diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ip_pool.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ip_pool.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ip_pool.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ip_pool.h 2004-04-19 10:08:36.000000000 +0200 @@ -0,0 +1,64 @@ +#ifndef _IP_POOL_H +#define _IP_POOL_H + +/***************************************************************************/ +/* This program is free software; you can redistribute it and/or modify */ +/* it under the terms of the GNU General Public License as published by */ +/* the Free Software Foundation; either version 2 of the License, or */ +/* (at your option) any later version. */ +/* */ +/* This program is distributed in the hope that it will be useful, */ +/* but WITHOUT ANY WARRANTY; without even the implied warranty of */ +/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the */ +/* GNU General Public License for more details. */ +/* */ +/* You should have received a copy of the GNU General Public License */ +/* along with this program; if not, write to the Free Software */ +/* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA*/ +/***************************************************************************/ + +/* A sockopt of such quality has hardly ever been seen before on the open + * market! This little beauty, hardly ever used: above 64, so it's + * traditionally used for firewalling, not touched (even once!) by the + * 2.0, 2.2 and 2.4 kernels! + * + * Comes with its own certificate of authenticity, valid anywhere in the + * Free world! + * + * Rusty, 19.4.2000 + */ +#define SO_IP_POOL 81 + +typedef int ip_pool_t; /* pool index */ +#define IP_POOL_NONE ((ip_pool_t)-1) + +struct ip_pool_request { + int op; + ip_pool_t index; + u_int32_t addr; + u_int32_t addr2; +}; + +/* NOTE: I deliberately break the first cut ippool utility. Nobody uses it. */ + +#define IP_POOL_BAD001 0x00000010 + +#define IP_POOL_FLUSH 0x00000011 /* req.index, no arguments */ +#define IP_POOL_INIT 0x00000012 /* from addr to addr2 incl. */ +#define IP_POOL_DESTROY 0x00000013 /* req.index, no arguments */ +#define IP_POOL_ADD_ADDR 0x00000014 /* add addr to pool */ +#define IP_POOL_DEL_ADDR 0x00000015 /* del addr from pool */ +#define IP_POOL_HIGH_NR 0x00000016 /* result in req.index */ +#define IP_POOL_LOOKUP 0x00000017 /* result in addr and addr2 */ +#define IP_POOL_USAGE 0x00000018 /* result in addr */ +#define IP_POOL_TEST_ADDR 0x00000019 /* result (0/1) returned */ + +#ifdef __KERNEL__ + +/* NOTE: ip_pool_match() and ip_pool_mod() expect ADDR to be host byte order */ +extern int ip_pool_match(ip_pool_t pool, u_int32_t addr); +extern int ip_pool_mod(ip_pool_t pool, u_int32_t addr, int isdel); + +#endif + +#endif /*_IP_POOL_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_TTL.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_TTL.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_TTL.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_TTL.h 2004-04-19 10:08:28.000000000 +0200 @@ -0,0 +1,21 @@ +/* TTL modification module for IP tables + * (C) 2000 by Harald Welte */ + +#ifndef _IPT_TTL_H +#define _IPT_TTL_H + +enum { + IPT_TTL_SET = 0, + IPT_TTL_INC, + IPT_TTL_DEC +}; + +#define IPT_TTL_MAXMODE IPT_TTL_DEC + +struct ipt_TTL_info { + u_int8_t mode; + u_int8_t ttl; +}; + + +#endif diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_connlimit.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_connlimit.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_connlimit.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_connlimit.h 2004-04-19 10:08:29.000000000 +0200 @@ -0,0 +1,12 @@ +#ifndef _IPT_CONNLIMIT_H +#define _IPT_CONNLIMIT_H + +struct ipt_connlimit_data; + +struct ipt_connlimit_info { + int limit; + int inverse; + u_int32_t mask; + struct ipt_connlimit_data *data; +}; +#endif /* _IPT_CONNLIMIT_H */ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_dstlimit.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_dstlimit.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_dstlimit.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_dstlimit.h 2004-04-19 10:08:30.000000000 +0200 @@ -0,0 +1,39 @@ +#ifndef _IPT_DSTLIMIT_H +#define _IPT_DSTLIMIT_H + +/* timings are in milliseconds. */ +#define IPT_DSTLIMIT_SCALE 10000 +/* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 + seconds, or one every 59 hours. */ + +/* details of this structure hidden by the implementation */ +struct ipt_dstlimit_htable; + +#define IPT_DSTLIMIT_HASH_DIP 0x0001 +#define IPT_DSTLIMIT_HASH_DPT 0x0002 +#define IPT_DSTLIMIT_HASH_SIP 0x0004 + +struct dstlimit_cfg { + u_int32_t mode; /* bitmask of IPT_DSTLIMIT_HASH_* */ + u_int32_t avg; /* Average secs between packets * scale */ + u_int32_t burst; /* Period multiplier for upper limit. */ + + /* user specified */ + u_int32_t size; /* how many buckets */ + u_int32_t max; /* max number of entries */ + u_int32_t gc_interval; /* gc interval */ + u_int32_t expire; /* when do entries expire? */ +}; + +struct ipt_dstlimit_info { + char name [IFNAMSIZ]; /* name */ + struct dstlimit_cfg cfg; + struct ipt_dstlimit_htable *hinfo; + + /* Used internally by the kernel */ + union { + void *ptr; + struct ipt_dstlimit_info *master; + } u; +}; +#endif /*_IPT_DSTLIMIT_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_fuzzy.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_fuzzy.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_fuzzy.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_fuzzy.h 2004-04-19 10:08:31.000000000 +0200 @@ -0,0 +1,21 @@ +#ifndef _IPT_FUZZY_H +#define _IPT_FUZZY_H + +#include +#include + +#define MAXFUZZYRATE 10000000 +#define MINFUZZYRATE 3 + +struct ipt_fuzzy_info { + u_int32_t minimum_rate; + u_int32_t maximum_rate; + u_int32_t packets_total; + u_int32_t bytes_total; + u_int32_t previous_time; + u_int32_t present_time; + u_int32_t mean_rate; + u_int8_t acceptance_rate; +}; + +#endif /*_IPT_FUZZY_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_ipv4options.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_ipv4options.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_ipv4options.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_ipv4options.h 2004-04-19 10:08:32.000000000 +0200 @@ -0,0 +1,21 @@ +#ifndef __ipt_ipv4options_h_included__ +#define __ipt_ipv4options_h_included__ + +#define IPT_IPV4OPTION_MATCH_SSRR 0x01 /* For strict source routing */ +#define IPT_IPV4OPTION_MATCH_LSRR 0x02 /* For loose source routing */ +#define IPT_IPV4OPTION_DONT_MATCH_SRR 0x04 /* any source routing */ +#define IPT_IPV4OPTION_MATCH_RR 0x08 /* For Record route */ +#define IPT_IPV4OPTION_DONT_MATCH_RR 0x10 +#define IPT_IPV4OPTION_MATCH_TIMESTAMP 0x20 /* For timestamp request */ +#define IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP 0x40 +#define IPT_IPV4OPTION_MATCH_ROUTER_ALERT 0x80 /* For router-alert */ +#define IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT 0x100 +#define IPT_IPV4OPTION_MATCH_ANY_OPT 0x200 /* match packet with any option */ +#define IPT_IPV4OPTION_DONT_MATCH_ANY_OPT 0x400 /* match packet with no option */ + +struct ipt_ipv4options_info { + u_int16_t options; +}; + + +#endif /* __ipt_ipv4options_h_included__ */ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_mport.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_mport.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_mport.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_mport.h 2004-04-19 10:08:33.000000000 +0200 @@ -0,0 +1,24 @@ +#ifndef _IPT_MPORT_H +#define _IPT_MPORT_H +#include + +#define IPT_MPORT_SOURCE (1<<0) +#define IPT_MPORT_DESTINATION (1<<1) +#define IPT_MPORT_EITHER (IPT_MPORT_SOURCE|IPT_MPORT_DESTINATION) + +#define IPT_MULTI_PORTS 15 + +/* Must fit inside union ipt_matchinfo: 32 bytes */ +/* every entry in ports[] except for the last one has one bit in pflags + * associated with it. If this bit is set, the port is the first port of + * a portrange, with the next entry being the last. + * End of list is marked with pflags bit set and port=65535. + * If 14 ports are used (last one does not have a pflag), the last port + * is repeated to fill the last entry in ports[] */ +struct ipt_mport +{ + u_int8_t flags:2; /* Type of comparison */ + u_int16_t pflags:14; /* Port flags */ + u_int16_t ports[IPT_MULTI_PORTS]; /* Ports */ +}; +#endif /*_IPT_MPORT_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_nth.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_nth.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_nth.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_nth.h 2004-04-19 10:08:34.000000000 +0200 @@ -0,0 +1,19 @@ +#ifndef _IPT_NTH_H +#define _IPT_NTH_H + +#include +#include + +#ifndef IPT_NTH_NUM_COUNTERS +#define IPT_NTH_NUM_COUNTERS 16 +#endif + +struct ipt_nth_info { + u_int8_t every; + u_int8_t not; + u_int8_t startat; + u_int8_t counter; + u_int8_t packet; +}; + +#endif /*_IPT_NTH_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_osf.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_osf.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_osf.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_osf.h 2004-04-19 10:08:35.000000000 +0200 @@ -0,0 +1,148 @@ +/* + * ipt_osf.h + * + * Copyright (c) 2003 Evgeniy Polyakov + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef _IPT_OSF_H +#define _IPT_OSF_H + +#define MAXGENRELEN 32 +#define MAXDETLEN 64 + +#define IPT_OSF_GENRE 1 +#define IPT_OSF_SMART 2 +#define IPT_OSF_LOG 4 +#define IPT_OSF_NETLINK 8 + +#define IPT_OSF_LOGLEVEL_ALL 0 +#define IPT_OSF_LOGLEVEL_FIRST 1 + +#include + +#ifndef __KERNEL__ +#include +#include + +struct list_head +{ + struct list_head *prev, *next; +}; +#endif + +struct ipt_osf_info +{ + char genre[MAXGENRELEN]; + int len; + unsigned long flags; + int loglevel; + int invert; /* UNSUPPORTED */ +}; + +struct osf_wc +{ + char wc; + unsigned long val; +}; + +/* This struct represents IANA options + * http://www.iana.org/assignments/tcp-parameters + */ +struct osf_opt +{ + unsigned char kind; + unsigned char length; + struct osf_wc wc; +}; + +struct osf_finger +{ + struct list_head flist; + struct osf_wc wss; + unsigned char ttl; + unsigned char df; + unsigned long ss; + unsigned char genre[MAXGENRELEN]; + unsigned char version[MAXGENRELEN], subtype[MAXGENRELEN]; + + /* Not needed, but for consistency with original table from Michal Zalewski */ + unsigned char details[MAXDETLEN]; + + int opt_num; + struct osf_opt opt[MAX_IPOPTLEN]; /* In case it is all NOP or EOL */ + +}; + +struct ipt_osf_nlmsg +{ + struct osf_finger f; + struct iphdr ip; + struct tcphdr tcp; +}; + +#ifdef __KERNEL__ + +/* Defines for IANA option kinds */ + +#define OSFOPT_EOL 0 /* End of options */ +#define OSFOPT_NOP 1 /* NOP */ +#define OSFOPT_MSS 2 /* Maximum segment size */ +#define OSFOPT_WSO 3 /* Window scale option */ +#define OSFOPT_SACKP 4 /* SACK permitted */ +#define OSFOPT_SACK 5 /* SACK */ +#define OSFOPT_ECHO 6 +#define OSFOPT_ECHOREPLY 7 +#define OSFOPT_TS 8 /* Timestamp option */ +#define OSFOPT_POCP 9 /* Partial Order Connection Permitted */ +#define OSFOPT_POSP 10 /* Partial Order Service Profile */ +/* Others are not used in current OSF */ + +static struct osf_opt IANA_opts[] = +{ + {0, 1,}, + {1, 1,}, + {2, 4,}, + {3, 3,}, + {4, 2,}, + {5, 1 ,}, /* SACK length is not defined */ + {6, 6,}, + {7, 6,}, + {8, 10,}, + {9, 2,}, + {10, 3,}, + {11, 1,}, /* CC: Suppose 1 */ + {12, 1,}, /* the same */ + {13, 1,}, /* and here too */ + {14, 3,}, + {15, 1,}, /* TCP Alternate Checksum Data. Length is not defined */ + {16, 1,}, + {17, 1,}, + {18, 3,}, + {19, 18,}, + {20, 1,}, + {21, 1,}, + {22, 1,}, + {23, 1,}, + {24, 1,}, + {25, 1,}, + {26, 1,}, +}; + +#endif /* __KERNEL__ */ + +#endif /* _IPT_OSF_H */ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_pool.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_pool.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_pool.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_pool.h 2004-04-19 10:08:36.000000000 +0200 @@ -0,0 +1,25 @@ +#ifndef _IPT_POOL_H +#define _IPT_POOL_H + +#include + +#define IPT_POOL_INV_SRC 0x00000001 +#define IPT_POOL_INV_DST 0x00000002 +#define IPT_POOL_DEL_SRC 0x00000004 +#define IPT_POOL_DEL_DST 0x00000008 +#define IPT_POOL_INV_MOD_SRC 0x00000010 +#define IPT_POOL_INV_MOD_DST 0x00000020 +#define IPT_POOL_MOD_SRC_ACCEPT 0x00000040 +#define IPT_POOL_MOD_DST_ACCEPT 0x00000080 +#define IPT_POOL_MOD_SRC_DROP 0x00000100 +#define IPT_POOL_MOD_DST_DROP 0x00000200 + +/* match info */ +struct ipt_pool_info +{ + ip_pool_t src; + ip_pool_t dst; + unsigned flags; +}; + +#endif /*_IPT_POOL_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_psd.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_psd.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_psd.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_psd.h 2004-04-19 10:08:37.000000000 +0200 @@ -0,0 +1,40 @@ +#ifndef _IPT_PSD_H +#define _IPT_PSD_H + +#include +#include + +/* + * High port numbers have a lower weight to reduce the frequency of false + * positives, such as from passive mode FTP transfers. + */ +#define PORT_WEIGHT_PRIV 3 +#define PORT_WEIGHT_HIGH 1 + +/* + * Port scan detection thresholds: at least COUNT ports need to be scanned + * from the same source, with no longer than DELAY ticks between ports. + */ +#define SCAN_MIN_COUNT 7 +#define SCAN_MAX_COUNT (SCAN_MIN_COUNT * PORT_WEIGHT_PRIV) +#define SCAN_WEIGHT_THRESHOLD SCAN_MAX_COUNT +#define SCAN_DELAY_THRESHOLD (HZ * 3) + +/* + * Keep track of up to LIST_SIZE source addresses, using a hash table of + * HASH_SIZE entries for faster lookups, but limiting hash collisions to + * HASH_MAX source addresses per the same hash value. + */ +#define LIST_SIZE 0x100 +#define HASH_LOG 9 +#define HASH_SIZE (1 << HASH_LOG) +#define HASH_MAX 0x10 + +struct ipt_psd_info { + unsigned int weight_threshold; + unsigned int delay_threshold; + unsigned short lo_ports_weight; + unsigned short hi_ports_weight; +}; + +#endif /*_IPT_PSD_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_quota.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_quota.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_quota.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_quota.h 2004-04-19 10:08:37.000000000 +0200 @@ -0,0 +1,11 @@ +#ifndef _IPT_QUOTA_H +#define _IPT_QUOTA_H + +/* print debug info in both kernel/netfilter module & iptable library */ +//#define DEBUG_IPT_QUOTA + +struct ipt_quota_info { + u_int64_t quota; +}; + +#endif /*_IPT_QUOTA_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_random.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_random.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_random.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_random.h 2004-04-19 10:08:38.000000000 +0200 @@ -0,0 +1,11 @@ +#ifndef _IPT_RAND_H +#define _IPT_RAND_H + +#include +#include + +struct ipt_rand_info { + u_int8_t average; +}; + +#endif /*_IPT_RAND_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_realm.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_realm.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_realm.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_realm.h 2004-04-19 10:08:39.000000000 +0200 @@ -0,0 +1,9 @@ +#ifndef _IPT_REALM_H +#define _IPT_REALM_H + +struct ipt_realm_info { + u_int32_t id; + u_int32_t mask; + u_int8_t invert; +}; +#endif /*_IPT_REALM_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_sctp.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_sctp.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_sctp.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_sctp.h 2004-04-19 10:08:40.000000000 +0200 @@ -0,0 +1,107 @@ +#ifndef _IPT_SCTP_H_ +#define _IPT_SCTP_H_ + +#define IPT_SCTP_SRC_PORTS 0x01 +#define IPT_SCTP_DEST_PORTS 0x02 +#define IPT_SCTP_CHUNK_TYPES 0x04 + +#define IPT_SCTP_VALID_FLAGS 0x07 + +#define ELEMCOUNT(x) (sizeof(x)/sizeof(x[0])) + + +struct ipt_sctp_flag_info { + u_int8_t chunktype; + u_int8_t flag; + u_int8_t flag_mask; +}; + +#define IPT_NUM_SCTP_FLAGS 4 + +struct ipt_sctp_info { + u_int16_t dpts[2]; /* Min, Max */ + u_int16_t spts[2]; /* Min, Max */ + + u_int32_t chunkmap[256 / sizeof (u_int32_t)]; /* Bit mask of chunks to be matched according to RFC 2960 */ + +#define SCTP_CHUNK_MATCH_ANY 0x01 /* Match if any of the chunk types are present */ +#define SCTP_CHUNK_MATCH_ALL 0x02 /* Match if all of the chunk types are present */ +#define SCTP_CHUNK_MATCH_ONLY 0x04 /* Match if these are the only chunk types present */ + + u_int32_t chunk_match_type; + struct ipt_sctp_flag_info flag_info[IPT_NUM_SCTP_FLAGS]; + int flag_count; + + u_int32_t flags; + u_int32_t invflags; +}; + +#define bytes(type) (sizeof(type) * 8) + +#define SCTP_CHUNKMAP_SET(chunkmap, type) \ + do { \ + chunkmap[type / bytes(u_int32_t)] |= \ + 1 << (type % bytes(u_int32_t)); \ + } while (0) + +#define SCTP_CHUNKMAP_CLEAR(chunkmap, type) \ + do { \ + chunkmap[type / bytes(u_int32_t)] &= \ + ~(1 << (type % bytes(u_int32_t))); \ + } while (0) + +#define SCTP_CHUNKMAP_IS_SET(chunkmap, type) \ +({ \ + (chunkmap[type / bytes (u_int32_t)] & \ + (1 << (type % bytes (u_int32_t)))) ? 1: 0; \ +}) + +#define SCTP_CHUNKMAP_RESET(chunkmap) \ + do { \ + int i; \ + for (i = 0; i < ELEMCOUNT(chunkmap); i++) \ + chunkmap[i] = 0; \ + } while (0) + +#define SCTP_CHUNKMAP_SET_ALL(chunkmap) \ + do { \ + int i; \ + for (i = 0; i < ELEMCOUNT(chunkmap); i++) \ + chunkmap[i] = ~0; \ + } while (0) + +#define SCTP_CHUNKMAP_COPY(destmap, srcmap) \ + do { \ + int i; \ + for (i = 0; i < ELEMCOUNT(chunkmap); i++) \ + destmap[i] = srcmap[i]; \ + } while (0) + +#define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) \ +({ \ + int i; \ + int flag = 1; \ + for (i = 0; i < ELEMCOUNT(chunkmap); i++) { \ + if (chunkmap[i]) { \ + flag = 0; \ + break; \ + } \ + } \ + flag; \ +}) + +#define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) \ +({ \ + int i; \ + int flag = 1; \ + for (i = 0; i < ELEMCOUNT(chunkmap); i++) { \ + if (chunkmap[i] != ~0) { \ + flag = 0; \ + break; \ + } \ + } \ + flag; \ +}) + +#endif /* _IPT_SCTP_H_ */ + diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_time.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_time.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_time.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_time.h 2004-04-19 10:08:41.000000000 +0200 @@ -0,0 +1,13 @@ +#ifndef __ipt_time_h_included__ +#define __ipt_time_h_included__ + + +struct ipt_time_info { + u_int8_t days_match; /* 1 bit per day. -SMTWTFS */ + u_int16_t time_start; /* 0 < time_start < 23*60+59 = 1439 */ + u_int16_t time_stop; /* 0:0 < time_stat < 23:59 */ + u_int8_t kerneltime; /* ignore skb time (and use kerneltime) or not. */ +}; + + +#endif /* __ipt_time_h_included__ */ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_u32.h linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_u32.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv4/ipt_u32.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv4/ipt_u32.h 2004-04-19 10:08:46.000000000 +0200 @@ -0,0 +1,40 @@ +#ifndef _IPT_U32_H +#define _IPT_U32_H +#include + +enum ipt_u32_ops +{ + IPT_U32_AND, + IPT_U32_LEFTSH, + IPT_U32_RIGHTSH, + IPT_U32_AT +}; + +struct ipt_u32_location_element +{ + u_int32_t number; + u_int8_t nextop; +}; +struct ipt_u32_value_element +{ + u_int32_t min; + u_int32_t max; +}; +/* *** any way to allow for an arbitrary number of elements? + for now I settle for a limit of 10 of each */ +#define U32MAXSIZE 10 +struct ipt_u32_test +{ + u_int8_t nnums; + struct ipt_u32_location_element location[U32MAXSIZE+1]; + u_int8_t nvalues; + struct ipt_u32_value_element value[U32MAXSIZE+1]; +}; + +struct ipt_u32 +{ + u_int8_t ntests; + struct ipt_u32_test tests[U32MAXSIZE+1]; +}; + +#endif /*_IPT_U32_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_HL.h linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_HL.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_HL.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_HL.h 2004-04-19 10:08:26.000000000 +0200 @@ -0,0 +1,22 @@ +/* Hop Limit modification module for ip6tables + * Maciej Soltysiak + * Based on HW's TTL module */ + +#ifndef _IP6T_HL_H +#define _IP6T_HL_H + +enum { + IP6T_HL_SET = 0, + IP6T_HL_INC, + IP6T_HL_DEC +}; + +#define IP6T_HL_MAXMODE IP6T_HL_DEC + +struct ip6t_HL_info { + u_int8_t mode; + u_int8_t hop_limit; +}; + + +#endif diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_REJECT.h linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_REJECT.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_REJECT.h 2004-04-15 03:33:49.000000000 +0200 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_REJECT.h 2004-04-19 10:08:28.000000000 +0200 @@ -2,15 +2,17 @@ #define _IP6T_REJECT_H enum ip6t_reject_with { - IP6T_ICMP_NET_UNREACHABLE, - IP6T_ICMP_HOST_UNREACHABLE, - IP6T_ICMP_PROT_UNREACHABLE, - IP6T_ICMP_PORT_UNREACHABLE, - IP6T_ICMP_ECHOREPLY + IP6T_ICMP6_NO_ROUTE, + IP6T_ICMP6_ADM_PROHIBITED, + IP6T_ICMP6_NOT_NEIGHBOUR, + IP6T_ICMP6_ADDR_UNREACH, + IP6T_ICMP6_PORT_UNREACH, + IP6T_ICMP6_ECHOREPLY, + IP6T_TCP_RESET }; struct ip6t_reject_info { enum ip6t_reject_with with; /* reject type */ }; -#endif /*_IPT_REJECT_H*/ +#endif /*_IP6T_REJECT_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_fuzzy.h linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_fuzzy.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_fuzzy.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_fuzzy.h 2004-04-19 10:08:31.000000000 +0200 @@ -0,0 +1,21 @@ +#ifndef _IP6T_FUZZY_H +#define _IP6T_FUZZY_H + +#include +#include + +#define MAXFUZZYRATE 10000000 +#define MINFUZZYRATE 3 + +struct ip6t_fuzzy_info { + u_int32_t minimum_rate; + u_int32_t maximum_rate; + u_int32_t packets_total; + u_int32_t bytes_total; + u_int32_t previous_time; + u_int32_t present_time; + u_int32_t mean_rate; + u_int8_t acceptance_rate; +}; + +#endif /*_IP6T_FUZZY_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_nth.h linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_nth.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_nth.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_nth.h 2004-04-19 10:08:34.000000000 +0200 @@ -0,0 +1,19 @@ +#ifndef _IP6T_NTH_H +#define _IP6T_NTH_H + +#include +#include + +#ifndef IP6T_NTH_NUM_COUNTERS +#define IP6T_NTH_NUM_COUNTERS 16 +#endif + +struct ip6t_nth_info { + u_int8_t every; + u_int8_t not; + u_int8_t startat; + u_int8_t counter; + u_int8_t packet; +}; + +#endif /*_IP6T_NTH_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_random.h linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_random.h --- linux-2.6.6-rc1.org/include/linux/netfilter_ipv6/ip6t_random.h 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/include/linux/netfilter_ipv6/ip6t_random.h 2004-04-19 10:08:38.000000000 +0200 @@ -0,0 +1,11 @@ +#ifndef _IP6T_RAND_H +#define _IP6T_RAND_H + +#include +#include + +struct ip6t_rand_info { + u_int8_t average; +}; + +#endif /*_IP6T_RAND_H*/ diff -Nur linux-2.6.6-rc1.org/include/linux/skbuff.h linux-2.6.6-rc1/include/linux/skbuff.h --- linux-2.6.6-rc1.org/include/linux/skbuff.h 2004-04-15 03:35:04.000000000 +0200 +++ linux-2.6.6-rc1/include/linux/skbuff.h 2004-04-19 10:08:24.000000000 +0200 @@ -1201,6 +1201,14 @@ if (nfct) atomic_inc(&nfct->master->use); } +static inline void nf_reset(struct sk_buff *skb) +{ + nf_conntrack_put(skb->nfct); + skb->nfct = NULL; +#ifdef CONFIG_NETFILTER_DEBUG + skb->nf_debug = 0; +#endif +} #ifdef CONFIG_BRIDGE_NETFILTER static inline void nf_bridge_put(struct nf_bridge_info *nf_bridge) @@ -1213,9 +1221,10 @@ if (nf_bridge) atomic_inc(&nf_bridge->use); } -#endif - -#endif +#endif /* CONFIG_BRIDGE_NETFILTER */ +#else /* CONFIG_NETFILTER */ +static inline void nf_reset(struct sk_buff *skb) {} +#endif /* CONFIG_NETFILTER */ #endif /* __KERNEL__ */ #endif /* _LINUX_SKBUFF_H */ diff -Nur linux-2.6.6-rc1.org/net/ipv4/ip_gre.c linux-2.6.6-rc1/net/ipv4/ip_gre.c --- linux-2.6.6-rc1.org/net/ipv4/ip_gre.c 2004-04-15 03:35:20.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/ip_gre.c 2004-04-19 10:08:24.000000000 +0200 @@ -643,13 +643,7 @@ skb->dev = tunnel->dev; dst_release(skb->dst); skb->dst = NULL; -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); ipgre_ecn_decapsulate(iph, skb); netif_rx(skb); read_unlock(&ipgre_lock); @@ -877,13 +871,7 @@ } } -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); IPTUNNEL_XMIT(); tunnel->recursion--; diff -Nur linux-2.6.6-rc1.org/net/ipv4/ip_input.c linux-2.6.6-rc1/net/ipv4/ip_input.c --- linux-2.6.6-rc1.org/net/ipv4/ip_input.c 2004-04-15 03:33:53.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/ip_input.c 2004-04-19 10:08:24.000000000 +0200 @@ -202,17 +202,13 @@ #ifdef CONFIG_NETFILTER_DEBUG nf_debug_ip_local_deliver(skb); - skb->nf_debug = 0; #endif /*CONFIG_NETFILTER_DEBUG*/ __skb_pull(skb, ihl); -#ifdef CONFIG_NETFILTER /* Free reference early: we don't need it any more, and it may hold ip_conntrack module loaded indefinitely. */ - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#endif /*CONFIG_NETFILTER*/ + nf_reset(skb); /* Point into the IP datagram, just past the header. */ skb->h.raw = skb->data; diff -Nur linux-2.6.6-rc1.org/net/ipv4/ipip.c linux-2.6.6-rc1/net/ipv4/ipip.c --- linux-2.6.6-rc1.org/net/ipv4/ipip.c 2004-04-15 03:36:03.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/ipip.c 2004-04-19 10:08:24.000000000 +0200 @@ -496,13 +496,7 @@ skb->dev = tunnel->dev; dst_release(skb->dst); skb->dst = NULL; -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); ipip_ecn_decapsulate(iph, skb); netif_rx(skb); read_unlock(&ipip_lock); @@ -647,13 +641,7 @@ if ((iph->ttl = tiph->ttl) == 0) iph->ttl = old_iph->ttl; -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); IPTUNNEL_XMIT(); tunnel->recursion--; diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/Kconfig linux-2.6.6-rc1/net/ipv4/netfilter/Kconfig --- linux-2.6.6-rc1.org/net/ipv4/netfilter/Kconfig 2004-04-19 09:59:32.