
if (opts == nopts)
break;
}
if (opts == 0 && nopts == nopts_sve)
return 1;
else
return 0;
}
static int
iptos_match(struct ip *ip, struct ip_fw *f)
{
u_int flags = (ip->ip_tos & 0x1f);
u_char opts, nopts, nopts_sve;
opts = (f->fw_iptos & 0x1f);
nopts = nopts_sve = f->fw_ipntos;
while (flags != 0) {
u_int flag;
flag = 1 << (ffs(flags) -1);
opts &= ~flag;
nopts &= ~flag;
flags &= ~flag;
}
if (opts == 0 && nopts == nopts_sve)
return 1;
else
return 0;
}
static int
tcpopts_match(struct tcphdr *tcp, struct ip_fw *f)
{
register u_char *cp;
int opt, optlen, cnt;
u_char opts, nopts, nopts_sve;
cp = (u_char *)(tcp + 1);
cnt = (tcp->th_off << 2) - sizeof (struct tcphdr);
opts = f->fw_tcpopt;
nopts = nopts_sve = f->fw_tcpnopt;
for (; cnt > 0; cnt -= optlen, cp += optlen) {
opt = cp[0];
if (opt == TCPOPT_EOL)
break;
if (opt == TCPOPT_NOP)
optlen = 1;
else {
optlen = cp[1];
if (optlen <= 0)
break;
}
switch (opt) {
default:
break;
case TCPOPT_MAXSEG:
opts &= ~IP_FW_TCPOPT_MSS;
nopts &= ~IP_FW_TCPOPT_MSS;
break;
case TCPOPT_WINDOW:
opts &= ~IP_FW_TCPOPT_WINDOW;
nopts &= ~IP_FW_TCPOPT_WINDOW;
break;
case TCPOPT_SACK_PERMITTED:
case TCPOPT_SACK:
opts &= ~IP_FW_TCPOPT_SACK;
nopts &= ~IP_FW_TCPOPT_SACK;
break;
case TCPOPT_TIMESTAMP:
opts &= ~IP_FW_TCPOPT_TS;
nopts &= ~IP_FW_TCPOPT_TS;
break;
case TCPOPT_CC:
case TCPOPT_CCNEW:
case TCPOPT_CCECHO:
opts &= ~IP_FW_TCPOPT_CC;
nopts &= ~IP_FW_TCPOPT_CC;
break;
}
if (opts == nopts)
break;
}
if (opts == 0 && nopts == nopts_sve)
return 1;
else
return 0;
}
static int
iface_match(struct ifnet *ifp, union ip_fw_if *ifu, int byname)
{
/* Check by name or by IP address */
if (byname) {
/* Check unit number (-1 is wildcard) */
if (ifu->fu_via_if.unit != -1
&& ifp->if_unit != ifu->fu_via_if.unit)
return(0);
/* Check name */
if (strncmp(ifp->if_name, ifu->fu_via_if.name, FW_IFNLEN))
return(0);
return(1);
} else if (ifu->fu_via_ip.s_addr != 0) { /* Zero == wildcard */
struct ifaddr *ia;
TAILQ_FOREACH(ia, &ifp->if_addrhead, ifa_link) {
if (ia->ifa_addr == NULL)
continue;
if (ia->ifa_addr->sa_family != AF_INET)
continue;
if (ifu->fu_via_ip.s_addr != ((struct sockaddr_in *)
(ia->ifa_addr))->sin_addr.s_addr)
continue;
return(1);
}
return(0);
}
return(1);
}
static void
ipfw_report(struct ip_fw *f, struct ip *ip, int ip_off, int ip_len,
struct ifnet *rif, struct ifnet *oif)
{
struct tcphdr *const tcp = (struct tcphdr *) ((u_int32_t *) ip+ ip->ip_hl);
struct udphdr *const udp = (struct udphdr *) ((u_int32_t *) ip+ ip->ip_hl);
struct icmp *const icmp = (struct icmp *) ((u_int32_t *) ip + ip->ip_hl);
u_int64_t count;
char *action;
char action2[32], proto[47], name[18], fragment[27];
int len;
int offset = ip_off & IP_OFFMASK;
count = f ? f->fw_pcnt : ++counter;
if ((f == NULL && fw_verbose_limit != 0 && count > fw_verbose_limit) ||
(f && f->fw_logamount != 0 && count > f->fw_loghighest))
return;
/* Print command name */
snprintf(SNPARGS(name, 0), "ipfw: %d", f ? f->fw_number : -1);
action = action2;
if (!f)
action = "Refuse";
else {
switch (f->fw_flg & IP_FW_F_COMMAND) {
case IP_FW_F_DENY:
action = "Deny";
break;
case IP_FW_F_REJECT:
if (f->fw_reject_code == IP_FW_REJECT_RST)
action = "Reset";
else
action = "Unreach";
break;
case IP_FW_F_ACCEPT:
action = "Accept";
break;
case IP_FW_F_COUNT:
action = "Count";
break;
#ifdef IPDIVERT
case IP_FW_F_DIVERT:
snprintf(SNPARGS(action2, 0), "Divert %d",
f->fw_divert_port);
break;
case IP_FW_F_TEE:
snprintf(SNPARGS(action2, 0), "Tee %d",
f->fw_divert_port);
break;
#endif
case IP_FW_F_SKIPTO:
snprintf(SNPARGS(action2, 0), "SkipTo %d",
f->fw_skipto_rule);
break;
case IP_FW_F_PIPE:
snprintf(SNPARGS(action2, 0), "Pipe %d",
f->fw_skipto_rule);
break;
2000-06-08 09:45:23 +00:00
case IP_FW_F_QUEUE:
snprintf(SNPARGS(action2, 0), "Queue %d",
f->fw_skipto_rule);
break;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
case IP_FW_F_FWD:
if (f->fw_fwd_ip.sin_port)
snprintf(SNPARGS(action2, 0),
"Forward to %s:%d",
inet_ntoa(f->fw_fwd_ip.sin_addr),
f->fw_fwd_ip.sin_port);
else
snprintf(SNPARGS(action2, 0), "Forward to %s",
inet_ntoa(f->fw_fwd_ip.sin_addr));
break;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
default:
action = "UNKNOWN";
break;
}
}
switch (ip->ip_p) {
case IPPROTO_TCP:
len = snprintf(SNPARGS(proto, 0), "TCP %s",
inet_ntoa(ip->ip_src));
if (offset == 0)
len += snprintf(SNPARGS(proto, len), ":%d ",
ntohs(tcp->th_sport));
else
len += snprintf(SNPARGS(proto, len), " ");
len += snprintf(SNPARGS(proto, len), "%s",
inet_ntoa(ip->ip_dst));
if (offset == 0)
snprintf(SNPARGS(proto, len), ":%d",
ntohs(tcp->th_dport));
break;
case IPPROTO_UDP:
len = snprintf(SNPARGS(proto, 0), "UDP %s",
inet_ntoa(ip->ip_src));
if (offset == 0)
len += snprintf(SNPARGS(proto, len), ":%d ",
ntohs(udp->uh_sport));
else
len += snprintf(SNPARGS(proto, len), " ");
len += snprintf(SNPARGS(proto, len), "%s",
inet_ntoa(ip->ip_dst));
if (offset == 0)
snprintf(SNPARGS(proto, len), ":%d",
ntohs(udp->uh_dport));
break;
case IPPROTO_ICMP:
if (offset == 0)
len = snprintf(SNPARGS(proto, 0), "ICMP:%u.%u ",
icmp->icmp_type, icmp->icmp_code);
else
len = snprintf(SNPARGS(proto, 0), "ICMP ");
len += snprintf(SNPARGS(proto, len), "%s",
inet_ntoa(ip->ip_src));
snprintf(SNPARGS(proto, len), " %s", inet_ntoa(ip->ip_dst));
break;
default:
len = snprintf(SNPARGS(proto, 0), "P:%d %s", ip->ip_p,
inet_ntoa(ip->ip_src));
snprintf(SNPARGS(proto, len), " %s", inet_ntoa(ip->ip_dst));
break;
}
if (ip_off & (IP_MF | IP_OFFMASK))
2001-12-14 19:32:00 +00:00
snprintf(SNPARGS(fragment, 0), " (frag %d:%d@%d%s)",
ntohs(ip->ip_id), ip_len - (ip->ip_hl << 2),
offset << 3,
(ip_off & IP_MF) ? "+" : "");
else
fragment[0] = '\0';
if (oif)
log(LOG_SECURITY | LOG_INFO, "%s %s %s out via %s%d%s\n",
name, action, proto, oif->if_name, oif->if_unit, fragment);
else if (rif)
log(LOG_SECURITY | LOG_INFO, "%s %s %s in via %s%d%s\n", name,
action, proto, rif->if_name, rif->if_unit, fragment);
else
log(LOG_SECURITY | LOG_INFO, "%s %s %s%s\n", name, action,
proto, fragment);
if ((f ? f->fw_logamount != 0 : 1) &&
count == (f ? f->fw_loghighest : fw_verbose_limit))
log(LOG_SECURITY | LOG_NOTICE,
"ipfw: limit %d reached on entry %d\n",
f ? f->fw_logamount : fw_verbose_limit,
f ? f->fw_number : -1);
}
static __inline int
hash_packet(struct ipfw_flow_id *id)
{
u_int32_t i ;
i = (id->dst_ip) ^ (id->src_ip) ^ (id->dst_port) ^ (id->src_port);
i &= (curr_dyn_buckets - 1) ;
return i ;
}
/**
* unlink a dynamic rule from a chain. prev is a pointer to
* the previous one, q is a pointer to the rule to delete,
* head is a pointer to the head of the queue.
