diff options
author | Linus Torvalds <torvalds@g5.osdl.org> | 2006-01-13 15:28:10 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@g5.osdl.org> | 2006-01-13 15:28:10 -0800 |
commit | 69eebed240b32377b9034551f1b8c4ae2774ceb6 (patch) | |
tree | 6955dbf8110c6f03d64747fe55e1feb1d81e951f /net | |
parent | 67daf5f11f06b9b15f8320de1d237ccc2e74fe43 (diff) | |
parent | dca80b962a811a0454d779ab6250b59446a8c86c (diff) |
Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Diffstat (limited to 'net')
164 files changed, 21020 insertions, 3512 deletions
diff --git a/net/Kconfig b/net/Kconfig index 60f6f321bd7..9296b269d67 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -159,6 +159,7 @@ source "net/ipx/Kconfig" source "drivers/net/appletalk/Kconfig" source "net/x25/Kconfig" source "net/lapb/Kconfig" +source "net/tipc/Kconfig" config NET_DIVERT bool "Frame Diverter (EXPERIMENTAL)" diff --git a/net/Makefile b/net/Makefile index f5141b9d4f3..065796f5fb1 100644 --- a/net/Makefile +++ b/net/Makefile @@ -45,6 +45,7 @@ obj-$(CONFIG_VLAN_8021Q) += 8021q/ obj-$(CONFIG_IP_DCCP) += dccp/ obj-$(CONFIG_IP_SCTP) += sctp/ obj-$(CONFIG_IEEE80211) += ieee80211/ +obj-$(CONFIG_TIPC) += tipc/ ifeq ($(CONFIG_NET),y) obj-$(CONFIG_SYSCTL) += sysctl_net.o diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c index 9f6e0193ae1..a29c1232c42 100644 --- a/net/bridge/netfilter/ebt_log.c +++ b/net/bridge/netfilter/ebt_log.c @@ -15,6 +15,7 @@ #include <linux/netfilter.h> #include <linux/module.h> #include <linux/ip.h> +#include <linux/in.h> #include <linux/if_arp.h> #include <linux/spinlock.h> diff --git a/net/core/filter.c b/net/core/filter.c index 9eb9d0017a0..a52665f7522 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -287,7 +287,9 @@ load_b: * no references or jumps that are out of range, no illegal * instructions, and must end with a RET instruction. * - * Returns 0 if the rule set is legal or a negative errno code if not. + * All jumps are forward as they are not signed. + * + * Returns 0 if the rule set is legal or -EINVAL if not. */ int sk_chk_filter(struct sock_filter *filter, int flen) { @@ -299,7 +301,6 @@ int sk_chk_filter(struct sock_filter *filter, int flen) /* check the filter code now */ for (pc = 0; pc < flen; pc++) { - /* all jumps are forward as they are not signed */ ftest = &filter[pc]; /* Only allow valid instructions */ @@ -383,11 +384,6 @@ int sk_chk_filter(struct sock_filter *filter, int flen) } } - /* - * The program must end with a return. We don't care where they - * jumped within the script (its always forwards) but in the end - * they _will_ hit this. - */ return (BPF_CLASS(filter[flen - 1].code) == BPF_RET) ? 0 : -EINVAL; } diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index a9893ec03e0..db783036e4d 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -182,6 +182,7 @@ config IP_NF_QUEUE config IP_NF_IPTABLES tristate "IP tables support (required for filtering/masq/NAT)" + depends on NETFILTER_XTABLES help iptables is a general, extensible packet identification framework. The packet filtering and full NAT (masquerading, port forwarding, @@ -191,16 +192,6 @@ config IP_NF_IPTABLES To compile it as a module, choose M here. If unsure, say N. # The matches. -config IP_NF_MATCH_LIMIT - tristate "limit match support" - depends on IP_NF_IPTABLES - help - limit matching allows you to control the rate at which a rule can be - matched: mainly useful in combination with the LOG target ("LOG - target support", below) and to avoid some Denial of Service attacks. - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_IPRANGE tristate "IP range match support" depends on IP_NF_IPTABLES @@ -210,37 +201,6 @@ config IP_NF_MATCH_IPRANGE To compile it as a module, choose M here. If unsure, say N. -config IP_NF_MATCH_MAC - tristate "MAC address match support" - depends on IP_NF_IPTABLES - help - MAC matching allows you to match packets based on the source - Ethernet address of the packet. - - To compile it as a module, choose M here. If unsure, say N. - -config IP_NF_MATCH_PKTTYPE - tristate "Packet type match support" - depends on IP_NF_IPTABLES - help - Packet type matching allows you to match a packet by - its "class", eg. BROADCAST, MULTICAST, ... - - Typical usage: - iptables -A INPUT -m pkttype --pkt-type broadcast -j LOG - - To compile it as a module, choose M here. If unsure, say N. - -config IP_NF_MATCH_MARK - tristate "netfilter MARK match support" - depends on IP_NF_IPTABLES - help - Netfilter mark matching allows you to match packets based on the - `nfmark' value in the packet. This can be set by the MARK target - (see below). - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_MULTIPORT tristate "Multiple port match support" depends on IP_NF_IPTABLES @@ -301,15 +261,6 @@ config IP_NF_MATCH_AH_ESP To compile it as a module, choose M here. If unsure, say N. -config IP_NF_MATCH_LENGTH - tristate "LENGTH match support" - depends on IP_NF_IPTABLES - help - This option allows you to match the length of a packet against a - specific value or range of values. - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_TTL tristate "TTL match support" depends on IP_NF_IPTABLES @@ -319,50 +270,6 @@ config IP_NF_MATCH_TTL To compile it as a module, choose M here. If unsure, say N. -config IP_NF_MATCH_TCPMSS - tristate "tcpmss match support" - depends on IP_NF_IPTABLES - help - This option adds a `tcpmss' match, which allows you to examine the - MSS value of TCP SYN packets, which control the maximum packet size - for that connection. - - To compile it as a module, choose M here. If unsure, say N. - -config IP_NF_MATCH_HELPER - tristate "Helper match support" - depends on IP_NF_IPTABLES - depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 - help - Helper matching allows you to match packets in dynamic connections - tracked by a conntrack-helper, ie. ip_conntrack_ftp - - To compile it as a module, choose M here. If unsure, say Y. - -config IP_NF_MATCH_STATE - tristate "Connection state match support" - depends on IP_NF_IPTABLES - depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 - help - Connection state matching allows you to match packets based on their - relationship to a tracked connection (ie. previous packets). This - is a powerful tool for packet classification. - - To compile it as a module, choose M here. If unsure, say N. - -config IP_NF_MATCH_CONNTRACK - tristate "Connection tracking match support" - depends on IP_NF_IPTABLES - depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 - help - This is a general conntrack match module, a superset of the state match. - - It allows matching on additional conntrack information, which is - useful in complex configurations, such as NAT gateways with multiple - internet links or tunnels. - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_OWNER tristate "Owner match support" depends on IP_NF_IPTABLES @@ -372,15 +279,6 @@ config IP_NF_MATCH_OWNER To compile it as a module, choose M here. If unsure, say N. -config IP_NF_MATCH_PHYSDEV - tristate "Physdev match support" - depends on IP_NF_IPTABLES && BRIDGE_NETFILTER - help - Physdev packet matching matches against the physical bridge ports - the IP packet arrived on or will leave by. - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_ADDRTYPE tristate 'address type match support' depends on IP_NF_IPTABLES @@ -391,75 +289,6 @@ config IP_NF_MATCH_ADDRTYPE If you want to compile it as a module, say M here and read <file:Documentation/modules.txt>. If unsure, say `N'. -config IP_NF_MATCH_REALM - tristate 'realm match support' - depends on IP_NF_IPTABLES - select NET_CLS_ROUTE - help - This option adds a `realm' match, which allows you to use the realm - key from the routing subsystem inside iptables. - - This match pretty much resembles the CONFIG_NET_CLS_ROUTE4 option - in tc world. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - -config IP_NF_MATCH_SCTP - tristate 'SCTP protocol match support' - depends on IP_NF_IPTABLES - help - With this option enabled, you will be able to use the iptables - `sctp' match in order to match on SCTP source/destination ports - and SCTP chunk types. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - -config IP_NF_MATCH_DCCP - tristate 'DCCP protocol match support' - depends on IP_NF_IPTABLES - help - With this option enabled, you will be able to use the iptables - `dccp' match in order to match on DCCP source/destination ports - and DCCP flags. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - -config IP_NF_MATCH_COMMENT - tristate 'comment match support' - depends on IP_NF_IPTABLES - help - This option adds a `comment' dummy-match, which allows you to put - comments in your iptables ruleset. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - -config IP_NF_MATCH_CONNMARK - tristate 'Connection mark match support' - depends on IP_NF_IPTABLES - depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) - help - This option adds a `connmark' match, which allows you to match the - connection mark value previously set for the session by `CONNMARK'. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. The module will be called - ipt_connmark.o. If unsure, say `N'. - -config IP_NF_MATCH_CONNBYTES - tristate 'Connection byte/packet counter match support' - depends on IP_NF_IPTABLES - depends on (IP_NF_CONNTRACK && IP_NF_CT_ACCT) || (NF_CT_ACCT && NF_CONNTRACK_IPV4) - help - This option adds a `connbytes' match, which allows you to match the - number of bytes and/or packets for each direction within a connection. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - config IP_NF_MATCH_HASHLIMIT tristate 'hashlimit match support' depends on IP_NF_IPTABLES @@ -474,19 +303,6 @@ config IP_NF_MATCH_HASHLIMIT destination IP' or `500pps from any given source IP' with a single IPtables rule. -config IP_NF_MATCH_STRING - tristate 'string match support' - depends on IP_NF_IPTABLES - select TEXTSEARCH - select TEXTSEARCH_KMP - select TEXTSEARCH_BM - select TEXTSEARCH_FSM - help - This option adds a `string' match, which allows you to look for - pattern matchings in packets. - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_MATCH_POLICY tristate "IPsec policy match support" depends on IP_NF_IPTABLES && XFRM @@ -572,17 +388,6 @@ config IP_NF_TARGET_TCPMSS To compile it as a module, choose M here. If unsure, say N. -config IP_NF_TARGET_NFQUEUE - tristate "NFQUEUE Target Support" - depends on IP_NF_IPTABLES - help - This Target replaced the old obsolete QUEUE target. - - As opposed to QUEUE, it supports 65535 different queues, - not just one. - - To compile it as a module, choose M here. If unsure, say N. - # NAT + specific targets config IP_NF_NAT tristate "Full NAT" @@ -735,31 +540,6 @@ config IP_NF_TARGET_DSCP To compile it as a module, choose M here. If unsure, say N. -config IP_NF_TARGET_MARK - tristate "MARK target support" - depends on IP_NF_MANGLE - help - This option adds a `MARK' target, which allows you to create rules - in the `mangle' table which alter the netfilter mark (nfmark) field - associated with the packet prior to routing. This can change - the routing method (see `Use netfilter MARK value as routing - key') and can also be used by other subsystems to change their - behavior. - - To compile it as a module, choose M here. If unsure, say N. - -config IP_NF_TARGET_CLASSIFY - tristate "CLASSIFY target support" - depends on IP_NF_MANGLE - help - This option adds a `CLASSIFY' target, which enables the user to set - the priority of a packet. Some qdiscs can use this value for - classification, among these are: - - atm, cbq, dsmark, pfifo_fast, htb, prio - - To compile it as a module, choose M here. If unsure, say N. - config IP_NF_TARGET_TTL tristate 'TTL target support' depends on IP_NF_MANGLE @@ -774,19 +554,6 @@ config IP_NF_TARGET_TTL To compile it as a module, choose M here. If unsure, say N. -config IP_NF_TARGET_CONNMARK - tristate 'CONNMARK target support' - depends on IP_NF_MANGLE - depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) - help - This option adds a `CONNMARK' target, which allows one to manipulate - the connection mark value. Similar to the MARK target, but - affects the connection mark value rather than the packet mark value. - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. The module will be called - ipt_CONNMARK.o. If unsure, say `N'. - config IP_NF_TARGET_CLUSTERIP tristate "CLUSTERIP target support (EXPERIMENTAL)" depends on IP_NF_MANGLE && EXPERIMENTAL @@ -810,23 +577,10 @@ config IP_NF_RAW If you want to compile it as a module, say M here and read <file:Documentation/modules.txt>. If unsure, say `N'. -config IP_NF_TARGET_NOTRACK - tristate 'NOTRACK target support' - depends on IP_NF_RAW - depends on IP_NF_CONNTRACK || NF_CONNTRACK_IPV4 - help - The NOTRACK target allows a select rule to specify - which packets *not* to enter the conntrack/NAT - subsystem with all the consequences (no ICMP error tracking, - no protocol helpers for the selected packets). - - If you want to compile it as a module, say M here and read - <file:Documentation/modules.txt>. If unsure, say `N'. - - # ARP tables config IP_NF_ARPTABLES tristate "ARP tables support" + depends on NETFILTER_XTABLES help arptables is a general, extensible packet identification framework. The ARP packet filtering and mangling (manipulation)subsystems diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 549b01a648b..bcefe64b931 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -47,14 +47,8 @@ obj-$(CONFIG_IP_NF_RAW) += iptable_raw.o # matches obj-$(CONFIG_IP_NF_MATCH_HELPER) += ipt_helper.o -obj-$(CONFIG_IP_NF_MATCH_LIMIT) += ipt_limit.o obj-$(CONFIG_IP_NF_MATCH_HASHLIMIT) += ipt_hashlimit.o -obj-$(CONFIG_IP_NF_MATCH_SCTP) += ipt_sctp.o -obj-$(CONFIG_IP_NF_MATCH_DCCP) += ipt_dccp.o -obj-$(CONFIG_IP_NF_MATCH_MARK) += ipt_mark.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-$(CONFIG_IP_NF_MATCH_OWNER) += ipt_owner.o obj-$(CONFIG_IP_NF_MATCH_TOS) += ipt_tos.o @@ -62,40 +56,25 @@ obj-$(CONFIG_IP_NF_MATCH_RECENT) += ipt_recent.o obj-$(CONFIG_IP_NF_MATCH_ECN) += ipt_ecn.o obj-$(CONFIG_IP_NF_MATCH_DSCP) += ipt_dscp.o obj-$(CONFIG_IP_NF_MATCH_AH_ESP) += ipt_ah.o ipt_esp.o -obj-$(CONFIG_IP_NF_MATCH_LENGTH) += ipt_length.o obj-$(CONFIG_IP_NF_MATCH_TTL) += ipt_ttl.o -obj-$(CONFIG_IP_NF_MATCH_STATE) += ipt_state.o -obj-$(CONFIG_IP_NF_MATCH_CONNMARK) += ipt_connmark.o -obj-$(CONFIG_IP_NF_MATCH_CONNTRACK) += ipt_conntrack.o -obj-$(CONFIG_IP_NF_MATCH_CONNBYTES) += ipt_connbytes.o -obj-$(CONFIG_IP_NF_MATCH_TCPMSS) += ipt_tcpmss.o -obj-$(CONFIG_IP_NF_MATCH_REALM) += ipt_realm.o obj-$(CONFIG_IP_NF_MATCH_ADDRTYPE) += ipt_addrtype.o -obj-$(CONFIG_IP_NF_MATCH_PHYSDEV) += ipt_physdev.o obj-$(CONFIG_IP_NF_MATCH_POLICY) += ipt_policy.o -obj-$(CONFIG_IP_NF_MATCH_COMMENT) += ipt_comment.o -obj-$(CONFIG_IP_NF_MATCH_STRING) += ipt_string.o # targets obj-$(CONFIG_IP_NF_TARGET_REJECT) += ipt_REJECT.o obj-$(CONFIG_IP_NF_TARGET_TOS) += ipt_TOS.o obj-$(CONFIG_IP_NF_TARGET_ECN) += ipt_ECN.o obj-$(CONFIG_IP_NF_TARGET_DSCP) += ipt_DSCP.o -obj-$(CONFIG_IP_NF_TARGET_MARK) += ipt_MARK.o obj-$(CONFIG_IP_NF_TARGET_MASQUERADE) += ipt_MASQUERADE.o obj-$(CONFIG_IP_NF_TARGET_REDIRECT) += ipt_REDIRECT.o obj-$(CONFIG_IP_NF_TARGET_NETMAP) += ipt_NETMAP.o obj-$(CONFIG_IP_NF_TARGET_SAME) += ipt_SAME.o -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_CONNMARK) += ipt_CONNMARK.o obj-$(CONFIG_IP_NF_TARGET_ULOG) += ipt_ULOG.o obj-$(CONFIG_IP_NF_TARGET_TCPMSS) += ipt_TCPMSS.o -obj-$(CONFIG_IP_NF_TARGET_NOTRACK) += ipt_NOTRACK.o obj-$(CONFIG_IP_NF_TARGET_CLUSTERIP) += ipt_CLUSTERIP.o obj-$(CONFIG_IP_NF_TARGET_TTL) += ipt_TTL.o -obj-$(CONFIG_IP_NF_TARGET_NFQUEUE) += ipt_NFQUEUE.o # generic ARP tables obj-$(CONFIG_IP_NF_ARPTABLES) += arp_tables.o diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index b6d5284c802..afe3d8f8177 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -24,6 +24,7 @@ #include <asm/uaccess.h> #include <asm/semaphore.h> +#include <linux/netfilter/x_tables.h> #include <linux/netfilter_arp/arp_tables.h> MODULE_LICENSE("GPL"); @@ -55,28 +56,9 @@ do { \ #else #define ARP_NF_ASSERT(x) #endif -#define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) -static DECLARE_MUTEX(arpt_mutex); - -#define ASSERT_READ_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0) -#define ASSERT_WRITE_LOCK(x) ARP_NF_ASSERT(down_trylock(&arpt_mutex) != 0) #include <linux/netfilter_ipv4/listhelp.h> -struct arpt_table_info { - unsigned int size; - unsigned int number; - unsigned int initial_entries; - unsigned int hook_entry[NF_ARP_NUMHOOKS]; - unsigned int underflow[NF_ARP_NUMHOOKS]; - void *entries[NR_CPUS]; -}; - -static LIST_HEAD(arpt_target); -static LIST_HEAD(arpt_tables); -#define SET_COUNTER(c,b,p) do { (c).bcnt = (b); (c).pcnt = (p); } while(0) -#define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0) - static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap, char *hdr_addr, int len) { @@ -223,9 +205,9 @@ static inline int arp_checkentry(const struct arpt_arp *arp) } static unsigned int arpt_error(struct sk_buff **pskb, - unsigned int hooknum, const struct net_device *in, const struct net_device *out, + unsigned int hooknum, const void *targinfo, void *userinfo) { @@ -254,6 +236,7 @@ unsigned int arpt_do_table(struct sk_buff **pskb, struct arpt_entry *e, *back; const char *indev, *outdev; void *table_base; + struct xt_table_info *private = table->private; /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ if (!pskb_may_pull((*pskb), (sizeof(struct arphdr) + @@ -265,9 +248,9 @@ unsigned int arpt_do_table(struct sk_buff **pskb, outdev = out ? out->name : nulldevname; read_lock_bh(&table->lock); - table_base = (void *)table->private->entries[smp_processor_id()]; - e = get_entry(table_base, table->private->hook_entry[hook]); - back = get_entry(table_base, table->private->underflow[hook]); + table_base = (void *)private->entries[smp_processor_id()]; + e = get_entry(table_base, private->hook_entry[hook]); + back = get_entry(table_base, private->underflow[hook]); arp = (*pskb)->nh.arph; do { @@ -315,8 +298,8 @@ unsigned int arpt_do_table(struct sk_buff **pskb, * abs. verdicts */ verdict = t->u.kernel.target->target(pskb, - hook, in, out, + hook, t->data, userdata); @@ -341,106 +324,6 @@ unsigned int arpt_do_table(struct sk_buff **pskb, return verdict; } -/* - * These are weird, but module loading must not be done with mutex - * held (since they will register), and we have to have a single - * function to use try_then_request_module(). - */ - -/* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ -static inline struct arpt_table *find_table_lock(const char *name) -{ - struct arpt_table *t; - - if (down_interruptible(&arpt_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &arpt_tables, list) - if (strcmp(t->name, name) == 0 && try_module_get(t->me)) - return t; - up(&arpt_mutex); - return NULL; -} - - -/* Find target, grabs ref. Returns ERR_PTR() on error. */ -static inline struct arpt_target *find_target(const char *name, u8 revision) -{ - struct arpt_target *t; - int err = 0; - - if (down_interruptible(&arpt_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &arpt_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision == revision) { - if (try_module_get(t->me)) { - up(&arpt_mutex); - return t; - } - } else - err = -EPROTOTYPE; /* Found something. */ - } - } - up(&arpt_mutex); - return ERR_PTR(err); -} - -struct arpt_target *arpt_find_target(const char *name, u8 revision) -{ - struct arpt_target *target; - - target = try_then_request_module(find_target(name, revision), - "arpt_%s", name); - if (IS_ERR(target) || !target) - return NULL; - return target; -} - -static int target_revfn(const char *name, u8 revision, int *bestp) -{ - struct arpt_target *t; - int have_rev = 0; - - list_for_each_entry(t, &arpt_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision > *bestp) - *bestp = t->revision; - if (t->revision == revision) - have_rev =1; - } - } - return have_rev; -} - -/* Returns true or false (if no such extension at all) */ -static inline int find_revision(const char *name, u8 revision, - int (*revfn)(const char *, u8, int *), - int *err) -{ - int have_rev, best = -1; - - if (down_interruptible(&arpt_mutex) != 0) { - *err = -EINTR; - return 1; - } - have_rev = revfn(name, revision, &best); - up(&arpt_mutex); - - /* Nothing at all? Return 0 to try loading module. */ - if (best == -1) { - *err = -ENOENT; - return 0; - } - - *err = best; - if (!have_rev) - *err = -EPROTONOSUPPORT; - return 1; -} - - /* All zeroes == unconditional rule. */ static inline int unconditional(const struct arpt_arp *arp) { @@ -456,7 +339,7 @@ static inline int unconditional(const struct arpt_arp *arp) /* Figures out from what hook each rule can be called: returns 0 if * there are loops. Puts hook bitmask in comefrom. */ -static int mark_source_chains(struct arpt_table_info *newinfo, +static int mark_source_chains(struct xt_table_info *newinfo, unsigned int valid_hooks, void *entry0) { unsigned int hook; @@ -587,8 +470,8 @@ static inline int check_entry(struct arpt_entry *e, const char *name, unsigned i } t = arpt_get_target(e); - target = try_then_request_module(find_target(t->u.user.name, - t->u.user.revision), + target = try_then_request_module(xt_find_target(NF_ARP, t->u.user.name, + t->u.user.revision), "arpt_%s", t->u.user.name); if (IS_ERR(target) || !target) { duprintf("check_entry: `%s' not found\n", t->u.user.name); @@ -622,7 +505,7 @@ out: } static inline int check_entry_size_and_hooks(struct arpt_entry *e, - struct arpt_table_info *newinfo, + struct xt_table_info *newinfo, unsigned char *base, unsigned char *limit, const unsigned int *hook_entries, @@ -656,7 +539,7 @@ static inline int check_entry_size_and_hooks(struct arpt_entry *e, < 0 (not ARPT_RETURN). --RR */ /* Clear counters and comefrom */ - e->counters = ((struct arpt_counters) { 0, 0 }); + e->counters = ((struct xt_counters) { 0, 0 }); e->comefrom = 0; (*i)++; @@ -683,7 +566,7 @@ static inline int cleanup_entry(struct arpt_entry *e, unsigned int *i) */ static int translate_table(const char *name, unsigned int valid_hooks, - struct arpt_table_info *newinfo, + struct xt_table_info *newinfo, void *entry0, unsigned int size, unsigned int number, @@ -764,34 +647,9 @@ static int translate_table(const char *name, return ret; } -static struct arpt_table_info *replace_table(struct arpt_table *table, - unsigned int num_counters, - struct arpt_table_info *newinfo, - int *error) -{ - struct arpt_table_info *oldinfo; - - /* Do the substitution. */ - write_lock_bh(&table->lock); - /* Check inside lock: is the old number correct? */ - if (num_counters != table->private->number) { - duprintf("num_counters != table->private->number (%u/%u)\n", - num_counters, table->private->number); - write_unlock_bh(&table->lock); - *error = -EAGAIN; - return NULL; - } - oldinfo = table->private; - table->private = newinfo; - newinfo->initial_entries = oldinfo->initial_entries; - write_unlock_bh(&table->lock); - - return oldinfo; -} - /* Gets counters. */ static inline int add_entry_to_counter(const struct arpt_entry *e, - struct arpt_counters total[], + struct xt_counters total[], unsigned int *i) { ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt); @@ -801,7 +659,7 @@ static inline int add_entry_to_counter(const struct arpt_entry *e, } static inline int set_entry_to_counter(const struct arpt_entry *e, - struct arpt_counters total[], + struct xt_counters total[], unsigned int *i) { SET_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt); @@ -810,8 +668,8 @@ static inline int set_entry_to_counter(const struct arpt_entry *e, return 0; } -static void get_counters(const struct arpt_table_info *t, - struct arpt_counters counters[]) +static void get_counters(const struct xt_table_info *t, + struct xt_counters counters[]) { unsigned int cpu; unsigned int i; @@ -849,7 +707,8 @@ static int copy_entries_to_user(unsigned int total_size, { unsigned int off, num, countersize; struct arpt_entry *e; - struct arpt_counters *counters; + struct xt_counters *counters; + struct xt_table_info *private = table->private; int ret = 0; void *loc_cpu_entry; @@ -857,18 +716,18 @@ static int copy_entries_to_user(unsigned int total_size, * (other than comefrom, which userspace doesn't care * about). */ - countersize = sizeof(struct arpt_counters) * table->private->number; - counters = vmalloc(countersize); + countersize = sizeof(struct xt_counters) * private->number; + counters = vmalloc_node(countersize, numa_node_id()); if (counters == NULL) return -ENOMEM; /* First, sum counters... */ write_lock_bh(&table->lock); - get_counters(table->private, counters); + get_counters(private, counters); write_unlock_bh(&table->lock); - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; /* ... then copy entire thing ... */ if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; @@ -911,75 +770,34 @@ static int get_entries(const struct arpt_get_entries *entries, int ret; struct arpt_table *t; - t = find_table_lock(entries->name); + t = xt_find_table_lock(NF_ARP, entries->name); if (t || !IS_ERR(t)) { + struct xt_table_info *private = t->private; duprintf("t->private->number = %u\n", - t->private->number); - if (entries->size == t->private->size) - ret = copy_entries_to_user(t->private->size, + private->number); + if (entries->size == private->size) + ret = copy_entries_to_user(private->size, t, uptr->entrytable); else { duprintf("get_entries: I've got %u not %u!\n", - t->private->size, - entries->size); + private->size, entries->size); ret = -EINVAL; } module_put(t->me); - up(&arpt_mutex); + xt_table_unlock(t); } else ret = t ? PTR_ERR(t) : -ENOENT; return ret; } -static void free_table_info(struct arpt_table_info *info) -{ - int cpu; - for_each_cpu(cpu) { - if (info->size <= PAGE_SIZE) - kfree(info->entries[cpu]); - else - vfree(info->entries[cpu]); - } - kfree(info); -} - -static struct arpt_table_info *alloc_table_info(unsigned int size) -{ - struct arpt_table_info *newinfo; - int cpu; - - newinfo = kzalloc(sizeof(struct arpt_table_info), GFP_KERNEL); - if (!newinfo) - return NULL; - - newinfo->size = size; - - for_each_cpu(cpu) { - if (size <= PAGE_SIZE) - newinfo->entries[cpu] = kmalloc_node(size, - GFP_KERNEL, - cpu_to_node(cpu)); - else - newinfo->entries[cpu] = vmalloc_node(size, - cpu_to_node(cpu)); - - if (newinfo->entries[cpu] == NULL) { - free_table_info(newinfo); - return NULL; - } - } - - return newinfo; -} - static int do_replace(void __user *user, unsigned int len) { int ret; struct arpt_replace tmp; struct arpt_table *t; - struct arpt_table_info *newinfo, *oldinfo; - struct arpt_counters *counters; + struct xt_table_info *newinfo, *oldinfo; + struct xt_counters *counters; void *loc_cpu_entry, *loc_cpu_old_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) @@ -989,11 +807,7 @@ static int do_replace(void __user *user, unsigned int len) if (len != sizeof(tmp) + tmp.size) return -ENOPROTOOPT; - /* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */ - if ((SMP_ALIGN(tmp.size) >> PAGE_SHIFT) + 2 > num_physpages) - return -ENOMEM; - - newinfo = alloc_table_info(tmp.size); + newinfo = xt_alloc_table_info(tmp.size); if (!newinfo) return -ENOMEM; @@ -1005,7 +819,7 @@ static int do_replace(void __user *user, unsigned int len) goto free_newinfo; } - counters = vmalloc(tmp.num_counters * sizeof(struct arpt_counters)); + counters = vmalloc(tmp.num_counters * sizeof(struct xt_counters)); if (!counters) { ret = -ENOMEM; goto free_newinfo; @@ -1019,7 +833,7 @@ static int do_replace(void __user *user, unsigned int len) duprintf("arp_tables: Translated table\n"); - t = try_then_request_module(find_table_lock(tmp.name), + t = try_then_request_module(xt_find_table_lock(NF_ARP, tmp.name), "arptable_%s", tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; @@ -1034,7 +848,7 @@ static int do_replace(void __user *user, unsigned int len) goto put_module; } - oldinfo = replace_table(t, tmp.num_counters, newinfo, &ret); + oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret); if (!oldinfo) goto put_module; @@ -1054,23 +868,23 @@ static int do_replace(void __user *user, unsigned int len) loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; ARPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL); - free_table_info(oldinfo); + xt_free_table_info(oldinfo); if (copy_to_user(tmp.counters, counters, - sizeof(struct arpt_counters) * tmp.num_counters) != 0) + sizeof(struct xt_counters) * tmp.num_counters) != 0) ret = -EFAULT; vfree(counters); - up(&arpt_mutex); + xt_table_unlock(t); return ret; put_module: module_put(t->me); - up(&arpt_mutex); + xt_table_unlock(t); free_newinfo_counters_untrans: ARPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry, NULL); free_newinfo_counters: vfree(counters); free_newinfo: - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } @@ -1078,7 +892,7 @@ static int do_replace(void __user *user, unsigned int len) * and everything is OK. */ static inline int add_counter_to_entry(struct arpt_entry *e, - const struct arpt_counters addme[], + const struct xt_counters addme[], unsigned int *i) { @@ -1091,15 +905,16 @@ static inline int add_counter_to_entry(struct arpt_entry *e, static int do_add_counters(void __user *user, unsigned int len) { unsigned int i; - struct arpt_counters_info tmp, *paddc; + struct xt_counters_info tmp, *paddc; struct arpt_table *t; + struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; - if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct arpt_counters)) + if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters)) return -EINVAL; paddc = vmalloc(len); @@ -1111,29 +926,30 @@ static int do_add_counters(void __user *user, unsigned int len) goto free; } - t = find_table_lock(tmp.name); + t = xt_find_table_lock(NF_ARP, tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; goto free; } write_lock_bh(&t->lock); - if (t->private->number != paddc->num_counters) { + private = t->private; + if (private->number != paddc->num_counters) { ret = -EINVAL; goto unlock_up_free; } i = 0; /* Choose the copy that is on our node */ - loc_cpu_entry = t->private->entries[smp_processor_id()]; + loc_cpu_entry = private->entries[smp_processor_id()]; ARPT_ENTRY_ITERATE(loc_cpu_entry, - t->private->size, + private->size, add_counter_to_entry, paddc->counters, &i); unlock_up_free: write_unlock_bh(&t->lock); - up(&arpt_mutex); + xt_table_unlock(t); module_put(t->me); free: vfree(paddc); @@ -1190,25 +1006,26 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len } name[ARPT_TABLE_MAXNAMELEN-1] = '\0'; - t = try_then_request_module(find_table_lock(name), + t = try_then_request_module(xt_find_table_lock(NF_ARP, name), "arptable_%s", name); if (t && !IS_ERR(t)) { struct arpt_getinfo info; + struct xt_table_info *private = t->private; info.valid_hooks = t->valid_hooks; - memcpy(info.hook_entry, t->private->hook_entry, + memcpy(info.hook_entry, private->hook_entry, sizeof(info.hook_entry)); - memcpy(info.underflow, t->private->underflow, + memcpy(info.underflow, private->underflow, sizeof(info.underflow)); - info.num_entries = t->private->number; - info.size = t->private->size; + info.num_entries = private->number; + info.size = private->size; strcpy(info.name, name); if (copy_to_user(user, &info, *len) != 0) ret = -EFAULT; else ret = 0; - up(&arpt_mutex); + xt_table_unlock(t); module_put(t->me); } else ret = t ? PTR_ERR(t) : -ENOENT; @@ -1233,7 +1050,7 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len } case ARPT_SO_GET_REVISION_TARGET: { - struct arpt_get_revision rev; + struct xt_get_revision rev; if (*len != sizeof(rev)) { ret = -EINVAL; @@ -1244,8 +1061,8 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len break; } - try_then_request_module(find_revision(rev.name, rev.revision, - target_revfn, &ret), + try_then_request_module(xt_find_revision(NF_ARP, rev.name, + rev.revision, 1, &ret), "arpt_%s", rev.name); break; } @@ -1258,38 +1075,16 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len return ret; } -/* Registration hooks for targets. */ -int arpt_register_target(struct arpt_target *target) -{ - int ret; - - ret = down_interruptible(&arpt_mutex); - if (ret != 0) - return ret; - - list_add(&target->list, &arpt_target); - up(&arpt_mutex); - - return ret; -} - -void arpt_unregister_target(struct arpt_target *target) -{ - down(&arpt_mutex); - LIST_DELETE(&arpt_target, target); - up(&arpt_mutex); -} - int arpt_register_table(struct arpt_table *table, const struct arpt_replace *repl) { int ret; - struct arpt_table_info *newinfo; - static struct arpt_table_info bootstrap + struct xt_table_info *newinfo; + static struct xt_table_info bootstrap = { 0, 0, 0, { 0 }, { 0 }, { } }; void *loc_cpu_entry; - newinfo = alloc_table_info(repl->size); + newinfo = xt_alloc_table_info(repl->size); if (!newinfo) { ret = -ENOMEM; return ret; @@ -1304,60 +1099,33 @@ int arpt_register_table(struct arpt_table *table, repl->num_entries, repl->hook_entry, repl->underflow); + duprintf("arpt_register_table: translate table gives %d\n", ret); if (ret != 0) { - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } - ret = down_interruptible(&arpt_mutex); - if (ret != 0) { - free_table_info(newinfo); + if (xt_register_table(table, &bootstrap, newinfo) != 0) { + xt_free_table_info(newinfo); return ret; } - /* Don't autoload: we'd eat our tail... */ - if (list_named_find(&arpt_tables, table->name)) { - ret = -EEXIST; - goto free_unlock; - } - - /* Simplifies replace_table code. */ - table->private = &bootstrap; - if (!replace_table(table, 0, newinfo, &ret)) - goto free_unlock; - - duprintf("table->private->number = %u\n", - table->private->number); - - /* save number of initial entries */ - table->private->initial_entries = table->private->number; - - rwlock_init(&table->lock); - list_prepend(&arpt_tables, table); - - unlock: - up(&arpt_mutex); - return ret; - - free_unlock: - free_table_info(newinfo); - goto unlock; + return 0; } void arpt_unregister_table(struct arpt_table *table) { + struct xt_table_info *private; void *loc_cpu_entry; - down(&arpt_mutex); - LIST_DELETE(&arpt_tables, table); - up(&arpt_mutex); + private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; - ARPT_ENTRY_ITERATE(loc_cpu_entry, table->private->size, + loc_cpu_entry = private->entries[raw_smp_processor_id()]; + ARPT_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL); - free_table_info(table->private); + xt_free_table_info(private); } /* The built-in targets: standard (NULL) and error. */ @@ -1380,52 +1148,15 @@ static struct nf_sockopt_ops arpt_sockopts = { .get = do_arpt_get_ctl, }; -#ifdef CONFIG_PROC_FS -static inline int print_name(const struct arpt_table *t, - off_t start_offset, char *buffer, int length, - off_t *pos, unsigned int *count) -{ - if ((*count)++ >= start_offset) { - unsigned int namelen; - - namelen = sprintf(buffer + *pos, "%s\n", t->name); - if (*pos + namelen > length) { - /* Stop iterating */ - return 1; - } - *pos += namelen; - } - return 0; -} - -static int arpt_get_tables(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&arpt_mutex) != 0) - return 0; - - LIST_FIND(&arpt_tables, print_name, struct arpt_table *, - offset, buffer, length, &pos, &count); - - up(&arpt_mutex); - - /* `start' hack - see fs/proc/generic.c line ~105 */ - *start=(char *)((unsigned long)count-offset); - return pos; -} -#endif /*CONFIG_PROC_FS*/ - static int __init init(void) { int ret; + xt_proto_init(NF_ARP); + /* Noone else will be downing sem now, so we won't sleep */ - down(&arpt_mutex); - list_append(&arpt_target, &arpt_standard_target); - list_append(&arpt_target, &arpt_error_target); - up(&arpt_mutex); + xt_register_target(NF_ARP, &arpt_standard_target); + xt_register_target(NF_ARP, &arpt_error_target); /* Register setsockopt */ ret = nf_register_sockopt(&arpt_sockopts); @@ -1434,19 +1165,6 @@ static int __init init(void) return ret; } -#ifdef CONFIG_PROC_FS - { - struct proc_dir_entry *proc; - - proc = proc_net_create("arp_tables_names", 0, arpt_get_tables); - if (!proc) { - nf_unregister_sockopt(&arpt_sockopts); - return -ENOMEM; - } - proc->owner = THIS_MODULE; - } -#endif - printk("arp_tables: (C) 2002 David S. Miller\n"); return 0; } @@ -1454,16 +1172,12 @@ static int __init init(void) static void __exit fini(void) { nf_unregister_sockopt(&arpt_sockopts); -#ifdef CONFIG_PROC_FS - proc_net_remove("arp_tables_names"); -#endif + xt_proto_fini(NF_ARP); } EXPORT_SYMBOL(arpt_register_table); EXPORT_SYMBOL(arpt_unregister_table); EXPORT_SYMBOL(arpt_do_table); -EXPORT_SYMBOL(arpt_register_target); -EXPORT_SYMBOL(arpt_unregister_target); module_init(init); module_exit(fini); diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index 3e592ec8648..c97650a16a5 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -8,8 +8,9 @@ MODULE_AUTHOR("Bart De Schuymer <bdschuym@pandora.be>"); MODULE_DESCRIPTION("arptables arp payload mangle target"); static unsigned int -target(struct sk_buff **pskb, unsigned int hooknum, const struct net_device *in, - const struct net_device *out, const void *targinfo, void *userinfo) +target(struct sk_buff **pskb, const struct net_device *in, + const struct net_device *out, unsigned int hooknum, const void *targinfo, + void *userinfo) { const struct arpt_mangle *mangle = targinfo; struct arphdr *arp; @@ -64,7 +65,7 @@ target(struct sk_buff **pskb, unsigned int hooknum, const struct net_device *in, } static int -checkentry(const char *tablename, const struct arpt_entry *e, void *targinfo, +checkentry(const char *tablename, const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { const struct arpt_mangle *mangle = targinfo; diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 0d759f5a4ef..f6ab45f4868 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -145,6 +145,7 @@ static struct arpt_table packet_filter = { .lock = RW_LOCK_UNLOCKED, .private = NULL, .me = THIS_MODULE, + .af = NF_ARP, }; /* The work comes in here from netfilter.c */ diff --git a/net/ipv4/netfilter/ip_conntrack_standalone.c b/net/ipv4/netfilter/ip_conntrack_standalone.c index 9dec1293f67..833fcb4be5e 100644 --- a/net/ipv4/netfilter/ip_conntrack_standalone.c +++ b/net/ipv4/netfilter/ip_conntrack_standalone.c @@ -944,7 +944,7 @@ module_exit(fini); /* Some modules need us, but don't depend directly on any symbol. They should call this. */ -void need_ip_conntrack(void) +void need_conntrack(void) { } @@ -962,7 +962,7 @@ EXPORT_SYMBOL(ip_ct_get_tuple); EXPORT_SYMBOL(invert_tuplepr); EXPORT_SYMBOL(ip_conntrack_alter_reply); EXPORT_SYMBOL(ip_conntrack_destroyed); -EXPORT_SYMBOL(need_ip_conntrack); +EXPORT_SYMBOL(need_conntrack); EXPORT_SYMBOL(ip_conntrack_helper_register); EXPORT_SYMBOL(ip_conntrack_helper_unregister); EXPORT_SYMBOL(ip_ct_iterate_cleanup); diff --git a/net/ipv4/netfilter/ip_nat_rule.c b/net/ipv4/netfilter/ip_nat_rule.c index cb66b8bddeb..1de86282d23 100644 --- a/net/ipv4/netfilter/ip_nat_rule.c +++ b/net/ipv4/netfilter/ip_nat_rule.c @@ -95,6 +95,7 @@ static struct ipt_table nat_table = { .valid_hooks = NAT_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, .me = THIS_MODULE, + .af = AF_INET, }; /* Source NAT */ @@ -168,7 +169,7 @@ static unsigned int ipt_dnat_target(struct sk_buff **pskb, } static int ipt_snat_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) @@ -201,7 +202,7 @@ static int ipt_snat_checkentry(const char *tablename, } static int ipt_dnat_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index 8b8a1f00bbf..ad438fb185b 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -364,7 +364,7 @@ static int init_or_cleanup(int init) { int ret = 0; - need_ip_conntrack(); + need_conntrack(); if (!init) goto cleanup; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 877bc96d333..2371b2062c2 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -2,7 +2,7 @@ * Packet matching code. * * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling - * Copyright (C) 2000-2004 Netfilter Core Team <coreteam@netfilter.org> + * Copyright (C) 2000-2005 Netfilter Core Team <coreteam@netfilter.org> * * 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 @@ -11,6 +11,8 @@ * 19 Jan 2002 Harald Welte <laforge@gnumonks.org> * - increase module usage count as soon as we have rules inside * a table + * 08 Oct 2005 Harald Welte <lafore@netfilter.org> + * - Generalize into "x_tables" layer and "{ip,ip6,arp}_tables" */ #include <linux/config.h> #include <linux/cache.h> @@ -20,8 +22,6 @@ #include <linux/vmalloc.h> #include <linux/netdevice.h> #include <linux/module.h> -#include <linux/tcp.h> -#include <linux/udp.h> #include <linux/icmp.h> #include <net/ip.h> #include <asm/uaccess.h> @@ -30,6 +30,7 @@ #include <linux/err.h> #include <linux/cpumask.h> +#include <linux/netfilter/x_tables.h> #include <linux/netfilter_ipv4/ip_tables.h> MODULE_LICENSE("GPL"); @@ -62,14 +63,6 @@ do { \ #else #define IP_NF_ASSERT(x) #endif -#define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) - -static DECLARE_MUTEX(ipt_mutex); - -/* Must have mutex */ -#define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0) -#define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ipt_mutex) != 0) -#include <linux/netfilter_ipv4/listhelp.h> #if 0 /* All the better to debug you with... */ @@ -86,36 +79,6 @@ static DECLARE_MUTEX(ipt_mutex); Hence the start of any table is given by get_table() below. */ -/* The table itself */ -struct ipt_table_info -{ - /* Size per table */ - unsigned int size; - /* Number of entries: FIXME. --RR */ - unsigned int number; - /* Initial number of entries. Needed for module usage count */ - unsigned int initial_entries; - - /* Entry points and underflows */ - unsigned int hook_entry[NF_IP_NUMHOOKS]; - unsigned int underflow[NF_IP_NUMHOOKS]; - - /* ipt_entry tables: one per CPU */ - void *entries[NR_CPUS]; -}; - -static LIST_HEAD(ipt_target); -static LIST_HEAD(ipt_match); -static LIST_HEAD(ipt_tables); -#define SET_COUNTER(c,b,p) do { (c).bcnt = (b); (c).pcnt = (p); } while(0) -#define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0) - -#if 0 -#define down(x) do { printk("DOWN:%u:" #x "\n", __LINE__); down(x); } while(0) -#define down_interruptible(x) ({ int __r; printk("DOWNi:%u:" #x "\n", __LINE__); __r = down_interruptible(x); if (__r != 0) printk("ABORT-DOWNi:%u\n", __LINE__); __r; }) -#define up(x) do { printk("UP:%u:" #x "\n", __LINE__); up(x); } while(0) -#endif - /* Returns whether matches rule or not. */ static inline int ip_packet_match(const struct iphdr *ip, @@ -234,7 +197,8 @@ int do_match(struct ipt_entry_match *m, int *hotdrop) { /* Stop iteration if it doesn't match */ - if (!m->u.kernel.match->match(skb, in, out, m->data, offset, hotdrop)) + if (!m->u.kernel.match->match(skb, in, out, m->data, offset, + skb->nh.iph->ihl*4, hotdrop)) return 1; else return 0; @@ -265,6 +229,7 @@ ipt_do_table(struct sk_buff **pskb, const char *indev, *outdev; void *table_base; struct ipt_entry *e, *back; + struct xt_table_info *private = table->private; /* Initialization */ ip = (*pskb)->nh.iph; @@ -281,24 +246,11 @@ ipt_do_table(struct sk_buff **pskb, read_lock_bh(&table->lock); IP_NF_ASSERT(table->valid_hooks & (1 << hook)); - table_base = (void *)table->private->entries[smp_processor_id()]; - e = get_entry(table_base, table->private->hook_entry[hook]); - -#ifdef CONFIG_NETFILTER_DEBUG - /* Check noone else using our table */ - if (((struct ipt_entry *)table_base)->comefrom != 0xdead57ac - && ((struct ipt_entry *)table_base)->comefrom != 0xeeeeeeec) { - printk("ASSERT: CPU #%u, %s comefrom(%p) = %X\n", - smp_processor_id(), - table->name, - &((struct ipt_entry *)table_base)->comefrom, - ((struct ipt_entry *)table_base)->comefrom); - } - ((struct ipt_entry *)table_base)->comefrom = 0x57acc001; -#endif + table_base = (void *)private->entries[smp_processor_id()]; + e = get_entry(table_base, private->hook_entry[hook]); /* For return from builtin chain */ - back = get_entry(table_base, table->private->underflow[hook]); + back = get_entry(table_base, private->underflow[hook]); do { IP_NF_ASSERT(e); @@ -384,9 +336,6 @@ ipt_do_table(struct sk_buff **pskb, } } while (!hotdrop); -#ifdef CONFIG_NETFILTER_DEBUG - ((struct ipt_entry *)table_base)->comefrom = 0xdead57ac; -#endif read_unlock_bh(&table->lock); #ifdef DEBUG_ALLOW_ALL @@ -398,145 +347,6 @@ ipt_do_table(struct sk_buff **pskb, #endif } -/* - * These are weird, but module loading must not be done with mutex - * held (since they will register), and we have to have a single - * function to use try_then_request_module(). - */ - -/* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ -static inline struct ipt_table *find_table_lock(const char *name) -{ - struct ipt_table *t; - - if (down_interruptible(&ipt_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &ipt_tables, list) - if (strcmp(t->name, name) == 0 && try_module_get(t->me)) - return t; - up(&ipt_mutex); - return NULL; -} - -/* Find match, grabs ref. Returns ERR_PTR() on error. */ -static inline struct ipt_match *find_match(const char *name, u8 revision) -{ - struct ipt_match *m; - int err = 0; - - if (down_interruptible(&ipt_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(m, &ipt_match, list) { - if (strcmp(m->name, name) == 0) { - if (m->revision == revision) { - if (try_module_get(m->me)) { - up(&ipt_mutex); - return m; - } - } else - err = -EPROTOTYPE; /* Found something. */ - } - } - up(&ipt_mutex); - return ERR_PTR(err); -} - -/* Find