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/netfilter/Kconfig 2004-04-19 10:08:46.000000000 +0200 @@ -603,5 +603,94 @@ . If unsure, say `N'. help +config IP_NF_TARGET_IPV4OPTSSTRIP + tristate 'IPV4OPTSSTRIP target support' + depends on IP_NF_MANGLE + help + +config IP_NF_TARGET_TTL + tristate 'TTL target support' + depends on IP_NF_MANGLE + help + +config IP_NF_MATCH_CONNLIMIT + tristate 'Connections/IP limit match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_DSTLIMIT + tristate 'dstlimit match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_FUZZY + tristate 'fuzzy match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_IPV4OPTIONS + tristate 'IPV4OPTIONS match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_MPORT + tristate 'Multiple port with ranges match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_NTH + tristate 'Nth match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_OSF + tristate 'OSF match support' + depends on IP_NF_IPTABLES + help + +config IP_POOL_STATISTICS + bool 'enable statistics on pool usage' + depends on IP_NF_POOL!=n + +config IP_NF_POOL + tristate 'IP address pool support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_PSD + tristate 'psd match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_QUOTA + tristate 'quota match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_RANDOM + tristate 'random match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_REALM + tristate 'realm match support' + depends on IP_NF_IPTABLES && NET_CLS_ROUTE + help + +config IP_NF_MATCH_SCTP + tristate 'SCTP protocol match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_TIME + tristate 'TIME match support' + depends on IP_NF_IPTABLES + help + +config IP_NF_MATCH_U32 + tristate 'U32 match support' + depends on IP_NF_IPTABLES + help + endmenu diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/Makefile linux-2.6.6-rc1/net/ipv4/netfilter/Makefile --- linux-2.6.6-rc1.org/net/ipv4/netfilter/Makefile 2004-04-19 09:59:32.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/netfilter/Makefile 2004-04-19 10:08:46.000000000 +0200 @@ -43,15 +43,39 @@ # matches obj-$(CONFIG_IP_NF_MATCH_HELPER) += ipt_helper.o obj-$(CONFIG_IP_NF_MATCH_LIMIT) += ipt_limit.o +obj-$(CONFIG_IP_NF_MATCH_SCTP) += ipt_sctp.o +obj-$(CONFIG_IP_NF_MATCH_QUOTA) += ipt_quota.o +obj-$(CONFIG_IP_NF_MATCH_DSTLIMIT) += ipt_dstlimit.o obj-$(CONFIG_IP_NF_MATCH_MARK) += ipt_mark.o +obj-$(CONFIG_IP_NF_POOL) += ipt_pool.o ipt_POOL.o ip_pool.o obj-$(CONFIG_IP_NF_MATCH_MAC) += ipt_mac.o obj-$(CONFIG_IP_NF_MATCH_IPRANGE) += ipt_iprange.o obj-$(CONFIG_IP_NF_MATCH_PKTTYPE) += ipt_pkttype.o obj-$(CONFIG_IP_NF_MATCH_MULTIPORT) += ipt_multiport.o + +obj-$(CONFIG_IP_NF_MATCH_MPORT) += ipt_mport.o + obj-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o +obj-$(CONFIG_IP_NF_MATCH_TIME) += ipt_time.o + + +obj-$(CONFIG_IP_NF_MATCH_RANDOM) += ipt_random.o + +obj-$(CONFIG_IP_NF_MATCH_PSD) += ipt_psd.o + +obj-$(CONFIG_IP_NF_MATCH_OSF) += ipt_osf.o + + +obj-$(CONFIG_IP_NF_MATCH_NTH) += ipt_nth.o + +obj-$(CONFIG_IP_NF_MATCH_IPV4OPTIONS) += ipt_ipv4options.o + + +obj-$(CONFIG_IP_NF_MATCH_FUZZY) += ipt_fuzzy.o + obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o @@ -60,10 +84,15 @@ obj-$(CONFIG_IP_NF_MATCH_LENGTH) += ipt_length.o +obj-$(CONFIG_IP_NF_MATCH_U32) += ipt_u32.o + + obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o obj-$(CONFIG_IP_NF_MATCH_STATE) += ipt_state.o +obj-$(CONFIG_IP_NF_MATCH_CONNLIMIT) += ipt_connlimit.o obj-$(CONFIG_IP_NF_MATCH_CONNTRACK) += ipt_conntrack.o obj-$(CONFIG_IP_NF_MATCH_TCPMSS) += ipt_tcpmss.o +obj-$(CONFIG_IP_NF_MATCH_REALM) += ipt_realm.o obj-$(CONFIG_IP_NF_MATCH_PHYSDEV) += ipt_physdev.o @@ -80,6 +109,8 @@ obj-$(CONFIG_IP_NF_TARGET_CLASSIFY) += ipt_CLASSIFY.o obj-$(CONFIG_IP_NF_NAT_SNMP_BASIC) += ip_nat_snmp_basic.o obj-$(CONFIG_IP_NF_TARGET_LOG) += ipt_LOG.o +obj-$(CONFIG_IP_NF_TARGET_TTL) += ipt_TTL.o +obj-$(CONFIG_IP_NF_TARGET_IPV4OPTSSTRIP) += ipt_IPV4OPTSSTRIP.o obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o obj-$(CONFIG_IP_NF_TARGET_TCPMSS) += ipt_TCPMSS.o obj-$(CONFIG_IP_NF_TARGET_NOTRACK) += ipt_NOTRACK.o diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_conntrack_core.c linux-2.6.6-rc1/net/ipv4/netfilter/ip_conntrack_core.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_conntrack_core.c 2004-04-19 09:59:32.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ip_conntrack_core.c 2004-04-19 10:08:22.000000000 +0200 @@ -692,42 +692,50 @@ struct ip_conntrack_expect *, tuple); READ_UNLOCK(&ip_conntrack_expect_tuple_lock); - /* If master is not in hash table yet (ie. packet hasn't left - this machine yet), how can other end know about expected? - Hence these are not the droids you are looking for (if - master ct never got confirmed, we'd hold a reference to it - and weird things would happen to future packets). */ - if (expected && !is_confirmed(expected->expectant)) - expected = NULL; - - /* Look up the conntrack helper for master connections only */ - if (!expected) - conntrack->helper = ip_ct_find_helper(&repl_tuple); - - /* If the expectation is dying, then this is a loser. */ - if (expected - && expected->expectant->helper->timeout - && ! del_timer(&expected->timeout)) - expected = NULL; - if (expected) { - DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n", - conntrack, expected); - /* Welcome, Mr. Bond. We've been expecting you... */ - IP_NF_ASSERT(master_ct(conntrack)); - __set_bit(IPS_EXPECTED_BIT, &conntrack->status); - conntrack->master = expected; - expected->sibling = conntrack; - LIST_DELETE(&ip_conntrack_expect_list, expected); - expected->expectant->expecting--; - nf_conntrack_get(&master_ct(conntrack)->infos[0]); - } - atomic_inc(&ip_conntrack_count); + /* If master is not in hash table yet (ie. packet hasn't left + this machine yet), how can other end know about expected? + Hence these are not the droids you are looking for (if + master ct never got confirmed, we'd hold a reference to it + and weird things would happen to future packets). */ + if (!is_confirmed(expected->expectant)) { + + conntrack->helper = ip_ct_find_helper(&repl_tuple); + goto end; + } + + /* Expectation is dying... */ + if (expected->expectant->helper->timeout + && ! del_timer(&expected->timeout)) { + goto end; + } + + DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n", + conntrack, expected); + /* Welcome, Mr. Bond. We've been expecting you... */ + IP_NF_ASSERT(master_ct(conntrack)); + __set_bit(IPS_EXPECTED_BIT, &conntrack->status); + conntrack->master = expected; + expected->sibling = conntrack; + LIST_DELETE(&ip_conntrack_expect_list, expected); + expected->expectant->expecting--; + nf_conntrack_get(&master_ct(conntrack)->infos[0]); + + /* this is a braindead... --pablo */ + atomic_inc(&ip_conntrack_count); + WRITE_UNLOCK(&ip_conntrack_lock); + + if (expected->expectfn) + expected->expectfn(conntrack); + + goto ret; + } else + conntrack->helper = ip_ct_find_helper(&repl_tuple); + +end: atomic_inc(&ip_conntrack_count); WRITE_UNLOCK(&ip_conntrack_lock); - if (expected && expected->expectfn) - expected->expectfn(conntrack); - return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; +ret: return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; } /* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_conntrack_core.c.orig linux-2.6.6-rc1/net/ipv4/netfilter/ip_conntrack_core.c.orig --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_conntrack_core.c.orig 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ip_conntrack_core.c.orig 2004-04-19 10:02:28.000000000 +0200 @@ -0,0 +1,1467 @@ +/* Connection state tracking for netfilter. This is separated from, + but required by, the NAT layer; it can also be used by an iptables + extension. */ + +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * 23 Apr 2001: Harald Welte + * - new API and handling of conntrack/nat helpers + * - now capable of multiple expectations for one master + * 16 Jul 2002: Harald Welte + * - add usage/reference counts to ip_conntrack_expect + * - export ip_conntrack[_expect]_{find_get,put} functions + * */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +/* For ERR_PTR(). Yeah, I know... --RR */ +#include + +/* This rwlock protects the main hash table, protocol/helper/expected + registrations, conntrack timers*/ +#define ASSERT_READ_LOCK(x) MUST_BE_READ_LOCKED(&ip_conntrack_lock) +#define ASSERT_WRITE_LOCK(x) MUST_BE_WRITE_LOCKED(&ip_conntrack_lock) + +#include +#include +#include +#include +#include + +#define IP_CONNTRACK_VERSION "2.1" + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_RWLOCK(ip_conntrack_lock); +DECLARE_RWLOCK(ip_conntrack_expect_tuple_lock); + +void (*ip_conntrack_destroyed)(struct ip_conntrack *conntrack) = NULL; +LIST_HEAD(ip_conntrack_expect_list); +LIST_HEAD(protocol_list); +static LIST_HEAD(helpers); +unsigned int ip_conntrack_htable_size = 0; +int ip_conntrack_max; +static atomic_t ip_conntrack_count = ATOMIC_INIT(0); +struct list_head *ip_conntrack_hash; +static kmem_cache_t *ip_conntrack_cachep; +struct ip_conntrack ip_conntrack_untracked; + +extern struct ip_conntrack_protocol ip_conntrack_generic_protocol; + +static inline int proto_cmpfn(const struct ip_conntrack_protocol *curr, + u_int8_t protocol) +{ + return protocol == curr->proto; +} + +struct ip_conntrack_protocol *__ip_ct_find_proto(u_int8_t protocol) +{ + struct ip_conntrack_protocol *p; + + MUST_BE_READ_LOCKED(&ip_conntrack_lock); + p = LIST_FIND(&protocol_list, proto_cmpfn, + struct ip_conntrack_protocol *, protocol); + if (!p) + p = &ip_conntrack_generic_protocol; + + return p; +} + +struct ip_conntrack_protocol *ip_ct_find_proto(u_int8_t protocol) +{ + struct ip_conntrack_protocol *p; + + READ_LOCK(&ip_conntrack_lock); + p = __ip_ct_find_proto(protocol); + READ_UNLOCK(&ip_conntrack_lock); + return p; +} + +inline void +ip_conntrack_put(struct ip_conntrack *ct) +{ + IP_NF_ASSERT(ct); + IP_NF_ASSERT(ct->infos[0].master); + /* nf_conntrack_put wants to go via an info struct, so feed it + one at random. */ + nf_conntrack_put(&ct->infos[0]); +} + +static int ip_conntrack_hash_rnd_initted; +static unsigned int ip_conntrack_hash_rnd; + +static u_int32_t +hash_conntrack(const struct ip_conntrack_tuple *tuple) +{ +#if 0 + dump_tuple(tuple); +#endif + return (jhash_3words(tuple->src.ip, + (tuple->dst.ip ^ tuple->dst.protonum), + (tuple->src.u.all | (tuple->dst.u.all << 16)), + ip_conntrack_hash_rnd) % ip_conntrack_htable_size); +} + +int +get_tuple(const struct iphdr *iph, + const struct sk_buff *skb, + unsigned int dataoff, + struct ip_conntrack_tuple *tuple, + const struct ip_conntrack_protocol *protocol) +{ + /* Never happen */ + if (iph->frag_off & htons(IP_OFFSET)) { + printk("ip_conntrack_core: Frag of proto %u.\n", + iph->protocol); + return 0; + } + + tuple->src.ip = iph->saddr; + tuple->dst.ip = iph->daddr; + tuple->dst.protonum = iph->protocol; + + return protocol->pkt_to_tuple(skb, dataoff, tuple); +} + +static int +invert_tuple(struct ip_conntrack_tuple *inverse, + const struct ip_conntrack_tuple *orig, + const struct ip_conntrack_protocol *protocol) +{ + inverse->src.ip = orig->dst.ip; + inverse->dst.ip = orig->src.ip; + inverse->dst.protonum = orig->dst.protonum; + + return protocol->invert_tuple(inverse, orig); +} + + +/* ip_conntrack_expect helper functions */ + +/* Compare tuple parts depending on mask. */ +static inline int expect_cmp(const struct ip_conntrack_expect *i, + const struct ip_conntrack_tuple *tuple) +{ + MUST_BE_READ_LOCKED(&ip_conntrack_expect_tuple_lock); + return ip_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask); +} + +static void +destroy_expect(struct ip_conntrack_expect *exp) +{ + DEBUGP("destroy_expect(%p) use=%d\n", exp, atomic_read(&exp->use)); + IP_NF_ASSERT(atomic_read(&exp->use)); + IP_NF_ASSERT(!timer_pending(&exp->timeout)); + + kfree(exp); +} + + +inline void ip_conntrack_expect_put(struct ip_conntrack_expect *exp) +{ + IP_NF_ASSERT(exp); + + if (atomic_dec_and_test(&exp->use)) { + /* usage count dropped to zero */ + destroy_expect(exp); + } +} + +static inline struct ip_conntrack_expect * +__ip_ct_expect_find(const struct ip_conntrack_tuple *tuple) +{ + MUST_BE_READ_LOCKED(&ip_conntrack_lock); + MUST_BE_READ_LOCKED(&ip_conntrack_expect_tuple_lock); + return LIST_FIND(&ip_conntrack_expect_list, expect_cmp, + struct ip_conntrack_expect *, tuple); +} + +/* Find a expectation corresponding to a tuple. */ +struct ip_conntrack_expect * +ip_conntrack_expect_find_get(const struct ip_conntrack_tuple *tuple) +{ + struct ip_conntrack_expect *exp; + + READ_LOCK(&ip_conntrack_lock); + READ_LOCK(&ip_conntrack_expect_tuple_lock); + exp = __ip_ct_expect_find(tuple); + if (exp) + atomic_inc(&exp->use); + READ_UNLOCK(&ip_conntrack_expect_tuple_lock); + READ_UNLOCK(&ip_conntrack_lock); + + return exp; +} + +/* remove one specific expectation from all lists and drop refcount, + * does _NOT_ delete the timer. */ +static void __unexpect_related(struct ip_conntrack_expect *expect) +{ + DEBUGP("unexpect_related(%p)\n", expect); + MUST_BE_WRITE_LOCKED(&ip_conntrack_lock); + + /* we're not allowed to unexpect a confirmed expectation! */ + IP_NF_ASSERT(!expect->sibling); + + /* delete from global and local lists */ + list_del(&expect->list); + list_del(&expect->expected_list); + + /* decrement expect-count of master conntrack */ + if (expect->expectant) + expect->expectant->expecting--; + + ip_conntrack_expect_put(expect); +} + +/* remove one specific expecatation from all lists, drop refcount + * and expire timer. + * This function can _NOT_ be called for confirmed expects! */ +static void unexpect_related(struct ip_conntrack_expect *expect) +{ + IP_NF_ASSERT(expect->expectant); + IP_NF_ASSERT(expect->expectant->helper); + /* if we are supposed to have a timer, but we can't delete + * it: race condition. __unexpect_related will + * be calledd by timeout function */ + if (expect->expectant->helper->timeout + && !del_timer(&expect->timeout)) + return; + + __unexpect_related(expect); +} + +/* delete all unconfirmed expectations for this conntrack */ +static void remove_expectations(struct ip_conntrack *ct, int drop_refcount) +{ + struct list_head *exp_entry, *next; + struct ip_conntrack_expect *exp; + + DEBUGP("remove_expectations(%p)\n", ct); + + list_for_each_safe(exp_entry, next, &ct->sibling_list) { + exp = list_entry(exp_entry, struct ip_conntrack_expect, + expected_list); + + /* we skip established expectations, as we want to delete + * the un-established ones only */ + if (exp->sibling) { + DEBUGP("remove_expectations: skipping established %p of %p\n", exp->sibling, ct); + if (drop_refcount) { + /* Indicate that this expectations parent is dead */ + ip_conntrack_put(exp->expectant); + exp->expectant = NULL; + } + continue; + } + + IP_NF_ASSERT(list_inlist(&ip_conntrack_expect_list, exp)); + IP_NF_ASSERT(exp->expectant == ct); + + /* delete expectation from global and private lists */ + unexpect_related(exp); + } +} + +static void +clean_from_lists(struct ip_conntrack *ct) +{ + unsigned int ho, hr; + + DEBUGP("clean_from_lists(%p)\n", ct); + MUST_BE_WRITE_LOCKED(&ip_conntrack_lock); + + ho = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + hr = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + LIST_DELETE(&ip_conntrack_hash[ho], &ct->tuplehash[IP_CT_DIR_ORIGINAL]); + LIST_DELETE(&ip_conntrack_hash[hr], &ct->tuplehash[IP_CT_DIR_REPLY]); + + /* Destroy all un-established, pending expectations */ + remove_expectations(ct, 1); +} + +static void +destroy_conntrack(struct nf_conntrack *nfct) +{ + struct ip_conntrack *ct = (struct ip_conntrack *)nfct, *master = NULL; + struct ip_conntrack_protocol *proto; + + DEBUGP("destroy_conntrack(%p)\n", ct); + IP_NF_ASSERT(atomic_read(&nfct->use) == 0); + IP_NF_ASSERT(!timer_pending(&ct->timeout)); + + /* To make sure we don't get any weird locking issues here: + * destroy_conntrack() MUST NOT be called with a write lock + * to ip_conntrack_lock!!! -HW */ + proto = ip_ct_find_proto(ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + if (proto && proto->destroy) + proto->destroy(ct); + + if (ip_conntrack_destroyed) + ip_conntrack_destroyed(ct); + + WRITE_LOCK(&ip_conntrack_lock); + /* Delete us from our own list to prevent corruption later */ + list_del(&ct->sibling_list); + + /* Delete our master expectation */ + if (ct->master) { + if (ct->master->expectant) { + /* can't call __unexpect_related here, + * since it would screw up expect_list */ + list_del(&ct->master->expected_list); + master = ct->master->expectant; + } + kfree(ct->master); + } + WRITE_UNLOCK(&ip_conntrack_lock); + + if (master) + ip_conntrack_put(master); + + DEBUGP("destroy_conntrack: returning ct=%p to slab\n", ct); + kmem_cache_free(ip_conntrack_cachep, ct); + atomic_dec(&ip_conntrack_count); +} + +static void death_by_timeout(unsigned long ul_conntrack) +{ + struct ip_conntrack *ct = (void *)ul_conntrack; + + WRITE_LOCK(&ip_conntrack_lock); + clean_from_lists(ct); + WRITE_UNLOCK(&ip_conntrack_lock); + ip_conntrack_put(ct); +} + +static inline int +conntrack_tuple_cmp(const struct ip_conntrack_tuple_hash *i, + const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack *ignored_conntrack) +{ + MUST_BE_READ_LOCKED(&ip_conntrack_lock); + return i->ctrack != ignored_conntrack + && ip_ct_tuple_equal(tuple, &i->tuple); +} + +static struct ip_conntrack_tuple_hash * +__ip_conntrack_find(const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack *ignored_conntrack) +{ + struct ip_conntrack_tuple_hash *h; + unsigned int hash = hash_conntrack(tuple); + + MUST_BE_READ_LOCKED(&ip_conntrack_lock); + h = LIST_FIND(&ip_conntrack_hash[hash], + conntrack_tuple_cmp, + struct ip_conntrack_tuple_hash *, + tuple, ignored_conntrack); + return h; +} + +/* Find a connection corresponding to a tuple. */ +struct ip_conntrack_tuple_hash * +ip_conntrack_find_get(const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack *ignored_conntrack) +{ + struct ip_conntrack_tuple_hash *h; + + READ_LOCK(&ip_conntrack_lock); + h = __ip_conntrack_find(tuple, ignored_conntrack); + if (h) + atomic_inc(&h->ctrack->ct_general.use); + READ_UNLOCK(&ip_conntrack_lock); + + return h; +} + +static inline struct ip_conntrack * +__ip_conntrack_get(struct nf_ct_info *nfct, enum ip_conntrack_info *ctinfo) +{ + struct ip_conntrack *ct + = (struct ip_conntrack *)nfct->master; + + /* ctinfo is the index of the nfct inside the conntrack */ + *ctinfo = nfct - ct->infos; + IP_NF_ASSERT(*ctinfo >= 0 && *ctinfo < IP_CT_NUMBER); + return ct; +} + +/* Return conntrack and conntrack_info given skb->nfct->master */ +struct ip_conntrack * +ip_conntrack_get(struct sk_buff *skb, enum ip_conntrack_info *ctinfo) +{ + if (skb->nfct) + return __ip_conntrack_get(skb->nfct, ctinfo); + return NULL; +} + +/* Confirm a connection given skb->nfct; places it in hash table */ +int +__ip_conntrack_confirm(struct nf_ct_info *nfct) +{ + unsigned int hash, repl_hash; + struct ip_conntrack *ct; + enum ip_conntrack_info ctinfo; + + ct = __ip_conntrack_get(nfct, &ctinfo); + + /* ipt_REJECT uses ip_conntrack_attach to attach related + ICMP/TCP RST packets in other direction. Actual packet + which created connection will be IP_CT_NEW or for an + expected connection, IP_CT_RELATED. */ + if (CTINFO2DIR(ctinfo) != IP_CT_DIR_ORIGINAL) + return NF_ACCEPT; + + hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + repl_hash = hash_conntrack(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + + /* We're not in hash table, and we refuse to set up related + connections for unconfirmed conns. But packet copies and + REJECT will give spurious warnings here. */ + /* IP_NF_ASSERT(atomic_read(&ct->ct_general.use) == 1); */ + + /* No external references means noone else could have + confirmed us. */ + IP_NF_ASSERT(!is_confirmed(ct)); + DEBUGP("Confirming conntrack %p\n", ct); + + WRITE_LOCK(&ip_conntrack_lock); + /* See if there's one in the list already, including reverse: + NAT could have grabbed it without realizing, since we're + not in the hash. If there is, we lost race. */ + if (!LIST_FIND(&ip_conntrack_hash[hash], + conntrack_tuple_cmp, + struct ip_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple, NULL) + && !LIST_FIND(&ip_conntrack_hash[repl_hash], + conntrack_tuple_cmp, + struct ip_conntrack_tuple_hash *, + &ct->tuplehash[IP_CT_DIR_REPLY].tuple, NULL)) { + list_prepend(&ip_conntrack_hash[hash], + &ct->tuplehash[IP_CT_DIR_ORIGINAL]); + list_prepend(&ip_conntrack_hash[repl_hash], + &ct->tuplehash[IP_CT_DIR_REPLY]); + /* Timer relative to confirmation time, not original + setting time, otherwise we'd get timer wrap in + weird delay cases. */ + ct->timeout.expires += jiffies; + add_timer(&ct->timeout); + atomic_inc(&ct->ct_general.use); + set_bit(IPS_CONFIRMED_BIT, &ct->status); + WRITE_UNLOCK(&ip_conntrack_lock); + return NF_ACCEPT; + } + + WRITE_UNLOCK(&ip_conntrack_lock); + return NF_DROP; +} + +/* Returns true if a connection correspondings to the tuple (required + for NAT). */ +int +ip_conntrack_tuple_taken(const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack *ignored_conntrack) +{ + struct ip_conntrack_tuple_hash *h; + + READ_LOCK(&ip_conntrack_lock); + h = __ip_conntrack_find(tuple, ignored_conntrack); + READ_UNLOCK(&ip_conntrack_lock); + + return h != NULL; +} + +/* Returns conntrack if it dealt with ICMP, and filled in skb fields */ +struct ip_conntrack * +icmp_error_track(struct sk_buff *skb, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct ip_conntrack_tuple innertuple, origtuple; + struct { + struct icmphdr icmp; + struct iphdr ip; + } inside; + struct ip_conntrack_protocol *innerproto; + struct ip_conntrack_tuple_hash *h; + int dataoff; + + IP_NF_ASSERT(skb->nfct == NULL); + + /* Not enough header? */ + if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &inside, sizeof(inside))!=0) + return NULL; + + if (inside.icmp.type != ICMP_DEST_UNREACH + && inside.icmp.type != ICMP_SOURCE_QUENCH + && inside.icmp.type != ICMP_TIME_EXCEEDED + && inside.icmp.type != ICMP_PARAMETERPROB + && inside.icmp.type != ICMP_REDIRECT) + return NULL; + + /* Ignore ICMP's containing fragments (shouldn't happen) */ + if (inside.ip.frag_off & htons(IP_OFFSET)) { + DEBUGP("icmp_error_track: fragment of proto %u\n", + inside.ip.protocol); + return NULL; + } + + innerproto = ip_ct_find_proto(inside.ip.protocol); + dataoff = skb->nh.iph->ihl*4 + sizeof(inside.icmp) + inside.ip.ihl*4; + /* Are they talking about one of our connections? */ + if (!get_tuple(&inside.ip, skb, dataoff, &origtuple, innerproto)) { + DEBUGP("icmp_error: ! get_tuple p=%u", inside.ip.protocol); + return NULL; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!invert_tuple(&innertuple, &origtuple, innerproto)) { + DEBUGP("icmp_error_track: Can't invert tuple\n"); + return NULL; + } + + *ctinfo = IP_CT_RELATED; + + h = ip_conntrack_find_get(&innertuple, NULL); + if (!h) { + /* Locally generated ICMPs will match inverted if they + haven't been SNAT'ed yet */ + /* FIXME: NAT code has to handle half-done double NAT --RR */ + if (hooknum == NF_IP_LOCAL_OUT) + h = ip_conntrack_find_get(&origtuple, NULL); + + if (!h) { + DEBUGP("icmp_error_track: no match\n"); + return NULL; + } + /* Reverse direction from that found */ + if (DIRECTION(h) != IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } else { + if (DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &h->ctrack->infos[*ctinfo]; + return h->ctrack; +} + +/* There's a small race here where we may free a just-assured + connection. Too bad: we're in trouble anyway. */ +static inline int unreplied(const struct ip_conntrack_tuple_hash *i) +{ + return !(test_bit(IPS_ASSURED_BIT, &i->ctrack->status)); +} + +static int early_drop(struct list_head *chain) +{ + /* Traverse backwards: gives us oldest, which is roughly LRU */ + struct ip_conntrack_tuple_hash *h; + int dropped = 0; + + READ_LOCK(&ip_conntrack_lock); + h = LIST_FIND_B(chain, unreplied, struct ip_conntrack_tuple_hash *); + if (h) + atomic_inc(&h->ctrack->ct_general.use); + READ_UNLOCK(&ip_conntrack_lock); + + if (!h) + return dropped; + + if (del_timer(&h->ctrack->timeout)) { + death_by_timeout((unsigned long)h->ctrack); + dropped = 1; + } + ip_conntrack_put(h->ctrack); + return dropped; +} + +static inline int helper_cmp(const struct ip_conntrack_helper *i, + const struct ip_conntrack_tuple *rtuple) +{ + return ip_ct_tuple_mask_cmp(rtuple, &i->tuple, &i->mask); +} + +struct ip_conntrack_helper *ip_ct_find_helper(const struct ip_conntrack_tuple *tuple) +{ + return LIST_FIND(&helpers, helper_cmp, + struct ip_conntrack_helper *, + tuple); +} + +/* Allocate a new conntrack: we return -ENOMEM if classification + failed due to stress. Otherwise it really is unclassifiable. */ +static struct ip_conntrack_tuple_hash * +init_conntrack(const struct ip_conntrack_tuple *tuple, + struct ip_conntrack_protocol *protocol, + struct sk_buff *skb) +{ + struct ip_conntrack *conntrack; + struct ip_conntrack_tuple repl_tuple; + size_t hash; + struct ip_conntrack_expect *expected; + int i; + static unsigned int drop_next; + + if (!ip_conntrack_hash_rnd_initted) { + get_random_bytes(&ip_conntrack_hash_rnd, 4); + ip_conntrack_hash_rnd_initted = 1; + } + + hash = hash_conntrack(tuple); + + if (ip_conntrack_max && + atomic_read(&ip_conntrack_count) >= ip_conntrack_max) { + /* Try dropping from random chain, or else from the + chain about to put into (in case they're trying to + bomb one hash chain). */ + unsigned int next = (drop_next++)%ip_conntrack_htable_size; + + if (!early_drop(&ip_conntrack_hash[next]) + && !early_drop(&ip_conntrack_hash[hash])) { + if (net_ratelimit()) + printk(KERN_WARNING + "ip_conntrack: table full, dropping" + " packet.\n"); + return ERR_PTR(-ENOMEM); + } + } + + if (!invert_tuple(&repl_tuple, tuple, protocol)) { + DEBUGP("Can't invert tuple.\n"); + return NULL; + } + + conntrack = kmem_cache_alloc(ip_conntrack_cachep, GFP_ATOMIC); + if (!conntrack) { + DEBUGP("Can't allocate conntrack.