* Modifies q and potentially also head.
*/
#define UNLINK_DYN_RULE(prev, head, q) { \
struct ipfw_dyn_rule *old_q = q; \
\
/* remove a refcount to the parent */ \
if (q->dyn_type == DYN_LIMIT) \
q->parent->count--; \
DEB(printf("-- unlink entry 0x%08x %d -> 0x%08x %d, %d left\n", \
(q->id.src_ip), (q->id.src_port), \
(q->id.dst_ip), (q->id.dst_port), dyn_count-1 ); ) \
if (prev != NULL) \
prev->next = q = q->next ; \
else \
ipfw_dyn_v[i] = q = q->next ; \
dyn_count-- ; \
free(old_q, M_IPFW); }
#define TIME_LEQ(a,b) ((int)((a)-(b)) <= 0)
/**
* Remove all dynamic rules pointing to a given rule, or all
* rules if rule == NULL. Second parameter is 1 if we want to
* delete unconditionally, otherwise only expired rules are removed.
*/
static void
remove_dyn_rule(struct ip_fw *rule, int force)
{
struct ipfw_dyn_rule *prev, *q;
int i, pass, max_pass ;
static u_int32_t last_remove = 0 ;
if (ipfw_dyn_v == NULL || dyn_count == 0)
return ;
/* do not expire more than once per second, it is useless */
if (force == 0 && last_remove == time_second)
return ;
last_remove = time_second ;
/*
* because DYN_LIMIT refer to parent rules, during the first pass only
* remove child and mark any pending LIMIT_PARENT, and remove
* them in a second pass.
*/
for (pass = max_pass = 0; pass <= max_pass ; pass++ ) {
for (i = 0 ; i < curr_dyn_buckets ; i++) {
for (prev=NULL, q = ipfw_dyn_v[i] ; q ; ) {
/*
* logic can become complex here, so we split tests.
* First, test if we match any rule,
* then make sure the rule is expired or we want to kill it,
* and possibly more in the future.
*/
int zap = ( rule == NULL || rule == q->rule);
if (zap)
zap = force || TIME_LEQ( q->expire , time_second );
/* do not zap parent in first pass, record we need a second pass */
if (q->dyn_type == DYN_LIMIT_PARENT) {
max_pass = 1; /* we need a second pass */
if (zap == 1 && (pass == 0 || q->count != 0) ) {
zap = 0 ;
if (pass == 1) /* should not happen */
printf("OUCH! cannot remove rule, count %d\n",
q->count);
}
}
if (zap) {
UNLINK_DYN_RULE(prev, ipfw_dyn_v[i], q);
} else {
prev = q ;
q = q->next ;
}
}
}
}
}
#define EXPIRE_DYN_CHAIN(rule) remove_dyn_rule(rule, 0 /* expired ones */)
#define EXPIRE_DYN_CHAINS() remove_dyn_rule(NULL, 0 /* expired ones */)
#define DELETE_DYN_CHAIN(rule) remove_dyn_rule(rule, 1 /* force removal */)
#define DELETE_DYN_CHAINS() remove_dyn_rule(NULL, 1 /* force removal */)
/**
* lookup a dynamic rule.
*/
static struct ipfw_dyn_rule *
lookup_dyn_rule(struct ipfw_flow_id *pkt, int *match_direction)
{
/*
* stateful ipfw extensions.
* Lookup into dynamic session queue
*/
struct ipfw_dyn_rule *prev, *q ;
int i, dir = 0;
#define MATCH_FORWARD 1
if (ipfw_dyn_v == NULL)
return NULL ;
i = hash_packet( pkt );
for (prev=NULL, q = ipfw_dyn_v[i] ; q != NULL ; ) {
if (q->dyn_type == DYN_LIMIT_PARENT)
goto next;
if (TIME_LEQ( q->expire , time_second ) ) { /* expire entry */
UNLINK_DYN_RULE(prev, ipfw_dyn_v[i], q);
continue;
}
if ( pkt->proto == q->id.proto) {
if (pkt->src_ip == q->id.src_ip &&
pkt->dst_ip == q->id.dst_ip &&
pkt->src_port == q->id.src_port &&
pkt->dst_port == q->id.dst_port ) {
dir = MATCH_FORWARD ;
goto found ;
}
if (pkt->src_ip == q->id.dst_ip &&
pkt->dst_ip == q->id.src_ip &&
pkt->src_port == q->id.dst_port &&
pkt->dst_port == q->id.src_port ) {
dir = 0 ; /* reverse match */
goto found ;
}
}
next:
prev = q ;
q = q->next ;
}
return NULL ; /* clearly not found */
found:
if ( prev != NULL) { /* found and not in front */
prev->next = q->next ;
q->next = ipfw_dyn_v[i] ;
ipfw_dyn_v[i] = q ;
}
if (pkt->proto == IPPROTO_TCP) {
/* update state according to flags */
u_char flags = pkt->flags & (TH_FIN|TH_SYN|TH_RST);
q->state |= (dir == MATCH_FORWARD ) ? flags : (flags << 8);
switch (q->state) {
case TH_SYN :
/* opening */
q->expire = time_second + dyn_syn_lifetime ;
break ;
case TH_SYN | (TH_SYN << 8) :
/* move to established */
q->expire = time_second + dyn_ack_lifetime ;
break ;
case TH_SYN | (TH_SYN << 8) | TH_FIN :
case TH_SYN | (TH_SYN << 8) | (TH_FIN << 8) :
/* one side tries to close */
q->expire = time_second + dyn_ack_lifetime ;
break ;
case TH_SYN | (TH_SYN << 8) | TH_FIN | (TH_FIN << 8) :
/* both sides closed */
q->expire = time_second + dyn_fin_lifetime ;
break ;
default:
2000-05-02 15:39:36 +00:00
#if 0
2000-06-08 09:45:23 +00:00
/*
* reset or some invalid combination, but can also
* occur if we use keep-state the wrong way.
*/
if ( (q->state & ((TH_RST << 8)|TH_RST)) == 0)
printf("invalid state: 0x%x\n", q->state);
2000-05-02 15:39:36 +00:00
#endif
q->expire = time_second + dyn_rst_lifetime ;
break ;
}
} else if (pkt->proto == IPPROTO_UDP) {
q->expire = time_second + dyn_udp_lifetime ;
} else {
/* other protocols */
q->expire = time_second + dyn_short_lifetime ;
}
if (match_direction)
*match_direction = dir ;
return q ;
}
/**
* Install state of type 'type' for a dynamic session.