target, grabs ref. Returns ERR_PTR() on error. */ -static inline struct ipt_target *find_target(const char *name, u8 revision) -{ - struct ipt_target *t; - int err = 0; - - if (down_interruptible(&ipt_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &ipt_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision == revision) { - if (try_module_get(t->me)) { - up(&ipt_mutex); - return t; - } - } else - err = -EPROTOTYPE; /* Found something. */ - } - } - up(&ipt_mutex); - return ERR_PTR(err); -} - -struct ipt_target *ipt_find_target(const char *name, u8 revision) -{ - struct ipt_target *target; - - target = try_then_request_module(find_target(name, revision), - "ipt_%s", name); - if (IS_ERR(target) || !target) - return NULL; - return target; -} - -static int match_revfn(const char *name, u8 revision, int *bestp) -{ - struct ipt_match *m; - int have_rev = 0; - - list_for_each_entry(m, &ipt_match, list) { - if (strcmp(m->name, name) == 0) { - if (m->revision > *bestp) - *bestp = m->revision; - if (m->revision == revision) - have_rev = 1; - } - } - return have_rev; -} - -static int target_revfn(const char *name, u8 revision, int *bestp) -{ - struct ipt_target *t; - int have_rev = 0; - - list_for_each_entry(t, &ipt_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision > *bestp) - *bestp = t->revision; - if (t->revision == revision) - have_rev = 1; - } - } - return have_rev; -} - -/* Returns true or false (if no such extension at all) */ -static inline int find_revision(const char *name, u8 revision, - int (*revfn)(const char *, u8, int *), - int *err) -{ - int have_rev, best = -1; - - if (down_interruptible(&ipt_mutex) != 0) { - *err = -EINTR; - return 1; - } - have_rev = revfn(name, revision, &best); - up(&ipt_mutex); - - /* Nothing at all? Return 0 to try loading module. */ - if (best == -1) { - *err = -ENOENT; - return 0; - } - - *err = best; - if (!have_rev) - *err = -EPROTONOSUPPORT; - return 1; -} - - /* All zeroes == unconditional rule. */ static inline int unconditional(const struct ipt_ip *ip) @@ -553,7 +363,7 @@ unconditional(const struct ipt_ip *ip) /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int -mark_source_chains(struct ipt_table_info *newinfo, +mark_source_chains(struct xt_table_info *newinfo, unsigned int valid_hooks, void *entry0) { unsigned int hook; @@ -699,7 +509,7 @@ check_match(struct ipt_entry_match *m, { struct ipt_match *match; - match = try_then_request_module(find_match(m->u.user.name, + match = try_then_request_module(xt_find_match(AF_INET, m->u.user.name, m->u.user.revision), "ipt_%s", m->u.user.name); if (IS_ERR(match) || !match) { @@ -744,7 +554,8 @@ check_entry(struct ipt_entry *e, const char *name, unsigned int size, goto cleanup_matches; t = ipt_get_target(e); - target = try_then_request_module(find_target(t->u.user.name, + target = try_then_request_module(xt_find_target(AF_INET, + t->u.user.name, t->u.user.revision), "ipt_%s", t->u.user.name); if (IS_ERR(target) || !target) { @@ -781,7 +592,7 @@ check_entry(struct ipt_entry *e, const char *name, unsigned int size, static inline int check_entry_size_and_hooks(struct ipt_entry *e, - struct ipt_table_info *newinfo, + struct xt_table_info *newinfo, unsigned char *base, unsigned char *limit, const unsigned int *hook_entries, @@ -815,7 +626,7 @@ check_entry_size_and_hooks(struct ipt_entry *e, < 0 (not IPT_RETURN). --RR */ /* Clear counters and comefrom */ - e->counters = ((struct ipt_counters) { 0, 0 }); + e->counters = ((struct xt_counters) { 0, 0 }); e->comefrom = 0; (*i)++; @@ -845,7 +656,7 @@ cleanup_entry(struct ipt_entry *e, unsigned int *i) static int translate_table(const char *name, unsigned int valid_hooks, - struct ipt_table_info *newinfo, + struct xt_table_info *newinfo, void *entry0, unsigned int size, unsigned int number, @@ -922,48 +733,10 @@ translate_table(const char *name, return ret; } -static struct ipt_table_info * -replace_table(struct ipt_table *table, - unsigned int num_counters, - struct ipt_table_info *newinfo, - int *error) -{ - struct ipt_table_info *oldinfo; - -#ifdef CONFIG_NETFILTER_DEBUG - { - int cpu; - - for_each_cpu(cpu) { - struct ipt_entry *table_base = newinfo->entries[cpu]; - if (table_base) - table_base->comefrom = 0xdead57ac; - } - } -#endif - - /* Do the substitution. */ - write_lock_bh(&table->lock); - /* Check inside lock: is the old number correct? */ - if (num_counters != table->private->number) { - duprintf("num_counters != table->private->number (%u/%u)\n", - num_counters, table->private->number); - write_unlock_bh(&table->lock); - *error = -EAGAIN; - return NULL; - } - oldinfo = table->private; - table->private = newinfo; - newinfo->initial_entries = oldinfo->initial_entries; - write_unlock_bh(&table->lock); - - return oldinfo; -} - /* Gets counters. */ static inline int add_entry_to_counter(const struct ipt_entry *e, - struct ipt_counters total[], + struct xt_counters total[], unsigned int *i) { ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt); @@ -984,8 +757,8 @@ set_entry_to_counter(const struct ipt_entry *e, } static void -get_counters(const struct ipt_table_info *t, - struct ipt_counters counters[]) +get_counters(const struct xt_table_info *t, + struct xt_counters counters[]) { unsigned int cpu; unsigned int i; @@ -1024,14 +797,15 @@ copy_entries_to_user(unsigned int total_size, { unsigned int off, num, countersize; struct ipt_entry *e; - struct ipt_counters *counters; + struct xt_counters *counters; + struct xt_table_info *private = table->private; int ret = 0; void *loc_cpu_entry; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care about). */ - countersize = sizeof(struct ipt_counters) * table->private->number; + countersize = sizeof(struct xt_counters) * private->number; counters = vmalloc_node(countersize, numa_node_id()); if (counters == NULL) @@ -1039,14 +813,14 @@ copy_entries_to_user(unsigned int total_size, /* First, sum counters... */ write_lock_bh(&table->lock); - get_counters(table->private, counters); + get_counters(private, counters); write_unlock_bh(&table->lock); /* choose the copy that is on our node/cpu, ... * This choice is lazy (because current thread is * allowed to migrate to another cpu) */ - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; /* ... then copy entire thing ... */ if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; @@ -1108,74 +882,36 @@ get_entries(const struct ipt_get_entries *entries, int ret; struct ipt_table *t; - t = find_table_lock(entries->name); + t = xt_find_table_lock(AF_INET, entries->name); if (t && !IS_ERR(t)) { + struct xt_table_info *private = t->private; duprintf("t->private->number = %u\n", - t->private->number); - if (entries->size == t->private->size) - ret = copy_entries_to_user(t->private->size, + private->number); + if (entries->size == private->size) + ret = copy_entries_to_user(private->size, t, uptr->entrytable); else { duprintf("get_entries: I've got %u not %u!\n", - t->private->size, + private->size, entries->size); ret = -EINVAL; } module_put(t->me); - up(&ipt_mutex); + xt_table_unlock(t); } else ret = t ? PTR_ERR(t) : -ENOENT; return ret; } -static void free_table_info(struct ipt_table_info *info) -{ - int cpu; - for_each_cpu(cpu) { - if (info->size <= PAGE_SIZE) - kfree(info->entries[cpu]); - else - vfree(info->entries[cpu]); - } - kfree(info); -} - -static struct ipt_table_info *alloc_table_info(unsigned int size) -{ - struct ipt_table_info *newinfo; - int cpu; - - newinfo = kzalloc(sizeof(struct ipt_table_info), GFP_KERNEL); - if (!newinfo) - return NULL; - - newinfo->size = size; - - for_each_cpu(cpu) { - if (size <= PAGE_SIZE) - newinfo->entries[cpu] = kmalloc_node(size, - GFP_KERNEL, - cpu_to_node(cpu)); - else - newinfo->entries[cpu] = vmalloc_node(size, cpu_to_node(cpu)); - if (newinfo->entries[cpu] == 0) { - free_table_info(newinfo); - return NULL; - } - } - - return newinfo; -} - static int do_replace(void __user *user, unsigned int len) { int ret; struct ipt_replace tmp; struct ipt_table *t; - struct ipt_table_info *newinfo, *oldinfo; - struct ipt_counters *counters; + struct xt_table_info *newinfo, *oldinfo; + struct xt_counters *counters; void *loc_cpu_entry, *loc_cpu_old_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) @@ -1185,11 +921,7 @@ do_replace(void __user *user, unsigned int len) if (len != sizeof(tmp) + tmp.size) return -ENOPROTOOPT; - /* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */ - if ((SMP_ALIGN(tmp.size) >> PAGE_SHIFT) + 2 > num_physpages) - return -ENOMEM; - - newinfo = alloc_table_info(tmp.size); + newinfo = xt_alloc_table_info(tmp.size); if (!newinfo) return -ENOMEM; @@ -1201,7 +933,7 @@ do_replace(void __user *user, unsigned int len) goto free_newinfo; } - counters = vmalloc(tmp.num_counters * sizeof(struct ipt_counters)); + counters = vmalloc(tmp.num_counters * sizeof(struct xt_counters)); if (!counters) { ret = -ENOMEM; goto free_newinfo; @@ -1215,7 +947,7 @@ do_replace(void __user *user, unsigned int len) duprintf("ip_tables: Translated table\n"); - t = try_then_request_module(find_table_lock(tmp.name), + t = try_then_request_module(xt_find_table_lock(AF_INET, tmp.name), "iptable_%s", tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; @@ -1230,7 +962,7 @@ do_replace(void __user *user, unsigned int len) goto put_module; } - oldinfo = replace_table(t, tmp.num_counters, newinfo, &ret); + oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret); if (!oldinfo) goto put_module; @@ -1249,23 +981,23 @@ do_replace(void __user *user, unsigned int len) /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; IPT_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL); - free_table_info(oldinfo); + xt_free_table_info(oldinfo); if (copy_to_user(tmp.counters, counters, - sizeof(struct ipt_counters) * tmp.num_counters) != 0) + sizeof(struct xt_counters) * tmp.num_counters) != 0) ret = -EFAULT; vfree(counters); - up(&ipt_mutex); + xt_table_unlock(t); return ret; put_module: module_put(t->me); - up(&ipt_mutex); + xt_table_unlock(t); free_newinfo_counters_untrans: IPT_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry,NULL); free_newinfo_counters: vfree(counters); free_newinfo: - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } @@ -1273,7 +1005,7 @@ do_replace(void __user *user, unsigned int len) * and everything is OK. */ static inline int add_counter_to_entry(struct ipt_entry *e, - const struct ipt_counters addme[], + const struct xt_counters addme[], unsigned int *i) { #if 0 @@ -1295,15 +1027,16 @@ static int do_add_counters(void __user *user, unsigned int len) { unsigned int i; - struct ipt_counters_info tmp, *paddc; + struct xt_counters_info tmp, *paddc; struct ipt_table *t; + struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; - if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct ipt_counters)) + if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters)) return -EINVAL; paddc = vmalloc_node(len, numa_node_id()); @@ -1315,29 +1048,30 @@ do_add_counters(void __user *user, unsigned int len) goto free; } - t = find_table_lock(tmp.name); + t = xt_find_table_lock(AF_INET, tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; goto free; } write_lock_bh(&t->lock); - if (t->private->number != paddc->num_counters) { + private = t->private; + if (private->number != paddc->num_counters) { ret = -EINVAL; goto unlock_up_free; } i = 0; /* Choose