\n"); + return ERR_PTR(-ENOMEM); + } + + memset(conntrack, 0, sizeof(*conntrack)); + atomic_set(&conntrack->ct_general.use, 1); + conntrack->ct_general.destroy = destroy_conntrack; + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].tuple = *tuple; + conntrack->tuplehash[IP_CT_DIR_ORIGINAL].ctrack = conntrack; + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = repl_tuple; + conntrack->tuplehash[IP_CT_DIR_REPLY].ctrack = conntrack; + for (i=0; i < IP_CT_NUMBER; i++) + conntrack->infos[i].master = &conntrack->ct_general; + + if (!protocol->new(conntrack, skb)) { + kmem_cache_free(ip_conntrack_cachep, conntrack); + return NULL; + } + /* Don't set timer yet: wait for confirmation */ + init_timer(&conntrack->timeout); + conntrack->timeout.data = (unsigned long)conntrack; + conntrack->timeout.function = death_by_timeout; + + INIT_LIST_HEAD(&conntrack->sibling_list); + + WRITE_LOCK(&ip_conntrack_lock); + /* Need finding and deleting of expected ONLY if we win race */ + READ_LOCK(&ip_conntrack_expect_tuple_lock); + expected = LIST_FIND(&ip_conntrack_expect_list, expect_cmp, + struct ip_conntrack_expect *, tuple); + READ_UNLOCK(&ip_conntrack_expect_tuple_lock); + + /* If master is not in hash table yet (ie. packet hasn't left + this machine yet), how can other end know about expected? + Hence these are not the droids you are looking for (if + master ct never got confirmed, we'd hold a reference to it + and weird things would happen to future packets). */ + if (expected && !is_confirmed(expected->expectant)) + expected = NULL; + + /* Look up the conntrack helper for master connections only */ + if (!expected) + conntrack->helper = ip_ct_find_helper(&repl_tuple); + + /* If the expectation is dying, then this is a loser. */ + if (expected + && expected->expectant->helper->timeout + && ! del_timer(&expected->timeout)) + expected = NULL; + + if (expected) { + DEBUGP("conntrack: expectation arrives ct=%p exp=%p\n", + conntrack, expected); + /* Welcome, Mr. Bond. We've been expecting you... */ + IP_NF_ASSERT(master_ct(conntrack)); + __set_bit(IPS_EXPECTED_BIT, &conntrack->status); + conntrack->master = expected; + expected->sibling = conntrack; + LIST_DELETE(&ip_conntrack_expect_list, expected); + expected->expectant->expecting--; + nf_conntrack_get(&master_ct(conntrack)->infos[0]); + } + atomic_inc(&ip_conntrack_count); + WRITE_UNLOCK(&ip_conntrack_lock); + + if (expected && expected->expectfn) + expected->expectfn(conntrack); + return &conntrack->tuplehash[IP_CT_DIR_ORIGINAL]; +} + +/* On success, returns conntrack ptr, sets skb->nfct and ctinfo */ +static inline struct ip_conntrack * +resolve_normal_ct(struct sk_buff *skb, + struct ip_conntrack_protocol *proto, + int *set_reply, + unsigned int hooknum, + enum ip_conntrack_info *ctinfo) +{ + struct ip_conntrack_tuple tuple; + struct ip_conntrack_tuple_hash *h; + + IP_NF_ASSERT((skb->nh.iph->frag_off & htons(IP_OFFSET)) == 0); + + if (!get_tuple(skb->nh.iph, skb, skb->nh.iph->ihl*4, &tuple, proto)) + return NULL; + + /* look for tuple match */ + h = ip_conntrack_find_get(&tuple, NULL); + if (!h) { + h = init_conntrack(&tuple, proto, skb); + if (!h) + return NULL; + if (IS_ERR(h)) + return (void *)h; + } + + /* It exists; we have (non-exclusive) reference. */ + if (DIRECTION(h) == IP_CT_DIR_REPLY) { + *ctinfo = IP_CT_ESTABLISHED + IP_CT_IS_REPLY; + /* Please set reply bit if this packet OK */ + *set_reply = 1; + } else { + /* Once we've had two way comms, always ESTABLISHED. */ + if (test_bit(IPS_SEEN_REPLY_BIT, &h->ctrack->status)) { + DEBUGP("ip_conntrack_in: normal packet for %p\n", + h->ctrack); + *ctinfo = IP_CT_ESTABLISHED; + } else if (test_bit(IPS_EXPECTED_BIT, &h->ctrack->status)) { + DEBUGP("ip_conntrack_in: related packet for %p\n", + h->ctrack); + *ctinfo = IP_CT_RELATED; + } else { + DEBUGP("ip_conntrack_in: new packet for %p\n", + h->ctrack); + *ctinfo = IP_CT_NEW; + } + *set_reply = 0; + } + skb->nfct = &h->ctrack->infos[*ctinfo]; + return h->ctrack; +} + +/* Netfilter hook itself. */ +unsigned int ip_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct ip_conntrack *ct; + enum ip_conntrack_info ctinfo; + struct ip_conntrack_protocol *proto; + int set_reply; + int ret; + + /* Never happen */ + if ((*pskb)->nh.iph->frag_off & htons(IP_OFFSET)) { + if (net_ratelimit()) { + printk(KERN_ERR "ip_conntrack_in: Frag of proto %u (hook=%u)\n", + (*pskb)->nh.iph->protocol, hooknum); + } + return NF_DROP; + } + + /* FIXME: Do this right please. --RR */ + (*pskb)->nfcache |= NFC_UNKNOWN; + +/* Doesn't cover locally-generated broadcast, so not worth it. */ +#if 0 + /* Ignore broadcast: no `connection'. */ + if ((*pskb)->pkt_type == PACKET_BROADCAST) { + printk("Broadcast packet!\n"); + return NF_ACCEPT; + } else if (((*pskb)->nh.iph->daddr & htonl(0x000000FF)) + == htonl(0x000000FF)) { + printk("Should bcast: %u.%u.%u.%u->%u.%u.%u.%u (sk=%p, ptype=%u)\n", + NIPQUAD((*pskb)->nh.iph->saddr), + NIPQUAD((*pskb)->nh.iph->daddr), + (*pskb)->sk, (*pskb)->pkt_type); + } +#endif + + /* Previously seen (loopback or untracked)? Ignore. */ + if ((*pskb)->nfct) + return NF_ACCEPT; + + proto = ip_ct_find_proto((*pskb)->nh.iph->protocol); + + /* It may be an icmp error... */ + if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP + && icmp_error_track(*pskb, &ctinfo, hooknum)) + return NF_ACCEPT; + + if (!(ct = resolve_normal_ct(*pskb, proto,&set_reply,hooknum,&ctinfo))) + /* Not valid part of a connection */ + return NF_ACCEPT; + + if (IS_ERR(ct)) + /* Too stressed to deal. */ + return NF_DROP; + + IP_NF_ASSERT((*pskb)->nfct); + + ret = proto->packet(ct, *pskb, ctinfo); + if (ret == -1) { + /* Invalid */ + nf_conntrack_put((*pskb)->nfct); + (*pskb)->nfct = NULL; + return NF_ACCEPT; + } + + if (ret != NF_DROP && ct->helper) { + ret = ct->helper->help(*pskb, ct, ctinfo); + if (ret == -1) { + /* Invalid */ + nf_conntrack_put((*pskb)->nfct); + (*pskb)->nfct = NULL; + return NF_ACCEPT; + } + } + if (set_reply) + set_bit(IPS_SEEN_REPLY_BIT, &ct->status); + + return ret; +} + +int invert_tuplepr(struct ip_conntrack_tuple *inverse, + const struct ip_conntrack_tuple *orig) +{ + return invert_tuple(inverse, orig, ip_ct_find_proto(orig->dst.protonum)); +} + +static inline int resent_expect(const struct ip_conntrack_expect *i, + const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack_tuple *mask) +{ + DEBUGP("resent_expect\n"); + DEBUGP(" tuple: "); DUMP_TUPLE(&i->tuple); + DEBUGP("ct_tuple: "); DUMP_TUPLE(&i->ct_tuple); + DEBUGP("test tuple: "); DUMP_TUPLE(tuple); + return (((i->ct_tuple.dst.protonum == 0 && ip_ct_tuple_equal(&i->tuple, tuple)) + || (i->ct_tuple.dst.protonum && ip_ct_tuple_equal(&i->ct_tuple, tuple))) + && ip_ct_tuple_equal(&i->mask, mask)); +} + +/* Would two expected things clash? */ +static inline int expect_clash(const struct ip_conntrack_expect *i, + const struct ip_conntrack_tuple *tuple, + const struct ip_conntrack_tuple *mask) +{ + /* Part covered by intersection of masks must be unequal, + otherwise they clash */ + struct ip_conntrack_tuple intersect_mask + = { { i->mask.src.ip & mask->src.ip, + { i->mask.src.u.all & mask->src.u.all } }, + { i->mask.dst.ip & mask->dst.ip, + { i->mask.dst.u.all & mask->dst.u.all }, + i->mask.dst.protonum & mask->dst.protonum } }; + + return ip_ct_tuple_mask_cmp(&i->tuple, tuple, &intersect_mask); +} + +inline void ip_conntrack_unexpect_related(struct ip_conntrack_expect *expect) +{ + WRITE_LOCK(&ip_conntrack_lock); + unexpect_related(expect); + WRITE_UNLOCK(&ip_conntrack_lock); +} + +static void expectation_timed_out(unsigned long ul_expect) +{ + struct ip_conntrack_expect *expect = (void *) ul_expect; + + DEBUGP("expectation %p timed out\n", expect); + WRITE_LOCK(&ip_conntrack_lock); + __unexpect_related(expect); + WRITE_UNLOCK(&ip_conntrack_lock); +} + +struct ip_conntrack_expect * +ip_conntrack_expect_alloc() +{ + struct ip_conntrack_expect *new; + + new = (struct ip_conntrack_expect *) + kmalloc(sizeof(struct ip_conntrack_expect), GFP_ATOMIC); + if (!new) { + DEBUGP("expect_related: OOM allocating expect\n"); + return NULL; + } + + /* tuple_cmp compares whole union, we have to initialized cleanly */ + memset(new, 0, sizeof(struct ip_conntrack_expect)); + + return new; +} + +static void +ip_conntrack_expect_insert(struct ip_conntrack_expect *new, + struct ip_conntrack *related_to) +{ + DEBUGP("new expectation %p of conntrack %p\n", new, related_to); + new->expectant = related_to; + new->sibling = NULL; + atomic_set(&new->use, 1); + + /* add to expected list for this connection */ + list_add(&new->expected_list, &related_to->sibling_list); + /* add to global list of expectations */ + + list_prepend(&ip_conntrack_expect_list, &new->list); + /* add and start timer if required */ + if (related_to->helper->timeout) { + init_timer(&new->timeout); + new->timeout.data = (unsigned long)new; + new->timeout.function = expectation_timed_out; + new->timeout.expires = jiffies + + related_to->helper->timeout * HZ; + add_timer(&new->timeout); + } + related_to->expecting++; +} + +/* Add a related connection. */ +int ip_conntrack_expect_related(struct ip_conntrack_expect *expect, + struct ip_conntrack *related_to) +{ + struct ip_conntrack_expect *old; + int ret = 0; + + WRITE_LOCK(&ip_conntrack_lock); + /* Because of the write lock, no reader can walk the lists, + * so there is no need to use the tuple lock too */ + + DEBUGP("ip_conntrack_expect_related %p\n", related_to); + DEBUGP("tuple: "); DUMP_TUPLE(&expect->tuple); + DEBUGP("mask: "); DUMP_TUPLE(&expect->mask); + + old = LIST_FIND(&ip_conntrack_expect_list, resent_expect, + struct ip_conntrack_expect *, &expect->tuple, + &expect->mask); + if (old) { + /* Helper private data may contain offsets but no pointers + pointing into the payload - otherwise we should have to copy + the data filled out by the helper over the old one */ + DEBUGP("expect_related: resent packet\n"); + if (related_to->helper->timeout) { + if (!del_timer(&old->timeout)) { + /* expectation is dying. Fall through */ + goto out; + } else { + old->timeout.expires = jiffies + + related_to->helper->timeout * HZ; + add_timer(&old->timeout); + } + } + + WRITE_UNLOCK(&ip_conntrack_lock); + kfree(expect); + return -EEXIST; + + } else if (related_to->helper->max_expected && + related_to->expecting >= related_to->helper->max_expected) { + struct list_head *cur_item; + /* old == NULL */ + if (!(related_to->helper->flags & + IP_CT_HELPER_F_REUSE_EXPECT)) { + WRITE_UNLOCK(&ip_conntrack_lock); + if (net_ratelimit()) + printk(KERN_WARNING + "ip_conntrack: max number of expected " + "connections %i of %s reached for " + "%u.%u.%u.%u->%u.%u.%u.%u\n", + related_to->helper->max_expected, + related_to->helper->name, + NIPQUAD(related_to->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip), + NIPQUAD(related_to->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip)); + kfree(expect); + return -EPERM; + } + DEBUGP("ip_conntrack: max number of expected " + "connections %i of %s reached for " + "%u.%u.%u.%u->%u.%u.%u.%u, reusing\n", + related_to->helper->max_expected, + related_to->helper->name, + NIPQUAD(related_to->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip), + NIPQUAD(related_to->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip)); + + /* choose the the oldest expectation to evict */ + list_for_each(cur_item, &related_to->sibling_list) { + struct ip_conntrack_expect *cur; + + cur = list_entry(cur_item, + struct ip_conntrack_expect, + expected_list); + if (cur->sibling == NULL) { + old = cur; + break; + } + } + + /* (!old) cannot happen, since related_to->expecting is the + * number of unconfirmed expects */ + IP_NF_ASSERT(old); + + /* newnat14 does not reuse the real allocated memory + * structures but rather unexpects the old and + * allocates a new. unexpect_related will decrement + * related_to->expecting. + */ + unexpect_related(old); + ret = -EPERM; + } else if (LIST_FIND(&ip_conntrack_expect_list, expect_clash, + struct ip_conntrack_expect *, &expect->tuple, + &expect->mask)) { + WRITE_UNLOCK(&ip_conntrack_lock); + DEBUGP("expect_related: busy!\n"); + + kfree(expect); + return -EBUSY; + } + +out: ip_conntrack_expect_insert(expect, related_to); + + WRITE_UNLOCK(&ip_conntrack_lock); + + return ret; +} + +/* Change tuple in an existing expectation */ +int ip_conntrack_change_expect(struct ip_conntrack_expect *expect, + struct ip_conntrack_tuple *newtuple) +{ + int ret; + + MUST_BE_READ_LOCKED(&ip_conntrack_lock); + WRITE_LOCK(&ip_conntrack_expect_tuple_lock); + + DEBUGP("change_expect:\n"); + DEBUGP("exp tuple: "); DUMP_TUPLE(&expect->tuple); + DEBUGP("exp mask: "); DUMP_TUPLE(&expect->mask); + DEBUGP("newtuple: "); DUMP_TUPLE(newtuple); + if (expect->ct_tuple.dst.protonum == 0) { + /* Never seen before */ + DEBUGP("change expect: never seen before\n"); + if (!ip_ct_tuple_equal(&expect->tuple, newtuple) + && LIST_FIND(&ip_conntrack_expect_list, expect_clash, + struct ip_conntrack_expect *, newtuple, &expect->mask)) { + /* Force NAT to find an unused tuple */ + ret = -1; + } else { + memcpy(&expect->ct_tuple, &expect->tuple, sizeof(expect->tuple)); + memcpy(&expect->tuple, newtuple, sizeof(expect->tuple)); + ret = 0; + } + } else { + /* Resent packet */ + DEBUGP("change expect: resent packet\n"); + if (ip_ct_tuple_equal(&expect->tuple, newtuple)) { + ret = 0; + } else { + /* Force NAT to choose again the same port */ + ret = -1; + } + } + WRITE_UNLOCK(&ip_conntrack_expect_tuple_lock); + + return ret; +} + +/* Alter reply tuple (maybe alter helper). If it's already taken, + return 0 and don't do alteration. */ +int ip_conntrack_alter_reply(struct ip_conntrack *conntrack, + const struct ip_conntrack_tuple *newreply) +{ + WRITE_LOCK(&ip_conntrack_lock); + if (__ip_conntrack_find(newreply, conntrack)) { + WRITE_UNLOCK(&ip_conntrack_lock); + return 0; + } + /* Should be unconfirmed, so not in hash table yet */ + IP_NF_ASSERT(!is_confirmed(conntrack)); + + DEBUGP("Altering reply tuple of %p to ", conntrack); + DUMP_TUPLE(newreply); + + conntrack->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; + if (!conntrack->master) + conntrack->helper = LIST_FIND(&helpers, helper_cmp, + struct ip_conntrack_helper *, + newreply); + WRITE_UNLOCK(&ip_conntrack_lock); + + return 1; +} + +int ip_conntrack_helper_register(struct ip_conntrack_helper *me) +{ + WRITE_LOCK(&ip_conntrack_lock); + list_prepend(&helpers, me); + WRITE_UNLOCK(&ip_conntrack_lock); + + return 0; +} + +static inline int unhelp(struct ip_conntrack_tuple_hash *i, + const struct ip_conntrack_helper *me) +{ + if (i->ctrack->helper == me) { + /* Get rid of any expected. */ + remove_expectations(i->ctrack, 0); + /* And *then* set helper to NULL */ + i->ctrack->helper = NULL; + } + return 0; +} + +void ip_conntrack_helper_unregister(struct ip_conntrack_helper *me) +{ + unsigned int i; + + /* Need write lock here, to delete helper. */ + WRITE_LOCK(&ip_conntrack_lock); + LIST_DELETE(&helpers, me); + + /* Get rid of expecteds, set helpers to NULL. */ + for (i = 0; i < ip_conntrack_htable_size; i++) + LIST_FIND_W(&ip_conntrack_hash[i], unhelp, + struct ip_conntrack_tuple_hash *, me); + WRITE_UNLOCK(&ip_conntrack_lock); + + /* Someone could be still looking at the helper in a bh. */ + synchronize_net(); +} + +/* Refresh conntrack for this many jiffies. */ +void ip_ct_refresh(struct ip_conntrack *ct, unsigned long extra_jiffies) +{ + IP_NF_ASSERT(ct->timeout.data == (unsigned long)ct); + + /* If not in hash table, timer will not be active yet */ + if (!is_confirmed(ct)) + ct->timeout.expires = extra_jiffies; + else { + WRITE_LOCK(&ip_conntrack_lock); + /* Need del_timer for race avoidance (may already be dying). */ + if (del_timer(&ct->timeout)) { + ct->timeout.expires = jiffies + extra_jiffies; + add_timer(&ct->timeout); + } + WRITE_UNLOCK(&ip_conntrack_lock); + } +} + +/* Returns new sk_buff, or NULL */ +struct sk_buff * +ip_ct_gather_frags(struct sk_buff *skb) +{ + struct sock *sk = skb->sk; +#ifdef CONFIG_NETFILTER_DEBUG + unsigned int olddebug = skb->nf_debug; +#endif + if (sk) { + sock_hold(sk); + skb_orphan(skb); + } + + local_bh_disable(); + skb = ip_defrag(skb); + local_bh_enable(); + + if (!skb) { + if (sk) + sock_put(sk); + return skb; + } + + if (sk) { + skb_set_owner_w(skb, sk); + sock_put(sk); + } + + ip_send_check(skb->nh.iph); + skb->nfcache |= NFC_ALTERED; +#ifdef CONFIG_NETFILTER_DEBUG + /* Packet path as if nothing had happened. */ + skb->nf_debug = olddebug; +#endif + return skb; +} + +/* Used by ipt_REJECT. */ +static void ip_conntrack_attach(struct sk_buff *nskb, struct nf_ct_info *nfct) +{ + struct ip_conntrack *ct; + enum ip_conntrack_info ctinfo; + + ct = __ip_conntrack_get(nfct, &ctinfo); + + /* This ICMP is in reverse direction to the packet which + caused it */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_ORIGINAL) + ctinfo = IP_CT_RELATED + IP_CT_IS_REPLY; + else + ctinfo = IP_CT_RELATED; + + /* Attach new skbuff, and increment count */ + nskb->nfct = &ct->infos[ctinfo]; + atomic_inc(&ct->ct_general.use); +} + +static inline int +do_kill(const struct ip_conntrack_tuple_hash *i, + int (*kill)(const struct ip_conntrack *i, void *data), + void *data) +{ + return kill(i->ctrack, data); +} + +/* Bring out ya dead! */ +static struct ip_conntrack_tuple_hash * +get_next_corpse(int (*kill)(const struct ip_conntrack *i, void *data), + void *data, unsigned int *bucket) +{ + struct ip_conntrack_tuple_hash *h = NULL; + + READ_LOCK(&ip_conntrack_lock); + for (; !h && *bucket < ip_conntrack_htable_size; (*bucket)++) { + h = LIST_FIND(&ip_conntrack_hash[*bucket], do_kill, + struct ip_conntrack_tuple_hash *, kill, data); + } + if (h) + atomic_inc(&h->ctrack->ct_general.use); + READ_UNLOCK(&ip_conntrack_lock); + + return h; +} + +void +ip_ct_selective_cleanup(int (*kill)(const struct ip_conntrack *i, void *data), + void *data) +{ + struct ip_conntrack_tuple_hash *h; + unsigned int bucket = 0; + + while ((h = get_next_corpse(kill, data, &bucket)) != NULL) { + /* Time to push up daises... */ + if (del_timer(&h->ctrack->timeout)) + death_by_timeout((unsigned long)h->ctrack); + /* ... else the timer will get him soon. */ + + ip_conntrack_put(h->ctrack); + } +} + +/* Fast function for those who don't want to parse /proc (and I don't + blame them). */ +/* Reversing the socket's dst/src point of view gives us the reply + mapping. */ +static int +getorigdst(struct sock *sk, int optval, void *user, int *len) +{ + struct inet_opt *inet = inet_sk(sk); + struct ip_conntrack_tuple_hash *h; + struct ip_conntrack_tuple tuple; + + IP_CT_TUPLE_U_BLANK(&tuple); + tuple.src.ip = inet->rcv_saddr; + tuple.src.u.tcp.port = inet->sport; + tuple.dst.ip = inet->daddr; + tuple.dst.u.tcp.port = inet->dport; + tuple.dst.protonum = IPPROTO_TCP; + + /* We only do TCP at the moment: is there a better way? */ + if (strcmp(sk->sk_prot->name, "TCP")) { + DEBUGP("SO_ORIGINAL_DST: Not a TCP socket\n"); + return -ENOPROTOOPT; + } + + if ((unsigned int) *len < sizeof(struct sockaddr_in)) { + DEBUGP("SO_ORIGINAL_DST: len %u not %u\n", + *len, sizeof(struct sockaddr_in)); + return -EINVAL; + } + + h = ip_conntrack_find_get(&tuple, NULL); + if (h) { + struct sockaddr_in sin; + + sin.sin_family = AF_INET; + sin.sin_port = h->ctrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.u.tcp.port; + sin.sin_addr.s_addr = h->ctrack->tuplehash[IP_CT_DIR_ORIGINAL] + .tuple.dst.ip; + + DEBUGP("SO_ORIGINAL_DST: %u.%u.%u.%u %u\n", + NIPQUAD(sin.sin_addr.s_addr), ntohs(sin.sin_port)); + ip_conntrack_put(h->ctrack); + if (copy_to_user(user, &sin, sizeof(sin)) != 0) + return -EFAULT; + else + return 0; + } + DEBUGP("SO_ORIGINAL_DST: Can't find %u.%u.%u.%u/%u-%u.%u.%u.%u/%u.\n", + NIPQUAD(tuple.src.ip), ntohs(tuple.src.u.tcp.port), + NIPQUAD(tuple.dst.ip), ntohs(tuple.dst.u.tcp.port)); + return -ENOENT; +} + +static struct nf_sockopt_ops so_getorigdst = { + .pf = PF_INET, + .get_optmin = SO_ORIGINAL_DST, + .get_optmax = SO_ORIGINAL_DST+1, + .get = &getorigdst, +}; + +static int kill_all(const struct ip_conntrack *i, void *data) +{ + return 1; +} + +/* Mishearing the voices in his head, our hero wonders how he's + supposed to kill the mall. */ +void ip_conntrack_cleanup(void) +{ + ip_ct_attach = NULL; + /* This makes sure all current packets have passed through + netfilter framework. Roll on, two-stage module + delete... */ + synchronize_net(); + + i_see_dead_people: + ip_ct_selective_cleanup(kill_all, NULL); + if (atomic_read(&ip_conntrack_count) != 0) { + schedule(); + goto i_see_dead_people; + } + + kmem_cache_destroy(ip_conntrack_cachep); + vfree(ip_conntrack_hash); + nf_unregister_sockopt(&so_getorigdst); +} + +static int hashsize; +MODULE_PARM(hashsize, "i"); + +int __init ip_conntrack_init(void) +{ + unsigned int i; + int ret; + + /* Idea from tcp.c: use 1/16384 of memory. On i386: 32MB + * machine has 256 buckets. >= 1GB machines have 8192 buckets. */ + if (hashsize) { + ip_conntrack_htable_size = hashsize; + } else { + ip_conntrack_htable_size + = (((num_physpages << PAGE_SHIFT) / 16384) + / sizeof(struct list_head)); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + ip_conntrack_htable_size = 8192; + if (ip_conntrack_htable_size < 16) + ip_conntrack_htable_size = 16; + } + ip_conntrack_max = 8 * ip_conntrack_htable_size; + + printk("ip_conntrack version %s (%u buckets, %d max)" + " - %Zd bytes per conntrack\n", IP_CONNTRACK_VERSION, + ip_conntrack_htable_size, ip_conntrack_max, + sizeof(struct ip_conntrack)); + + ret = nf_register_sockopt(&so_getorigdst); + if (ret != 0) { + printk(KERN_ERR "Unable to register netfilter socket option\n"); + return ret; + } + + ip_conntrack_hash = vmalloc(sizeof(struct list_head) + * ip_conntrack_htable_size); + if (!ip_conntrack_hash) { + printk(KERN_ERR "Unable to create ip_conntrack_hash\n"); + goto err_unreg_sockopt; + } + + ip_conntrack_cachep = kmem_cache_create("ip_conntrack", + sizeof(struct ip_conntrack), 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + if (!ip_conntrack_cachep) { + printk(KERN_ERR "Unable to create ip_conntrack slab cache\n"); + goto err_free_hash; + } + /* Don't NEED lock here, but good form anyway. */ + WRITE_LOCK(&ip_conntrack_lock); + /* Sew in builtin protocols. */ + list_append(&protocol_list, &ip_conntrack_protocol_tcp); + list_append(&protocol_list, &ip_conntrack_protocol_udp); + list_append(&protocol_list, &ip_conntrack_protocol_icmp); + WRITE_UNLOCK(&ip_conntrack_lock); + + for (i = 0; i < ip_conntrack_htable_size; i++) + INIT_LIST_HEAD(&ip_conntrack_hash[i]); + + /* For use by ipt_REJECT */ + ip_ct_attach = ip_conntrack_attach; + + /* Set up fake conntrack: + - to never be deleted, not in any hashes */ + atomic_set(&ip_conntrack_untracked.ct_general.use, 1); + /* - and look it like as a confirmed connection */ + set_bit(IPS_CONFIRMED_BIT, &ip_conntrack_untracked.status); + /* - and prepare the ctinfo field for REJECT & NAT. */ + ip_conntrack_untracked.infos[IP_CT_NEW].master = + ip_conntrack_untracked.infos[IP_CT_RELATED].master = + ip_conntrack_untracked.infos[IP_CT_RELATED + IP_CT_IS_REPLY].master = + &ip_conntrack_untracked.ct_general; + + return ret; + +err_free_hash: + vfree(ip_conntrack_hash); +err_unreg_sockopt: + nf_unregister_sockopt(&so_getorigdst); + + return -ENOMEM; +} diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_pool.c linux-2.6.6-rc1/net/ipv4/netfilter/ip_pool.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ip_pool.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ip_pool.c 2004-04-19 10:08:36.000000000 +0200 @@ -0,0 +1,334 @@ +/* Kernel module for IP pool management */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DP printk +#else +#define DP(format, args...) +#endif + +MODULE_LICENSE("GPL"); + +#define NR_POOL 16 +static int nr_pool = NR_POOL;/* overwrite this when loading module */ + +struct ip_pool { + u_int32_t first_ip; /* host byte order, included in range */ + u_int32_t last_ip; /* host byte order, included in range */ + void *members; /* the bitmap proper */ + int nr_use; /* total nr. of tests through this */ + int nr_match; /* total nr. of matches through this */ + rwlock_t lock; +}; + +static struct ip_pool *POOL; + +static inline struct ip_pool *lookup(ip_pool_t index) +{ + if (index < 0 || index >= nr_pool) { + DP("ip_pool:lookup: bad index %d\n", index); + return 0; + } + return POOL+index; +} + +int ip_pool_match(ip_pool_t index, u_int32_t addr) +{ + struct ip_pool *pool = lookup(index); + int res = 0; + + if (!pool || !pool->members) + return 0; + read_lock_bh(&pool->lock); + if (pool->members) { + if (addr >= pool->first_ip && addr <= pool->last_ip) { + addr -= pool->first_ip; + if (test_bit(addr, pool->members)) { + res = 1; +#ifdef CONFIG_IP_POOL_STATISTICS + pool->nr_match++; +#endif + } + } +#ifdef CONFIG_IP_POOL_STATISTICS + pool->nr_use++; +#endif + } + read_unlock_bh(&pool->lock); + return res; +} +EXPORT_SYMBOL(ip_pool_match); + +static int pool_change(ip_pool_t index, u_int32_t addr, int isdel) +{ + struct ip_pool *pool; + int res = -1; + + pool = lookup(index); + if ( !pool || !pool->members + || addr < pool->first_ip || addr > pool->last_ip) + return -1; + read_lock_bh(&pool->lock); + if (pool->members && addr >= pool->first_ip && addr <= pool->last_ip) { + addr -= pool->first_ip; + res = isdel + ? (0 != test_and_clear_bit(addr, pool->members)) + : (0 != test_and_set_bit(addr, pool->members)); + } + read_unlock_bh(&pool->lock); + return res; +} + +int ip_pool_mod(ip_pool_t index, u_int32_t addr, int isdel) +{ + int res = pool_change(index,addr,isdel); + + if (!isdel) res = !res; + return res; +} +EXPORT_SYMBOL(ip_pool_mod); + +static inline int bitmap_bytes(u_int32_t a, u_int32_t b) +{ + return 4*((((b-a+8)/8)+3)/4); +} + +static inline int poolbytes(ip_pool_t index) +{ + struct ip_pool *pool = lookup(index); + + return pool ? bitmap_bytes(pool->first_ip, pool->last_ip) : 0; +} + +static int setpool( + struct sock *sk, + int optval, + void *user, + unsigned int len +) { + struct ip_pool_request req; + + DP("ip_pool:setpool: optval=%d, user=%p, len=%d\n", optval, user, len); + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + if (optval != SO_IP_POOL) + return -EBADF; + if (len != sizeof(req)) + return -EINVAL; + if (copy_from_user(&req, user, sizeof(req)) != 0) + return -EFAULT; + printk("obsolete op - upgrade your ippool(8) utility.\n"); + return -EINVAL; +} + +static int getpool( + struct sock *sk, + int optval, + void *user, + int *len +) { + struct ip_pool_request req; + struct ip_pool *pool; + ip_pool_t i; + int newbytes; + void *newmembers; + int res; + + DP("ip_pool:getpool: optval=%d, user=%p\n", optval, user); + if (!capable(CAP_NET_ADMIN)) + return -EINVAL; + if (optval != SO_IP_POOL) + return -EINVAL; + if (*len != sizeof(req)) { + return -EFAULT; + } + if (copy_from_user(&req, user, sizeof(req)) != 0) + return -EFAULT; + DP("ip_pool:getpool op=%d, index=%d\n", req.op, req.index); + if (req.op < IP_POOL_BAD001) { + printk("obsolete op - upgrade your ippool(8) utility.