* The hash table contains two type of rules:
* - regular rules (DYN_KEEP_STATE)
* - rules for sessions with limited number of sess per user
* (DYN_LIMIT). When they are created, the parent is
* increased by 1, and decreased on delete. In this case,
* the third parameter is the parent rule and not the chain.
* - "parent" rules for the above (DYN_LIMIT_PARENT).
*/
static struct ipfw_dyn_rule *
add_dyn_rule(struct ipfw_flow_id *id, u_int8_t dyn_type, struct ip_fw *rule)
{
struct ipfw_dyn_rule *r ;
int i ;
if (ipfw_dyn_v == NULL ||
(dyn_count == 0 && dyn_buckets != curr_dyn_buckets)) {
/* try reallocation, make sure we have a power of 2 */
u_int32_t i = dyn_buckets ;
while ( i > 0 && (i & 1) == 0 )
i >>= 1 ;
if (i != 1) /* not a power of 2 */
dyn_buckets = curr_dyn_buckets ; /* reset */
else {
curr_dyn_buckets = dyn_buckets ;
if (ipfw_dyn_v != NULL)
free(ipfw_dyn_v, M_IPFW);
ipfw_dyn_v = malloc(curr_dyn_buckets * sizeof r,
M_IPFW, M_DONTWAIT | M_ZERO);
if (ipfw_dyn_v == NULL)
return NULL; /* failed ! */
}
}
i = hash_packet(id);
r = malloc(sizeof *r, M_IPFW, M_NOWAIT | M_ZERO);
if (r == NULL) {
printf ("sorry cannot allocate state\n");
return NULL ;
}
/* increase refcount on parent, and set pointer */
if (dyn_type == DYN_LIMIT) {
struct ipfw_dyn_rule *parent = (struct ipfw_dyn_rule *)rule;
if ( parent->dyn_type != DYN_LIMIT_PARENT)
panic("invalid parent");
parent->count++ ;
r->parent = parent ;
rule = parent->rule;
}
r->id = *id ;
r->expire = time_second + dyn_syn_lifetime ;
r->rule = rule ;
r->dyn_type = dyn_type ;
r->pcnt = r->bcnt = 0 ;
r->count = 0 ;
r->bucket = i ;
r->next = ipfw_dyn_v[i] ;
ipfw_dyn_v[i] = r ;
dyn_count++ ;
DEB(printf("-- add entry 0x%08x %d -> 0x%08x %d, total %d\n",
(r->id.src_ip), (r->id.src_port),
(r->id.dst_ip), (r->id.dst_port),
dyn_count ); )
return r;
}
/**
* lookup dynamic parent rule using pkt and rule as search keys.
* If the lookup fails, then install one.
*/
static struct ipfw_dyn_rule *
lookup_dyn_parent(struct ipfw_flow_id *pkt, struct ip_fw *rule)
{
struct ipfw_dyn_rule *q;
int i;
if (ipfw_dyn_v) {
i = hash_packet( pkt );
for (q = ipfw_dyn_v[i] ; q != NULL ; q=q->next)
if (q->dyn_type == DYN_LIMIT_PARENT && rule == q->rule &&
pkt->proto == q->id.proto &&
pkt->src_ip == q->id.src_ip &&
pkt->dst_ip == q->id.dst_ip &&
pkt->src_port == q->id.src_port &&
pkt->dst_port == q->id.dst_port) {
q->expire = time_second + dyn_short_lifetime ;
DEB(printf("lookup_dyn_parent found 0x%p\n", q);)
return q;
}
}
return add_dyn_rule(pkt, DYN_LIMIT_PARENT, rule);
}
/*
* Install dynamic state.
* There are different types of dynamic rules which can be installed.
* The type is in rule->dyn_type.
* Type 0 (default) is a bidirectional rule
*
* Returns 1 (failure) if state is not installed because of errors or because
* session limitations are enforced.
*/
static int
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
install_state(struct ip_fw *rule, struct ip_fw_args *args)
{
struct ipfw_dyn_rule *q ;
2000-06-08 09:45:23 +00:00
static int last_log ;
u_int8_t type = rule->dyn_type ;
DEB(printf("-- install state type %d 0x%08x %u -> 0x%08x %u\n",
type,
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
(args->f_id.src_ip), (args->f_id.src_port),
(args->f_id.dst_ip), (args->f_id.dst_port) );)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
q = lookup_dyn_rule(&args->f_id, NULL) ;
if (q != NULL) { /* should never occur */
if (last_log != time_second) {
last_log = time_second ;
printf(" entry already present, done\n");
}
return 0 ;
}
if (dyn_count >= dyn_max) /* try remove old ones... */
EXPIRE_DYN_CHAINS();
if (dyn_count >= dyn_max) {
if (last_log != time_second) {
last_log = time_second ;
printf(" Too many dynamic rules, sorry\n");
}
return 1; /* cannot install, notify caller */
}
switch (type) {
case DYN_KEEP_STATE: /* bidir rule */
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
add_dyn_rule(&args->f_id, DYN_KEEP_STATE, rule);
break ;
case DYN_LIMIT: /* limit number of sessions */
{
u_int16_t limit_mask = rule->limit_mask ;
u_int16_t conn_limit = rule->conn_limit ;
struct ipfw_flow_id id;
struct ipfw_dyn_rule *parent;
DEB(printf("installing dyn-limit rule %d\n", conn_limit);)
id.dst_ip = id.src_ip = 0;
id.dst_port = id.src_port = 0 ;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
id.proto = args->f_id.proto ;
if (limit_mask & DYN_SRC_ADDR)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
id.src_ip = args->f_id.src_ip;
if (limit_mask & DYN_DST_ADDR)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
id.dst_ip = args->f_id.dst_ip;
if (limit_mask & DYN_SRC_PORT)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
id.src_port = args->f_id.src_port;
if (limit_mask & DYN_DST_PORT)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
id.dst_port = args->f_id.dst_port;
parent = lookup_dyn_parent(&id, rule);
if (parent == NULL) {
printf("add parent failed\n");
return 1;
}
if (parent->count >= conn_limit) {
EXPIRE_DYN_CHAIN(rule); /* try to expire some */
if (parent->count >= conn_limit) {
printf("drop session, too many entries\n");
return 1;
}
}
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
add_dyn_rule(&args->f_id, DYN_LIMIT, (struct ip_fw *)parent);
}
break ;
default:
printf("unknown dynamic rule type %u\n", type);
return 1 ;
}
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
lookup_dyn_rule(&args->f_id, NULL) ; /* XXX just set the lifetime */
return 0;
}
/*
* given an ip_fw *, lookup_next_rule will return a pointer
* of the same type to the next one. This can be either the jump
* target (for skipto instructions) or the next one in the list (in
* all other cases including a missing jump target).
* Backward jumps are not allowed, so start looking from the next
* rule...
*/
static struct ip_fw * lookup_next_rule(struct ip_fw *me);
static struct ip_fw *
lookup_next_rule(struct ip_fw *me)
{
struct ip_fw *rule ;
int rulenum = me->fw_skipto_rule ; /* guess... */
if ( (me->fw_flg & IP_FW_F_COMMAND) == IP_FW_F_SKIPTO )
for (rule = LIST_NEXT(me,next); rule ; rule = LIST_NEXT(rule,next))
if (rule->fw_number >= rulenum)
return rule ;
return LIST_NEXT(me,next) ; /* failure or not a skipto */
}
/*
* Parameters:
*
* *m The packet; we set to NULL when/if we nuke it.
* oif Outgoing interface, or NULL if packet is incoming
* *cookie Skip up to the first rule past this rule number;
* upon return, non-zero port number for divert or tee.
* Special case: cookie == NULL on input for bridging.
* *flow_id pointer to the last matching rule (in/out)
* *next_hop socket we are forwarding to (in/out).
* For bridged packets, this is a pointer to the MAC header.
*
* Return value:
*
* IP_FW_PORT_DENY_FLAG the packet must be dropped.
* 0 The packet is to be accepted and routed normally OR
* the packet was denied/rejected and has been dropped;
* in the latter case, *m is equal to NULL upon return.