the copy that is on our node */ - loc_cpu_entry = t->private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; IPT_ENTRY_ITERATE(loc_cpu_entry, - t->private->size, + private->size, add_counter_to_entry, paddc->counters, &i); unlock_up_free: write_unlock_bh(&t->lock); - up(&ipt_mutex); + xt_table_unlock(t); module_put(t->me); free: vfree(paddc); @@ -1396,25 +1130,26 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) } name[IPT_TABLE_MAXNAMELEN-1] = '\0'; - t = try_then_request_module(find_table_lock(name), + t = try_then_request_module(xt_find_table_lock(AF_INET, name), "iptable_%s", name); if (t && !IS_ERR(t)) { struct ipt_getinfo info; + struct xt_table_info *private = t->private; info.valid_hooks = t->valid_hooks; - memcpy(info.hook_entry, t->private->hook_entry, + memcpy(info.hook_entry, private->hook_entry, sizeof(info.hook_entry)); - memcpy(info.underflow, t->private->underflow, + memcpy(info.underflow, private->underflow, sizeof(info.underflow)); - info.num_entries = t->private->number; - info.size = t->private->size; + info.num_entries = private->number; + info.size = private->size; memcpy(info.name, name, sizeof(info.name)); if (copy_to_user(user, &info, *len) != 0) ret = -EFAULT; else ret = 0; - up(&ipt_mutex); + xt_table_unlock(t); module_put(t->me); } else ret = t ? PTR_ERR(t) : -ENOENT; @@ -1441,7 +1176,7 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) case IPT_SO_GET_REVISION_MATCH: case IPT_SO_GET_REVISION_TARGET: { struct ipt_get_revision rev; - int (*revfn)(const char *, u8, int *); + int target; if (*len != sizeof(rev)) { ret = -EINVAL; @@ -1453,12 +1188,13 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) } if (cmd == IPT_SO_GET_REVISION_TARGET) - revfn = target_revfn; + target = 1; else - revfn = match_revfn; + target = 0; - try_then_request_module(find_revision(rev.name, rev.revision, - revfn, &ret), + try_then_request_module(xt_find_revision(AF_INET, rev.name, + rev.revision, + target, &ret), "ipt_%s", rev.name); break; } @@ -1471,60 +1207,15 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) return ret; } -/* Registration hooks for targets. */ -int -ipt_register_target(struct ipt_target *target) +int ipt_register_table(struct xt_table *table, const struct ipt_replace *repl) { int ret; - - ret = down_interruptible(&ipt_mutex); - if (ret != 0) - return ret; - list_add(&target->list, &ipt_target); - up(&ipt_mutex); - return ret; -} - -void -ipt_unregister_target(struct ipt_target *target) -{ - down(&ipt_mutex); - LIST_DELETE(&ipt_target, target); - up(&ipt_mutex); -} - -int -ipt_register_match(struct ipt_match *match) -{ - int ret; - - ret = down_interruptible(&ipt_mutex); - if (ret != 0) - return ret; - - list_add(&match->list, &ipt_match); - up(&ipt_mutex); - - return ret; -} - -void -ipt_unregister_match(struct ipt_match *match) -{ - down(&ipt_mutex); - LIST_DELETE(&ipt_match, match); - up(&ipt_mutex); -} - -int ipt_register_table(struct ipt_table *table, const struct ipt_replace *repl) -{ - int ret; - struct ipt_table_info *newinfo; - static struct ipt_table_info bootstrap + struct xt_table_info *newinfo; + static struct xt_table_info bootstrap = { 0, 0, 0, { 0 }, { 0 }, { } }; void *loc_cpu_entry; - newinfo = alloc_table_info(repl->size); + newinfo = xt_alloc_table_info(repl->size); if (!newinfo) return -ENOMEM; @@ -1540,246 +1231,29 @@ int ipt_register_table(struct ipt_table *table, const struct ipt_replace *repl) repl->hook_entry, repl->underflow); if (ret != 0) { - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } - ret = down_interruptible(&ipt_mutex); - if (ret != 0) { - free_table_info(newinfo); + if (xt_register_table(table, &bootstrap, newinfo) != 0) { + xt_free_table_info(newinfo); return ret; } - /* Don't autoload: we'd eat our tail... */ - if (list_named_find(&ipt_tables, table->name)) { - ret = -EEXIST; - goto free_unlock; - } - - /* Simplifies replace_table code. */ - table->private = &bootstrap; - if (!replace_table(table, 0, newinfo, &ret)) - goto free_unlock; - - duprintf("table->private->number = %u\n", - table->private->number); - - /* save number of initial entries */ - table->private->initial_entries = table->private->number; - - rwlock_init(&table->lock); - list_prepend(&ipt_tables, table); - - unlock: - up(&ipt_mutex); - return ret; - - free_unlock: - free_table_info(newinfo); - goto unlock; + return 0; } void ipt_unregister_table(struct ipt_table *table) { + struct xt_table_info *private; void *loc_cpu_entry; - down(&ipt_mutex); - LIST_DELETE(&ipt_tables, table); - up(&ipt_mutex); + private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; - IPT_ENTRY_ITERATE(loc_cpu_entry, table->private->size, - cleanup_entry, NULL); - free_table_info(table->private); -} - -/* Returns 1 if the port is matched by the range, 0 otherwise */ -static inline int -port_match(u_int16_t min, u_int16_t max, u_int16_t port, int invert) -{ - int ret; - - ret = (port >= min && port <= max) ^ invert; - return ret; -} - -static int -tcp_find_option(u_int8_t option, - const struct sk_buff *skb, - unsigned int optlen, - int invert, - int *hotdrop) -{ - /* tcp.doff is only 4 bits, ie. max 15 * 4 bytes */ - u_int8_t _opt[60 - sizeof(struct tcphdr)], *op; - unsigned int i; - - duprintf("tcp_match: finding option\n"); - - if (!optlen) - return invert; - - /* If we don't have the whole header, drop packet. */ - op = skb_header_pointer(skb, - skb->nh.iph->ihl*4 + sizeof(struct tcphdr), - optlen, _opt); - if (op == NULL) { - *hotdrop = 1; - return 0; - } - - for (i = 0; i < optlen; ) { - if (op[i] == option) return !invert; - if (op[i] < 2) i++; - else i += op[i+1]?:1; - } - - return invert; -} - -static int -tcp_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - int *hotdrop) -{ - struct tcphdr _tcph, *th; - const struct ipt_tcp *tcpinfo = matchinfo; - - if (offset) { - /* To quote Alan: - - Don't allow a fragment of TCP 8 bytes in. Nobody normal - causes this. Its a cracker trying to break in by doing a - flag overwrite to pass the direction checks. - */ - if (offset == 1) { - duprintf("Dropping evil TCP offset=1 frag.\n"); - *hotdrop = 1; - } - /* Must not be a fragment. */ - return 0; - } - -#define FWINVTCP(bool,invflg) ((bool) ^ !!(tcpinfo->invflags & invflg)) - - th = skb_header_pointer(skb, skb->nh.iph->ihl*4, - sizeof(_tcph), &_tcph); - if (th == NULL) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - duprintf("Dropping evil TCP offset=0 tinygram.\n"); - *hotdrop = 1; - return 0; - } - - if (!port_match(tcpinfo->spts[0], tcpinfo->spts[1], - ntohs(th->source), - !!(tcpinfo->invflags & IPT_TCP_INV_SRCPT))) - return 0; - if (!port_match(tcpinfo->dpts[0], tcpinfo->dpts[1], - ntohs(th->dest), - !!(tcpinfo->invflags & IPT_TCP_INV_DSTPT))) - return 0; - if (!FWINVTCP((((unsigned char *)th)[13] & tcpinfo->flg_mask) - == tcpinfo->flg_cmp, - IPT_TCP_INV_FLAGS)) - return 0; - if (tcpinfo->option) { - if (th->doff * 4 < sizeof(_tcph)) { - *hotdrop = 1; - return 0; - } - if (!tcp_find_option(tcpinfo->option, skb, - th->doff*4 - sizeof(_tcph), - tcpinfo->invflags & IPT_TCP_INV_OPTION, - hotdrop)) - return 0; - } - return 1; -} - -/* Called when user tries to insert an entry of this type. */ -static int -tcp_checkentry(const char *tablename, - const struct ipt_ip *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - const struct ipt_tcp *tcpinfo = matchinfo; - - /* Must specify proto == TCP, and no unknown invflags */ - return ip->proto == IPPROTO_TCP - && !(ip->invflags & IPT_INV_PROTO) - && matchsize == IPT_ALIGN(sizeof(struct ipt_tcp)) - && !(tcpinfo->invflags & ~IPT_TCP_INV_MASK); -} - -static int -udp_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - int *hotdrop) -{ - struct udphdr _udph, *uh; - const struct ipt_udp *udpinfo = matchinfo; - - /* Must not be a fragment. */ - if (offset) - return 0; - - uh = skb_header_pointer(skb, skb->nh.iph->ihl*4, - sizeof(_udph), &_udph); - if (uh == NULL) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - duprintf("Dropping evil UDP tinygram.\n"); - *hotdrop = 1; - return 0; - } - - return port_match(udpinfo->spts[0], udpinfo->spts[1], - ntohs(uh->source), - !!(udpinfo->invflags & IPT_UDP_INV_SRCPT)) - && port_match(udpinfo->dpts[0], udpinfo->dpts[1], - ntohs(uh->dest), - !!(udpinfo->invflags & IPT_UDP_INV_DSTPT)); -} - -/* Called when user tries to insert an entry of this type. */ -static int -udp_checkentry(const char *tablename, - const struct ipt_ip *ip, - void *matchinfo, - unsigned int matchinfosize, - unsigned int hook_mask) -{ - const struct ipt_udp *udpinfo = matchinfo; - - /* Must specify proto == UDP, and no unknown invflags */ - if (ip->proto != IPPROTO_UDP || (ip->invflags & IPT_INV_PROTO)) { - duprintf("ipt_udp: Protocol %u != %u\n", ip->proto, - IPPROTO_UDP); - return 0; - } - if (matchinfosize != IPT_ALIGN(sizeof(struct ipt_udp))) { - duprintf("ipt_udp: matchsize %u != %u\n", - matchinfosize, IPT_ALIGN(sizeof(struct ipt_udp))); - return 0; - } - if (udpinfo->invflags & ~IPT_UDP_INV_MASK) { - duprintf("ipt_udp: unknown flags %X\n", - udpinfo->invflags); - return 0; - } - - return 1; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; + IPT_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL); + xt_free_table_info(private); } /* Returns 1 if the type and code is matched by the range, 0 otherwise */ @@ -1798,6 +1272,7 @@ icmp_match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { struct icmphdr _icmph, *ic; @@ -1807,8 +1282,7 @@ icmp_match(const struct sk_buff *skb, if (offset) return 0; - ic = skb_header_pointer(skb, skb->nh.iph->ihl*4, - sizeof(_icmph), &_icmph); + ic = skb_header_pointer(skb, protoff, sizeof(_icmph), &_icmph); if (ic == NULL) { /* We've been asked to examine this packet, and we * can't. Hence, no choice but to drop. @@ -1828,11 +1302,12 @@ icmp_match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int icmp_checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *info, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { + const struct ipt_ip *ip = info; const struct ipt_icmp *icmpinfo = matchinfo; /* Must specify proto == ICMP, and no unknown invflags */ @@ -1862,123 +1337,22 @@ static struct nf_sockopt_ops ipt_sockopts = { .get = do_ipt_get_ctl, }; -static struct ipt_match tcp_matchstruct = { - .name = "tcp", - .match = &tcp_match, - .checkentry = &tcp_checkentry, -}; - -static struct ipt_match udp_matchstruct = { - .name = "udp", - .match = &udp_match, - .checkentry = &udp_checkentry, -}; - static struct ipt_match icmp_matchstruct = { .name = "icmp", .match = &icmp_match, .checkentry = &icmp_checkentry, }; -#ifdef CONFIG_PROC_FS -static inline int print_name(const char *i, - off_t start_offset, char *buffer, int length, - off_t *pos, unsigned int *count) -{ - if ((*count)++ >= start_offset) { - unsigned int namelen; - - namelen = sprintf(buffer + *pos, "%s\n", - i + sizeof(struct list_head)); - if (*pos + namelen > length) { - /* Stop iterating */ - return 1; - } - *pos += namelen; - } - return 0; -} - -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_tables(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ipt_mutex) != 0) - return 0; - - LIST_FIND(&ipt_tables, print_name, void *, - offset, buffer, length, &pos, &count); - - up(&ipt_mutex); - - /* `start' hack - see fs/proc/generic.c line ~105 */ - *start=(char *)((unsigned long)count-offset); - return pos; -} - -static int