\n"); + return -EFAULT; + } + switch(req.op) { + case IP_POOL_HIGH_NR: + DP("ip_pool HIGH_NR\n"); + req.index = IP_POOL_NONE; + for (i=0; imembers) + return -EBADF; + req.addr = htonl(pool->first_ip); + req.addr2 = htonl(pool->last_ip); + return copy_to_user(user, &req, sizeof(req)); + case IP_POOL_USAGE: + DP("ip_pool USE\n"); + pool = lookup(req.index); + if (!pool) + return -EINVAL; + if (!pool->members) + return -EBADF; + req.addr = pool->nr_use; + req.addr2 = pool->nr_match; + return copy_to_user(user, &req, sizeof(req)); + case IP_POOL_TEST_ADDR: + DP("ip_pool TEST 0x%08x\n", req.addr); + pool = lookup(req.index); + if (!pool) + return -EINVAL; + res = 0; + read_lock_bh(&pool->lock); + if (!pool->members) { + DP("ip_pool TEST_ADDR no members in pool\n"); + res = -EBADF; + goto unlock_and_return_res; + } + req.addr = ntohl(req.addr); + if (req.addr < pool->first_ip) { + DP("ip_pool TEST_ADDR address < pool bounds\n"); + res = -ERANGE; + goto unlock_and_return_res; + } + if (req.addr > pool->last_ip) { + DP("ip_pool TEST_ADDR address > pool bounds\n"); + res = -ERANGE; + goto unlock_and_return_res; + } + req.addr = (0 != test_bit((req.addr - pool->first_ip), + pool->members)); + read_unlock_bh(&pool->lock); + return copy_to_user(user, &req, sizeof(req)); + case IP_POOL_FLUSH: + DP("ip_pool FLUSH not yet implemented.\n"); + return -EBUSY; + case IP_POOL_DESTROY: + DP("ip_pool DESTROY not yet implemented.\n"); + return -EBUSY; + case IP_POOL_INIT: + DP("ip_pool INIT 0x%08x-0x%08x\n", req.addr, req.addr2); + pool = lookup(req.index); + if (!pool) + return -EINVAL; + req.addr = ntohl(req.addr); + req.addr2 = ntohl(req.addr2); + if (req.addr > req.addr2) { + DP("ip_pool INIT bad ip range\n"); + return -EINVAL; + } + newbytes = bitmap_bytes(req.addr, req.addr2); + newmembers = kmalloc(newbytes, GFP_KERNEL); + if (!newmembers) { + DP("ip_pool INIT out of mem for %d bytes\n", newbytes); + return -ENOMEM; + } + memset(newmembers, 0, newbytes); + write_lock_bh(&pool->lock); + if (pool->members) { + DP("ip_pool INIT pool %d exists\n", req.index); + kfree(newmembers); + res = -EBUSY; + goto unlock_and_return_res; + } + pool->first_ip = req.addr; + pool->last_ip = req.addr2; + pool->nr_use = 0; + pool->nr_match = 0; + pool->members = newmembers; + write_unlock_bh(&pool->lock); + return 0; + case IP_POOL_ADD_ADDR: + DP("ip_pool ADD_ADDR 0x%08x\n", req.addr); + req.addr = pool_change(req.index, ntohl(req.addr), 0); + return copy_to_user(user, &req, sizeof(req)); + case IP_POOL_DEL_ADDR: + DP("ip_pool DEL_ADDR 0x%08x\n", req.addr); + req.addr = pool_change(req.index, ntohl(req.addr), 1); + return copy_to_user(user, &req, sizeof(req)); + default: + DP("ip_pool:getpool bad op %d\n", req.op); + return -EINVAL; + } + return -EINVAL; + +unlock_and_return_res: + if (pool) + read_unlock_bh(&pool->lock); + return res; +} + +static struct nf_sockopt_ops so_pool += { { NULL, NULL }, PF_INET, + SO_IP_POOL, SO_IP_POOL+1, &setpool, + SO_IP_POOL, SO_IP_POOL+1, &getpool, + 0, NULL }; + +MODULE_PARM(nr_pool, "i"); + +static int __init init(void) +{ + ip_pool_t i; + int res; + + if (nr_pool < 1) { + printk("ip_pool module init: bad nr_pool %d\n", nr_pool); + return -EINVAL; + } + POOL = kmalloc(nr_pool * sizeof(*POOL), GFP_KERNEL); + if (!POOL) { + printk("ip_pool module init: out of memory for nr_pool %d\n", + nr_pool); + return -ENOMEM; + } + for (i=0; i= start_offset) { unsigned int namelen; @@ -1752,6 +1752,15 @@ return pos; } +static inline int print_target(const struct ipt_target *t, + off_t start_offset, char *buffer, int length, + off_t *pos, unsigned int *count) +{ + if (t != &ipt_standard_target && t != &ipt_error_target) + return 0; + return print_name((char *)t, start_offset, buffer, length, pos, count); +} + static int ipt_get_targets(char *buffer, char **start, off_t offset, int length) { off_t pos = 0; @@ -1760,7 +1769,7 @@ if (down_interruptible(&ipt_mutex) != 0) return 0; - LIST_FIND(&ipt_target, print_name, void *, + LIST_FIND(&ipt_target, print_target, struct ipt_target *, offset, buffer, length, &pos, &count); up(&ipt_mutex); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_IPV4OPTSSTRIP.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_IPV4OPTSSTRIP.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_IPV4OPTSSTRIP.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_IPV4OPTSSTRIP.c 2004-04-19 10:08:26.000000000 +0200 @@ -0,0 +1,89 @@ +/** + * Strip all IP options in the IP packet header. + * + * (C) 2001 by Fabrice MARIE + * This software is distributed under GNU GPL v2, 1991 + */ + +#include +#include +#include +#include + +#include + +MODULE_AUTHOR("Fabrice MARIE "); +MODULE_DESCRIPTION("Strip all options in IPv4 packets"); +MODULE_LICENSE("GPL"); + +static unsigned int +target(struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + unsigned int hooknum, + const void *targinfo, + void *userinfo) +{ + struct iphdr *iph; + struct sk_buff *skb; + struct ip_options *opt; + unsigned char *optiph; + int l; + + if (!skb_ip_make_writable(pskb, (*pskb)->len)) + return NF_DROP; + + skb = (*pskb); + iph = (*pskb)->nh.iph; + optiph = skb->nh.raw; + l = ((struct ip_options *)(&(IPCB(skb)->opt)))->optlen; + + /* if no options in packet then nothing to clear. */ + if (iph->ihl * 4 == sizeof(struct iphdr)) + return IPT_CONTINUE; + + /* else clear all options */ + memset(&(IPCB(skb)->opt), 0, sizeof(struct ip_options)); + memset(optiph+sizeof(struct iphdr), IPOPT_NOOP, l); + opt = &(IPCB(skb)->opt); + opt->is_data = 0; + opt->optlen = l; + + skb->nfcache |= NFC_ALTERED; + + return IPT_CONTINUE; +} + +static int +checkentry(const char *tablename, + const struct ipt_entry *e, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + if (strcmp(tablename, "mangle")) { + printk(KERN_WARNING "IPV4OPTSSTRIP: can only be called from \"mangle\" table, not \"%s\"\n", tablename); + return 0; + } + /* nothing else to check because no parameters */ + return 1; +} + +static struct ipt_target ipt_ipv4optsstrip_reg = { + .name = "IPV4OPTSSTRIP", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE }; + +static int __init init(void) +{ + return ipt_register_target(&ipt_ipv4optsstrip_reg); +} + +static void __exit fini(void) +{ + ipt_unregister_target(&ipt_ipv4optsstrip_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_POOL.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_POOL.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_POOL.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_POOL.c 2004-04-19 10:08:36.000000000 +0200 @@ -0,0 +1,116 @@ +/* ipt_POOL.c - netfilter target to manipulate IP pools + * + * This target can be used almost everywhere. It acts on some specified + * IP pool, adding or deleting some IP address in the pool. The address + * can be either the source (--addsrc, --delsrc), or destination (--add/deldst) + * of the packet under inspection. + * + * The target normally returns IPT_CONTINUE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +/*** NOTE NOTE NOTE NOTE *** +** +** By sheer luck, I get away with using the "struct ipt_pool_info", as defined +** in , both as the match and target info. +** Here, in the target implementation, ipt_pool_info.src, if not IP_POOL_NONE, +** is modified for the source IP address of the packet under inspection. +** The same way, the ipt_pool_info.dst pool is modified for the destination. +** +** The address is added to the pool normally. However, if IPT_POOL_DEL_dir +** flag is set in ipt_pool_info.flags, the address is deleted from the pool. +** +** If a modification was done to the pool, we possibly return ACCEPT or DROP, +** if the right IPT_POOL_MOD_dir_ACCEPT or _MOD_dir_DROP flags are set. +** The IPT_POOL_INV_MOD_dir flag inverts the sense of the check (i.e. the +** ACCEPT and DROP flags are evaluated when the pool was not modified.) +*/ + +static int +do_check(const char *tablename, + const struct ipt_entry *e, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + const struct ipt_pool_info *ipi = targinfo; + + if (targinfosize != IPT_ALIGN(sizeof(*ipi))) { + DEBUGP("POOL_check: size %u.\n", targinfosize); + return 0; + } + DEBUGP("ipt_POOL:do_check(%d,%d,%d)\n",ipi->src,ipi->dst,ipi->flags); + return 1; +} + +static unsigned int +do_target(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + const void *targinfo, + void *userinfo) +{ + const struct ipt_pool_info *ipi = targinfo; + int modified; + unsigned int verdict = IPT_CONTINUE; + + if (ipi->src != IP_POOL_NONE) { + modified = ip_pool_mod(ipi->src, ntohl((*pskb)->nh.iph->saddr), + ipi->flags & IPT_POOL_DEL_SRC); + if (!!modified ^ !!(ipi->flags & IPT_POOL_INV_MOD_SRC)) { + if (ipi->flags & IPT_POOL_MOD_SRC_ACCEPT) + verdict = NF_ACCEPT; + else if (ipi->flags & IPT_POOL_MOD_SRC_DROP) + verdict = NF_DROP; + } + } + if (verdict == IPT_CONTINUE && ipi->dst != IP_POOL_NONE) { + modified = ip_pool_mod(ipi->dst, ntohl((*pskb)->nh.iph->daddr), + ipi->flags & IPT_POOL_DEL_DST); + if (!!modified ^ !!(ipi->flags & IPT_POOL_INV_MOD_DST)) { + if (ipi->flags & IPT_POOL_MOD_DST_ACCEPT) + verdict = NF_ACCEPT; + else if (ipi->flags & IPT_POOL_MOD_DST_DROP) + verdict = NF_DROP; + } + } + return verdict; +} + +static struct ipt_target pool_reg += { { NULL, NULL }, "POOL", do_target, do_check, NULL, THIS_MODULE }; + +static int __init init(void) +{ + DEBUGP("init ipt_POOL\n"); + return ipt_register_target(&pool_reg); +} + +static void __exit fini(void) +{ + DEBUGP("fini ipt_POOL\n"); + ipt_unregister_target(&pool_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_TTL.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_TTL.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_TTL.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_TTL.c 2004-04-19 10:08:28.000000000 +0200 @@ -0,0 +1,120 @@ +/* TTL modification target for IP tables + * (C) 2000 by Harald Welte + * + * Version: $Revision$ + * + * This software is distributed under the terms of GNU GPL + */ + +#include +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("IP tables TTL modification module"); +MODULE_LICENSE("GPL"); + +static unsigned int +ipt_ttl_target(struct sk_buff **pskb, const struct net_device *in, + const struct net_device *out, unsigned int hooknum, + const void *targinfo, void *userinfo) +{ + struct iphdr *iph; + const struct ipt_TTL_info *info = targinfo; + u_int16_t diffs[2]; + int new_ttl; + + if (!skb_ip_make_writable(pskb, (*pskb)->len)) + return NF_DROP; + + iph = (*pskb)->nh.iph; + + switch (info->mode) { + case IPT_TTL_SET: + new_ttl = info->ttl; + break; + case IPT_TTL_INC: + new_ttl = iph->ttl + info->ttl; + if (new_ttl > 255) + new_ttl = 255; + break; + case IPT_TTL_DEC: + new_ttl = iph->ttl + info->ttl; + if (new_ttl < 0) + new_ttl = 0; + break; + default: + new_ttl = iph->ttl; + break; + } + + if (new_ttl != iph->ttl) { + diffs[0] = htons(((unsigned)iph->ttl) << 8) ^ 0xFFFF; + iph->ttl = new_ttl; + diffs[1] = htons(((unsigned)iph->ttl) << 8); + iph->check = csum_fold(csum_partial((char *)diffs, + sizeof(diffs), + iph->check^0xFFFF)); + (*pskb)->nfcache |= NFC_ALTERED; + } + + return IPT_CONTINUE; +} + +static int ipt_ttl_checkentry(const char *tablename, + const struct ipt_entry *e, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + struct ipt_TTL_info *info = targinfo; + + if (targinfosize != IPT_ALIGN(sizeof(struct ipt_TTL_info))) { + printk(KERN_WARNING "TTL: targinfosize %u != %Zu\n", + targinfosize, + IPT_ALIGN(sizeof(struct ipt_TTL_info))); + return 0; + } + + if (strcmp(tablename, "mangle")) { + printk(KERN_WARNING "TTL: can only be called from \"mangle\" table, not \"%s\"\n", tablename); + return 0; + } + + if (info->mode > IPT_TTL_MAXMODE) { + printk(KERN_WARNING "TTL: invalid or unknown Mode %u\n", + info->mode); + return 0; + } + + if ((info->mode != IPT_TTL_SET) && (info->ttl == 0)) { + printk(KERN_WARNING "TTL: increment/decrement doesn't make sense with value 0\n"); + return 0; + } + + return 1; +} + +static struct ipt_target ipt_TTL = { + .name = "TTL", + .target = ipt_ttl_target, + .checkentry = ipt_ttl_checkentry, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_target(&ipt_TTL); +} + +static void __exit fini(void) +{ + ipt_unregister_target(&ipt_TTL); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_connlimit.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_connlimit.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_connlimit.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_connlimit.c 2004-04-19 10:08:29.000000000 +0200 @@ -0,0 +1,230 @@ +/* + * netfilter module to limit the number of parallel tcp + * connections per IP address. + * (c) 2000 Gerd Knorr + * Nov 2002: Martin Bene : + * only ignore TIME_WAIT or gone connections + * + * based on ... + * + * Kernel module to match connection tracking information. + * GPL (C) 1999 Rusty Russell (rusty@rustcorp.com.au). + */ +#include +#include +#include +#include +#include +#include +#include +#include + +#define DEBUG 0 + +MODULE_LICENSE("GPL"); + +/* we'll save the tuples of all connections we care about */ +struct ipt_connlimit_conn +{ + struct list_head list; + struct ip_conntrack_tuple tuple; +}; + +struct ipt_connlimit_data { + spinlock_t lock; + struct list_head iphash[256]; +}; + +static int ipt_iphash(u_int32_t addr) +{ + int hash; + + hash = addr & 0xff; + hash ^= (addr >> 8) & 0xff; + hash ^= (addr >> 16) & 0xff; + hash ^= (addr >> 24) & 0xff; + return hash; +} + +static int count_them(struct ipt_connlimit_data *data, + u_int32_t addr, u_int32_t mask, + struct ip_conntrack *ct) +{ +#if DEBUG + const static char *tcp[] = { "none", "established", "syn_sent", "syn_recv", + "fin_wait", "time_wait", "close", "close_wait", + "last_ack", "listen" }; +#endif + int addit = 1, matches = 0; + struct ip_conntrack_tuple tuple; + struct ip_conntrack_tuple_hash *found; + struct ipt_connlimit_conn *conn; + struct list_head *hash,*lh; + + spin_lock(&data->lock); + tuple = ct->tuplehash[0].tuple; + hash = &data->iphash[ipt_iphash(addr & mask)]; + + /* check the saved connections */ + for (lh = hash->next; lh != hash; lh = lh->next) { + conn = list_entry(lh,struct ipt_connlimit_conn,list); + found = ip_conntrack_find_get(&conn->tuple,ct); + if (0 == memcmp(&conn->tuple,&tuple,sizeof(tuple)) && + found != NULL && + found->ctrack->proto.tcp.state != TCP_CONNTRACK_TIME_WAIT) { + /* Just to be sure we have it only once in the list. + We should'nt see tuples twice unless someone hooks this + into a table without "-p tcp --syn" */ + addit = 0; + } +#if DEBUG + printk("ipt_connlimit [%d]: src=%u.%u.%u.%u:%d dst=%u.%u.%u.%u:%d %s\n", + ipt_iphash(addr & mask), + NIPQUAD(conn->tuple.src.ip), ntohs(conn->tuple.src.u.tcp.port), + NIPQUAD(conn->tuple.dst.ip), ntohs(conn->tuple.dst.u.tcp.port), + (NULL != found) ? tcp[found->ctrack->proto.tcp.state] : "gone"); +#endif + if (NULL == found) { + /* this one is gone */ + lh = lh->prev; + list_del(lh->next); + kfree(conn); + continue; + } + if (found->ctrack->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT) { + /* we don't care about connections which are + closed already -> ditch it */ + lh = lh->prev; + list_del(lh->next); + kfree(conn); + nf_conntrack_put(&found->ctrack->infos[0]); + continue; + } + if ((addr & mask) == (conn->tuple.src.ip & mask)) { + /* same source IP address -> be counted! */ + matches++; + } + nf_conntrack_put(&found->ctrack->infos[0]); + } + if (addit) { + /* save the new connection in our list */ +#if DEBUG + printk("ipt_connlimit [%d]: src=%u.%u.%u.%u:%d dst=%u.%u.%u.%u:%d new\n", + ipt_iphash(addr & mask), + NIPQUAD(tuple.src.ip), ntohs(tuple.src.u.tcp.port), + NIPQUAD(tuple.dst.ip), ntohs(tuple.dst.u.tcp.port)); +#endif + conn = kmalloc(sizeof(*conn),GFP_ATOMIC); + if (NULL == conn) + return -1; + memset(conn,0,sizeof(*conn)); + INIT_LIST_HEAD(&conn->list); + conn->tuple = tuple; + list_add(&conn->list,hash); + matches++; + } + spin_unlock(&data->lock); + return matches; +} + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_connlimit_info *info = matchinfo; + int connections, match; + struct ip_conntrack *ct; + enum ip_conntrack_info ctinfo; + + ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo); + if (NULL == ct) { + printk("ipt_connlimit: Oops: invalid ct state ?\n"); + *hotdrop = 1; + return 0; + } + connections = count_them(info->data,skb->nh.iph->saddr,info->mask,ct); + if (-1 == connections) { + printk("ipt_connlimit: Hmm, kmalloc failed :-(\n"); + *hotdrop = 1; /* let's free some memory :-) */ + return 0; + } + match = (info->inverse) ? (connections <= info->limit) : (connections > info->limit); +#if DEBUG + printk("ipt_connlimit: src=%u.%u.%u.%u mask=%u.%u.%u.%u " + "connections=%d limit=%d match=%s\n", + NIPQUAD(skb->nh.iph->saddr), NIPQUAD(info->mask), + connections, info->limit, match ? "yes" : "no"); +#endif + + return match; +} + +static int check(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + struct ipt_connlimit_info *info = matchinfo; + int i; + + /* verify size */ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_connlimit_info))) + return 0; + + /* refuse anything but tcp */ + if (ip->proto != IPPROTO_TCP) + return 0; + + /* init private data */ + info->data = kmalloc(sizeof(struct ipt_connlimit_data),GFP_KERNEL); + spin_lock_init(&(info->data->lock)); + for (i = 0; i < 256; i++) + INIT_LIST_HEAD(&(info->data->iphash[i])); + + return 1; +} + +static void destroy(void *matchinfo, unsigned int matchinfosize) +{ + struct ipt_connlimit_info *info = matchinfo; + struct ipt_connlimit_conn *conn; + struct list_head *hash; + int i; + + /* cleanup */ + for (i = 0; i < 256; i++) { + hash = &(info->data->iphash[i]); + while (hash != hash->next) { + conn = list_entry(hash->next,struct ipt_connlimit_conn,list); + list_del(hash->next); + kfree(conn); + } + } + kfree(info->data); +} + +static struct ipt_match connlimit_match = { + .name = "connlimit", + .match = &match, + .checkentry = &check, + .destroy = &destroy, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&connlimit_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&connlimit_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_dstlimit.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_dstlimit.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_dstlimit.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_dstlimit.c 2004-04-19 10:08:30.000000000 +0200 @@ -0,0 +1,690 @@ +/* iptables match extension to limit the number of packets per second + * seperately for each destination. + * + * (C) 2003 by Harald Welte + * + * $Id$ + * + * Development of this code was funded by Astaro AG, http://www.astaro.com/ + * + * based on ipt_limit.c by: + * Jérôme de Vivie + * Hervé Eychenne + * Rusty Russell + * + * The general idea is to create a hash table for every dstip and have a + * seperate limit counter per tuple. This way you can do something like 'limit + * the number of syn packets for each of my internal addresses. + * + * Ideally this would just be implemented as a general 'hash' match, which would + * allow us to attach any iptables target to it's hash buckets. But this is + * not possible in the current iptables architecture. As always, pkttables for + * 2.7.x will help ;) + */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define ASSERT_READ_LOCK(x) +#define ASSERT_WRITE_LOCK(x) +#include +#include + +#include +#include + +/* FIXME: this is just for IP_NF_ASSERRT */ +#include + +#define MS2JIFFIES(x) ((x*HZ)/1000) + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte "); +MODULE_DESCRIPTION("iptables match for limiting per destination"); + +/* need to declare this at the top */ +static struct proc_dir_entry *dstlimit_procdir; +static struct file_operations dl_file_ops; + +/* hash table crap */ + +struct dsthash_dst { + u_int32_t src_ip; + u_int32_t dst_ip; + u_int16_t port; +}; + +struct dsthash_ent { + /* static / read-only parts in the beginning */ + struct list_head list; + struct dsthash_dst dst; + + /* modified structure members in the end */ + unsigned long expires; /* precalculated expiry time */ + struct { + unsigned long prev; /* last modification */ + u_int32_t credit; + u_int32_t credit_cap, cost; + } rateinfo; +}; + +struct ipt_dstlimit_htable { + struct list_head list; /* global list of all htables */ + atomic_t use; + + struct dstlimit_cfg cfg; /* config */ + + /* used internally */ + spinlock_t lock; /* lock for list_head */ + u_int32_t rnd; /* random seed for hash */ + struct timer_list timer; /* timer for gc */ + atomic_t count; /* number entries in table */ + + /* seq_file stuff */ + struct proc_dir_entry *pde; + + struct list_head hash[0]; /* hashtable itself */ +}; + +DECLARE_RWLOCK(dstlimit_lock); /* protects htables list */ +static LIST_HEAD(dstlimit_htables); +static kmem_cache_t *dstlimit_cachep; + +static inline int dst_cmp(const struct dsthash_ent *ent, struct dsthash_dst *b) +{ + return (ent->dst.dst_ip == b->dst_ip + && ent->dst.port == b->port + && ent->dst.src_ip == b->src_ip); +} + +static inline u_int32_t +hash_dst(const struct ipt_dstlimit_htable *ht, const struct dsthash_dst *dst) +{ + return (jhash_3words(dst->dst_ip, dst->port, + dst->src_ip, ht->rnd) % ht->cfg.size); +} + +static inline struct dsthash_ent * +__dsthash_find(const struct ipt_dstlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + u_int32_t hash = hash_dst(ht, dst); + MUST_BE_LOCKED(&ht->lock); + ent = LIST_FIND(&ht->hash[hash], dst_cmp, struct dsthash_ent *, dst); + return ent; +} + +/* allocate dsthash_ent, initialize dst, put in htable and lock it */ +static struct dsthash_ent * +__dsthash_alloc_init(struct ipt_dstlimit_htable *ht, struct dsthash_dst *dst) +{ + struct dsthash_ent *ent; + + /* initialize hash with random val at the time we allocate + * the first hashtable entry */ + if (!ht->rnd) + get_random_bytes(&ht->rnd, 4); + + if (ht->cfg.max && + atomic_read(&ht->count) >= ht->cfg.max) { + /* FIXME: do something. question is what.. */ + if (net_ratelimit()) + printk(KERN_WARNING + "ipt_dstlimit: max count of %u reached\n", + ht->cfg.max); + return NULL; + } + + ent = kmem_cache_alloc(dstlimit_cachep, GFP_ATOMIC); + if (!ent) { + if (net_ratelimit()) + printk(KERN_ERR + "ipt_dstlimit: can't allocate dsthash_ent\n"); + return NULL; + } + + atomic_inc(&ht->count); + + ent->dst.dst_ip = dst->dst_ip; + ent->dst.port = dst->port; + ent->dst.src_ip = dst->src_ip; + + list_add(&ent->list, &ht->hash[hash_dst(ht, dst)]); + + return ent; +} + +static inline void +__dsthash_free(struct ipt_dstlimit_htable *ht, struct dsthash_ent *ent) +{ + MUST_BE_LOCKED(&ht->lock); + + list_del(&ent->list); + kmem_cache_free(dstlimit_cachep, ent); + atomic_dec(&ht->count); +} +static void htable_gc(unsigned long htlong); + +static int htable_create(struct ipt_dstlimit_info *minfo) +{ + int i; + unsigned int size; + struct ipt_dstlimit_htable *hinfo; + + if (minfo->cfg.size) + size = minfo->cfg.size; + else { + size = (((num_physpages << PAGE_SHIFT) / 16384) + / sizeof(struct list_head)); + if (num_physpages > (1024 * 1024 * 1024 / PAGE_SIZE)) + size = 8192; + if (size < 16) + size = 16; + } + /* FIXME: don't use vmalloc() here or anywhere else -HW */ + hinfo = vmalloc(sizeof(struct ipt_dstlimit_htable) + + (sizeof(struct list_head) * size)); + if (!hinfo) { + printk(KERN_ERR "ipt_dstlimit: Unable to create hashtable\n"); + return -1; + } + minfo->hinfo = hinfo; + + /* copy match config into hashtable config */ + memcpy(&hinfo->cfg, &minfo->cfg, sizeof(hinfo->cfg)); + hinfo->cfg.size = size; + if (!hinfo->cfg.max) + hinfo->cfg.max = 8 * hinfo->cfg.size; + else if (hinfo->cfg.max < hinfo->cfg.size) + hinfo->cfg.max = hinfo->cfg.size; + + for (i = 0; i < hinfo->cfg.size; i++) + INIT_LIST_HEAD(&hinfo->hash[i]); + + atomic_set(&hinfo->count, 0); + atomic_set(&hinfo->use, 1); + hinfo->rnd = 0; + hinfo->lock = SPIN_LOCK_UNLOCKED; + hinfo->pde = create_proc_entry(minfo->name, 0, dstlimit_procdir); + if (!hinfo->pde) { + vfree(hinfo); + return -1; + } + hinfo->pde->proc_fops = &dl_file_ops; + hinfo->pde->data = hinfo; + + init_timer(&hinfo->timer); + hinfo->timer.expires = jiffies + MS2JIFFIES(hinfo->cfg.gc_interval); + hinfo->timer.data = (unsigned long )hinfo; + hinfo->timer.function = htable_gc; + add_timer(&hinfo->timer); + + WRITE_LOCK(&dstlimit_lock); + list_add(&hinfo->list, &dstlimit_htables); + WRITE_UNLOCK(&dstlimit_lock); + + return 0; +} + +static int select_all(struct ipt_dstlimit_htable *ht, struct dsthash_ent *he) +{ + return 1; +} + +static int select_gc(struct ipt_dstlimit_htable *ht, struct dsthash_ent *he) +{ + return (jiffies >= he->expires); +} + +static void htable_selective_cleanup(struct ipt_dstlimit_htable *ht, + int (*select)(struct ipt_dstlimit_htable *ht, + struct dsthash_ent *he)) +{ + int i; + + IP_NF_ASSERT(ht->cfg.size && ht->cfg.max); + + /* lock hash table and iterate over it */ + LOCK_BH(&ht->lock); + for (i = 0; i < ht->cfg.size; i++) { + struct dsthash_ent *dh, *n; + list_for_each_entry_safe(dh, n, &ht->hash[i], list) { + if ((*select)(ht, dh)) + __dsthash_free(ht, dh); + } + } + UNLOCK_BH(&ht->lock); +} + +/* hash table garbage collector, run by timer */ +static void htable_gc(unsigned long htlong) +{ + struct ipt_dstlimit_htable *ht = (struct ipt_dstlimit_htable *)htlong; + + htable_selective_cleanup(ht, select_gc); + + /* re-add the timer accordingly */ + ht->timer.expires = jiffies + MS2JIFFIES(ht->cfg.gc_interval); + add_timer(&ht->timer); +} + +static void htable_destroy(struct ipt_dstlimit_htable *hinfo) +{ + /* remove timer, if it is pending */ + if (timer_pending(&hinfo->timer)) + del_timer(&hinfo->timer); + + /* remove proc entry */ + remove_proc_entry(hinfo->pde->name, dstlimit_procdir); + + htable_selective_cleanup(hinfo, select_all); + vfree(hinfo); +} + +static struct ipt_dstlimit_htable *htable_find_get(char *name) +{ + struct ipt_dstlimit_htable *hinfo; + + READ_LOCK(&dstlimit_lock); + list_for_each_entry(hinfo, &dstlimit_htables, list) { + if (!strcmp(name, hinfo->pde->name)) { + atomic_inc(&hinfo->use); + READ_UNLOCK(&dstlimit_lock); + return hinfo; + } + } + READ_UNLOCK(&dstlimit_lock); + + return