* port Divert the packet to port, with these caveats:
*
* - If IP_FW_PORT_TEE_FLAG is set, tee the packet instead
* of diverting it (ie, 'ipfw tee').
*
* - If IP_FW_PORT_DYNT_FLAG is set, interpret the lower
* 16 bits as a dummynet pipe number instead of diverting
*/
static int
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
ip_fw_chk(struct ip_fw_args *args)
#if 0 /* the old interface was this: */
struct mbuf **m, struct ifnet *oif, u_int16_t *cookie,
struct ip_fw **flow_id, struct sockaddr_in **next_hop)
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
#endif
{
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
/*
* grab things into variables to minimize diffs.
* XXX this has to be cleaned up later.
*/
struct mbuf **m = &(args->m);
struct ifnet *oif = args->oif;
u_int16_t *cookie = &(args->divert_rule);
struct ip_fw **flow_id = &(args->rule);
struct sockaddr_in **next_hop = &(args->next_hop);
struct ip_fw *f = NULL; /* matching rule */
struct ip *ip = mtod(*m, struct ip *);
struct ifnet *const rif = (*m)->m_pkthdr.rcvif;
struct ifnet *tif;
u_int hlen = ip->ip_hl << 2;
struct ether_header * eh = NULL;
u_short ip_off=0, offset = 0 ;
/* local copy of addresses for faster matching */
u_short src_port = 0, dst_port = 0;
struct in_addr src_ip, dst_ip;
u_int8_t proto= 0, flags = 0;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
u_int16_t skipto;
u_int16_t ip_len=0;
int dyn_checked = 0 ; /* set after dyn.rules have been checked. */
int direction = MATCH_FORWARD ; /* dirty trick... */
struct ipfw_dyn_rule *q = NULL ;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
#define BRIDGED (args->eh != NULL)
if (BRIDGED) { /* this is a bridged packet */
eh = args->eh;
if ( (*m)->m_pkthdr.len >= sizeof(struct ip) &&
ntohs(eh->ether_type) == ETHERTYPE_IP)
hlen = ip->ip_hl << 2;
} else
hlen = ip->ip_hl << 2;
/* Grab and reset cookie */
skipto = *cookie;
*cookie = 0;
/*
* Collect parameters into local variables for faster matching.
*/
if (hlen > 0) { /* this is an IP packet */
proto = ip->ip_p;
src_ip = ip->ip_src;
dst_ip = ip->ip_dst;
if (BRIDGED) { /* bridged packets are as on the wire */
ip_off = ntohs(ip->ip_off);
ip_len = ntohs(ip->ip_len);
} else {
ip_off = ip->ip_off;
ip_len = ip->ip_len;
}
offset = ip_off & IP_OFFMASK;
if (offset == 0) {
#define PULLUP_TO(len) \
do { \
if ((*m)->m_len < (len)) { \
*m = m_pullup(*m, (len)); \
if (*m == 0) \
goto bogusfrag; \
ip = mtod(*m, struct ip *); \
} \
} while (0)
switch (proto) {
case IPPROTO_TCP : {
struct tcphdr *tcp;
PULLUP_TO(hlen + sizeof(struct tcphdr));
tcp =(struct tcphdr *)((u_int32_t *)ip + ip->ip_hl);
dst_port = tcp->th_dport ;
src_port = tcp->th_sport ;
flags = tcp->th_flags ;
}
break ;
case IPPROTO_UDP : {
struct udphdr *udp;
PULLUP_TO(hlen + sizeof(struct udphdr));
udp =(struct udphdr *)((u_int32_t *)ip + ip->ip_hl);
dst_port = udp->uh_dport ;
src_port = udp->uh_sport ;
}
break;
case IPPROTO_ICMP:
PULLUP_TO(hlen + 4); /* type, code and checksum. */
flags = ((struct icmp *)
((u_int32_t *)ip + ip->ip_hl))->icmp_type ;
break ;
default :
break;
}
#undef PULLUP_TO
}
}
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
args->f_id.src_ip = ntohl(src_ip.s_addr);
args->f_id.dst_ip = ntohl(dst_ip.s_addr);
args->f_id.proto = proto;
args->f_id.src_port = ntohs(src_port);
args->f_id.dst_port = ntohs(dst_port);
args->f_id.flags = flags;
if (*flow_id) {
/*
* Packet has already been tagged. Look for the next rule
* to restart processing.
*/
if (fw_one_pass) /* just accept if fw_one_pass is set */
return 0;
f = (*flow_id)->next_rule_ptr ;
if (f == NULL)
f = (*flow_id)->next_rule_ptr = lookup_next_rule(*flow_id);
if (f == NULL)
goto dropit;
} else {
/*
* Go down the list, looking for enlightment.
* If we've been asked to start at a given rule, do so.
*/
f = LIST_FIRST(&ip_fw_chain_head);
if (skipto != 0) {
if (skipto >= IPFW_DEFAULT_RULE)
goto dropit;
while (f && f->fw_number <= skipto)
f = LIST_NEXT(f, next);
if (f == NULL)
goto dropit;
}
}
for (; f; f = LIST_NEXT(f, next)) {
again:
if (f->fw_number == IPFW_DEFAULT_RULE)
goto got_match ;
/* Check if rule only valid for bridged packets */
if ((f->fw_flg & IP_FW_BRIDGED) != 0 && !(BRIDGED))
continue;
if (oif) {
/* Check direction outbound */
if (!(f->fw_flg & IP_FW_F_OUT))
continue;
} else {
/* Check direction inbound */
if (!(f->fw_flg & IP_FW_F_IN))
continue;
}
if (f->fw_flg & IP_FW_F_MAC) {
u_int32_t *want, *mask, *hdr;
if (eh == NULL) /* header not available */
continue;
want = (void *)&(f->fw_mac_hdr);
mask = (void *)&(f->fw_mac_mask);
hdr = (void *)eh;
if ( want[0] != (hdr[0] & mask[0]) )
continue;
if ( want[1] != (hdr[1] & mask[1]) )
continue;
if ( want[2] != (hdr[2] & mask[2]) )
continue;
if (f->fw_flg & IP_FW_F_SRNG) {
u_int16_t type = ntohs(eh->ether_type);
if (type < (u_int16_t)(f->fw_mac_type) ||
type > (u_int16_t)(f->fw_mac_mask_type) )
continue;
} else {
if ((u_int16_t)(f->fw_mac_type) != (eh->ether_type &
(u_int16_t)(f->fw_mac_mask_type)) )
continue;
}
}
/* Interface check */
if ((f->fw_flg & IF_FW_F_VIAHACK) == IF_FW_F_VIAHACK) {
struct ifnet *const iface = oif ? oif : rif;
/* Backwards compatibility hack for "via" */
if (!iface || !iface_match(iface,
&f->fw_in_if, f->fw_flg & IP_FW_F_OIFNAME))
continue;
} else {
/* Check receive interface */
if ((f->fw_flg & IP_FW_F_IIFACE)
&& (!rif || !iface_match(rif,
&f->fw_in_if, f->fw_flg & IP_FW_F_IIFNAME)))
continue;
/* Check outgoing interface */
if ((f->fw_flg & IP_FW_F_OIFACE)
&& (!oif || !iface_match(oif,
&f->fw_out_if, f->fw_flg & IP_FW_F_OIFNAME)))
1996-02-23 20:11:37 +00:00
continue;
}
/*
* For packets which matched the MAC check, we do not need
* to continue, this is a valid match.
* For not-ip packets, the rule does not apply.
*/
if (f->fw_flg & IP_FW_F_MAC)
goto rnd_then_got_match;
if (hlen == 0)
continue;
/*
* dynamic rules are checked at the first keep-state or
* check-state occurrence.