ipt_get_targets(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ipt_mutex) != 0) - return 0; - - LIST_FIND(&ipt_target, print_target, struct ipt_target *, - offset, buffer, length, &pos, &count); - - up(&ipt_mutex); - - *start = (char *)((unsigned long)count - offset); - return pos; -} - -static int ipt_get_matches(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ipt_mutex) != 0) - return 0; - - LIST_FIND(&ipt_match, print_name, void *, - offset, buffer, length, &pos, &count); - - up(&ipt_mutex); - - *start = (char *)((unsigned long)count - offset); - return pos; -} - -static const struct { char *name; get_info_t *get_info; } ipt_proc_entry[] = -{ { "ip_tables_names", ipt_get_tables }, - { "ip_tables_targets", ipt_get_targets }, - { "ip_tables_matches", ipt_get_matches }, - { NULL, NULL} }; -#endif /*CONFIG_PROC_FS*/ - static int __init init(void) { int ret; + xt_proto_init(AF_INET); + /* Noone else will be downing sem now, so we won't sleep */ - down(&ipt_mutex); - list_append(&ipt_target, &ipt_standard_target); - list_append(&ipt_target, &ipt_error_target); - list_append(&ipt_match, &tcp_matchstruct); - list_append(&ipt_match, &udp_matchstruct); - list_append(&ipt_match, &icmp_matchstruct); - up(&ipt_mutex); + xt_register_target(AF_INET, &ipt_standard_target); + xt_register_target(AF_INET, &ipt_error_target); + xt_register_match(AF_INET, &icmp_matchstruct); /* Register setsockopt */ ret = nf_register_sockopt(&ipt_sockopts); @@ -1987,49 +1361,23 @@ static int __init init(void) return ret; } -#ifdef CONFIG_PROC_FS - { - struct proc_dir_entry *proc; - int i; - - for (i = 0; ipt_proc_entry[i].name; i++) { - proc = proc_net_create(ipt_proc_entry[i].name, 0, - ipt_proc_entry[i].get_info); - if (!proc) { - while (--i >= 0) - proc_net_remove(ipt_proc_entry[i].name); - nf_unregister_sockopt(&ipt_sockopts); - return -ENOMEM; - } - proc->owner = THIS_MODULE; - } - } -#endif - - printk("ip_tables: (C) 2000-2002 Netfilter core team\n"); + printk("ip_tables: (C) 2000-2006 Netfilter Core Team\n"); return 0; } static void __exit fini(void) { nf_unregister_sockopt(&ipt_sockopts); -#ifdef CONFIG_PROC_FS - { - int i; - for (i = 0; ipt_proc_entry[i].name; i++) - proc_net_remove(ipt_proc_entry[i].name); - } -#endif + + xt_unregister_match(AF_INET, &icmp_matchstruct); + xt_unregister_target(AF_INET, &ipt_error_target); + xt_unregister_target(AF_INET, &ipt_standard_target); + + xt_proto_fini(AF_INET); } EXPORT_SYMBOL(ipt_register_table); EXPORT_SYMBOL(ipt_unregister_table); -EXPORT_SYMBOL(ipt_register_match); -EXPORT_SYMBOL(ipt_unregister_match); EXPORT_SYMBOL(ipt_do_table); -EXPORT_SYMBOL(ipt_register_target); -EXPORT_SYMBOL(ipt_unregister_target); -EXPORT_SYMBOL(ipt_find_target); - module_init(init); module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 45c52d8f4d9..d9bc971f03a 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -379,12 +379,13 @@ target(struct sk_buff **pskb, static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { struct ipt_clusterip_tgt_info *cipinfo = targinfo; + const struct ipt_entry *e = e_void; struct clusterip_config *config; diff --git a/net/ipv4/netfilter/ipt_DSCP.c b/net/ipv4/netfilter/ipt_DSCP.c index 6e319570a28..898cdf79ce1 100644 --- a/net/ipv4/netfilter/ipt_DSCP.c +++ b/net/ipv4/netfilter/ipt_DSCP.c @@ -57,7 +57,7 @@ target(struct sk_buff **pskb, static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_ECN.c b/net/ipv4/netfilter/ipt_ECN.c index a1319693f64..706445426a6 100644 --- a/net/ipv4/netfilter/ipt_ECN.c +++ b/net/ipv4/netfilter/ipt_ECN.c @@ -113,12 +113,13 @@ target(struct sk_buff **pskb, static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { const struct ipt_ECN_info *einfo = (struct ipt_ECN_info *)targinfo; + const struct ipt_entry *e = e_void; if (targinfosize != IPT_ALIGN(sizeof(struct ipt_ECN_info))) { printk(KERN_WARNING "ECN: targinfosize %u != %Zu\n", diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c index 30be0f1dae3..6606ddb66a2 100644 --- a/net/ipv4/netfilter/ipt_LOG.c +++ b/net/ipv4/netfilter/ipt_LOG.c @@ -431,7 +431,7 @@ ipt_log_target(struct sk_buff **pskb, } static int ipt_log_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index 27860510ca6..12c56d3343c 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -40,7 +40,7 @@ static DEFINE_RWLOCK(masq_lock); /* FIXME: Multiple targets. --RR */ static int masquerade_check(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_NETMAP.c b/net/ipv4/netfilter/ipt_NETMAP.c index e6e7b609536..b074467fe67 100644 --- a/net/ipv4/netfilter/ipt_NETMAP.c +++ b/net/ipv4/netfilter/ipt_NETMAP.c @@ -31,7 +31,7 @@ MODULE_DESCRIPTION("iptables 1:1 NAT mapping of IP networks target"); static int check(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_NFQUEUE.c b/net/ipv4/netfilter/ipt_NFQUEUE.c deleted file mode 100644 index 3cedc9be880..00000000000 --- a/net/ipv4/netfilter/ipt_NFQUEUE.c +++ /dev/null @@ -1,70 +0,0 @@ -/* iptables module for using new netfilter netlink queue - * - * (C) 2005 by Harald Welte <laforge@netfilter.org> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> - -#include <linux/netfilter.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_NFQUEUE.h> - -MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); -MODULE_DESCRIPTION("iptables NFQUEUE target"); -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) -{ - const struct ipt_NFQ_info *tinfo = targinfo; - - return NF_QUEUE_NR(tinfo->queuenum); -} - -static int -checkentry(const char *tablename, - const struct ipt_entry *e, - void *targinfo, - unsigned int targinfosize, - unsigned int hook_mask) -{ - if (targinfosize != IPT_ALIGN(sizeof(struct ipt_NFQ_info))) { - printk(KERN_WARNING "NFQUEUE: targinfosize %u != %Zu\n", - targinfosize, - IPT_ALIGN(sizeof(struct ipt_NFQ_info))); - return 0; - } - - return 1; -} - -static struct ipt_target ipt_NFQ_reg = { - .name = "NFQUEUE", - .target = target, - .checkentry = checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ipt_register_target(&ipt_NFQ_reg); -} - -static void __exit fini(void) -{ - ipt_unregister_target(&ipt_NFQ_reg); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_REDIRECT.c b/net/ipv4/netfilter/ipt_REDIRECT.c index 5245bfd33d5..140be51f2f0 100644 --- a/net/ipv4/netfilter/ipt_REDIRECT.c +++ b/net/ipv4/netfilter/ipt_REDIRECT.c @@ -33,7 +33,7 @@ MODULE_DESCRIPTION("iptables REDIRECT target module"); /* FIXME: Take multiple ranges --RR */ static int redirect_check(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index 6693526ae12..3eb47aae78c 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -282,12 +282,13 @@ static unsigned int reject(struct sk_buff **pskb, } static int check(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { const struct ipt_reject_info *rejinfo = targinfo; + const struct ipt_entry *e = e_void; if (targinfosize != IPT_ALIGN(sizeof(struct ipt_reject_info))) { DEBUGP("REJECT: targinfosize %u != 0\n", targinfosize); diff --git a/net/ipv4/netfilter/ipt_SAME.c b/net/ipv4/netfilter/ipt_SAME.c index 7a0536d864a..a22de59bba0 100644 --- a/net/ipv4/netfilter/ipt_SAME.c +++ b/net/ipv4/netfilter/ipt_SAME.c @@ -49,7 +49,7 @@ MODULE_DESCRIPTION("iptables special SNAT module for consistent sourceip"); static int same_check(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_TCPMSS.c b/net/ipv4/netfilter/ipt_TCPMSS.c index 8db70d6908c..c122841e182 100644 --- a/net/ipv4/netfilter/ipt_TCPMSS.c +++ b/net/ipv4/netfilter/ipt_TCPMSS.c @@ -210,12 +210,13 @@ static inline int find_syn_match(const struct ipt_entry_match *m) /* Must specify -p tcp --syn/--tcp-flags SYN */ static int ipt_tcpmss_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { const struct ipt_tcpmss_info *tcpmssinfo = targinfo; + const struct ipt_entry *e = e_void; if (targinfosize != IPT_ALIGN(sizeof(struct ipt_tcpmss_info))) { DEBUGP("ipt_tcpmss_checkentry: targinfosize %u != %u\n", diff --git a/net/ipv4/netfilter/ipt_TOS.c b/net/ipv4/netfilter/ipt_TOS.c index deadb36d442..3a44a56db23 100644 --- a/net/ipv4/netfilter/ipt_TOS.c +++ b/net/ipv4/netfilter/ipt_TOS.c @@ -52,7 +52,7 @@ target(struct sk_buff **pskb, static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e_void, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_TTL.c b/net/ipv4/netfilter/ipt_TTL.c index b9ae6a9382f..b769eb23197 100644 --- a/net/ipv4/netfilter/ipt_TTL.c +++ b/net/ipv4/netfilter/ipt_TTL.c @@ -66,7 +66,7 @@ ipt_ttl_target(struct sk_buff **pskb, const struct net_device *in, } static int ipt_ttl_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 38641cd0612..641dbc47765 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -330,7 +330,7 @@ static void ipt_logfn(unsigned int pf, } static int ipt_ulog_checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hookmask) diff --git a/net/ipv4/netfilter/ipt_addrtype.c b/net/ipv4/netfilter/ipt_addrtype.c index e19c2a52d00..d6b83a97651 100644 --- a/net/ipv4/netfilter/ipt_addrtype.c +++ b/net/ipv4/netfilter/ipt_addrtype.c @@ -29,7 +29,7 @@ static inline int match_type(u_int32_t addr, u_int16_t mask) 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) + int offset, unsigned int protoff, int *hotdrop) { const struct ipt_addrtype_info *info = matchinfo; const struct iphdr *iph = skb->nh.iph; @@ -43,7 +43,7 @@ static int match(const struct sk_buff *skb, const struct net_device *in, return ret; } -static int checkentry(const char *tablename, const struct ipt_ip *ip, +static int checkentry(const char *tablename, const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { diff --git a/net/ipv4/netfilter/ipt_ah.c b/net/ipv4/netfilter/ipt_ah.c index a0fea847cb7..144adfec13c 100644 --- a/net/ipv4/netfilter/ipt_ah.c +++ b/net/ipv4/netfilter/ipt_ah.c @@ -41,6 +41,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { struct ip_auth_hdr _ahdr, *ah; @@ -50,7 +51,7 @@ match(const struct sk_buff *skb, if (offset) return 0; - ah = skb_header_pointer(skb, skb->nh.iph->ihl * 4, + ah = skb_header_pointer(skb, protoff, sizeof(_ahdr), &_ahdr); if (ah == NULL) { /* We've been asked to examine this packet, and we @@ -69,12 +70,13 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip_void, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) { const struct ipt_ah *ahinfo = matchinfo; + const struct ipt_ip *ip = ip_void; /* Must specify proto == AH, and no unknown invflags */ if (ip->proto != IPPROTO_AH || (ip->invflags & IPT_INV_PROTO)) { diff --git a/net/ipv4/netfilter/ipt_dscp.c b/net/ipv4/netfilter/ipt_dscp.c index 5df52a64a5d..92063b4f860 100644 --- a/net/ipv4/netfilter/ipt_dscp.c +++ b/net/ipv4/netfilter/ipt_dscp.c @@ -21,7 +21,7 @@ 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) + int offset, unsigned int protoff, int *hotdrop) { const struct ipt_dscp_info *info = matchinfo; const struct iphdr *iph = skb->nh.iph; @@ -31,7 +31,7 @@ static int match(const struct sk_buff *skb, const struct net_device *in, return ((iph->tos&IPT_DSCP_MASK) == sh_dscp) ^ info->invert; } -static int checkentry(const char *tablename, const struct ipt_ip *ip, +static int checkentry(const char *tablename, const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { diff --git a/net/ipv4/netfilter/ipt_ecn.c b/net/ipv4/netfilter/ipt_ecn.c index b6f7181e89c..e68b0c7981f 100644 --- a/net/ipv4/netfilter/ipt_ecn.c +++ b/net/ipv4/netfilter/ipt_ecn.c @@ -67,7 +67,7 @@ static inline int match_tcp(const struct sk_buff *skb, 