NULL; +} + +static void htable_put(struct ipt_dstlimit_htable *hinfo) +{ + if (atomic_dec_and_test(&hinfo->use)) { + WRITE_LOCK(&dstlimit_lock); + list_del(&hinfo->list); + WRITE_UNLOCK(&dstlimit_lock); + htable_destroy(hinfo); + } +} + + +/* The algorithm used is the Simple Token Bucket Filter (TBF) + * see net/sched/sch_tbf.c in the linux source tree + */ + +/* Rusty: This is my (non-mathematically-inclined) understanding of + this algorithm. The `average rate' in jiffies becomes your initial + amount of credit `credit' and the most credit you can ever have + `credit_cap'. The `peak rate' becomes the cost of passing the + test, `cost'. + + `prev' tracks the last packet hit: you gain one credit per jiffy. + If you get credit balance more than this, the extra credit is + discarded. Every time the match passes, you lose `cost' credits; + if you don't have that many, the test fails. + + See Alexey's formal explanation in net/sched/sch_tbf.c. + + To get the maximum range, we multiply by this factor (ie. you get N + credits per jiffy). We want to allow a rate as low as 1 per day + (slowest userspace tool allows), which means + CREDITS_PER_JIFFY*HZ*60*60*24 < 2^32 ie. +*/ +#define MAX_CPJ (0xFFFFFFFF / (HZ*60*60*24)) + +/* Repeated shift and or gives us all 1s, final shift and add 1 gives + * us the power of 2 below the theoretical max, so GCC simply does a + * shift. */ +#define _POW2_BELOW2(x) ((x)|((x)>>1)) +#define _POW2_BELOW4(x) (_POW2_BELOW2(x)|_POW2_BELOW2((x)>>2)) +#define _POW2_BELOW8(x) (_POW2_BELOW4(x)|_POW2_BELOW4((x)>>4)) +#define _POW2_BELOW16(x) (_POW2_BELOW8(x)|_POW2_BELOW8((x)>>8)) +#define _POW2_BELOW32(x) (_POW2_BELOW16(x)|_POW2_BELOW16((x)>>16)) +#define POW2_BELOW32(x) ((_POW2_BELOW32(x)>>1) + 1) + +#define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ) + +/* Precision saver. */ +static inline u_int32_t +user2credits(u_int32_t user) +{ + /* If multiplying would overflow... */ + if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) + /* Divide first. */ + return (user / IPT_DSTLIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; + + return (user * HZ * CREDITS_PER_JIFFY) / IPT_DSTLIMIT_SCALE; +} + +static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now) +{ + dh->rateinfo.credit += (now - xchg(&dh->rateinfo.prev, now)) + * CREDITS_PER_JIFFY; + if (dh->rateinfo.credit > dh->rateinfo.credit_cap) + dh->rateinfo.credit = dh->rateinfo.credit_cap; +} + +static int +dstlimit_match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + struct ipt_dstlimit_info *r = + ((struct ipt_dstlimit_info *)matchinfo)->u.master; + struct ipt_dstlimit_htable *hinfo = r->hinfo; + unsigned long now = jiffies; + struct dsthash_ent *dh; + struct dsthash_dst dst; + + memset(&dst, 0, sizeof(dst)); + + /* dest ip is always in hash */ + dst.dst_ip = skb->nh.iph->daddr; + + /* source ip only if respective hashmode, otherwise set to + * zero */ + if (hinfo->cfg.mode & IPT_DSTLIMIT_HASH_SIP) + dst.src_ip = skb->nh.iph->saddr; + + /* dest port only if respective mode */ + if (hinfo->cfg.mode & IPT_DSTLIMIT_HASH_DPT) { + u16 ports[2]; + + /* Must not be a fragment. */ + if (offset) + return 0; + + /* Must be big enough to read ports (both UDP and TCP have + them at the start). */ + if (skb_copy_bits(skb, skb->nh.iph->ihl*4, ports, sizeof(ports)) < 0) { + /* We've been asked to examine this packet, and we + can't. Hence, no choice but to drop. */ + *hotdrop = 1; + return 0; + } + + switch (skb->nh.iph->protocol) { + struct tcphdr *th; + struct udphdr *uh; + case IPPROTO_TCP: + th = (void *)skb->nh.iph+skb->nh.iph->ihl*4; + dst.port = th->dest; + break; + case IPPROTO_UDP: + uh = (void *)skb->nh.iph+skb->nh.iph->ihl*4; + dst.port = uh->dest; + break; + default: + break; + } + } + + LOCK_BH(&hinfo->lock); + dh = __dsthash_find(hinfo, &dst); + if (!dh) { + dh = __dsthash_alloc_init(hinfo, &dst); + + if (!dh) { + /* enomem... don't match == DROP */ + if (net_ratelimit()) + printk(KERN_ERR "%s: ENOMEM\n", __FUNCTION__); + UNLOCK_BH(&hinfo->lock); + return 0; + } + + dh->expires = jiffies + MS2JIFFIES(hinfo->cfg.expire); + + dh->rateinfo.prev = jiffies; + dh->rateinfo.credit = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg * + hinfo->cfg.burst); + dh->rateinfo.cost = user2credits(hinfo->cfg.avg); + + UNLOCK_BH(&hinfo->lock); + return 1; + } + + /* update expiration timeout */ + dh->expires = now + MS2JIFFIES(hinfo->cfg.expire); + + rateinfo_recalc(dh, now); + if (dh->rateinfo.credit >= dh->rateinfo.cost) { + /* We're underlimit. */ + dh->rateinfo.credit -= dh->rateinfo.cost; + UNLOCK_BH(&hinfo->lock); + return 1; + } + + UNLOCK_BH(&hinfo->lock); + + /* default case: we're overlimit, thus don't match */ + return 0; +} + +static int +dstlimit_checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + struct ipt_dstlimit_info *r = matchinfo; + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_dstlimit_info))) + return 0; + + /* Check for overflow. */ + if (r->cfg.burst == 0 + || user2credits(r->cfg.avg * r->cfg.burst) < + user2credits(r->cfg.avg)) { + printk(KERN_ERR "ipt_dstlimit: Overflow, try lower: %u/%u\n", + r->cfg.avg, r->cfg.burst); + return 0; + } + + if (r->cfg.mode == 0 + || r->cfg.mode > (IPT_DSTLIMIT_HASH_DPT + |IPT_DSTLIMIT_HASH_DIP + |IPT_DSTLIMIT_HASH_SIP)) + return 0; + + if (!r->cfg.gc_interval) + return 0; + + if (!r->cfg.expire) + return 0; + + r->hinfo = htable_find_get(r->name); + if (!r->hinfo && (htable_create(r) != 0)) { + return 0; + } + + /* Ugly hack: For SMP, we only want to use one set */ + r->u.master = r; + + return 1; +} + +static void +dstlimit_destroy(void *matchinfo, unsigned int matchsize) +{ + struct ipt_dstlimit_info *r = (struct ipt_dstlimit_info *) matchinfo; + + htable_put(r->hinfo); +} + +static struct ipt_match ipt_dstlimit = { + .list = { .prev = NULL, .next = NULL }, + .name = "dstlimit", + .match = dstlimit_match, + .checkentry = dstlimit_checkentry, + .destroy = dstlimit_destroy, + .me = THIS_MODULE +}; + +/* PROC stuff */ + +static void *dl_seq_start(struct seq_file *s, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct ipt_dstlimit_htable *htable = pde->data; + unsigned int *bucket; + + LOCK_BH(&htable->lock); + if (*pos >= htable->cfg.size) + return NULL; + + bucket = kmalloc(sizeof(unsigned int), GFP_KERNEL); + if (!bucket) + return ERR_PTR(-ENOMEM); + + *bucket = *pos; + return bucket; +} + +static void *dl_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct proc_dir_entry *pde = s->private; + struct ipt_dstlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + *pos = ++(*bucket); + if (*pos >= htable->cfg.size) { + kfree(v); + return NULL; + } + return bucket; +} + +static void dl_seq_stop(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct ipt_dstlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + kfree(bucket); + + UNLOCK_BH(&htable->lock); +} + +static inline int dl_seq_real_show(struct dsthash_ent *ent, struct seq_file *s) +{ + /* recalculate to show accurate numbers */ + rateinfo_recalc(ent, jiffies); + + return seq_printf(s, "%ld %u.%u.%u.%u->%u.%u.%u.%u:%u %u %u %u\n", + (ent->expires - jiffies)/HZ, + NIPQUAD(ent->dst.src_ip), + NIPQUAD(ent->dst.dst_ip), ntohs(ent->dst.port), + ent->rateinfo.credit, ent->rateinfo.credit_cap, + ent->rateinfo.cost); +} + +static int dl_seq_show(struct seq_file *s, void *v) +{ + struct proc_dir_entry *pde = s->private; + struct ipt_dstlimit_htable *htable = pde->data; + unsigned int *bucket = (unsigned int *)v; + + if (LIST_FIND_W(&htable->hash[*bucket], dl_seq_real_show, + struct dsthash_ent *, s)) { + /* buffer was filled and unable to print that tuple */ + return 1; + } + return 0; +} + +static struct seq_operations dl_seq_ops = { + .start = dl_seq_start, + .next = dl_seq_next, + .stop = dl_seq_stop, + .show = dl_seq_show +}; + +static int dl_proc_open(struct inode *inode, struct file *file) +{ + int ret = seq_open(file, &dl_seq_ops); + + if (!ret) { + struct seq_file *sf = file->private_data; + sf->private = PDE(inode); + } + return ret; +} + +static struct file_operations dl_file_ops = { + .owner = THIS_MODULE, + .open = dl_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release +}; + +static int init_or_fini(int fini) +{ + int ret = 0; + + if (fini) + goto cleanup; + + if (ipt_register_match(&ipt_dstlimit)) { + ret = -EINVAL; + goto cleanup_nothing; + } + + /* FIXME: do we really want HWCACHE_ALIGN since our objects are + * quite small ? */ + dstlimit_cachep = kmem_cache_create("ipt_dstlimit", + sizeof(struct dsthash_ent), 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + if (!dstlimit_cachep) { + printk(KERN_ERR "Unable to create ipt_dstlimit slab cache\n"); + ret = -ENOMEM; + goto cleanup_unreg_match; + } + + dstlimit_procdir = proc_mkdir("ipt_dstlimit", proc_net); + if (!dstlimit_procdir) { + printk(KERN_ERR "Unable to create proc dir entry\n"); + ret = -ENOMEM; + goto cleanup_free_slab; + } + + return ret; + +cleanup: + remove_proc_entry("ipt_dstlimit", proc_net); +cleanup_free_slab: + kmem_cache_destroy(dstlimit_cachep); +cleanup_unreg_match: + ipt_unregister_match(&ipt_dstlimit); +cleanup_nothing: + return ret; + +} + +static int __init init(void) +{ + return init_or_fini(0); +} + +static void __exit fini(void) +{ + init_or_fini(1); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_fuzzy.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_fuzzy.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_fuzzy.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_fuzzy.c 2004-04-19 10:08:31.000000000 +0200 @@ -0,0 +1,185 @@ +/* + * This module implements a simple TSK FLC + * (Takagi-Sugeno-Kang Fuzzy Logic Controller) that aims + * to limit , in an adaptive and flexible way , the packet rate crossing + * a given stream . It serves as an initial and very simple (but effective) + * example of how Fuzzy Logic techniques can be applied to defeat DoS attacks. + * As a matter of fact , Fuzzy Logic can help us to insert any "behavior" + * into our code in a precise , adaptive and efficient manner. + * The goal is very similar to that of "limit" match , but using techniques of + * Fuzzy Control , that allow us to shape the transfer functions precisely , + * avoiding over and undershoots - and stuff like that . + * + * + * 2002-08-10 Hime Aguiar e Oliveira Jr. : Initial version. + * 2002-08-17 : Changed to eliminate floating point operations . + * 2002-08-23 : Coding style changes . +*/ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* + Packet Acceptance Rate - LOW and Packet Acceptance Rate - HIGH + Expressed in percentage +*/ + +#define PAR_LOW 1/100 +#define PAR_HIGH 1 + +static spinlock_t fuzzy_lock = SPIN_LOCK_UNLOCKED ; + +MODULE_AUTHOR("Hime Aguiar e Oliveira Junior "); +MODULE_DESCRIPTION("IP tables Fuzzy Logic Controller match module"); +MODULE_LICENSE("GPL"); + +static u_int8_t mf_high(u_int32_t tx,u_int32_t mini,u_int32_t maxi) +{ + if (tx >= maxi) + return 100; + + if (tx <= mini) + return 0; + + return ( (100*(tx-mini)) / (maxi-mini) ); +} + +static u_int8_t mf_low(u_int32_t tx,u_int32_t mini,u_int32_t maxi) +{ + if (tx <= mini) + return 100; + + if (tx >= maxi) + return 0; + + return ( (100*( maxi - tx )) / ( maxi - mini ) ); +} + +static int +ipt_fuzzy_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + /* From userspace */ + + struct ipt_fuzzy_info *info = (struct ipt_fuzzy_info *) matchinfo; + + u_int8_t random_number; + unsigned long amount; + u_int8_t howhigh, howlow; + + + spin_lock_bh(&fuzzy_lock); /* Rise the lock */ + + info->bytes_total += pskb->len; + info->packets_total++; + + info->present_time = jiffies; + + if (info->present_time >= info->previous_time) + amount = info->present_time - info->previous_time; + else { + /* There was a transition : I choose to re-sample + and keep the old acceptance rate... + */ + + amount = 0; + info->previous_time = info->present_time; + info->bytes_total = info->packets_total = 0; + }; + + if (amount > HZ/10) /* More than 100 ms elapsed ... */ + { + + info->mean_rate = (u_int32_t) ((HZ*info->packets_total) \ + / amount ); + + info->previous_time = info->present_time; + info->bytes_total = info->packets_total = 0; + + howhigh = mf_high(info->mean_rate,info->minimum_rate,info->maximum_rate); + howlow = mf_low(info->mean_rate,info->minimum_rate,info->maximum_rate); + + info->acceptance_rate = (u_int8_t) \ + (howhigh*PAR_LOW + PAR_HIGH*howlow); + + /* In fact , the above defuzzification would require a denominator + proportional to (howhigh+howlow) but , in this particular case , + that expression is constant . + An imediate consequence is that it isn't necessary to call + both mf_high and mf_low - but to keep things understandable , + I did so . */ + + } + + spin_unlock_bh(&fuzzy_lock); /* Release the lock */ + + + if ( info->acceptance_rate < 100 ) + { + get_random_bytes((void *)(&random_number), 1); + + /* If within the acceptance , it can pass => don't match */ + if (random_number <= (255 * info->acceptance_rate) / 100) + return 0; + else + return 1; /* It can't pass ( It matches ) */ + } ; + + return 0; /* acceptance_rate == 100 % => Everything passes ... */ + +} + +static int +ipt_fuzzy_checkentry(const char *tablename, + const struct ipt_ip *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + + const struct ipt_fuzzy_info *info = matchinfo; + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_fuzzy_info))) { + printk("ipt_fuzzy: matchsize %u != %u\n", matchsize, + IPT_ALIGN(sizeof(struct ipt_fuzzy_info))); + return 0; + } + + if ((info->minimum_rate < MINFUZZYRATE ) || (info->maximum_rate > MAXFUZZYRATE) + || (info->minimum_rate >= info->maximum_rate )) { + printk("ipt_fuzzy: BAD limits , please verify !!!\n"); + return 0; + } + + return 1; +} + +static struct ipt_match ipt_fuzzy_reg = { + .name = "fuzzy", + .match = ipt_fuzzy_match, + .checkentry = ipt_fuzzy_checkentry, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&ipt_fuzzy_reg); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&ipt_fuzzy_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_helper.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_helper.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_helper.c 2004-04-15 03:35:20.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_helper.c 2004-04-19 10:08:23.000000000 +0200 @@ -41,17 +41,17 @@ struct ip_conntrack_expect *exp; struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; - int ret = 0; + int ret = info->invert; ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo); if (!ct) { DEBUGP("ipt_helper: Eek! invalid conntrack?\n"); - return 0; + return ret; } if (!ct->master) { DEBUGP("ipt_helper: conntrack %p has no master\n", ct); - return 0; + return ret; } exp = ct->master; @@ -71,8 +71,8 @@ DEBUGP("master's name = %s , info->name = %s\n", exp->expectant->helper->name, info->name); - ret = !strncmp(exp->expectant->helper->name, info->name, - strlen(exp->expectant->helper->name)) ^ info->invert; + ret ^= !strncmp(exp->expectant->helper->name, info->name, + strlen(exp->expectant->helper->name)); out_unlock: READ_UNLOCK(&ip_conntrack_lock); return ret; @@ -108,7 +108,6 @@ static int __init init(void) { - need_ip_conntrack(); return ipt_register_match(&helper_match); } diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_ipv4options.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_ipv4options.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_ipv4options.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_ipv4options.c 2004-04-19 10:08:32.000000000 +0200 @@ -0,0 +1,172 @@ +/* + This is a module which is used to match ipv4 options. + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 11-mars-2001 Fabrice MARIE : initial development. + 12-july-2001 Fabrice MARIE : added router-alert otions matching. Fixed a bug with no-srr + 12-august-2001 Imran Patel : optimization of the match. + 18-november-2001 Fabrice MARIE : added [!] 'any' option match. + 19-february-2004 Harald Welte : merge with 2.6.x +*/ + +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Fabrice Marie "); + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_ipv4options_info *info = matchinfo; /* match info for rule */ + const struct iphdr *iph = skb->nh.iph; + const struct ip_options *opt; + + if (iph->ihl * 4 == sizeof(struct iphdr)) { + /* No options, so we match only the "DONTs" and the "IGNOREs" */ + + if (((info->options & IPT_IPV4OPTION_MATCH_ANY_OPT) == IPT_IPV4OPTION_MATCH_ANY_OPT) || + ((info->options & IPT_IPV4OPTION_MATCH_SSRR) == IPT_IPV4OPTION_MATCH_SSRR) || + ((info->options & IPT_IPV4OPTION_MATCH_LSRR) == IPT_IPV4OPTION_MATCH_LSRR) || + ((info->options & IPT_IPV4OPTION_MATCH_RR) == IPT_IPV4OPTION_MATCH_RR) || + ((info->options & IPT_IPV4OPTION_MATCH_TIMESTAMP) == IPT_IPV4OPTION_MATCH_TIMESTAMP) || + ((info->options & IPT_IPV4OPTION_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_MATCH_ROUTER_ALERT)) + return 0; + return 1; + } + else { + if ((info->options & IPT_IPV4OPTION_MATCH_ANY_OPT) == IPT_IPV4OPTION_MATCH_ANY_OPT) + /* there are options, and we don't need to care which one */ + return 1; + else { + if ((info->options & IPT_IPV4OPTION_DONT_MATCH_ANY_OPT) == IPT_IPV4OPTION_DONT_MATCH_ANY_OPT) + /* there are options but we don't want any ! */ + return 0; + } + } + + opt = &(IPCB(skb)->opt); + + /* source routing */ + if ((info->options & IPT_IPV4OPTION_MATCH_SSRR) == IPT_IPV4OPTION_MATCH_SSRR) { + if (!((opt->srr) & (opt->is_strictroute))) + return 0; + } + else if ((info->options & IPT_IPV4OPTION_MATCH_LSRR) == IPT_IPV4OPTION_MATCH_LSRR) { + if (!((opt->srr) & (!opt->is_strictroute))) + return 0; + } + else if ((info->options & IPT_IPV4OPTION_DONT_MATCH_SRR) == IPT_IPV4OPTION_DONT_MATCH_SRR) { + if (opt->srr) + return 0; + } + /* record route */ + if ((info->options & IPT_IPV4OPTION_MATCH_RR) == IPT_IPV4OPTION_MATCH_RR) { + if (!opt->rr) + return 0; + } + else if ((info->options & IPT_IPV4OPTION_DONT_MATCH_RR) == IPT_IPV4OPTION_DONT_MATCH_RR) { + if (opt->rr) + return 0; + } + /* timestamp */ + if ((info->options & IPT_IPV4OPTION_MATCH_TIMESTAMP) == IPT_IPV4OPTION_MATCH_TIMESTAMP) { + if (!opt->ts) + return 0; + } + else if ((info->options & IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP) == IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP) { + if (opt->ts) + return 0; + } + /* router-alert option */ + if ((info->options & IPT_IPV4OPTION_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_MATCH_ROUTER_ALERT) { + if (!opt->router_alert) + return 0; + } + else if ((info->options & IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT) { + if (opt->router_alert) + return 0; + } + + /* we match ! */ + return 1; +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ipt_ipv4options_info *info = matchinfo; /* match info for rule */ + /* Check the size */ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_ipv4options_info))) + return 0; + /* Now check the coherence of the data ... */ + if (((info->options & IPT_IPV4OPTION_MATCH_ANY_OPT) == IPT_IPV4OPTION_MATCH_ANY_OPT) && + (((info->options & IPT_IPV4OPTION_DONT_MATCH_SRR) == IPT_IPV4OPTION_DONT_MATCH_SRR) || + ((info->options & IPT_IPV4OPTION_DONT_MATCH_RR) == IPT_IPV4OPTION_DONT_MATCH_RR) || + ((info->options & IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP) == IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP) || + ((info->options & IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT) || + ((info->options & IPT_IPV4OPTION_DONT_MATCH_ANY_OPT) == IPT_IPV4OPTION_DONT_MATCH_ANY_OPT))) + return 0; /* opposites */ + if (((info->options & IPT_IPV4OPTION_DONT_MATCH_ANY_OPT) == IPT_IPV4OPTION_DONT_MATCH_ANY_OPT) && + (((info->options & IPT_IPV4OPTION_MATCH_LSRR) == IPT_IPV4OPTION_MATCH_LSRR) || + ((info->options & IPT_IPV4OPTION_MATCH_SSRR) == IPT_IPV4OPTION_MATCH_SSRR) || + ((info->options & IPT_IPV4OPTION_MATCH_RR) == IPT_IPV4OPTION_MATCH_RR) || + ((info->options & IPT_IPV4OPTION_MATCH_TIMESTAMP) == IPT_IPV4OPTION_MATCH_TIMESTAMP) || + ((info->options & IPT_IPV4OPTION_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_MATCH_ROUTER_ALERT) || + ((info->options & IPT_IPV4OPTION_MATCH_ANY_OPT) == IPT_IPV4OPTION_MATCH_ANY_OPT))) + return 0; /* opposites */ + if (((info->options & IPT_IPV4OPTION_MATCH_SSRR) == IPT_IPV4OPTION_MATCH_SSRR) && + ((info->options & IPT_IPV4OPTION_MATCH_LSRR) == IPT_IPV4OPTION_MATCH_LSRR)) + return 0; /* cannot match in the same time loose and strict source routing */ + if ((((info->options & IPT_IPV4OPTION_MATCH_SSRR) == IPT_IPV4OPTION_MATCH_SSRR) || + ((info->options & IPT_IPV4OPTION_MATCH_LSRR) == IPT_IPV4OPTION_MATCH_LSRR)) && + ((info->options & IPT_IPV4OPTION_DONT_MATCH_SRR) == IPT_IPV4OPTION_DONT_MATCH_SRR)) + return 0; /* opposites */ + if (((info->options & IPT_IPV4OPTION_MATCH_RR) == IPT_IPV4OPTION_MATCH_RR) && + ((info->options & IPT_IPV4OPTION_DONT_MATCH_RR) == IPT_IPV4OPTION_DONT_MATCH_RR)) + return 0; /* opposites */ + if (((info->options & IPT_IPV4OPTION_MATCH_TIMESTAMP) == IPT_IPV4OPTION_MATCH_TIMESTAMP) && + ((info->options & IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP) == IPT_IPV4OPTION_DONT_MATCH_TIMESTAMP)) + return 0; /* opposites */ + if (((info->options & IPT_IPV4OPTION_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_MATCH_ROUTER_ALERT) && + ((info->options & IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT) == IPT_IPV4OPTION_DONT_MATCH_ROUTER_ALERT)) + return 0; /* opposites */ + + /* everything looks ok. */ + return 1; +} + +static struct ipt_match ipv4options_match = { + .name = "ipv4options", + .match = match, + .checkentry = checkentry, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&ipv4options_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&ipv4options_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_mport.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_mport.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_mport.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_mport.c 2004-04-19 10:08:33.000000000 +0200 @@ -0,0 +1,116 @@ +/* Kernel module to match one of a list of TCP/UDP ports: ports are in + the same place so we can treat them as equal. */ +#include +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); + +#if 0 +#define duprintf(format, args...) printk(format , ## args) +#else +#define duprintf(format, args...) +#endif + +/* Returns 1 if the port is matched by the test, 0 otherwise. */ +static inline int +ports_match(const struct ipt_mport *minfo, u_int16_t src, u_int16_t dst) +{ + unsigned int i; + unsigned int m; + u_int16_t pflags = minfo->pflags; + for (i=0, m=1; iports[i] == 65535) + return 0; + + s = minfo->ports[i]; + + if (pflags & m) { + e = minfo->ports[++i]; + m <<= 1; + } else + e = s; + + if (minfo->flags & IPT_MPORT_SOURCE + && src >= s && src <= e) + return 1; + + if (minfo->flags & IPT_MPORT_DESTINATION + && dst >= s && dst <= e) + return 1; + } + + return 0; +} + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + u16 ports[2]; + const struct ipt_mport *minfo = matchinfo; + + if (offset) + return 0; + + /* Must be big enough to read ports (both UDP and TCP have + them at the start). */ + if (skb_copy_bits(skb, skb->nh.iph->ihl*4, ports, sizeof(ports)) < 0) { + /* We've been asked to examine this packet, and we + can't. Hence, no choice but to drop. */ + duprintf("ipt_multiport:" + " Dropping evil offset=0 tinygram.\n"); + *hotdrop = 1; + return 0; + } + + return ports_match(minfo, ntohs(ports[0]), ntohs(ports[1])); +} + +/* Called when user tries to insert an entry of this type. */ +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_mport))) + return 0; + + /* Must specify proto == TCP/UDP, no unknown flags or bad count */ + return (ip->proto == IPPROTO_TCP || ip->proto == IPPROTO_UDP) + && !(ip->invflags & IPT_INV_PROTO) + && matchsize == IPT_ALIGN(sizeof(struct ipt_mport)); +} + +static struct ipt_match mport_match = { + .name = "mport", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&mport_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&mport_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_nth.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_nth.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_nth.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_nth.c 2004-04-19 10:08:34.000000000 +0200 @@ -0,0 +1,166 @@ +/* + This is a module which is used for match support for every Nth packet + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 2001-07-18 Fabrice MARIE : initial implementation. + 2001-09-20 Richard Wagner (rwagner@cloudnet.com) + * added support for multiple counters + * added support for matching on individual packets + in the counter cycle + 2004-02-19 Harald Welte + * port to 2.6.x + +*/ + +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Fabrice Marie "); + +/* + * State information. + */ +struct state { + spinlock_t lock; + u_int16_t number; +}; + +static struct state states[IPT_NTH_NUM_COUNTERS]; + +static int +ipt_nth_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + /* Parameters from userspace */ + const struct ipt_nth_info *info = matchinfo; + unsigned counter = info->counter; + if((counter < 0) || (counter >= IPT_NTH_NUM_COUNTERS)) + { + printk(KERN_WARNING "nth: invalid counter %u. counter between 0 and %u\n", counter, IPT_NTH_NUM_COUNTERS-1); + return 0; + }; + + spin_lock(&states[counter].lock); + + /* Are we matching every nth packet?