*/
if (f->fw_flg & (IP_FW_F_KEEP_S|IP_FW_F_CHECK_S) &&
dyn_checked == 0 ) {
dyn_checked = 1 ;
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
q = lookup_dyn_rule(&args->f_id, &direction);
if (q != NULL) {
DEB(printf("-- dynamic match 0x%08x %d %s 0x%08x %d\n",
(q->id.src_ip), (q->id.src_port),
(direction == MATCH_FORWARD ? "-->" : "<--"),
(q->id.dst_ip), (q->id.dst_port) ); )
f = q->rule ;
q->pcnt++ ;
q->bcnt += ip_len;
goto got_match ; /* random not allowed here */
}
/* if this was a check-only rule, continue with next */
if (f->fw_flg & IP_FW_F_CHECK_S)
continue ;
}
/* Fragments */
if ((f->fw_flg & IP_FW_F_FRAG) && offset == 0 )
continue;
/*
* For matching addresses, tif != NULL means we matched
* the address we requested (either "me" or addr/mask).
* Then the check for "xxx" or "not xxx" can be done
* with an XOR.
*/
/* source address -- mandatory */
if (f->fw_flg & IP_FW_F_SME) {
INADDR_TO_IFP(src_ip, tif);
} else
(int)tif = f->fw_src.s_addr ==
(src_ip.s_addr & f->fw_smsk.s_addr);
if ( ((f->fw_flg & IP_FW_F_INVSRC) != 0) ^ (tif == NULL) )
continue;
/* dst address -- mandatory */
if (f->fw_flg & IP_FW_F_DME) {
INADDR_TO_IFP(dst_ip, tif);
} else
(int)tif = f->fw_dst.s_addr ==
(dst_ip.s_addr & f->fw_dmsk.s_addr);
if ( ((f->fw_flg & IP_FW_F_INVDST) != 0) ^ (tif == NULL) )
continue;
/* Check IP header values */
if (f->fw_ipflg & IP_FW_IF_IPOPT && !ipopts_match(ip, f))
continue;
if (f->fw_ipflg & IP_FW_IF_IPLEN && f->fw_iplen != ip_len)
continue;
if (f->fw_ipflg & IP_FW_IF_IPID && f->fw_ipid != ntohs(ip->ip_id))
continue;
if (f->fw_ipflg & IP_FW_IF_IPPRE &&
(f->fw_iptos & 0xe0) != (ip->ip_tos & 0xe0))
continue;
if (f->fw_ipflg & IP_FW_IF_IPTOS && !iptos_match(ip, f))
continue;
if (f->fw_ipflg & IP_FW_IF_IPTTL && f->fw_ipttl != ip->ip_ttl)
continue;
if (f->fw_ipflg & IP_FW_IF_IPVER && f->fw_ipver != ip->ip_v)
continue;
/* Check protocol; if wildcard, and no [ug]id, match */
if (f->fw_prot == IPPROTO_IP) {
if (!(f->fw_flg & (IP_FW_F_UID|IP_FW_F_GID)))
goto rnd_then_got_match;
} else
/* If different, don't match */
if (proto != f->fw_prot)
continue;
/* Protocol specific checks for uid only */
if (f->fw_flg & (IP_FW_F_UID|IP_FW_F_GID)) {
switch (proto) {
case IPPROTO_TCP:
{
struct inpcb *P;
if (offset == 1) /* cf. RFC 1858 */
goto bogusfrag;
if (offset != 0)
continue;
if (oif)
P = in_pcblookup_hash(&tcbinfo, dst_ip,
dst_port, src_ip, src_port, 0,
oif);
else
P = in_pcblookup_hash(&tcbinfo, src_ip,
src_port, dst_ip, dst_port, 0,
NULL);
if (P && P->inp_socket) {
if (f->fw_flg & IP_FW_F_UID) {
if (socheckuid(P->inp_socket, f->fw_uid))
continue;
} else if (!groupmember(f->fw_gid,
P->inp_socket->so_cred))
continue;
} else
continue;
break;
}
case IPPROTO_UDP:
{
struct inpcb *P;
if (offset != 0)
continue;
if (oif)
P = in_pcblookup_hash(&udbinfo, dst_ip,
dst_port, src_ip, src_port, 1,
oif);
else
P = in_pcblookup_hash(&udbinfo, src_ip,
src_port, dst_ip, dst_port, 1,
NULL);
if (P && P->inp_socket) {
if (f->fw_flg & IP_FW_F_UID) {
if (socheckuid(P->inp_socket, f->fw_uid))
continue;
} else if (!groupmember(f->fw_gid,
P->inp_socket->so_cred))
continue;
} else
continue;
break;
}
default:
continue;
}
}
/* Protocol specific checks */
switch (proto) {
case IPPROTO_TCP:
{
struct tcphdr *tcp;
if (offset == 1) /* cf. RFC 1858 */
goto bogusfrag;
if (offset != 0) {
/*
* TCP flags and ports aren't available in this
* packet -- if this rule specified either one,
* we consider the rule a non-match.
*/
if (IP_FW_HAVEPORTS(f) != 0 ||
f->fw_ipflg & IP_FW_IF_TCPMSK)
continue;
break;
}
tcp = (struct tcphdr *) ((u_int32_t *)ip + ip->ip_hl);
if (f->fw_ipflg & IP_FW_IF_TCPOPT && !tcpopts_match(tcp, f))
continue;
if (((f->fw_ipflg & IP_FW_IF_TCPFLG) ||
(f->fw_ipflg & IP_FW_IF_TCPEST)) &&
!tcpflg_match(tcp, f))
continue;
if (f->fw_ipflg & IP_FW_IF_TCPSEQ && tcp->th_seq != f->fw_tcpseq)
continue;
if (f->fw_ipflg & IP_FW_IF_TCPACK && tcp->th_ack != f->fw_tcpack)
continue;
if (f->fw_ipflg & IP_FW_IF_TCPWIN && tcp->th_win != f->fw_tcpwin)
continue;
goto check_ports;
}
case IPPROTO_UDP:
if (offset != 0) {
/*
* Port specification is unavailable -- if this
* rule specifies a port, we consider the rule
* a non-match.
*/
if (IP_FW_HAVEPORTS(f) )
continue;
break;
}
check_ports:
if (!port_match(&f->fw_uar.fw_pts[0],
IP_FW_GETNSRCP(f), ntohs(src_port),
f->fw_flg & IP_FW_F_SRNG,
f->fw_flg & IP_FW_F_SMSK))
continue;
if (!port_match(&f->fw_uar.fw_pts[IP_FW_GETNSRCP(f)],
IP_FW_GETNDSTP(f), ntohs(dst_port),
f->fw_flg & IP_FW_F_DRNG,
f->fw_flg & IP_FW_F_DMSK))
continue;
break;
case IPPROTO_ICMP:
{
struct icmp *icmp;
if (offset != 0) /* Type isn't valid */
break;
icmp = (struct icmp *) ((u_int32_t *)ip + ip->ip_hl);
if (!icmptype_match(icmp, f))
continue;
break;
}
default:
break;
bogusfrag:
if (fw_verbose) {
if (*m != NULL)
ipfw_report(NULL, ip, ip_off, ip_len, rif, oif);
else
printf("pullup failed\n");
}
goto dropit;
}
rnd_then_got_match:
if ( f->dont_match_prob && random() < f->dont_match_prob )
continue ;
got_match:
/*
* If not a dynamic match (q == NULL) and keep-state, install
* a new dynamic entry.
*/
if (q == NULL && f->fw_flg & IP_FW_F_KEEP_S) {
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
if (install_state(f, args)) /* error or limit violation */
goto dropit;
}
/* Update statistics */
f->fw_pcnt += 1;
f->fw_bcnt += ip_len;
f->timestamp = time_second;
/* Log to console if desired */
if ((f->fw_flg & IP_FW_F_PRN) && fw_verbose && hlen >0)
ipfw_report(f, ip, ip_off, ip_len, rif, oif);
/* Take appropriate action */
switch (f->fw_flg & IP_FW_F_COMMAND) {
case IP_FW_F_ACCEPT:
return(0);
case IP_FW_F_COUNT:
continue;
#ifdef IPDIVERT
case IP_FW_F_DIVERT:
*cookie = f->fw_number;
return(f->fw_divert_port);
case IP_FW_F_TEE:
*cookie = f->fw_number;
return(f->fw_divert_port | IP_FW_PORT_TEE_FLAG);
#endif
case IP_FW_F_SKIPTO: /* XXX check */
if (f->next_rule_ptr == NULL)
f->next_rule_ptr = lookup_next_rule(f) ;
f = f->next_rule_ptr;
if (!f)
goto dropit;
goto again ;
case IP_FW_F_PIPE:
2000-06-08 09:45:23 +00:00
case IP_FW_F_QUEUE:
*flow_id = f; /* XXX set flow id */
return(f->fw_pipe_nr | IP_FW_PORT_DYNT_FLAG);
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
case IP_FW_F_FWD:
/* Change the next-hop address for this packet.
* Initially we'll only worry about directly
* reachable next-hop's, but ultimately
* we will work out for next-hops that aren't
* direct the route we would take for it. We
* [cs]ould leave this latter problem to
* ip_output.c. We hope to high [name the abode of
* your favourite deity] that ip_output doesn't modify
* the new value of next_hop (which is dst there)
* XXX warning-- there is a dangerous reference here
* from next_hop to a field within the rule. If the
* rule is deleted, weird things might occur.
*/
if (next_hop != NULL /* Make sure, first... */
&& (q == NULL || direction == MATCH_FORWARD) )
*next_hop = &(f->fw_fwd_ip);
return(0); /* Allow the packet */
Remove (almost all) global variables that were used to hold packet forwarding state ("annotations") during ip processing. The code is considerably cleaner now. The variables removed by this change are: ip_divert_cookie used by divert sockets ip_fw_fwd_addr used for transparent ip redirection last_pkt used by dynamic pipes in dummynet Removal of the first two has been done by carrying the annotations into volatile structs prepended to the mbuf chains, and adding appropriate code to add/remove annotations in the routines which make use of them, i.e. ip_input(), ip_output(), tcp_input(), bdg_forward(), ether_demux(), ether_output_frame(), div_output(). On passing, remove a bug in divert handling of fragmented packet. Now it is the fragment at offset 0 which sets the divert status of the whole packet, whereas formerly it was the last incoming fragment to decide. Removal of last_pkt required a change in the interface of ip_fw_chk() and dummynet_io(). On passing, use the same mechanism for dummynet annotations and for divert/forward annotations. option IPFIREWALL_FORWARD is effectively useless, the code to implement it is very small and is now in by default to avoid the obfuscation of conditionally compiled code. NOTES: * there is at least one global variable left, sro_fwd, in ip_output(). I am not sure if/how this can be removed. * I have deliberately avoided gratuitous style changes in this commit to avoid cluttering the diffs. Minor stule cleanup will likely be necessary * this commit only focused on the IP layer. I am sure there is a number of global variables used in the TCP and maybe UDP stack. * despite the number of files touched, there are absolutely no API's or data structures changed by this commit (except the interfaces of ip_fw_chk() and dummynet_io(), which are internal anyways), so an MFC is quite safe and unintrusive (and desirable, given the improved readability of the code). MFC after: 10 days
2002-06-22 11:51:02 +00:00
}
/* Deny/reject this packet using this rule */
break;
}
/* Rule IPFW_DEFAULT_RULE should always be there and match */
KASSERT(f != NULL, ("ip_fw: no chain"));
/*
* At this point, we're going to drop the packet.
* Send a reject notice if all of the following are true:
*
* - The packet matched a reject rule
* - The packet is not an ICMP packet, or is an ICMP query packet
* - The packet is not a multicast or broadcast packet
*/
if ((f->fw_flg & IP_FW_F_COMMAND) == IP_FW_F_REJECT
&& (proto != IPPROTO_ICMP || is_icmp_query(ip))
&& !((*m)->m_flags & (M_BCAST|M_MCAST))
&& !IN_MULTICAST(ntohl(ip->ip_dst.s_addr))) {
/* Must convert to host order for icmp_error() etc. */
if (BRIDGED) {
ip->ip_len = ntohs(ip->ip_len);
ip->ip_off = ntohs(ip->ip_off);
}
switch (f->fw_reject_code) {
case IP_FW_REJECT_RST:
{
/* XXX warning, this code writes into the mbuf */
struct tcphdr *const tcp =
(struct tcphdr *) ((u_int32_t *)ip + ip->ip_hl);
struct tcpiphdr ti, *const tip = (struct tcpiphdr *) ip;
if (offset != 0 || (tcp->th_flags & TH_RST))
break;
ti.ti_i = *((struct ipovly *) ip);
ti.ti_t = *tcp;
bcopy(&ti, ip, sizeof(ti));
tip->ti_seq = ntohl(tip->ti_seq);
tip->ti_ack = ntohl(tip->ti_ack);
tip->ti_len = ip_len - hlen - (tip->ti_off << 2);
if (tcp->th_flags & TH_ACK) {
tcp_respond(NULL, (void *)ip, tcp, *m,
(tcp_seq)0, tcp->th_ack, TH_RST);
} else {
if (tcp->th_flags & TH_SYN)
tip->ti_len++;
tcp_respond(NULL, (void *)ip, tcp, *m,
tip->ti_seq + tip->ti_len,
(tcp_seq)0, TH_RST|TH_ACK);
}
*m = NULL;
break;
}
default: /* Send an ICMP unreachable using code */
icmp_error(*m, ICMP_UNREACH,
f->fw_reject_code, 0L, 0);
*m = NULL;
break;
}
}
dropit:
/*
* Finally, drop the packet.
*/
return(IP_FW_PORT_DENY_FLAG);
#undef BRIDGED
}
/*
* when a rule is added/deleted, zero the direct pointers within
* all firewall rules. These will be reconstructed on the fly
* as packets are matched.
* Must be called at splimp().
*/
static void
flush_rule_ptrs()
{
struct ip_fw *fcp ;
LIST_FOREACH(fcp, &ip_fw_chain_head, next) {
fcp->next_rule_ptr = NULL ;
}
}
static int
add_entry(struct ip_fw_head *head, struct ip_fw *rule)
{
struct ip_fw *ftmp, *fcp, *fcpl;
u_short nbr = 0;
int s;
ftmp = malloc(sizeof *ftmp, M_IPFW, M_NOWAIT | M_ZERO);
if (!ftmp)
return (ENOSPC);
bcopy(rule, ftmp, sizeof(*ftmp));
ftmp->fw_in_if.fu_via_if.name[FW_IFNLEN - 1] = '\0';
ftmp->fw_pcnt = 0L;
ftmp->fw_bcnt = 0L;
ftmp->next_rule_ptr = NULL ;
ftmp->pipe_ptr = NULL ;
s = splimp();
if (LIST_FIRST(head) == 0) {
LIST_INSERT_HEAD(head, ftmp, next);
goto done;
}
/* If entry number is 0, find highest numbered rule and add 100 */
if (ftmp->fw_number == 0) {
LIST_FOREACH(fcp, head, next) {
if (fcp->fw_number != IPFW_DEFAULT_RULE)
nbr = fcp->fw_number;
else
break;
}
if (nbr < IPFW_DEFAULT_RULE - 100)
nbr += 100;
ftmp->fw_number = rule->fw_number = nbr;
}
/* Got a valid number; now insert it, keeping the list ordered */
fcpl = NULL ;
LIST_FOREACH(fcp, head, next) {
if (fcp->fw_number > ftmp->fw_number) {
if (fcpl) {
LIST_INSERT_AFTER(fcpl, ftmp, next);
} else {
LIST_INSERT_HEAD(head, ftmp, next);
}
break;
} else {
fcpl = fcp;
}
}
flush_rule_ptrs();
done:
static_count++;
splx(s);
DEB(printf("++ installed rule %d, static count now %d\n",
ftmp->fw_number, static_count);)
return (0);
}
/**
* free storage associated with a static rule entry (including
* dependent dynamic rules), and zeroes rule pointers to avoid
* dangling pointer dereferences.
* @return a pointer to the next entry.
* Must be called at splimp() and with a non-null argument.
*/
static struct ip_fw *
free_chain(struct ip_fw *fcp)
{
struct ip_fw *n;
n = LIST_NEXT(fcp, next);
DELETE_DYN_CHAIN(fcp);
LIST_REMOVE(fcp, next);
static_count--;
if (DUMMYNET_LOADED)
ip_dn_ruledel_ptr(fcp) ;
flush_rule_ptrs(); /* more efficient to do outside the loop */
free(fcp, M_IPFW);
return n;
}
/**
* remove all rules with given number.