*/ + if (info->packet == 0xFF) + { + /* We're matching every nth packet and only every nth packet*/ + /* Do we match or invert match? */ + if (info->not == 0) + { + if (states[counter].number == 0) + { + ++states[counter].number; + goto match; + } + if (states[counter].number >= info->every) + states[counter].number = 0; /* reset the counter */ + else + ++states[counter].number; + goto dontmatch; + } + else + { + if (states[counter].number == 0) + { + ++states[counter].number; + goto dontmatch; + } + if (states[counter].number >= info->every) + states[counter].number = 0; + else + ++states[counter].number; + goto match; + } + } + else + { + /* We're using the --packet, so there must be a rule for every value */ + if (states[counter].number == info->packet) + { + /* only increment the counter when a match happens */ + if (states[counter].number >= info->every) + states[counter].number = 0; /* reset the counter */ + else + ++states[counter].number; + goto match; + } + else + goto dontmatch; + } + + dontmatch: + /* don't match */ + spin_unlock(&states[counter].lock); + return 0; + + match: + spin_unlock(&states[counter].lock); + return 1; +} + +static int +ipt_nth_checkentry(const char *tablename, + const struct ipt_ip *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + /* Parameters from userspace */ + const struct ipt_nth_info *info = matchinfo; + unsigned counter = info->counter; + if((counter < 0) || (counter >= IPT_NTH_NUM_COUNTERS)) + { + printk(KERN_WARNING "nth: invalid counter %u. counter between 0 and %u\n", counter, IPT_NTH_NUM_COUNTERS-1); + return 0; + }; + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_nth_info))) { + printk("nth: matchsize %u != %u\n", matchsize, + IPT_ALIGN(sizeof(struct ipt_nth_info))); + return 0; + } + + states[counter].number = info->startat; + + return 1; +} + +static struct ipt_match ipt_nth_reg = { + .name = "nth", + .match = ipt_nth_match, + .checkentry = ipt_nth_checkentry, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + unsigned counter; + + memset(&states, 0, sizeof(states)); + for (counter = 0; counter < IPT_NTH_NUM_COUNTERS; counter++) + spin_lock_init(&(states[counter].lock)); + + return ipt_register_match(&ipt_nth_reg); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&ipt_nth_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_osf.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_osf.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_osf.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_osf.c 2004-04-19 10:08:35.000000000 +0200 @@ -0,0 +1,866 @@ +/* + * ipt_osf.c + * + * Copyright (c) 2003 Evgeniy Polyakov + * + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +/* + * OS fingerprint matching module. + * It simply compares various parameters from SYN packet with + * some hardcoded ones. + * + * Original table was created by Michal Zalewski + * for his p0f. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include + +#include + +#define OSF_DEBUG + +#ifdef OSF_DEBUG +#define log(x...) printk(KERN_INFO "ipt_osf: " x) +#define loga(x...) printk(x) +#else +#define log(x...) do {} while(0) +#define loga(x...) do {} while(0) +#endif + +#define FMATCH_WRONG 0 +#define FMATCH_OK 1 +#define FMATCH_OPT_WRONG 2 + +#define OPTDEL ',' +#define OSFPDEL ':' +#define MAXOPTSTRLEN 128 +#define OSFFLUSH "FLUSH" + +static rwlock_t osf_lock = RW_LOCK_UNLOCKED; +static spinlock_t ipt_osf_netlink_lock = SPIN_LOCK_UNLOCKED; +static struct list_head finger_list; +static int match(const struct sk_buff *, const struct net_device *, const struct net_device *, + const void *, int, + const void *, u_int16_t, + int *); +static int checkentry(const char *, const struct ipt_ip *, void *, + unsigned int, unsigned int); + +static unsigned long seq, ipt_osf_groups = 1; +static struct sock *nts; + +static struct ipt_match osf_match = +{ + { NULL, NULL }, + "osf", + &match, + &checkentry, + NULL, + THIS_MODULE +}; + +static void ipt_osf_nlsend(struct osf_finger *f, const struct sk_buff *sk) +{ + unsigned int size; + struct sk_buff *skb; + struct ipt_osf_nlmsg *data; + struct nlmsghdr *nlh; + + size = NLMSG_SPACE(sizeof(struct ipt_osf_nlmsg)); + + skb = alloc_skb(size, GFP_ATOMIC); + if (!skb) + { + log("skb_alloc() failed.\n"); + return; + } + + nlh = NLMSG_PUT(skb, 0, seq++, NLMSG_DONE, size - sizeof(*nlh)); + + data = (struct ipt_osf_nlmsg *)NLMSG_DATA(nlh); + + memcpy(&data->f, f, sizeof(struct osf_finger)); + memcpy(&data->ip, sk->nh.iph, sizeof(struct iphdr)); + memcpy(&data->tcp, (struct tcphdr *)((u_int32_t *)sk->nh.iph + sk->nh.iph->ihl), sizeof(struct tcphdr)); + + NETLINK_CB(skb).dst_groups = ipt_osf_groups; + netlink_broadcast(nts, skb, 0, ipt_osf_groups, GFP_ATOMIC); + +nlmsg_failure: + return; +} + +static inline int smart_dec(const struct sk_buff *skb, unsigned long flags, unsigned char f_ttl) +{ + struct iphdr *ip = skb->nh.iph; + + if (flags & IPT_OSF_SMART) + { + struct in_device *in_dev = in_dev_get(skb->dev); + + for_ifa(in_dev) + { + if (inet_ifa_match(ip->saddr, ifa)) + { + in_dev_put(in_dev); + return (ip->ttl == f_ttl); + } + } + endfor_ifa(in_dev); + + in_dev_put(in_dev); + return (ip->ttl <= f_ttl); + } + else + return (ip->ttl == f_ttl); +} + +static int +match(const struct sk_buff *skb, const struct net_device *in, const struct net_device *out, + const void *matchinfo, int offset, + const void *hdr, u_int16_t datalen, + int *hotdrop) +{ + struct ipt_osf_info *info = (struct ipt_osf_info *)matchinfo; + struct iphdr *ip = skb->nh.iph; + struct tcphdr *tcp; + int fmatch = FMATCH_WRONG, fcount = 0; + unsigned long totlen, optsize = 0, window; + unsigned char df, *optp = NULL, *_optp = NULL; + char check_WSS = 0; + struct list_head *ent; + struct osf_finger *f; + + if (!ip || !info) + return 0; + + tcp = (struct tcphdr *)((u_int32_t *)ip + ip->ihl); + + if (!tcp->syn) + return 0; + + totlen = ntohs(ip->tot_len); + df = ((ntohs(ip->frag_off) & IP_DF)?1:0); + window = ntohs(tcp->window); + + if (tcp->doff*4 > sizeof(struct tcphdr)) + { + _optp = optp = (char *)(tcp+1); + optsize = tcp->doff*4 - sizeof(struct tcphdr); + } + + + /* Actually we can create hash/table of all genres and search + * only in appropriate part, but here is initial variant, + * so will use slow path. + */ + read_lock(&osf_lock); + list_for_each(ent, &finger_list) + { + f = list_entry(ent, struct osf_finger, flist); + + if (!(info->flags & IPT_OSF_LOG) && strcmp(info->genre, f->genre)) + continue; + + optp = _optp; + fmatch = FMATCH_WRONG; + + if (totlen == f->ss && df == f->df && + smart_dec(skb, info->flags, f->ttl)) + { + unsigned long foptsize; + int optnum; + unsigned short mss = 0; + + check_WSS = 0; + + switch (f->wss.wc) + { + case 0: check_WSS = 0; break; + case 'S': check_WSS = 1; break; + case 'T': check_WSS = 2; break; + case '%': check_WSS = 3; break; + default: log("Wrong fingerprint wss.wc=%d, %s - %s\n", + f->wss.wc, f->genre, f->details); + check_WSS = 4; + break; + } + if (check_WSS == 4) + continue; + + /* Check options */ + + foptsize = 0; + for (optnum=0; optnumopt_num; ++optnum) + foptsize += f->opt[optnum].length; + + + if (foptsize > MAX_IPOPTLEN || optsize > MAX_IPOPTLEN || optsize != foptsize) + continue; + + if (!optp) + { + fmatch = FMATCH_OK; + loga("\tYEP : matching without options.\n"); + if ((info->flags & IPT_OSF_LOG) && + info->loglevel == IPT_OSF_LOGLEVEL_FIRST) + break; + else + continue; + } + + + for (optnum=0; optnumopt_num; ++optnum) + { + if (f->opt[optnum].kind == (*optp)) + { + unsigned char len = f->opt[optnum].length; + unsigned char *optend = optp + len; + int loop_cont = 0; + + fmatch = FMATCH_OK; + + + switch (*optp) + { + case OSFOPT_MSS: + mss = ntohs(*(unsigned short *)(optp+2)); + break; + case OSFOPT_TS: + loop_cont = 1; + break; + } + + if (loop_cont) + { + optp = optend; + continue; + } + + if (len != 1) + { + /* Skip kind and length fields*/ + optp += 2; + + if (f->opt[optnum].wc.val != 0) + { + unsigned long tmp = 0; + + /* Hmmm... It looks a bit ugly. :) */ + memcpy(&tmp, optp, + (len > sizeof(unsigned long)? + sizeof(unsigned long):len)); + /* 2 + 2: optlen(2 bytes) + + * kind(1 byte) + length(1 byte) */ + if (len == 4) + tmp = ntohs(tmp); + else + tmp = ntohl(tmp); + + if (f->opt[optnum].wc.wc == '%') + { + if ((tmp % f->opt[optnum].wc.val) != 0) + fmatch = FMATCH_OPT_WRONG; + } + else if (tmp != f->opt[optnum].wc.val) + fmatch = FMATCH_OPT_WRONG; + } + } + + optp = optend; + } + else + fmatch = FMATCH_OPT_WRONG; + + if (fmatch != FMATCH_OK) + break; + } + + if (fmatch != FMATCH_OPT_WRONG) + { + fmatch = FMATCH_WRONG; + + switch (check_WSS) + { + case 0: + if (f->wss.val == 0 || window == f->wss.val) + fmatch = FMATCH_OK; + break; + case 1: /* MSS */ +/* Lurked in OpenBSD */ +#define SMART_MSS 1460 + if (window == f->wss.val*mss || + window == f->wss.val*SMART_MSS) + fmatch = FMATCH_OK; + break; + case 2: /* MTU */ + if (window == f->wss.val*(mss+40) || + window == f->wss.val*(SMART_MSS+40)) + fmatch = FMATCH_OK; + break; + case 3: /* MOD */ + if ((window % f->wss.val) == 0) + fmatch = FMATCH_OK; + break; + } + } + + + if (fmatch == FMATCH_OK) + { + fcount++; + log("%s [%s:%s:%s] : %u.%u.%u.%u:%u -> %u.%u.%u.%u:%u hops=%d\n", + f->genre, f->version, + f->subtype, f->details, + NIPQUAD(ip->saddr), ntohs(tcp->source), + NIPQUAD(ip->daddr), ntohs(tcp->dest), + f->ttl - ip->ttl); + if (info->flags & IPT_OSF_NETLINK) + { + spin_lock_bh(&ipt_osf_netlink_lock); + ipt_osf_nlsend(f, skb); + spin_unlock_bh(&ipt_osf_netlink_lock); + } + if ((info->flags & IPT_OSF_LOG) && + info->loglevel == IPT_OSF_LOGLEVEL_FIRST) + break; + } + } + } + if (!fcount && (info->flags & (IPT_OSF_LOG | IPT_OSF_NETLINK))) + { + unsigned char opt[4 * 15 - sizeof(struct tcphdr)]; + unsigned int i, optsize; + struct osf_finger fg; + + memset(&fg, 0, sizeof(fg)); + + if ((info->flags & IPT_OSF_LOG)) + log("Unknown: %lu:%d:%d:%lu:", window, ip->ttl, df, totlen); + if (optp) + { + optsize = tcp->doff * 4 - sizeof(struct tcphdr); + if (skb_copy_bits(skb, ip->ihl*4 + sizeof(struct tcphdr), + opt, optsize) < 0) + { + if (info->flags & IPT_OSF_LOG) + loga("TRUNCATED"); + if (info->flags & IPT_OSF_NETLINK) + strcpy(fg.details, "TRUNCATED"); + } + else + { + for (i = 0; i < optsize; i++) + { + if (info->flags & IPT_OSF_LOG) + loga("%02X", opt[i]); + } + if (info->flags & IPT_OSF_NETLINK) + memcpy(fg.details, opt, MAXDETLEN); + } + } + if ((info->flags & IPT_OSF_LOG)) + loga(" %u.%u.%u.%u:%u -> %u.%u.%u.%u:%u\n", + NIPQUAD(ip->saddr), ntohs(tcp->source), + NIPQUAD(ip->daddr), ntohs(tcp->dest)); + + if (info->flags & IPT_OSF_NETLINK) + { + fg.wss.val = window; + fg.ttl = ip->ttl; + fg.df = df; + fg.ss = totlen; + strncpy(fg.genre, "Unknown", MAXGENRELEN); + + spin_lock_bh(&ipt_osf_netlink_lock); + ipt_osf_nlsend(&fg, skb); + spin_unlock_bh(&ipt_osf_netlink_lock); + } + } + + read_unlock(&osf_lock); + + return (fmatch == FMATCH_OK)?1:0; +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_osf_info))) + return 0; + if (ip->proto != IPPROTO_TCP) + return 0; + + return 1; +} + +static char * osf_strchr(char *ptr, char c) +{ + char *tmp; + + tmp = strchr(ptr, c); + + while (tmp && tmp+1 && isspace(*(tmp+1))) + tmp++; + + return tmp; +} + +static struct osf_finger * finger_alloc(void) +{ + struct osf_finger *f; + + f = kmalloc(sizeof(struct osf_finger), GFP_KERNEL); + if (f) + memset(f, 0, sizeof(struct osf_finger)); + + return f; +} + +static void finger_free(struct osf_finger *f) +{ + memset(f, 0, sizeof(struct osf_finger)); + kfree(f); +} + + +static void osf_parse_opt(struct osf_opt *opt, int *optnum, char *obuf, int olen) +{ + int i, op; + char *ptr, wc; + unsigned long val; + + ptr = &obuf[0]; + i = 0; + while (ptr != NULL && i < olen) + { + val = 0; + op = 0; + wc = 0; + switch (obuf[i]) + { + case 'N': + op = OSFOPT_NOP; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + *ptr = '\0'; + ptr++; + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + case 'S': + op = OSFOPT_SACKP; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + *ptr = '\0'; + ptr++; + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + case 'T': + op = OSFOPT_TS; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + *ptr = '\0'; + ptr++; + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + case 'W': + op = OSFOPT_WSO; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + switch (obuf[i+1]) + { + case '%': wc = '%'; break; + case 'S': wc = 'S'; break; + case 'T': wc = 'T'; break; + default: wc = 0; break; + } + + *ptr = '\0'; + ptr++; + if (wc) + val = simple_strtoul(&obuf[i+2], NULL, 10); + else + val = simple_strtoul(&obuf[i+1], NULL, 10); + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + case 'M': + op = OSFOPT_MSS; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + if (obuf[i+1] == '%') + wc = '%'; + *ptr = '\0'; + ptr++; + if (wc) + val = simple_strtoul(&obuf[i+2], NULL, 10); + else + val = simple_strtoul(&obuf[i+1], NULL, 10); + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + case 'E': + op = OSFOPT_EOL; + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + *ptr = '\0'; + ptr++; + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + default: + ptr = osf_strchr(&obuf[i], OPTDEL); + if (ptr) + { + ptr++; + i += (int)(ptr-&obuf[i]); + + } + else + i++; + break; + } + + opt[*optnum].kind = IANA_opts[op].kind; + opt[*optnum].length = IANA_opts[op].length; + opt[*optnum].wc.wc = wc; + opt[*optnum].wc.val = val; + + (*optnum)++; + } +} + +static int osf_proc_read(char *buf, char **start, off_t off, int count, int *eof, void *data) +{ + struct list_head *ent; + struct osf_finger *f = NULL; + int i; + + *eof = 1; + count = 0; + + read_lock_bh(&osf_lock); + list_for_each(ent, &finger_list) + { + f = list_entry(ent, struct osf_finger, flist); + + log("%s [%s]", f->genre, f->details); + + count += sprintf(buf+count, "%s - %s[%s] : %s", + f->genre, f->version, + f->subtype, f->details); + + if (f->opt_num) + { + loga(" OPT: "); + //count += sprintf(buf+count, " OPT: "); + for (i=0; iopt_num; ++i) + { + //count += sprintf(buf+count, "%d.%c%lu; ", + // f->opt[i].kind, (f->opt[i].wc.wc)?f->opt[i].wc.wc:' ', f->opt[i].wc.val); + loga("%d.%c%lu; ", + f->opt[i].kind, (f->opt[i].wc.wc)?f->opt[i].wc.wc:' ', f->opt[i].wc.val); + } + } + loga("\n"); + count += sprintf(buf+count, "\n"); + } + read_unlock_bh(&osf_lock); + + return count; +} + +static int osf_proc_write(struct file *file, const char *buffer, unsigned long count, void *data) +{ + int cnt; + unsigned long i; + char obuf[MAXOPTSTRLEN]; + struct osf_finger *finger; + struct list_head *ent, *n; + + char *pbeg, *pend; + + if (count == strlen(OSFFLUSH) && !strncmp(buffer, OSFFLUSH, strlen(OSFFLUSH))) + { + int i = 0; + write_lock_bh(&osf_lock); + list_for_each_safe(ent, n, &finger_list) + { + i++; + finger = list_entry(ent, struct osf_finger, flist); + list_del(&finger->flist); + finger_free(finger); + } + write_unlock_bh(&osf_lock); + + log("Flushed %d entries.\n", i); + + return count; + } + + + cnt = 0; + for (i=0; iwss.wc = 'S'; + if (pbeg[1] == '%') + finger->wss.val = simple_strtoul(pbeg+2, NULL, 10); + else if (pbeg[1] == '*') + finger->wss.val = 0; + else + finger->wss.val = simple_strtoul(pbeg+1, NULL, 10); + } + else if (pbeg[0] == 'T') + { + finger->wss.wc = 'T'; + if (pbeg[1] == '%') + finger->wss.val = simple_strtoul(pbeg+2, NULL, 10); + else if (pbeg[1] == '*') + finger->wss.val = 0; + else + finger->wss.val = simple_strtoul(pbeg+1, NULL, 10); + } + else if (pbeg[0] == '%') + { + finger->wss.wc = '%'; + finger->wss.val = simple_strtoul(pbeg+1, NULL, 10); + } + else if (isdigit(pbeg[0])) + { + finger->wss.wc = 0; + finger->wss.val = simple_strtoul(pbeg, NULL, 10); + } + + pbeg = pend+1; + } + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + finger->ttl = simple_strtoul(pbeg, NULL, 10); + pbeg = pend+1; + } + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + finger->df = simple_strtoul(pbeg, NULL, 10); + pbeg = pend+1; + } + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + finger->ss = simple_strtoul(pbeg, NULL, 10); + pbeg = pend+1; + } + + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + cnt = snprintf(obuf, sizeof(obuf), "%s", pbeg); + pbeg = pend+1; + } + + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + if (pbeg[0] == '@' || pbeg[0] == '*') + cnt = snprintf(finger->genre, sizeof(finger->genre), "%s", pbeg+1); + else + cnt = snprintf(finger->genre, sizeof(finger->genre), "%s", pbeg); + pbeg = pend+1; + } + + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + cnt = snprintf(finger->version, sizeof(finger->version), "%s", pbeg); + pbeg = pend+1; + } + + pend = osf_strchr(pbeg, OSFPDEL); + if (pend) + { + *pend = '\0'; + cnt = snprintf(finger->subtype, sizeof(finger->subtype), "%s", pbeg); + pbeg = pend+1; + } + + cnt = snprintf(finger->details, + ((count - (pbeg - buffer)+1) > MAXDETLEN)?MAXDETLEN:(count - (pbeg - buffer)+1), + "%s", pbeg); + + log("%s - %s[%s] : %s\n", + finger->genre, finger->version, + finger->subtype, finger->details); + + osf_parse_opt(finger->opt, &finger->opt_num, obuf, sizeof(obuf)); + + + write_lock_bh(&osf_lock); + list_add_tail(&finger->flist, &finger_list); + write_unlock_bh(&osf_lock); + + return count; +} + +static int __init osf_init(void) +{ + int err; + struct proc_dir_entry *p; + + log("Startng OS fingerprint matching module.\n"); + + INIT_LIST_HEAD(&finger_list); + + err = ipt_register_match(&osf_match); + if (err) + { + log("Failed to register OS fingerprint matching module.\n"); + return -ENXIO; + } + + p = create_proc_entry("sys/net/ipv4/osf", S_IFREG | 0644, NULL); + if (!p) + { + ipt_unregister_match(&osf_match); + return -ENXIO; + } + + p->write_proc = osf_proc_write; + p->read_proc = osf_proc_read; + + nts = netlink_kernel_create(NETLINK_NFLOG, NULL); + if (!nts) + { + log("netlink_kernel_create() failed\n"); + remove_proc_entry("sys/net/ipv4/osf", NULL); + ipt_unregister_match(&osf_match); + return -ENOMEM; + } + + return 0; +} + +static void __exit osf_fini(void) +{ + struct list_head *ent, *n; + struct osf_finger *f; + + remove_proc_entry("sys/net/ipv4/osf", NULL); + ipt_unregister_match(&osf_match); +// if (nts && nts->socket) +// sock_release(nts->socket); + + list_for_each_safe(ent, n, &finger_list) + { + f = list_entry(ent, struct osf_finger, flist); + list_del(&f->flist); + finger_free(f); + } + + log("OS fingerprint matching module finished.\n"); +} + +module_init(osf_init); +module_exit(osf_fini); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Evgeniy Polyakov "); +MODULE_DESCRIPTION("Passive OS fingerprint matching."); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_pool.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_pool.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_pool.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_pool.c 2004-04-19 10:08:36.000000000 +0200 @@ -0,0 +1,71 @@ +/* Kernel module to match an IP address pool. */ + +#include +#include +#include + +#include +#include +#include + +static inline int match_pool( + ip_pool_t index, + __u32 addr, + int inv +) { + if (ip_pool_match(index, ntohl(addr))) + inv = !inv; + return inv; +} + +static int match( + const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop +) { + const struct ipt_pool_info *info = matchinfo; + const struct iphdr *iph = skb->nh.iph; + + if (info->src != IP_POOL_NONE && !match_pool(info->src, iph->saddr, + info->flags&IPT_POOL_INV_SRC)) + return 0; + + if (info->dst != IP_POOL_NONE && !match_pool(info->dst, iph->daddr, + info->flags&IPT_POOL_INV_DST)) + return 0; + + return 1; +} + +static int checkentry( + const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask +) { + if (matchsize != IPT_ALIGN(sizeof(struct ipt_pool_info))) + return 0; + return 1; +} + +static struct ipt_match pool_match += { { NULL, NULL }, "pool", &match, &checkentry, NULL, THIS_MODULE }; + +static int __init init(void) +{ + return ipt_register_match(&pool_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&pool_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_psd.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_psd.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_psd.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_psd.c 2004-04-19 10:08:37.000000000 +0200 @@ -0,0 +1,361 @@ +/* + This is a module which is used for PSD (portscan detection) + Derived from scanlogd v2.1 written by Solar Designer + and LOG target module. + + Copyright (C) 2000,2001 astaro AG + + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 2000-05-04 Markus Hennig : initial + 2000-08-18 Dennis Koslowski : first release + 2000-12-01 Dennis Koslowski : UDP scans detection added + 2001-01-02 Dennis Koslowski : output modified + 2001-02-04 Jan Rekorajski : converted from target to match +*/ + +#include +#include +#include +#include +#include +#include +#include + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Dennis Koslowski "); + +#define HF_DADDR_CHANGING 0x01 +#define HF_SPORT_CHANGING 0x02 +#define HF_TOS_CHANGING 0x04 +#define HF_TTL_CHANGING 0x08 + +/* + * Information we keep per each target port + */ +struct port { + u_int16_t number; /* port number */ + u_int8_t proto; /* protocol number */ + u_int8_t and_flags; /* tcp ANDed flags */ + u_int8_t or_flags; /* tcp ORed flags */ +}; + +/* + * Information we keep per each source address. + */ +struct host { + struct host *next; /* Next entry with the same hash */ + clock_t timestamp; /* Last update time */ + struct in_addr src_addr; /* Source address */ + struct in_addr dest_addr; /* Destination address */ + unsigned short src_port; /* Source port */ + int count; /* Number of ports in the list */ + int weight; /* Total weight of ports in the list */ + struct port ports[SCAN_MAX_COUNT - 1]; /* List of ports */ + unsigned char tos; /* TOS */ + unsigned char ttl; /* TTL */ + unsigned char flags; /* HF_ flags bitmask */ +}; + +/* + * State information. + */ +static struct { + spinlock_t lock; + struct host list[LIST_SIZE]; /* List of source addresses */ + struct host *hash[HASH_SIZE]; /* Hash: pointers into the list */ + int index; /* Oldest entry to be replaced */ +} state; + +/* + * Convert an IP address into a hash table index. + */ +static inline int hashfunc(struct in_addr addr) +{ + unsigned int value; + int hash; + + value = addr.s_addr; + hash = 0; + do { + hash ^= value; + } while ((value >>= HASH_LOG)); + + return hash & (HASH_SIZE - 1); +} + +static int +ipt_psd_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + struct iphdr *ip_hdr; + struct tcphdr *tcp_hdr; + struct in_addr addr; + u_int16_t src_port,dest_port; + u_int8_t tcp_flags, proto; + clock_t now; + struct host *curr, *last, **head; + int hash, index, count; + + /* Parameters from userspace */ + const struct ipt_psd_info *psdinfo = matchinfo; + + /* IP header */ + ip_hdr = pskb->nh.iph; + + /* Sanity check */ + if (ntohs(ip_hdr->frag_off) & IP_OFFSET) { + DEBUGP("PSD: sanity check failed\n"); + return 0; + } + + /* TCP or UDP ? */ + proto = ip_hdr->protocol; + + if (proto != IPPROTO_TCP && proto != IPPROTO_UDP) { + DEBUGP("PSD: protocol not supported\n"); + return 0; + } + + /* Get the source address, source & destination ports, and TCP flags */ + + addr.s_addr = ip_hdr->saddr; + + tcp_hdr = (struct tcphdr*)((u_int32_t *)ip_hdr + ip_hdr->ihl); + + /* Yep, itīs dirty */ + src_port = tcp_hdr->source; + dest_port = tcp_hdr->dest; + + if (proto == IPPROTO_TCP) { + tcp_flags = *((u_int8_t*)tcp_hdr + 13); + } + else { + tcp_flags = 0x00; + } + + /* We're using IP address 0.0.0.0 for a special purpose here, so don't let + * them spoof us. [DHCP needs this feature - HW] */ + if (!addr.s_addr) { + DEBUGP("PSD: spoofed source address (0.0.0.0)\n"); + return 0; + } + + /* Use jiffies here not to depend on someone setting the time while we're + * running; we need to be careful with possible return value overflows. */ + now = jiffies; + + spin_lock(&state.lock); + + /* Do we know this source address already? */ + count = 0; + last = NULL; + if ((curr = *(head = &state.hash[hash = hashfunc(addr)]))) + do { + if (curr->src_addr.s_addr == addr.s_addr) break; + count++; + if (curr->next) last = curr; + } while ((curr = curr->next)); + + if (curr) { + + /* We know this address, and the entry isn't too old. Update it. */ + if (now - curr->timestamp <= (psdinfo->delay_threshold*HZ)/100 && + time_after_eq(now, curr->timestamp)) { + + /* Just update the appropriate list entry if we've seen this port already */ + for (index = 0; index < curr->count; index++) { + if (curr->ports[index].number == dest_port) { + curr->ports[index].proto = proto; + curr->ports[index].and_flags &= tcp_flags; + curr->ports[index].or_flags |= tcp_flags; + goto out_no_match; + } + } + + /* TCP/ACK and/or TCP/RST to a new port? This could be an outgoing connection. */ + if (proto == IPPROTO_TCP && (tcp_hdr->ack || tcp_hdr->rst)) + goto out_no_match; + + /* Packet to a new port, and not TCP/ACK: update the timestamp */ + curr->timestamp = now; + + /* Logged this scan already? Then drop the packet. */ + if (curr->weight >= psdinfo->weight_threshold) + goto out_match; + + /* Specify if destination address, source port, TOS or TTL are not fixed */ + if (curr->dest_addr.s_addr != ip_hdr->daddr) + curr->flags |= HF_DADDR_CHANGING; + if (curr->src_port != src_port) + curr->flags |= HF_SPORT_CHANGING; + if (curr->tos != ip_hdr->tos) + curr->flags |= HF_TOS_CHANGING; + if (curr->ttl != ip_hdr->ttl) + curr->flags |= HF_TTL_CHANGING; + + /* Update the total weight */ + curr->weight += (ntohs(dest_port) < 1024) ? + psdinfo->lo_ports_weight : psdinfo->hi_ports_weight; + + /* Got enough destination ports to decide that this is a scan? */ + /* Then log it and drop the packet. */ + if (curr->weight >= psdinfo->weight_threshold) + goto out_match; + + /* Remember the new port */ + if (curr->count < SCAN_MAX_COUNT) { + curr->ports[curr->count].number = dest_port; + curr->ports[curr->count].proto = proto; + curr->ports[curr->count].and_flags = tcp_flags; + curr->ports[curr->count].or_flags = tcp_flags; + curr->count++; + } + + goto out_no_match; + } + + /* We know this address, but the entry is outdated. Mark it unused, and + * remove from the hash table. We'll allocate a new entry instead since + * this one might get re-used too soon. */ + curr->src_addr.s_addr = 0; + if (last) + last->next = last->next->next; + else if (*head) + *head = (*head)->next; + last = NULL; + } + + /* We don't need an ACK from a new source address */ + if (proto == IPPROTO_TCP && tcp_hdr->ack) + goto out_no_match; + + /* Got too many source addresses with the same hash value? Then remove the + * oldest one from the hash table, so that they can't take too much of our + * CPU time even with carefully chosen spoofed IP addresses. */ + if (count >= HASH_MAX && last) last->next = NULL; + + /* We're going to re-use the oldest list entry, so remove it from the hash + * table first (if it is really already in use, and isn't removed from the + * hash table already because of the HASH_MAX check above). */ + + /* First, find it */ + if (state.list[state.index].src_addr.s_addr) + head = &state.hash[hashfunc(state.list[state.index].src_addr)]; + else + head = &last; + last = NULL; + if ((curr = *head)) + do { + if (curr == &state.list[state.index]) break; + last = curr; + } while ((curr = curr->next)); + + /* Then, remove it */ + if (curr) { + if (last) + last->next = last->next->next; + else if (*head) + *head = (*head)->next; + } + + /* Get our list entry */ + curr = &state.list[state.index++]; + if (state.index >= LIST_SIZE) state.index = 0; + + /* Link it into the hash table */ + head = &state.hash[hash]; + curr->next = *head; + *head = curr; + + /* And fill in the fields */ + curr->timestamp = now; + curr->src_addr = addr; + curr->dest_addr.s_addr = ip_hdr->daddr; + curr->src_port = src_port; + curr->count = 1; + curr->weight = (ntohs(dest_port) < 1024) ? + psdinfo->lo_ports_weight : psdinfo->hi_ports_weight; + curr->ports[0].number = dest_port; + curr->ports[0].proto = proto; + curr->ports[0].and_flags = tcp_flags; + curr->ports[0].or_flags = tcp_flags; + curr->tos = ip_hdr->tos; + curr->ttl = ip_hdr->ttl; + +out_no_match: + spin_unlock(&state.lock); + return 0; + +out_match: + spin_unlock(&state.lock); + return 1; +} + +static int ipt_psd_checkentry(const char *tablename, + const struct ipt_ip *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ +/* const struct ipt_psd_info *psdinfo = targinfo;*/ + + /* we accept TCP only */ +/* if (e->ip.proto != IPPROTO_TCP) { */ +/* DEBUGP("PSD: specified protocol may be TCP only\n"); */ +/* return 0; */ +/* } */ + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_psd_info))) { + DEBUGP("PSD: matchsize %u != %u\n", + matchsize, + IPT_ALIGN(sizeof(struct ipt_psd_info))); + return 0; + } + + return 1; +} + +static struct ipt_match ipt_psd_reg = { + {NULL, NULL}, + "psd", + ipt_psd_match, + ipt_psd_checkentry, + NULL, + THIS_MODULE }; + +static int __init init(void) +{ + if (ipt_register_match(&ipt_psd_reg)) + return -EINVAL; + + memset(&state, 0, sizeof(state)); + + spin_lock_init(&(state.lock)); + + printk("netfilter PSD loaded - (c) astaro AG\n"); + return 0; +} + +static void __exit fini(void) +{ + ipt_unregister_match(&ipt_psd_reg); + printk("netfilter PSD unloaded - (c) astaro AG\n"); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_quota.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_quota.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_quota.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_quota.c 2004-04-19 10:08:37.000000000 +0200 @@ -0,0 +1,91 @@ +/* + * netfilter module to enforce network quotas + * + * Sam Johnston + */ +#include +#include +#include +#include + +#include +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Sam Johnston "); + +static spinlock_t quota_lock = SPIN_LOCK_UNLOCKED; + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, int *hotdrop) +{ + struct ipt_quota_info *q = (struct ipt_quota_info *) matchinfo; + unsigned int datalen; + + if (skb->len < sizeof(struct iphdr)) + return NF_ACCEPT; + + datalen = skb->len - skb->nh.iph->ihl*4; + + spin_lock_bh("a_lock); + + if (q->quota >= datalen) { + /* we can afford this one */ + q->quota -= datalen; + spin_unlock_bh("a_lock); + +#ifdef DEBUG_IPT_QUOTA + printk("IPT Quota OK: %llu datlen %d \n", q->quota, datalen); +#endif + return 1; + } + + /* so we do not allow even small packets from now on */ + q->quota = 0; + +#ifdef DEBUG_IPT_QUOTA + printk("IPT Quota Failed: %llu datlen %d \n", q->quota, datalen); +#endif + + spin_unlock_bh("a_lock); + return 0; +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, unsigned int matchsize, unsigned int hook_mask) +{ + /* TODO: spinlocks? sanity checks? */ + if (matchsize != IPT_ALIGN(sizeof (struct ipt_quota_info))) + return 0; + + return 1; +} + +static struct ipt_match quota_match = { + .name = "quota", + .match = match, + .checkentry = checkentry, + .me = THIS_MODULE +}; + +static int __init +init(void) +{ + return ipt_register_match("a_match); +} + +static void __exit +fini(void) +{ + ipt_unregister_match("a_match); +} + +module_init(init); +module_exit(fini); + diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_random.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_random.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_random.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_random.c 2004-04-19 10:08:38.000000000 +0200 @@ -0,0 +1,96 @@ +/* + This is a module which is used for a "random" match support. + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 2001-10-14 Fabrice MARIE : initial implementation. +*/ + +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); + +static int +ipt_rand_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + /* Parameters from userspace */ + const struct ipt_rand_info *info = matchinfo; + u_int8_t random_number; + + /* get 1 random number from the kernel random number generation routine */ + get_random_bytes((void *)(&random_number), 1); + + /* Do we match ? */ + if (random_number <= info->average) + return 1; + else + return 0; +} + +static int +ipt_rand_checkentry(const char *tablename, + const struct ipt_ip *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + /* Parameters from userspace */ + const struct ipt_rand_info *info = matchinfo; + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_rand_info))) { + printk("ipt_random: matchsize %u != %u\n", matchsize, + IPT_ALIGN(sizeof(struct ipt_rand_info))); + return 0; + } + + /* must be 1 <= average % <= 99 */ + /* 1 x 2.55 = 2 */ + /* 99 x 2.55 = 252 */ + if ((info->average < 2) || (info->average > 252)) { + printk("ipt_random: invalid average %u\n", info->average); + return 0; + } + + return 1; +} + +static struct ipt_match ipt_rand_reg = { + {NULL, NULL}, + "random", + ipt_rand_match, + ipt_rand_checkentry, + NULL, + THIS_MODULE }; + +static int __init init(void) +{ + if (ipt_register_match(&ipt_rand_reg)) + return -EINVAL; + + printk("ipt_random match loaded\n"); + return 0; +} + +static void __exit fini(void) +{ + ipt_unregister_match(&ipt_rand_reg); + printk("ipt_random match unloaded\n"); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_realm.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_realm.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_realm.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_realm.c 2004-04-19 10:08:39.000000000 +0200 @@ -0,0 +1,78 @@ +/* IP tables module for matching the routing realm + * + * $Id$ + * + * (C) 2003 by Sampsa Ranta + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Sampsa Ranta "); +MODULE_LICENSE("GPL"); + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_realm_info *info = matchinfo; + struct dst_entry *dst = skb->dst; + + if (!dst) + return 0; + + return (info->id == (dst->tclassid & info->mask)) ^ info->invert; +} + +static int check(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + if (hook_mask + & ~((1 << NF_IP_POST_ROUTING) | (1 << NF_IP_FORWARD) | + (1 << NF_IP_LOCAL_OUT)| (1 << NF_IP_LOCAL_IN))) { + printk("ipt_realm: only valid for POST_ROUTING, LOCAL_OUT, " + "LOCAL_IN or FORWARD.\n"); + return 0; + } + + if (matchsize != IPT_ALIGN(sizeof(struct ipt_realm_info))) + return 0; + + return 1; +} + +static struct ipt_match realm_match = { + .name = "realm", + .match = match, + .checkentry = check, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&realm_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&realm_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_sctp.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_sctp.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_sctp.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_sctp.c 2004-04-19 10:08:40.000000000 +0200 @@ -0,0 +1,199 @@ +#include +#include +#include +#include + +#include +#include + +#if 0 +#define duprintf(format, args...) printk(format , ## args) +#else +#define duprintf(format, args...) +#endif + +#define SCCHECK(cond, option, flag, invflag) (!((flag) & (option)) \ + || (!!((invflag) & (option)) ^ (cond))) + +static int +match_flags(const struct ipt_sctp_flag_info *flag_info, + const int flag_count, + u_int8_t chunktype, + u_int8_t chunkflags) +{ + int i; + + for (i = 0; i < flag_count; i++) { + if (flag_info[i].chunktype == chunktype) { + return (chunkflags & flag_info[i].flag_mask) == flag_info[i].flag; + } + } + + return 1; +} + +static int +match_packet(const struct sk_buff *skb, + const u_int32_t *chunkmap, + int chunk_match_type, + const struct ipt_sctp_flag_info *flag_info, + const int flag_count, + int *hotdrop) +{ + int offset; + u_int32_t chunkmapcopy[256 / sizeof (u_int32_t)]; + sctp_chunkhdr_t sch; + + int i = 0; + + if (chunk_match_type == SCTP_CHUNK_MATCH_ALL) { + SCTP_CHUNKMAP_COPY(chunkmapcopy, chunkmap); + } + + offset = skb->nh.iph->ihl * 4 + sizeof (sctp_sctphdr_t); + do { + if (skb_copy_bits(skb, offset, &sch, sizeof(sch)) < 0) { + duprintf("Dropping invalid SCTP packet.\n"); + *hotdrop = 1; + return 0; + } + + duprintf("Chunk num: %d\toffset: %d\ttype: %d\tlength: %d\tflags: %x\n", + ++i, offset, sch.type, htons(sch.length), sch.flags); + + offset += (htons(sch.length) + 3) & ~3; + + duprintf("skb->len: %d\toffset: %d\n", skb->len, offset); + + if (SCTP_CHUNKMAP_IS_SET(chunkmap, sch.type)) { + switch (chunk_match_type) { + case SCTP_CHUNK_MATCH_ANY: + if (match_flags(flag_info, flag_count, + sch.type, sch.flags)) { + return 1; + } + break; + + case SCTP_CHUNK_MATCH_ALL: + if (match_flags(flag_info, flag_count, + sch.type, sch.flags)) { + SCTP_CHUNKMAP_CLEAR(chunkmapcopy, sch.type); + } + break; + + case SCTP_CHUNK_MATCH_ONLY: + if (!match_flags(flag_info, flag_count, + sch.type, sch.flags)) { + return 0; + } + break; + } + } else { + switch (chunk_match_type) { + case SCTP_CHUNK_MATCH_ONLY: + return 0; + } + } + } while (offset < skb->len); + + switch (chunk_match_type) { + case SCTP_CHUNK_MATCH_ALL: + return SCTP_CHUNKMAP_IS_CLEAR(chunkmap); + case SCTP_CHUNK_MATCH_ANY: + return 0; + case SCTP_CHUNK_MATCH_ONLY: + return 1; + } + + /* This will never be reached, but required to stop compiler whine */ + return 0; +} + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + int *hotdrop) +{ + const struct ipt_sctp_info *info; + sctp_sctphdr_t sh; + + info = (const struct ipt_sctp_info *)matchinfo; + + if (offset) { + duprintf("Dropping non-first fragment.. FIXME\n"); + return 0; + } + + if (skb_copy_bits(skb, skb->nh.iph->ihl*4, &sh, sizeof(sh)) < 0) { + duprintf("Dropping evil TCP offset=0 tinygram.\n"); + *hotdrop = 1; + return 0; + } + duprintf("spt: %d\tdpt: %d\n", ntohs(sh.source), ntohs(sh.dest)); + + return SCCHECK(((ntohs(sh.source) >= info->spts[0]) + && (ntohs(sh.source) <= info->spts[1])), + IPT_SCTP_SRC_PORTS, info->flags, info->invflags) + && SCCHECK(((ntohs(sh.dest) >= info->dpts[0]) + && (ntohs(sh.dest) <= info->dpts[1])), + IPT_SCTP_DEST_PORTS, info->flags, info->invflags) + && SCCHECK(match_packet(skb, info->chunkmap, info->chunk_match_type, + info->flag_info, info->flag_count, + hotdrop), + IPT_SCTP_CHUNK_TYPES, info->flags, info->invflags); +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ipt_sctp_info *info; + + info = (const struct ipt_sctp_info *)matchinfo; + + return ip->proto == IPPROTO_SCTP + && !(ip->invflags & IPT_INV_PROTO) + && matchsize == IPT_ALIGN(sizeof(struct ipt_sctp_info)) + && !(info->flags & ~IPT_SCTP_VALID_FLAGS) + && !(info->invflags & ~IPT_SCTP_VALID_FLAGS) + && !(info->invflags & ~info->flags) + && ((!(info->flags & IPT_SCTP_CHUNK_TYPES)) || + (info->chunk_match_type & + (SCTP_CHUNK_MATCH_ALL + | SCTP_CHUNK_MATCH_ANY + | SCTP_CHUNK_MATCH_ONLY))); +} + +static struct ipt_match sctp_match = +{ + .list = { NULL, NULL}, + .name = "sctp", + .match = &match, + .checkentry = &checkentry, + .destroy = NULL, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + return ipt_register_match(&sctp_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&sctp_match); +} + +module_init(init); +module_exit(fini); + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Kiran Kumar Immidi"); +MODULE_DESCRIPTION("Match for SCTP protocol packets"); + diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_time.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_time.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_time.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_time.c 2004-04-19 10:08:41.000000000 +0200 @@ -0,0 +1,185 @@ +/* + This is a module which is used for time matching + It is using some modified code from dietlibc (localtime() function) + that you can find at http://www.fefe.de/dietlibc/ + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: ftp://prep.ai.mit.edu/pub/gnu/GPL + 2001-05-04 Fabrice MARIE : initial development. + 2001-21-05 Fabrice MARIE : bug fix in the match code, + thanks to "Zeng Yu" for bug report. + 2001-26-09 Fabrice MARIE : force the match to be in LOCAL_IN or PRE_ROUTING only. + 2001-30-11 Fabrice : added the possibility to use the match in FORWARD/OUTPUT with a little hack, + added Nguyen Dang Phuoc Dong patch to support timezones. +*/ + +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Fabrice MARIE "); +MODULE_DESCRIPTION("Match arrival timestamp"); +MODULE_LICENSE("GPL"); + +struct tm +{ + int tm_sec; /* Seconds. [0-60] (1 leap second) */ + int tm_min; /* Minutes. [0-59] */ + int tm_hour; /* Hours. [0-23] */ + int tm_mday; /* Day. [1-31] */ + int tm_mon; /* Month. [0-11] */ + int tm_year; /* Year - 1900. */ + int tm_wday; /* Day of week. [0-6] */ + int tm_yday; /* Days in year.[0-365] */ + int tm_isdst; /* DST. [-1/0/1]*/ + + long int tm_gmtoff; /* we don't care, we count from GMT */ + const char *tm_zone; /* we don't care, we count from GMT */ +}; + +void +localtime(const time_t *timepr, struct tm *r); + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + const struct ipt_time_info *info = matchinfo; /* match info for rule */ + struct tm currenttime; /* time human readable */ + u_int8_t days_of_week[7] = {64, 32, 16, 8, 4, 2, 1}; + u_int16_t packet_time; + struct timeval kerneltimeval; + time_t packet_local_time; + + /* if kerneltime=1, we don't read the skb->timestamp but kernel time instead */ + if (info->kerneltime) + { + do_gettimeofday(&kerneltimeval); + packet_local_time = kerneltimeval.tv_sec; + } + else + packet_local_time = skb->stamp.tv_sec; + + /* Transform the timestamp of the packet, in a human readable form */ + localtime(&packet_local_time, ¤ttime); + + /* check if we match this timestamp, we start by the days... */ + if ((days_of_week[currenttime.tm_wday] & info->days_match) != days_of_week[currenttime.tm_wday]) + return 0; /* the day doesn't match */ + + /* ... check the time now */ + packet_time = (currenttime.tm_hour * 60) + currenttime.tm_min; + if ((packet_time < info->time_start) || (packet_time > info->time_stop)) + return 0; + + /* here we match ! */ + return 1; +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + struct ipt_time_info *info = matchinfo; /* match info for rule */ + + /* First, check that we are in the correct hook */ + /* PRE_ROUTING, LOCAL_IN or FROWARD */ + if (hook_mask + & ~((1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD) | (1 << NF_IP_LOCAL_OUT))) + { + printk("ipt_time: error, only valid for PRE_ROUTING, LOCAL_IN, FORWARD and OUTPUT)\n"); + return 0; + } + /* we use the kerneltime if we are in forward or output */ + info->kerneltime = 1; + if (hook_mask & ~((1 << NF_IP_FORWARD) | (1 << NF_IP_LOCAL_OUT))) + /* if not, we use the skb time */ + info->kerneltime = 0; + + /* Check the size */ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_time_info))) + return 0; + /* Now check the coherence of the data ... */ + if ((info->time_start > 1439) || /* 23*60+59 = 1439*/ + (info->time_stop > 1439)) + { + printk(KERN_WARNING "ipt_time: invalid argument\n"); + return 0; + } + + return 1; +} + +static struct ipt_match time_match += { { NULL, NULL }, "time", &match, &checkentry, NULL, THIS_MODULE }; + +static int __init init(void) +{ + printk("ipt_time loading\n"); + return ipt_register_match(&time_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&time_match); + printk("ipt_time unloaded\n"); +} + +module_init(init); +module_exit(fini); + + +/* The part below is borowed and modified from dietlibc */ + +/* seconds per day */ +#define SPD 24*60*60 + +void +localtime(const time_t *timepr, struct tm *r) { + time_t i; + time_t timep; + extern struct timezone sys_tz; + const unsigned int __spm[12] = + { 0, + (31), + (31+28), + (31+28+31), + (31+28+31+30), + (31+28+31+30+31), + (31+28+31+30+31+30), + (31+28+31+30+31+30+31), + (31+28+31+30+31+30+31+31), + (31+28+31+30+31+30+31+31+30), + (31+28+31+30+31+30+31+31+30+31), + (31+28+31+30+31+30+31+31+30+31+30), + }; + register time_t work; + + timep = (*timepr) - (sys_tz.tz_minuteswest * 60); + work=timep%(SPD); + r->tm_sec=work%60; work/=60; + r->tm_min=work%60; r->tm_hour=work/60; + work=timep/(SPD); + r->tm_wday=(4+work)%7; + for (i=1970; ; ++i) { + register time_t k= (!(i%4) && ((i%100) || !(i%400)))?366:365; + if (work>k) + work-=k; + else + break; + } + r->tm_year=i-1900; + for (i=11; i && __spm[i]>work; --i) ; + r->tm_mon=i; + r->tm_mday=work-__spm[i]+1; +} diff -Nur linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_u32.c linux-2.6.6-rc1/net/ipv4/netfilter/ipt_u32.c --- linux-2.6.6-rc1.org/net/ipv4/netfilter/ipt_u32.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv4/netfilter/ipt_u32.c 2004-04-19 10:08:46.000000000 +0200 @@ -0,0 +1,211 @@ +/* Kernel module to match u32 packet content. */ + +/* +U32 tests whether quantities of up to 4 bytes extracted from a packet +have specified values. The specification of what to extract is general +enough to find data at given offsets from tcp headers or payloads. + + --u32 tests + The argument amounts to a program in a small language described below. + tests := location = value | tests && location = value + value := range | value , range + range := number | number : number + a single number, n, is interpreted the same as n:n + n:m is interpreted as the range of numbers >=n and <=m + location := number | location operator number + operator := & | << | >> | @ + + The operators &, <<, >>, && mean the same as in c. The = is really a set + membership operator and the value syntax describes a set. The @ operator + is what allows moving to the next header and is described further below. + + *** Until I can find out how to avoid it, there are some artificial limits + on the size of the tests: + - no more than 10 ='s (and 9 &&'s) in the u32 argument + - no more than 10 ranges (and 9 commas) per value + - no more than 10 numbers (and 9 operators) per location + + To describe the meaning of location, imagine the following machine that + interprets it. There are three registers: + A is of type char*, initially the address of the IP header + B and C are unsigned 32 bit integers, initially zero + + The instructions are: + number B = number; + C = (*(A+B)<<24)+(*(A+B+1)<<16)+(*(A+B+2)<<8)+*(A+B+3) + &number C = C&number + <>number C = C>>number + @number A = A+C; then do the instruction number + Any access of memory outside [skb->head,skb->end] causes the match to fail. + Otherwise the result of the computation is the final value of C. + + Whitespace is allowed but not required in the tests. + However the characters that do occur there are likely to require + shell quoting, so it's a good idea to enclose the arguments in quotes. + +Example: + match IP packets with total length >= 256 + The IP header contains a total length field in bytes 2-3. + --u32 "0&0xFFFF=0x100:0xFFFF" + read bytes 0-3 + AND that with FFFF (giving bytes 2-3), + and test whether that's in the range [0x100:0xFFFF] + +Example: (more realistic, hence more complicated) + match icmp packets with icmp type 0 + First test that it's an icmp packet, true iff byte 9 (protocol) = 1 + --u32 "6&0xFF=1 && ... + read bytes 6-9, use & to throw away bytes 6-8 and compare the result to 1 + Next test that it's not a fragment. + (If so it might be part of such a packet but we can't always tell.) + n.b. This test is generally needed if you want to match anything + beyond the IP header. + The last 6 bits of byte 6 and all of byte 7 are 0 iff this is a complete + packet (not a fragment). Alternatively, you can allow first fragments + by only testing the last 5 bits of byte 6. + ... 4&0x3FFF=0 && ... + Last test: the first byte past the IP header (the type) is 0 + This is where we have to use the @syntax. The length of the IP header + (IHL) in 32 bit words is stored in the right half of byte 0 of the + IP header itself. + ... 0>>22&0x3C@0>>24=0" + The first 0 means read bytes 0-3, + >>22 means shift that 22 bits to the right. Shifting 24 bits would give + the first byte, so only 22 bits is four times that plus a few more bits. + &3C then eliminates the two extra bits on the right and the first four + bits of the first byte. + For instance, if IHL=5 then the IP header is 20 (4 x 5) bytes long. + In this case bytes 0-1 are (in binary) xxxx0101 yyzzzzzz, + >>22 gives the 10 bit value xxxx0101yy and &3C gives 010100. + @ means to use this number as a new offset into the packet, and read + four bytes starting from there. This is the first 4 bytes of the icmp + payload, of which byte 0 is the icmp type. Therefore we simply shift + the value 24 to the right to throw out all but the first byte and compare + the result with 0. + +Example: + tcp payload bytes 8-12 is any of 1, 2, 5 or 8 + First we test that the packet is a tcp packet (similar to icmp). + --u32 "6&0xFF=6 && ... + Next, test that it's not a fragment (same as above). + ... 0>>22&0x3C@12>>26&0x3C@8=1,2,5,8" + 0>>22&3C as above computes the number of bytes in the IP header. + @ makes this the new offset into the packet, which is the start of the + tcp header. The length of the tcp header (again in 32 bit words) is + the left half of byte 12 of the tcp header. The 12>>26&3C + computes this length in bytes (similar to the IP header before). + @ makes this the new offset, which is the start of the tcp payload. + Finally 8 reads bytes 8-12 of the payload and = checks whether the + result is any of 1, 2, 5 or 8 +*/ + +#include +#include + +#include +#include + +/* #include for timing */ + +MODULE_AUTHOR("Don Cohen "); +MODULE_DESCRIPTION("IP tables u32 matching module"); +MODULE_LICENSE("GPL"); + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + const struct ipt_u32 *data = matchinfo; + int testind, i; + unsigned char* origbase = (char*)skb->nh.iph; + unsigned char* base = origbase; + unsigned char* head = skb->head; + unsigned char* end = skb->end; + int nnums, nvals; + u_int32_t pos, val; + /* unsigned long long cycles1, cycles2, cycles3, cycles4; + cycles1 = get_cycles(); */ + + for (testind=0; testind < data->ntests; testind++) { + base = origbase; /* reset for each test */ + pos = data->tests[testind].location[0].number; + if (base+pos+3 > end || base+pos < head) + return 0; + val = (base[pos]<<24) + (base[pos+1]<<16) + + (base[pos+2]<<8) + base[pos+3]; + nnums = data->tests[testind].nnums; + for (i=1; i < nnums; i++) { + u_int32_t number = data->tests[testind].location[i].number; + switch (data->tests[testind].location[i].nextop) { + case IPT_U32_AND: + val = val & number; + break; + case IPT_U32_LEFTSH: + val = val << number; + break; + case IPT_U32_RIGHTSH: + val = val >> number; + break; + case IPT_U32_AT: + base = base + val; + pos = number; + if (base+pos+3 > end || base+pos < head) + return 0; + val = (base[pos]<<24) + (base[pos+1]<<16) + + (base[pos+2]<<8) + base[pos+3]; + break; + } + } + nvals = data->tests[testind].nvalues; + for (i=0; i < nvals; i++) { + if ((data->tests[testind].value[i].min <= val) && + (val <= data->tests[testind].value[i].max)) { + break; + } + } + if (i >= data->tests[testind].nvalues) { + /* cycles2 = get_cycles(); + printk("failed %d in %d cycles\n", testind, + cycles2-cycles1); */ + return 0; + } + } + /* cycles2 = get_cycles(); + printk("succeeded in %d cycles\n", cycles2-cycles1); */ + return 1; +} + +static int +checkentry(const char *tablename, + const struct ipt_ip *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + if (matchsize != IPT_ALIGN(sizeof(struct ipt_u32))) + return 0; + return 1; +} + +static struct ipt_match u32_match += { { NULL, NULL }, "u32", &match, &checkentry, NULL, THIS_MODULE }; + +static int __init init(void) +{ + return ipt_register_match(&u32_match); +} + +static void __exit fini(void) +{ + ipt_unregister_match(&u32_match); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/ip6_tunnel.c linux-2.6.6-rc1/net/ipv6/ip6_tunnel.c --- linux-2.6.6-rc1.org/net/ipv6/ip6_tunnel.c 2004-04-15 03:33:53.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv6/ip6_tunnel.c 2004-04-19 10:08:24.000000000 +0200 @@ -715,13 +715,7 @@ ipv6h->nexthdr = proto; ipv6_addr_copy(&ipv6h->saddr, &fl.fl6_src); ipv6_addr_copy(&ipv6h->daddr, &fl.fl6_dst); -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); pkt_len = skb->len; err = NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, skb, NULL, skb->dst->dev, dst_output); diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/Kconfig linux-2.6.6-rc1/net/ipv6/netfilter/Kconfig --- linux-2.6.6-rc1.org/net/ipv6/netfilter/Kconfig 2004-04-19 09:59:33.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv6/netfilter/Kconfig 2004-04-19 10:08:38.000000000 +0200 @@ -230,5 +230,30 @@ . If unsure, say `N'. help +config IP6_NF_TARGET_HL + tristate 'HL target support' + depends on IP6_NF_MANGLE + help + +config IP6_NF_TARGET_REJECT + tristate 'REJECT target support' + depends on IP6_NF_FILTER + help + +config IP6_NF_MATCH_FUZZY + tristate 'Fuzzy match support' + depends on IP6_NF_FILTER + help + +config IP6_NF_MATCH_NTH + tristate 'Nth match support' + depends on IP6_NF_IPTABLES + help + +config IP6_NF_MATCH_RANDOM + tristate 'Random match support' + depends on IP6_NF_IPTABLES + help + endmenu diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/Makefile linux-2.6.6-rc1/net/ipv6/netfilter/Makefile --- linux-2.6.6-rc1.org/net/ipv6/netfilter/Makefile 2004-04-19 09:59:33.