*/
static int
del_entry(struct ip_fw_head *chainptr, u_short number)
{
struct ip_fw *rule;
if (number != IPFW_DEFAULT_RULE) {
LIST_FOREACH(rule, chainptr, next) {
if (rule->fw_number == number) {
int s ;
s = splimp(); /* prevent access to rules while removing */
while (rule && rule->fw_number == number)
rule = free_chain(rule);
/* XXX could move flush_rule_ptrs() here */
splx(s);
return 0 ;
}
}
}
return (EINVAL);
}
/**
* Reset some or all counters on firewall rules.
* @arg frwl is null to clear all entries, or contains a specific
* rule number.
* @arg log_only is 1 if we only want to reset logs, zero otherwise.
*/
static int
zero_entry(struct ip_fw *frwl, int log_only)
{
struct ip_fw *rule;
int s;
u_short number = 0 ;
char *msg ;
if (frwl == 0) {
s = splimp();
LIST_FOREACH(rule, &ip_fw_chain_head, next) {
if (log_only == 0) {
rule->fw_bcnt = rule->fw_pcnt = 0;
rule->timestamp = 0;
}
rule->fw_loghighest = rule->fw_pcnt+rule->fw_logamount;
}
splx(s);
msg = log_only ? "ipfw: All logging counts cleared.\n" :
"ipfw: Accounting cleared.\n";
} else {
int cleared = 0;
number = frwl->fw_number ;
/*
* It is possible to insert multiple chain entries with the
* same number, so we don't stop after finding the first
* match if zeroing a specific entry.
*/
LIST_FOREACH(rule, &ip_fw_chain_head, next)
if (number == rule->fw_number) {
s = splimp();
while (rule && number == rule->fw_number) {
if (log_only == 0) {
rule->fw_bcnt = rule->fw_pcnt = 0;
rule->timestamp = 0;
}
rule->fw_loghighest = rule->fw_pcnt+ rule->fw_logamount;
rule = LIST_NEXT(rule, next);
}
splx(s);
cleared = 1;
break;
}
if (!cleared) /* we did not find any matching rules */
return (EINVAL);
msg = log_only ? "ipfw: Entry %d logging count reset.\n" :
"ipfw: Entry %d cleared.\n";
}
if (fw_verbose)
log(LOG_SECURITY | LOG_NOTICE, msg, number);
return (0);
}
static int
check_ipfw_struct(struct ip_fw *frwl)
{
/* Check for invalid flag bits */
if ((frwl->fw_flg & ~IP_FW_F_MASK) != 0) {
dprintf(("%s undefined flag bits set (flags=%x)\n",
err_prefix, frwl->fw_flg));
return (EINVAL);
}
if ( (frwl->fw_flg & IP_FW_F_MAC) ) { /* match MAC address */
return 0;
}
if (frwl->fw_flg == IP_FW_F_CHECK_S) {
2000-06-08 09:45:23 +00:00
/* check-state */
return 0 ;
}
/* Must apply to incoming or outgoing (or both) */
if (!(frwl->fw_flg & (IP_FW_F_IN | IP_FW_F_OUT))) {
dprintf(("%s neither in nor out\n", err_prefix));
return (EINVAL);
}
/* Empty interface name is no good */
if (((frwl->fw_flg & IP_FW_F_IIFNAME)
&& !*frwl->fw_in_if.fu_via_if.name)
|| ((frwl->fw_flg & IP_FW_F_OIFNAME)
&& !*frwl->fw_out_if.fu_via_if.name)) {
dprintf(("%s empty interface name\n", err_prefix));
return (EINVAL);
}
/* Sanity check interface matching */
if ((frwl->fw_flg & IF_FW_F_VIAHACK) == IF_FW_F_VIAHACK) {
; /* allow "via" backwards compatibility */
} else if ((frwl->fw_flg & IP_FW_F_IN)
&& (frwl->fw_flg & IP_FW_F_OIFACE)) {
dprintf(("%s outgoing interface check on incoming\n",
err_prefix));
return (EINVAL);
}
/* Sanity check port ranges */
if ((frwl->fw_flg & IP_FW_F_SRNG) && IP_FW_GETNSRCP(frwl) < 2) {
dprintf(("%s src range set but n_src_p=%d\n",
err_prefix, IP_FW_GETNSRCP(frwl)));
return (EINVAL);
}
if ((frwl->fw_flg & IP_FW_F_DRNG) && IP_FW_GETNDSTP(frwl) < 2) {
dprintf(("%s dst range set but n_dst_p=%d\n",
err_prefix, IP_FW_GETNDSTP(frwl)));
return (EINVAL);
}
if (IP_FW_GETNSRCP(frwl) + IP_FW_GETNDSTP(frwl) > IP_FW_MAX_PORTS) {
dprintf(("%s too many ports (%d+%d)\n",
err_prefix, IP_FW_GETNSRCP(frwl), IP_FW_GETNDSTP(frwl)));
return (EINVAL);
}
/*
* Protocols other than TCP/UDP don't use port range
*/
if ((frwl->fw_prot != IPPROTO_TCP) &&
(frwl->fw_prot != IPPROTO_UDP) &&
(IP_FW_GETNSRCP(frwl) || IP_FW_GETNDSTP(frwl))) {
dprintf(("%s port(s) specified for non TCP/UDP rule\n",
err_prefix));
return (EINVAL);
}
/*
* Rather than modify the entry to make such entries work,
* we reject this rule and require user level utilities
* to enforce whatever policy they deem appropriate.
*/
if ((frwl->fw_src.s_addr & (~frwl->fw_smsk.s_addr)) ||
(frwl->fw_dst.s_addr & (~frwl->fw_dmsk.s_addr))) {
dprintf(("%s rule never matches\n", err_prefix));
return (EINVAL);
}
if ((frwl->fw_flg & IP_FW_F_FRAG) &&
(frwl->fw_prot == IPPROTO_UDP || frwl->fw_prot == IPPROTO_TCP)) {
if (IP_FW_HAVEPORTS(frwl)) {
dprintf(("%s cannot mix 'frag' and ports\n", err_prefix));
return (EINVAL);
}
if (frwl->fw_prot == IPPROTO_TCP &&
frwl->fw_tcpf != frwl->fw_tcpnf) {
dprintf(("%s cannot mix 'frag' and TCP flags\n", err_prefix));
return (EINVAL);
}
}
if (frwl->fw_flg & (IP_FW_F_UID | IP_FW_F_GID)) {
2000-09-12 02:38:05 +00:00
if ((frwl->fw_prot != IPPROTO_TCP) &&
(frwl->fw_prot != IPPROTO_UDP) &&
(frwl->fw_prot != IPPROTO_IP)) {
dprintf(("%s cannot use uid/gid logic on non-TCP/UDP\n", err_prefix));
return (EINVAL);
}
}
/* Check command specific stuff */
switch (frwl->fw_flg & IP_FW_F_COMMAND) {
case IP_FW_F_REJECT:
if (frwl->fw_reject_code >= 0x100
&& !(frwl->fw_prot == IPPROTO_TCP
&& frwl->fw_reject_code == IP_FW_REJECT_RST)) {
dprintf(("%s unknown reject code\n", err_prefix));
return (EINVAL);
}
break;
#ifdef IPDIVERT
case IP_FW_F_DIVERT: /* Diverting to port zero is invalid */
case IP_FW_F_TEE:
#endif
case IP_FW_F_PIPE: /* pipe 0 is invalid */
case IP_FW_F_QUEUE: /* queue 0 is invalid */
if (frwl->fw_divert_port == 0) {
dprintf(("%s 0 is an invalid argument\n", err_prefix));
return (EINVAL);
}
break;
case IP_FW_F_DENY:
case IP_FW_F_ACCEPT:
case IP_FW_F_COUNT:
case IP_FW_F_SKIPTO:
case IP_FW_F_FWD:
break;
default:
dprintf(("%s invalid command\n", err_prefix));
return (EINVAL);
}
return 0;
}
static int
ip_fw_ctl(struct sockopt *sopt)
{
int error, s;
size_t size;
struct ip_fw *fcp;
struct ip_fw frwl, *bp , *buf;
/*
* Disallow modifications in really-really secure mode, but still allow
* the logging counters to be reset.