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv6/netfilter/Makefile 2004-04-19 10:08:38.000000000 +0200 @@ -8,6 +8,7 @@ obj-$(CONFIG_IP6_NF_MATCH_MARK) += ip6t_mark.o obj-$(CONFIG_IP6_NF_MATCH_LENGTH) += ip6t_length.o obj-$(CONFIG_IP6_NF_MATCH_MAC) += ip6t_mac.o +obj-$(CONFIG_IP6_NF_MATCH_FUZZY) += ip6t_fuzzy.o obj-$(CONFIG_IP6_NF_MATCH_RT) += ip6t_rt.o obj-$(CONFIG_IP6_NF_MATCH_OPTS) += ip6t_hbh.o ip6t_dst.o obj-$(CONFIG_IP6_NF_MATCH_IPV6HEADER) += ip6t_ipv6header.o @@ -19,7 +20,13 @@ obj-$(CONFIG_IP6_NF_FILTER) += ip6table_filter.o obj-$(CONFIG_IP6_NF_MANGLE) += ip6table_mangle.o obj-$(CONFIG_IP6_NF_TARGET_MARK) += ip6t_MARK.o +obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o obj-$(CONFIG_IP6_NF_QUEUE) += ip6_queue.o obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o + +obj-$(CONFIG_IP6_NF_MATCH_RANDOM) += ip6t_random.o + +obj-$(CONFIG_IP6_NF_MATCH_NTH) += ip6t_nth.o +obj-$(CONFIG_IP6_NF_TARGET_HL) += ip6t_HL.o obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_HL.c linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_HL.c --- linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_HL.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_HL.c 2004-04-19 10:08:26.000000000 +0200 @@ -0,0 +1,105 @@ +/* + * Hop Limit modification target for ip6tables + * Maciej Soltysiak + * Based on HW's TTL module + * + * This software is distributed under the terms of GNU GPL + */ + +#include +#include +#include + +#include +#include + +MODULE_AUTHOR("Maciej Soltysiak "); +MODULE_DESCRIPTION("IP tables Hop Limit modification module"); +MODULE_LICENSE("GPL"); + +static unsigned int ip6t_hl_target(struct sk_buff **pskb, unsigned int hooknum, + const struct net_device *in, const struct net_device *out, + const void *targinfo, void *userinfo) +{ + struct ipv6hdr *ip6h = (*pskb)->nh.ipv6h; + const struct ip6t_HL_info *info = targinfo; + u_int16_t diffs[2]; + int new_hl; + + switch (info->mode) { + case IP6T_HL_SET: + new_hl = info->hop_limit; + break; + case IP6T_HL_INC: + new_hl = ip6h->hop_limit + info->hop_limit; + if (new_hl > 255) + new_hl = 255; + break; + case IP6T_HL_DEC: + new_hl = ip6h->hop_limit + info->hop_limit; + if (new_hl < 0) + new_hl = 0; + break; + default: + new_hl = ip6h->hop_limit; + break; + } + + if (new_hl != ip6h->hop_limit) { + diffs[0] = htons(((unsigned)ip6h->hop_limit) << 8) ^ 0xFFFF; + ip6h->hop_limit = new_hl; + diffs[1] = htons(((unsigned)ip6h->hop_limit) << 8); + } + + return IP6T_CONTINUE; +} + +static int ip6t_hl_checkentry(const char *tablename, + const struct ip6t_entry *e, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + struct ip6t_HL_info *info = targinfo; + + if (targinfosize != IP6T_ALIGN(sizeof(struct ip6t_HL_info))) { + printk(KERN_WARNING "HL: targinfosize %u != %Zu\n", + targinfosize, + IP6T_ALIGN(sizeof(struct ip6t_HL_info))); + return 0; + } + + if (strcmp(tablename, "mangle")) { + printk(KERN_WARNING "HL: can only be called from \"mangle\" table, not \"%s\"\n", tablename); + return 0; + } + + if (info->mode > IP6T_HL_MAXMODE) { + printk(KERN_WARNING "HL: invalid or unknown Mode %u\n", + info->mode); + return 0; + } + + if ((info->mode != IP6T_HL_SET) && (info->hop_limit == 0)) { + printk(KERN_WARNING "HL: increment/decrement doesn't make sense with value 0\n"); + return 0; + } + + return 1; +} + +static struct ip6t_target ip6t_HL = { { NULL, NULL }, "HL", + ip6t_hl_target, ip6t_hl_checkentry, NULL, THIS_MODULE }; + +static int __init init(void) +{ + return ip6t_register_target(&ip6t_HL); +} + +static void __exit fini(void) +{ + ip6t_unregister_target(&ip6t_HL); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_REJECT.c linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_REJECT.c --- linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_REJECT.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_REJECT.c 2004-04-19 10:08:28.000000000 +0200 @@ -0,0 +1,458 @@ +/* + * IP6 tables REJECT target module + * Linux INET6 implementation + * + * Copyright (C)2003 USAGI/WIDE Project + * + * Authors: + * Yasuyuki Kozakai + * + * Based on net/ipv4/netfilter/ipt_REJECT.c + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_AUTHOR("Yasuyuki KOZAKAI "); +MODULE_DESCRIPTION("IP6 tables REJECT target module"); +MODULE_LICENSE("GPL"); + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#if 0 +static void connection_attach(struct sk_buff *new_skb, struct nf_ct_info *nfct) +{ + void (*attach)(struct sk_buff *, struct nf_ct_info *); + if (nfct && (attach = ip6_ct_attach) != NULL) { + mb(); + attach(new_skb, nfct); + } +} +#endif + +static int maybe_reroute(struct sk_buff *skb) +{ + if (skb->nfcache & NFC_ALTERED){ + if (ip6_route_me_harder(skb) != 0){ + kfree_skb(skb); + return -EINVAL; + } + } + + return dst_output(skb); +} + +/* Send RST reply */ +static void send_reset(struct sk_buff *oldskb) +{ + struct sk_buff *nskb; + struct tcphdr otcph, *tcph; + unsigned int otcplen, tcphoff, hh_len; + int needs_ack; + struct ipv6hdr *oip6h = oldskb->nh.ipv6h, *ip6h; + struct dst_entry *dst = NULL; + u8 proto; + struct flowi fl; + proto = oip6h->nexthdr; + int err; + + if ((!(ipv6_addr_type(&oip6h->saddr) & IPV6_ADDR_UNICAST)) || + (!(ipv6_addr_type(&oip6h->daddr) & IPV6_ADDR_UNICAST))) { + DEBUGP("ip6t_REJECT: addr is not unicast.\n"); + return; + } + + tcphoff = ipv6_skip_exthdr(oldskb, ((u8*)(oip6h+1) - oldskb->data), + &proto, oldskb->len - ((u8*)(oip6h+1) + - oldskb->data)); + + if ((tcphoff < 0) || (tcphoff > oldskb->len)) { + DEBUGP("ip6t_REJECT: Can't get TCP header.\n"); + return; + } + + otcplen = oldskb->len - tcphoff; + + /* IP header checks: fragment, too short. */ + if ((proto != IPPROTO_TCP) || (otcplen < sizeof(struct tcphdr))) { + DEBUGP("ip6t_REJECT: proto(%d) != IPPROTO_TCP, or too short. otcplen = %d\n", + proto, otcplen); + return; + } + + if (skb_copy_bits(oldskb, tcphoff, &otcph, sizeof(struct tcphdr))) { + if (net_ratelimit()) + printk("ip6t_REJECT: Can't copy tcp header\n"); + return; + } + + /* No RST for RST. */ + if (otcph.rst) { + DEBUGP("ip6t_REJECT: RST is set\n"); + return; + } + + /* Check checksum. */ + if (csum_ipv6_magic(&oip6h->saddr, &oip6h->daddr, otcplen, IPPROTO_TCP, + skb_checksum(oldskb, tcphoff, otcplen, 0))) { + DEBUGP("ip6t_REJECT: TCP checksum is invalid\n"); + return; + } + + memset(&fl, 0, sizeof(fl)); + fl.proto = IPPROTO_TCP; + ipv6_addr_copy(&fl.fl6_src, &oip6h->daddr); + ipv6_addr_copy(&fl.fl6_dst, &oip6h->saddr); + fl.fl_ip_sport = otcph.dest; + fl.fl_ip_dport = otcph.source; + err = ip6_dst_lookup(NULL, &dst, &fl); + if (err) { + if (net_ratelimit()) + printk("ip6t_REJECT: can't find dst. err = %d\n", err); + return; + } + + hh_len = (dst->dev->hard_header_len + 15)&~15; + nskb = alloc_skb(hh_len + 15 + dst->header_len + sizeof(struct ipv6hdr) + + sizeof(struct tcphdr) + dst->trailer_len, + GFP_ATOMIC); + + if (!nskb) { + if (net_ratelimit()) + printk("ip6t_REJECT: Can't alloc skb\n"); + dst_release(dst); + return; + } + + nskb->dst = dst; + dst_hold(dst); + + skb_reserve(nskb, hh_len + dst->header_len); + + ip6h = nskb->nh.ipv6h = (struct ipv6hdr *) + skb_put(nskb, sizeof(struct ipv6hdr)); + ip6h->version = 6; + ip6h->hop_limit = dst_metric(dst, RTAX_HOPLIMIT); + ip6h->nexthdr = IPPROTO_TCP; + ip6h->payload_len = htons(sizeof(struct tcphdr)); + ipv6_addr_copy(&ip6h->saddr, &oip6h->daddr); + ipv6_addr_copy(&ip6h->daddr, &oip6h->saddr); + + tcph = (struct tcphdr *)skb_put(nskb, sizeof(struct tcphdr)); + /* Truncate to length (no data) */ + tcph->doff = sizeof(struct tcphdr)/4; + tcph->source = otcph.dest; + tcph->dest = otcph.source; + + if (otcph.ack) { + needs_ack = 0; + tcph->seq = otcph.ack_seq; + tcph->ack_seq = 0; + } else { + needs_ack = 1; + tcph->ack_seq = htonl(ntohl(otcph.seq) + otcph.syn + otcph.fin + + otcplen - (otcph.doff<<2)); + tcph->seq = 0; + } + + /* Reset flags */ + ((u_int8_t *)tcph)[13] = 0; + tcph->rst = 1; + tcph->ack = needs_ack; + tcph->window = 0; + tcph->urg_ptr = 0; + tcph->check = 0; + + /* Adjust TCP checksum */ + tcph->check = csum_ipv6_magic(&nskb->nh.ipv6h->saddr, + &nskb->nh.ipv6h->daddr, + sizeof(struct tcphdr), IPPROTO_TCP, + csum_partial((char *)tcph, + sizeof(struct tcphdr), 0)); + +#if 0 + connection_attach(nskb, oldskb->nfct); +#endif + + NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, nskb, NULL, nskb->dst->dev, + maybe_reroute); + + dst_release(dst); +} + +static void send_unreach(struct sk_buff *skb_in, unsigned char code) +{ + struct ipv6hdr *ip6h, *hdr = skb_in->nh.ipv6h; + struct icmp6hdr *icmp6h; + struct dst_entry *dst = NULL; + struct rt6_info *rt; + int tmo; + __u32 csum; + unsigned int len, datalen, hh_len; + int saddr_type, daddr_type; + unsigned int ptr, ip6off; + u8 proto; + struct flowi fl; + struct sk_buff *nskb; + char *data; + + saddr_type = ipv6_addr_type(&hdr->saddr); + daddr_type = ipv6_addr_type(&hdr->daddr); + + if ((!(saddr_type & IPV6_ADDR_UNICAST)) || + (!(daddr_type & IPV6_ADDR_UNICAST))) { + DEBUGP("ip6t_REJECT: addr is not unicast.\n"); + return; + } + + ip6off = skb_in->nh.raw - skb_in->data; + proto = hdr->nexthdr; + ptr = ipv6_skip_exthdr(skb_in, ip6off + sizeof(struct ipv6hdr), &proto, + skb_in->len - ip6off); + + if ((ptr < 0) || (ptr > skb_in->len)) { + ptr = ip6off + sizeof(struct ipv6hdr); + proto = hdr->nexthdr; + } else if (proto == IPPROTO_ICMPV6) { + u8 type; + + if (skb_copy_bits(skb_in, ptr + offsetof(struct icmp6hdr, + icmp6_type), &type, 1)) { + DEBUGP("ip6t_REJECT: Can't get ICMPv6 type\n"); + return; + } + + if (!(type & ICMPV6_INFOMSG_MASK)) { + DEBUGP("ip6t_REJECT: no reply to icmp error\n"); + return; + } + } else if (proto == IPPROTO_UDP) { + int plen = skb_in->len - (ptr - ip6off); + uint16_t check; + + if (plen < sizeof(struct udphdr)) { + DEBUGP("ip6t_REJECT: too short\n"); + return; + } + + if (skb_copy_bits(skb_in, ptr + offsetof(struct udphdr, check), + &check, 2)) { + if (net_ratelimit()) + printk("ip6t_REJECT: can't get copy from skb"); + return; + } + + if (check && + csum_ipv6_magic(&hdr->saddr, &hdr->daddr, plen, + IPPROTO_UDP, + skb_checksum(skb_in, ptr, plen, 0))) { + DEBUGP("ip6t_REJECT: UDP checksum is invalid.\n"); + return; + } + } + + memset(&fl, 0, sizeof(fl)); + fl.proto = IPPROTO_ICMPV6; + ipv6_addr_copy(&fl.fl6_src, &hdr->daddr); + ipv6_addr_copy(&fl.fl6_dst, &hdr->saddr); + fl.fl_icmp_type = ICMPV6_DEST_UNREACH; + fl.fl_icmp_code = code; + + if (ip6_dst_lookup(NULL, &dst, &fl)) { + return; + } + + rt = (struct rt6_info *)dst; + tmo = 1*HZ; + + if (rt->rt6i_dst.plen < 128) + tmo >>= ((128 - rt->rt6i_dst.plen)>>5); + + if (!xrlim_allow(dst, tmo)) { + if (net_ratelimit()) + printk("ip6t_REJECT: rate limitted\n"); + goto dst_release_out; + } + + len = skb_in->len + sizeof(struct ipv6hdr) + sizeof(struct icmp6hdr); + + if (len > dst_pmtu(dst)) + len = dst_pmtu(dst); + if (len > IPV6_MIN_MTU) + len = IPV6_MIN_MTU; + + datalen = len - sizeof(struct ipv6hdr) - sizeof(struct icmp6hdr); + hh_len = (rt->u.dst.dev->hard_header_len + 15)&~15; + + nskb = alloc_skb(hh_len + 15 + dst->header_len + dst->trailer_len + len, + GFP_ATOMIC); + + if (!nskb) { + if (net_ratelimit()) + printk("ip6t_REJECT: can't alloc skb\n"); + goto dst_release_out; + } + + nskb->priority = 0; + nskb->dst = dst; + dst_hold(dst); + + skb_reserve(nskb, hh_len + dst->header_len); + + ip6h = nskb->nh.ipv6h = (struct ipv6hdr *) + skb_put(nskb, sizeof(struct ipv6hdr)); + ip6h->version = 6; + ip6h->hop_limit = dst_metric(dst, RTAX_HOPLIMIT); + ip6h->nexthdr = IPPROTO_ICMPV6; + ip6h->payload_len = htons(datalen + sizeof(struct icmp6hdr)); + ipv6_addr_copy(&ip6h->saddr, &hdr->daddr); + ipv6_addr_copy(&ip6h->daddr, &hdr->saddr); + + icmp6h = (struct icmp6hdr *) skb_put(nskb, sizeof(struct icmp6hdr)); + icmp6h->icmp6_type = ICMPV6_DEST_UNREACH; + icmp6h->icmp6_code = code; + icmp6h->icmp6_cksum = 0; + + data = skb_put(nskb, datalen); + + csum = csum_partial((unsigned char *)icmp6h, sizeof(struct icmp6hdr), 0); + csum = skb_copy_and_csum_bits(skb_in, ip6off, data, datalen, csum); + icmp6h->icmp6_cksum = csum_ipv6_magic(&hdr->saddr, &hdr->daddr, + datalen + sizeof(struct icmp6hdr), + IPPROTO_ICMPV6, csum); + +#if 0 + connection_attach(nskb, skb_in->nfct); +#endif + NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, nskb, NULL, nskb->dst->dev, + maybe_reroute); + +dst_release_out: + dst_release(dst); +} + +static unsigned int reject6_target(struct sk_buff **pskb, + unsigned int hooknum, + const struct net_device *in, + const struct net_device *out, + const void *targinfo, + void *userinfo) +{ + const struct ip6t_reject_info *reject = targinfo; + + DEBUGP(KERN_DEBUG "%s: medium point\n", __FUNCTION__); + /* WARNING: This code causes reentry within ip6tables. + This means that the ip6tables jump stack is now crap. We + must return an absolute verdict. --RR */ + switch (reject->with) { + case IP6T_ICMP6_NO_ROUTE: + send_unreach(*pskb, ICMPV6_NOROUTE); + break; + case IP6T_ICMP6_ADM_PROHIBITED: + send_unreach(*pskb, ICMPV6_ADM_PROHIBITED); + break; + case IP6T_ICMP6_NOT_NEIGHBOUR: + send_unreach(*pskb, ICMPV6_NOT_NEIGHBOUR); + break; + case IP6T_ICMP6_ADDR_UNREACH: + send_unreach(*pskb, ICMPV6_ADDR_UNREACH); + break; + case IP6T_ICMP6_PORT_UNREACH: + send_unreach(*pskb, ICMPV6_PORT_UNREACH); + break; + case IP6T_ICMP6_ECHOREPLY: + /* Do nothing */ + break; + case IP6T_TCP_RESET: + send_reset(*pskb); + break; + default: + if (net_ratelimit()) + printk(KERN_WARNING "ip6t_REJECT: case %u not handled yet\n", reject->with); + break; + } + + return NF_DROP; +} + +static int check(const char *tablename, + const struct ip6t_entry *e, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + const struct ip6t_reject_info *rejinfo = targinfo; + + if (targinfosize != IP6T_ALIGN(sizeof(struct ip6t_reject_info))) { + DEBUGP("ip6t_REJECT: targinfosize %u != 0\n", targinfosize); + return 0; + } + + /* Only allow these for packet filtering. */ + if (strcmp(tablename, "filter") != 0) { + DEBUGP("ip6t_REJECT: bad table `%s'.\n", tablename); + return 0; + } + + if ((hook_mask & ~((1 << NF_IP6_LOCAL_IN) + | (1 << NF_IP6_FORWARD) + | (1 << NF_IP6_LOCAL_OUT))) != 0) { + DEBUGP("ip6t_REJECT: bad hook mask %X\n", hook_mask); + return 0; + } + + if (rejinfo->with == IP6T_ICMP6_ECHOREPLY) { + printk("ip6t_REJECT: ECHOREPLY is not supported.\n"); + return 0; + } else if (rejinfo->with == IP6T_TCP_RESET) { + /* Must specify that it's a TCP packet */ + if (e->ipv6.proto != IPPROTO_TCP + || (e->ipv6.invflags & IP6T_INV_PROTO)) { + DEBUGP("ip6t_REJECT: TCP_RESET illegal for non-tcp\n"); + return 0; + } + } + + return 1; +} + +static struct ip6t_target ip6t_reject_reg = { + .name = "REJECT", + .target = reject6_target, + .checkentry = check, + .me = THIS_MODULE +}; + +static int __init init(void) +{ + if (ip6t_register_target(&ip6t_reject_reg)) + return -EINVAL; + return 0; +} + +static void __exit fini(void) +{ + ip6t_unregister_target(&ip6t_reject_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_fuzzy.c linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_fuzzy.c --- linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_fuzzy.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_fuzzy.c 2004-04-19 10:08:31.000000000 +0200 @@ -0,0 +1,189 @@ +/* + * This module implements a simple TSK FLC + * (Takagi-Sugeno-Kang Fuzzy Logic Controller) that aims + * to limit , in an adaptive and flexible way , the packet rate crossing + * a given stream . It serves as an initial and very simple (but effective) + * example of how Fuzzy Logic techniques can be applied to defeat DoS attacks. + * As a matter of fact , Fuzzy Logic can help us to insert any "behavior" + * into our code in a precise , adaptive and efficient manner. + * The goal is very similar to that of "limit" match , but using techniques of + * Fuzzy Control , that allow us to shape the transfer functions precisely , + * avoiding over and undershoots - and stuff like that . + * + * + * 2002-08-10 Hime Aguiar e Oliveira Jr. : Initial version. + * 2002-08-17 : Changed to eliminate floating point operations . + * 2002-08-23 : Coding style changes . + * 2003-04-08 Maciej Soltysiak : IPv6 Port + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +/* + Packet Acceptance Rate - LOW and Packet Acceptance Rate - HIGH + Expressed in percentage +*/ + +#define PAR_LOW 1/100 +#define PAR_HIGH 1 + +static spinlock_t fuzzy_lock = SPIN_LOCK_UNLOCKED; + +MODULE_AUTHOR("Hime Aguiar e Oliveira Junior "); +MODULE_DESCRIPTION("IP tables Fuzzy Logic Controller match module"); +MODULE_LICENSE("GPL"); + +static u_int8_t mf_high(u_int32_t tx,u_int32_t mini,u_int32_t maxi) +{ + if (tx >= maxi) return 100; + + if (tx <= mini) return 0; + + return ((100 * (tx-mini)) / (maxi-mini)); +} + +static u_int8_t mf_low(u_int32_t tx,u_int32_t mini,u_int32_t maxi) +{ + if (tx <= mini) return 100; + + if (tx >= maxi) return 0; + + return ((100 * (maxi - tx)) / (maxi - mini)); + +} + +static int +ip6t_fuzzy_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + /* From userspace */ + + struct ip6t_fuzzy_info *info = (struct ip6t_fuzzy_info *) matchinfo; + + u_int8_t random_number; + unsigned long amount; + u_int8_t howhigh, howlow; + + + spin_lock_bh(&fuzzy_lock); /* Rise the lock */ + + info->bytes_total += pskb->len; + info->packets_total++; + + info->present_time = jiffies; + + if (info->present_time >= info->previous_time) + amount = info->present_time - info->previous_time; + else { + /* There was a transition : I choose to re-sample + and keep the old acceptance rate... + */ + + amount = 0; + info->previous_time = info->present_time; + info->bytes_total = info->packets_total = 0; + }; + + if ( amount > HZ/10) {/* More than 100 ms elapsed ... */ + + info->mean_rate = (u_int32_t) ((HZ * info->packets_total) \ + / amount); + + info->previous_time = info->present_time; + info->bytes_total = info->packets_total = 0; + + howhigh = mf_high(info->mean_rate,info->minimum_rate,info->maximum_rate); + howlow = mf_low(info->mean_rate,info->minimum_rate,info->maximum_rate); + + info->acceptance_rate = (u_int8_t) \ + (howhigh * PAR_LOW + PAR_HIGH * howlow); + + /* In fact, the above defuzzification would require a denominator + * proportional to (howhigh+howlow) but, in this particular case, + * that expression is constant. + * An imediate consequence is that it is not necessary to call + * both mf_high and mf_low - but to keep things understandable, + * I did so. + */ + + } + + spin_unlock_bh(&fuzzy_lock); /* Release the lock */ + + + if (info->acceptance_rate < 100) + { + get_random_bytes((void *)(&random_number), 1); + + /* If within the acceptance , it can pass => don't match */ + if (random_number <= (255 * info->acceptance_rate) / 100) + return 0; + else + return 1; /* It can't pass (It matches) */ + }; + + return 0; /* acceptance_rate == 100 % => Everything passes ... */ + +} + +static int +ip6t_fuzzy_checkentry(const char *tablename, + const struct ip6t_ip6 *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + + const struct ip6t_fuzzy_info *info = matchinfo; + + if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_fuzzy_info))) { + printk("ip6t_fuzzy: matchsize %u != %u\n", matchsize, + IP6T_ALIGN(sizeof(struct ip6t_fuzzy_info))); + return 0; + } + + if ((info->minimum_rate < MINFUZZYRATE) || (info->maximum_rate > MAXFUZZYRATE) + || (info->minimum_rate >= info->maximum_rate)) { + printk("ip6t_fuzzy: BAD limits , please verify !!!\n"); + return 0; + } + + return 1; +} + +static struct ip6t_match ip6t_fuzzy_reg = { + {NULL, NULL}, + "fuzzy", + ip6t_fuzzy_match, + ip6t_fuzzy_checkentry, + NULL, + THIS_MODULE }; + +static int __init init(void) +{ + if (ip6t_register_match(&ip6t_fuzzy_reg)) + return -EINVAL; + + return 0; +} + +static void __exit fini(void) +{ + ip6t_unregister_match(&ip6t_fuzzy_reg); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_nth.c linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_nth.c --- linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_nth.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_nth.c 2004-04-19 10:08:34.000000000 +0200 @@ -0,0 +1,173 @@ +/* + This is a module which is used for match support for every Nth packet + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 2001-07-18 Fabrice MARIE : initial implementation. + 2001-09-20 Richard Wagner (rwagner@cloudnet.com) + * added support for multiple counters + * added support for matching on individual packets + in the counter cycle + 2003-04-30 Maciej Soltysiak : IPv6 Port + +*/ + +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); + +/* + * State information. + */ +struct state { + spinlock_t lock; + u_int16_t number; +}; + +static struct state states[IP6T_NTH_NUM_COUNTERS]; + +static int +ip6t_nth_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + /* Parameters from userspace */ + const struct ip6t_nth_info *info = matchinfo; + unsigned counter = info->counter; + if((counter < 0) || (counter >= IP6T_NTH_NUM_COUNTERS)) + { + printk(KERN_WARNING "nth: invalid counter %u. counter between 0 and %u\n", counter, IP6T_NTH_NUM_COUNTERS-1); + return 0; + }; + + spin_lock(&states[counter].lock); + + /* Are we matching every nth packet?*/ + if (info->packet == 0xFF) + { + /* We're matching every nth packet and only every nth packet*/ + /* Do we match or invert match? */ + if (info->not == 0) + { + if (states[counter].number == 0) + { + ++states[counter].number; + goto match; + } + if (states[counter].number >= info->every) + states[counter].number = 0; /* reset the counter */ + else + ++states[counter].number; + goto dontmatch; + } + else + { + if (states[counter].number == 0) + { + ++states[counter].number; + goto dontmatch; + } + if (states[counter].number >= info->every) + states[counter].number = 0; + else + ++states[counter].number; + goto match; + } + } + else + { + /* We're using the --packet, so there must be a rule for every value */ + if (states[counter].number == info->packet) + { + /* only increment the counter when a match happens */ + if (states[counter].number >= info->every) + states[counter].number = 0; /* reset the counter */ + else + ++states[counter].number; + goto match; + } + else + goto dontmatch; + } + + dontmatch: + /* don't match */ + spin_unlock(&states[counter].lock); + return 0; + + match: + spin_unlock(&states[counter].lock); + return 1; +} + +static int +ip6t_nth_checkentry(const char *tablename, + const struct ip6t_ip6 *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + /* Parameters from userspace */ + const struct ip6t_nth_info *info = matchinfo; + unsigned counter = info->counter; + if((counter < 0) || (counter >= IP6T_NTH_NUM_COUNTERS)) + { + printk(KERN_WARNING "nth: invalid counter %u. counter between 0 and %u\n", counter, IP6T_NTH_NUM_COUNTERS-1); + return 0; + }; + + if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_nth_info))) { + printk("nth: matchsize %u != %u\n", matchsize, + IP6T_ALIGN(sizeof(struct ip6t_nth_info))); + return 0; + } + + states[counter].number = info->startat; + + return 1; +} + +static struct ip6t_match ip6t_nth_reg = { + {NULL, NULL}, + "nth", + ip6t_nth_match, + ip6t_nth_checkentry, + NULL, + THIS_MODULE }; + +static int __init init(void) +{ + unsigned counter; + memset(&states, 0, sizeof(states)); + if (ip6t_register_match(&ip6t_nth_reg)) + return -EINVAL; + + for(counter = 0; counter < IP6T_NTH_NUM_COUNTERS; counter++) + { + spin_lock_init(&(states[counter].lock)); + }; + + printk("ip6t_nth match loaded\n"); + return 0; +} + +static void __exit fini(void) +{ + ip6t_unregister_match(&ip6t_nth_reg); + printk("ip6t_nth match unloaded\n"); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_random.c linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_random.c --- linux-2.6.6-rc1.org/net/ipv6/netfilter/ip6t_random.c 1970-01-01 01:00:00.000000000 +0100 +++ linux-2.6.6-rc1/net/ipv6/netfilter/ip6t_random.c 2004-04-19 10:08:38.000000000 +0200 @@ -0,0 +1,97 @@ +/* + This is a module which is used for a "random" match support. + This file is distributed under the terms of the GNU General Public + License (GPL). Copies of the GPL can be obtained from: + ftp://prep.ai.mit.edu/pub/gnu/GPL + + 2001-10-14 Fabrice MARIE : initial implementation. + 2003-04-30 Maciej Soltysiak : IPv6 Port +*/ + +#include +#include +#include +#include +#include +#include +#include +#include + +MODULE_LICENSE("GPL"); + +static int +ip6t_rand_match(const struct sk_buff *pskb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + const void *hdr, + u_int16_t datalen, + int *hotdrop) +{ + /* Parameters from userspace */ + const struct ip6t_rand_info *info = matchinfo; + u_int8_t random_number; + + /* get 1 random number from the kernel random number generation routine */ + get_random_bytes((void *)(&random_number), 1); + + /* Do we match ? */ + if (random_number <= info->average) + return 1; + else + return 0; +} + +static int +ip6t_rand_checkentry(const char *tablename, + const struct ip6t_ip6 *e, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + /* Parameters from userspace */ + const struct ip6t_rand_info *info = matchinfo; + + if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_rand_info))) { + printk("ip6t_random: matchsize %u != %u\n", matchsize, + IP6T_ALIGN(sizeof(struct ip6t_rand_info))); + return 0; + } + + /* must be 1 <= average % <= 99 */ + /* 1 x 2.55 = 2 */ + /* 99 x 2.55 = 252 */ + if ((info->average < 2) || (info->average > 252)) { + printk("ip6t_random: invalid average %u\n", info->average); + return 0; + } + + return 1; +} + +static struct ip6t_match ip6t_rand_reg = { + {NULL, NULL}, + "random", + ip6t_rand_match, + ip6t_rand_checkentry, + NULL, + THIS_MODULE }; + +static int __init init(void) +{ + if (ip6t_register_match(&ip6t_rand_reg)) + return -EINVAL; + + printk("ip6t_random match loaded\n"); + return 0; +} + +static void __exit fini(void) +{ + ip6t_unregister_match(&ip6t_rand_reg); + printk("ip6t_random match unloaded\n"); +} + +module_init(init); +module_exit(fini); diff -Nur linux-2.6.6-rc1.org/net/ipv6/sit.c linux-2.6.6-rc1/net/ipv6/sit.c --- linux-2.6.6-rc1.org/net/ipv6/sit.c 2004-04-15 03:35:37.000000000 +0200 +++ linux-2.6.6-rc1/net/ipv6/sit.c 2004-04-19 10:08:24.000000000 +0200 @@ -388,13 +388,7 @@ skb->dev = tunnel->dev; dst_release(skb->dst); skb->dst = NULL; -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); ipip6_ecn_decapsulate(iph, skb); netif_rx(skb); read_unlock(&ipip6_lock); @@ -580,13 +574,7 @@ if ((iph->ttl = tiph->ttl) == 0) iph->ttl = iph6->hop_limit; -#ifdef CONFIG_NETFILTER - nf_conntrack_put(skb->nfct); - skb->nfct = NULL; -#ifdef CONFIG_NETFILTER_DEBUG - skb->nf_debug = 0; -#endif -#endif + nf_reset(skb); IPTUNNEL_XMIT(); tunnel->recursion--;