*/
if (sopt->sopt_name == IP_FW_ADD ||
(sopt->sopt_dir == SOPT_SET && sopt->sopt_name != IP_FW_RESETLOG)) {
error = securelevel_ge(sopt->sopt_td->td_ucred, 3);
if (error)
return (error);
}
error = 0;
switch (sopt->sopt_name) {
case IP_FW_GET:
/*
* pass up a copy of the current rules. Static rules
* come first (the last of which has number 65535),
* followed by a possibly empty list of dynamic rule.
* The last dynamic rule has NULL in the "next" field.
*/
s = splimp();
/* size of static rules */
size = static_count * sizeof(struct ip_fw) ;
if (ipfw_dyn_v) /* add size of dyn.rules */
size += (dyn_count * sizeof(struct ipfw_dyn_rule));
/*
* XXX todo: if the user passes a short length to know how
* much room is needed, do not
* bother filling up the buffer, just jump to the
* sooptcopyout.
*/
buf = malloc(size, M_TEMP, M_WAITOK);
if (buf == 0) {
splx(s);
error = ENOBUFS;
break;
}
bp = buf ;
LIST_FOREACH(fcp, &ip_fw_chain_head, next) {
bcopy(fcp, bp, sizeof *fcp);
bp++;
}
if (ipfw_dyn_v) {
int i ;
struct ipfw_dyn_rule *p, *dst, *last = NULL ;
dst = (struct ipfw_dyn_rule *)bp ;
for (i = 0 ; i < curr_dyn_buckets ; i++ )
for ( p = ipfw_dyn_v[i] ; p != NULL ; p = p->next, dst++ ) {
bcopy(p, dst, sizeof *p);
(int)dst->rule = p->rule->fw_number ;
/*
* store a non-null value in "next". The userland
* code will interpret a NULL here as a marker
* for the last dynamic rule.
*/
dst->next = dst ;
last = dst ;
if (TIME_LEQ(dst->expire, time_second) )
dst->expire = 0 ;
else
dst->expire -= time_second ;
}
if (last != NULL)
last->next = NULL ; /* mark last dynamic rule */
}
splx(s);
error = sooptcopyout(sopt, buf, size);
free(buf, M_TEMP);
break;
case IP_FW_FLUSH:
/*
* Normally we cannot release the lock on each iteration.
* We could do it here only because we start from the head all
* the times so there is no risk of missing some entries.
* On the other hand, the risk is that we end up with
* a very inconsistent ruleset, so better keep the lock
* around the whole cycle.
*
* XXX this code can be improved by resetting the head of
* the list to point to the default rule, and then freeing
* the old list without the need for a lock.
*/
s = splimp();
while ( (fcp = LIST_FIRST(&ip_fw_chain_head)) &&
fcp->fw_number != IPFW_DEFAULT_RULE )
free_chain(fcp);
splx(s);
break;
case IP_FW_ADD:
error = sooptcopyin(sopt, &frwl, sizeof frwl, sizeof frwl);
if (error || (error = check_ipfw_struct(&frwl)))
break;
if (frwl.fw_number == IPFW_DEFAULT_RULE) {
dprintf(("%s can't add rule %u\n", err_prefix,
(unsigned)IPFW_DEFAULT_RULE));
error = EINVAL;
} else {
error = add_entry(&ip_fw_chain_head, &frwl);
if (!error && sopt->sopt_dir == SOPT_GET)
error = sooptcopyout(sopt, &frwl, sizeof frwl);
}
break;
case IP_FW_DEL:
error = sooptcopyin(sopt, &frwl, sizeof frwl, sizeof frwl);
if (error)
break;
if (frwl.fw_number == IPFW_DEFAULT_RULE) {
dprintf(("%s can't delete rule %u\n", err_prefix,
(unsigned)IPFW_DEFAULT_RULE));
error = EINVAL;
} else {
error = del_entry(&ip_fw_chain_head, frwl.fw_number);
}
break;
case IP_FW_ZERO:
case IP_FW_RESETLOG:
{
int cmd = (sopt->sopt_name == IP_FW_RESETLOG );
void *arg = NULL ;
if (sopt->sopt_val != 0) {
error = sooptcopyin(sopt, &frwl, sizeof frwl, sizeof frwl);
if (error)
break;
arg = &frwl ;
}
error = zero_entry(arg, cmd);
}
break;
default:
printf("ip_fw_ctl invalid option %d\n", sopt->sopt_name);
error = EINVAL ;
}
return (error);
}
/**
* dummynet needs a reference to the default rule, because rules can
* be deleted while packets hold a reference to them (e.g. to resume
* processing at the next rule). When this happens, dummynet changes
* the reference to the default rule (probably it could well be a
* NULL pointer, but this way we do not need to check for the special
* case, plus here he have info on the default behaviour.
*/
struct ip_fw *ip_fw_default_rule ;
void
ip_fw_init(void)
{
struct ip_fw default_rule;
ip_fw_chk_ptr = ip_fw_chk;
ip_fw_ctl_ptr = ip_fw_ctl;
LIST_INIT(&ip_fw_chain_head);
bzero(&default_rule, sizeof default_rule);
default_rule.fw_prot = IPPROTO_IP;
default_rule.fw_number = IPFW_DEFAULT_RULE;
#ifdef IPFIREWALL_DEFAULT_TO_ACCEPT
default_rule.fw_flg |= IP_FW_F_ACCEPT;
#else
default_rule.fw_flg |= IP_FW_F_DENY;
#endif
default_rule.fw_flg |= IP_FW_F_IN | IP_FW_F_OUT;
if (check_ipfw_struct(&default_rule) != 0 ||
add_entry(&ip_fw_chain_head, &default_rule))
1998-06-12 20:03:26 +00:00
panic("ip_fw_init");
ip_fw_default_rule = LIST_FIRST(&ip_fw_chain_head) ;
printf("IP packet filtering initialized, "
#ifdef IPDIVERT
"divert enabled, "
#else
"divert disabled, "
#endif
"rule-based forwarding enabled, "
#ifdef IPFIREWALL_DEFAULT_TO_ACCEPT
"default to accept, ");
#else
"default to deny, " );
#endif
#ifndef IPFIREWALL_VERBOSE
printf("logging disabled\n");
#else
if (fw_verbose_limit == 0)
printf("unlimited logging\n");
else
printf("logging limited to %d packets/entry by default\n",
fw_verbose_limit);
#endif
}
1998-12-21 22:40:54 +00:00
static int
ipfw_modevent(module_t mod, int type, void *unused)
{
int s;
int err = 0 ;
#if defined(KLD_MODULE)
struct ip_fw *fcp;
#endif
1998-12-21 22:40:54 +00:00
switch (type) {
case MOD_LOAD:
s = splimp();
if (IPFW_LOADED) {
splx(s);
printf("IP firewall already loaded\n");
err = EEXIST ;
} else {
ip_fw_init();
splx(s);
}
break ;
1998-12-21 22:40:54 +00:00
case MOD_UNLOAD:
#if !defined(KLD_MODULE)
printf("ipfw statically compiled, cannot unload\n");
err = EBUSY;
#else
s = splimp();
ip_fw_chk_ptr = NULL ;
ip_fw_ctl_ptr = NULL ;
while ( (fcp = LIST_FIRST(&ip_fw_chain_head)) != NULL)
free_chain(fcp);
1998-12-21 22:40:54 +00:00
splx(s);
printf("IP firewall unloaded\n");
#endif
break;
1998-12-21 22:40:54 +00:00
default:
break;
}
return err;
1998-12-21 22:40:54 +00:00
}
static moduledata_t ipfwmod = {
"ipfw",
ipfw_modevent,
0
};
DECLARE_MODULE(ipfw, ipfwmod, SI_SUB_PSEUDO, SI_ORDER_ANY);
MODULE_VERSION(ipfw, 1);