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) + int offset, unsigned int protoff, int *hotdrop) { const struct ipt_ecn_info *info = matchinfo; @@ -85,11 +85,12 @@ static int match(const struct sk_buff *skb, const struct net_device *in, return 1; } -static int checkentry(const char *tablename, const struct ipt_ip *ip, +static int checkentry(const char *tablename, const void *ip_void, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { const struct ipt_ecn_info *info = matchinfo; + const struct ipt_ip *ip = ip_void; if (matchsize != IPT_ALIGN(sizeof(struct ipt_ecn_info))) return 0; diff --git a/net/ipv4/netfilter/ipt_esp.c b/net/ipv4/netfilter/ipt_esp.c index e1d0dd31e11..9de191a8162 100644 --- a/net/ipv4/netfilter/ipt_esp.c +++ b/net/ipv4/netfilter/ipt_esp.c @@ -42,6 +42,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { struct ip_esp_hdr _esp, *eh; @@ -51,7 +52,7 @@ match(const struct sk_buff *skb, if (offset) return 0; - eh = skb_header_pointer(skb, skb->nh.iph->ihl * 4, + eh = skb_header_pointer(skb, protoff, sizeof(_esp), &_esp); if (eh == NULL) { /* We've been asked to examine this packet, and we @@ -70,12 +71,13 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip_void, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) { const struct ipt_esp *espinfo = matchinfo; + const struct ipt_ip *ip = ip_void; /* Must specify proto == ESP, and no unknown invflags */ if (ip->proto != IPPROTO_ESP || (ip->invflags & IPT_INV_PROTO)) { diff --git a/net/ipv4/netfilter/ipt_hashlimit.c b/net/ipv4/netfilter/ipt_hashlimit.c index 2dd1cccbdab..4fe48c1bd5f 100644 --- a/net/ipv4/netfilter/ipt_hashlimit.c +++ b/net/ipv4/netfilter/ipt_hashlimit.c @@ -429,6 +429,7 @@ hashlimit_match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { struct ipt_hashlimit_info *r = @@ -504,7 +505,7 @@ hashlimit_match(const struct sk_buff *skb, static int hashlimit_checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_iprange.c b/net/ipv4/netfilter/ipt_iprange.c index b835b7b2e56..13fb16fb789 100644 --- a/net/ipv4/netfilter/ipt_iprange.c +++ b/net/ipv4/netfilter/ipt_iprange.c @@ -28,7 +28,7 @@ match(const struct sk_buff *skb, const struct net_device *in, const struct net_device *out, const void *matchinfo, - int offset, int *hotdrop) + int offset, unsigned int protoff, int *hotdrop) { const struct ipt_iprange_info *info = matchinfo; const struct iphdr *iph = skb->nh.iph; @@ -63,7 +63,7 @@ match(const struct sk_buff *skb, } static int check(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_length.c b/net/ipv4/netfilter/ipt_length.c deleted file mode 100644 index 4eabcfbda9d..00000000000 --- a/net/ipv4/netfilter/ipt_length.c +++ /dev/null @@ -1,64 +0,0 @@ -/* Kernel module to match packet length. */ -/* (C) 1999-2001 James Morris <jmorros@intercode.com.au> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> - -#include <linux/netfilter_ipv4/ipt_length.h> -#include <linux/netfilter_ipv4/ip_tables.h> - -MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>"); -MODULE_DESCRIPTION("IP tables packet length 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, - int *hotdrop) -{ - const struct ipt_length_info *info = matchinfo; - u_int16_t pktlen = ntohs(skb->nh.iph->tot_len); - - return (pktlen >= info->min && pktlen <= info->max) ^ info->invert; -} - -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_length_info))) - return 0; - - return 1; -} - -static struct ipt_match length_match = { - .name = "length", - .match = &match, - .checkentry = &checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ipt_register_match(&length_match); -} - -static void __exit fini(void) -{ - ipt_unregister_match(&length_match); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_multiport.c b/net/ipv4/netfilter/ipt_multiport.c index 99e8188162e..2d52326553f 100644 --- a/net/ipv4/netfilter/ipt_multiport.c +++ b/net/ipv4/netfilter/ipt_multiport.c @@ -97,6 +97,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { u16 _ports[2], *pptr; @@ -105,7 +106,7 @@ match(const struct sk_buff *skb, if (offset) return 0; - pptr = skb_header_pointer(skb, skb->nh.iph->ihl * 4, + pptr = skb_header_pointer(skb, protoff, sizeof(_ports), _ports); if (pptr == NULL) { /* We've been asked to examine this packet, and we @@ -128,6 +129,7 @@ match_v1(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { u16 _ports[2], *pptr; @@ -136,7 +138,7 @@ match_v1(const struct sk_buff *skb, if (offset) return 0; - pptr = skb_header_pointer(skb, skb->nh.iph->ihl * 4, + pptr = skb_header_pointer(skb, protoff, sizeof(_ports), _ports); if (pptr == NULL) { /* We've been asked to examine this packet, and we @@ -154,7 +156,7 @@ match_v1(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) @@ -164,7 +166,7 @@ checkentry(const char *tablename, static int checkentry_v1(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_owner.c b/net/ipv4/netfilter/ipt_owner.c index 0cee2862ed8..4843d0c9734 100644 --- a/net/ipv4/netfilter/ipt_owner.c +++ b/net/ipv4/netfilter/ipt_owner.c @@ -27,6 +27,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { const struct ipt_owner_info *info = matchinfo; @@ -51,7 +52,7 @@ match(const struct sk_buff *skb, static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_physdev.c b/net/ipv4/netfilter/ipt_physdev.c deleted file mode 100644 index 03f554857a4..00000000000 --- a/net/ipv4/netfilter/ipt_physdev.c +++ /dev/null @@ -1,135 +0,0 @@ -/* Kernel module to match the bridge port in and - * out device for IP packets coming into contact with a bridge. */ - -/* (C) 2001-2003 Bart De Schuymer <bdschuym@pandora.be> - * - * 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 <linux/module.h> -#include <linux/netdevice.h> -#include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ipt_physdev.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_bridge.h> -#define MATCH 1 -#define NOMATCH 0 - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Bart De Schuymer <bdschuym@pandora.be>"); -MODULE_DESCRIPTION("iptables bridge physical device match module"); - -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) -{ - int i; - static const char nulldevname[IFNAMSIZ]; - const struct ipt_physdev_info *info = matchinfo; - unsigned int ret; - const char *indev, *outdev; - struct nf_bridge_info *nf_bridge; - - /* Not a bridged IP packet or no info available yet: - * LOCAL_OUT/mangle and LOCAL_OUT/nat don't know if - * the destination device will be a bridge. */ - if (!(nf_bridge = skb->nf_bridge)) { - /* Return MATCH if the invert flags of the used options are on */ - if ((info->bitmask & IPT_PHYSDEV_OP_BRIDGED) && - !(info->invert & IPT_PHYSDEV_OP_BRIDGED)) - return NOMATCH; - if ((info->bitmask & IPT_PHYSDEV_OP_ISIN) && - !(info->invert & IPT_PHYSDEV_OP_ISIN)) - return NOMATCH; - if ((info->bitmask & IPT_PHYSDEV_OP_ISOUT) && - !(info->invert & IPT_PHYSDEV_OP_ISOUT)) - return NOMATCH; - if ((info->bitmask & IPT_PHYSDEV_OP_IN) && - !(info->invert & IPT_PHYSDEV_OP_IN)) - return NOMATCH; - if ((info->bitmask & IPT_PHYSDEV_OP_OUT) && - !(info->invert & IPT_PHYSDEV_OP_OUT)) - return NOMATCH; - return MATCH; - } - - /* This only makes sense in the FORWARD and POSTROUTING chains */ - if ((info->bitmask & IPT_PHYSDEV_OP_BRIDGED) && - (!!(nf_bridge->mask & BRNF_BRIDGED) ^ - !(info->invert & IPT_PHYSDEV_OP_BRIDGED))) - return NOMATCH; - - if ((info->bitmask & IPT_PHYSDEV_OP_ISIN && - (!nf_bridge->physindev ^ !!(info->invert & IPT_PHYSDEV_OP_ISIN))) || - (info->bitmask & IPT_PHYSDEV_OP_ISOUT && - (!nf_bridge->physoutdev ^ !!(info->invert & IPT_PHYSDEV_OP_ISOUT)))) - return NOMATCH; - - if (!(info->bitmask & IPT_PHYSDEV_OP_IN)) - goto match_outdev; - indev = nf_bridge->physindev ? nf_bridge->physindev->name : nulldevname; - for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned int); i++) { - ret |= (((const unsigned int *)indev)[i] - ^ ((const unsigned int *)info->physindev)[i]) - & ((const unsigned int *)info->in_mask)[i]; - } - - if ((ret == 0) ^ !(info->invert & IPT_PHYSDEV_OP_IN)) - return NOMATCH; - -match_outdev: - if (!(info->bitmask & IPT_PHYSDEV_OP_OUT)) - return MATCH; - outdev = nf_bridge->physoutdev ? - nf_bridge->physoutdev->name : nulldevname; - for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned int); i++) { - ret |= (((const unsigned int *)outdev)[i] - ^ ((const unsigned int *)info->physoutdev)[i]) - & ((const unsigned int *)info->out_mask)[i]; - } - - return (ret != 0) ^ !(info->invert & IPT_PHYSDEV_OP_OUT); -} - -static int -checkentry(const char *tablename, - const struct ipt_ip *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - const struct ipt_physdev_info *info = matchinfo; - - if (matchsize != IPT_ALIGN(sizeof(struct ipt_physdev_info))) - return 0; - if (!(info->bitmask & IPT_PHYSDEV_OP_MASK) || - info->bitmask & ~IPT_PHYSDEV_OP_MASK) - return 0; - return 1; -} - -static struct ipt_match physdev_match = { - .name = "physdev", - .match = &match, - .checkentry = &checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ipt_register_match(&physdev_match); -} - -static void __exit fini(void) -{ - ipt_unregister_match(&physdev_match); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_recent.c b/net/ipv4/netfilter/ipt_recent.c index 5ddccb18c65..44611d6d14f 100644 --- a/net/ipv4/netfilter/ipt_recent.c +++ b/net/ipv4/netfilter/ipt_recent.c @@ -104,6 +104,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop); /* Function to hash a given address into the hash table of table_size size */ @@ -317,7 +318,7 @@ static int ip_recent_ctrl(struct file *file, const char __user *input, unsigned skb->nh.iph->daddr = 0; /* Clear ttl since we have no way of knowing it */ skb->nh.iph->ttl = 0; - match(skb,NULL,NULL,info,0,NULL); + match(skb,NULL,NULL,info,0,0,NULL); kfree(skb->nh.iph); out_free_skb: @@ -357,6 +358,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { int pkt_count, hits_found, ans; @@ -654,7 +656,7 @@ match(const struct sk_buff *skb, */ static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_tos.c b/net/ipv4/netfilter/ipt_tos.c index 086a1bb61e3..9ab765e126f 100644 --- a/net/ipv4/netfilter/ipt_tos.c +++ b/net/ipv4/netfilter/ipt_tos.c @@ -23,6 +23,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { const struct ipt_tos_info *info = matchinfo; @@ -32,7 +33,7 @@ match(const struct sk_buff *skb, static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv4/netfilter/ipt_ttl.c b/net/ipv4/netfilter/ipt_ttl.c index 219aa9de88c..82da53f430a 100644 --- a/net/ipv4/netfilter/ipt_ttl.c +++ b/net/ipv4/netfilter/ipt_ttl.c @@ -21,7 +21,7 @@ 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) + int offset, unsigned int protoff, int *hotdrop) { const struct ipt_ttl_info *info = matchinfo; @@ -47,7 +47,7 @@ static int match(const struct sk_buff *skb, const struct net_device *in, return 0; } -static int checkentry(const char *tablename, const struct ipt_ip *ip, +static int checkentry(const char *tablename, const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 260a4f0a2a9..212a3079085 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -78,7 +78,8 @@ static struct ipt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, - .me = THIS_MODULE + .me = THIS_MODULE, + .af = AF_INET, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index 160eb11b6e2..3212a5cc4b6 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -109,6 +109,7 @@ static struct ipt_table packet_mangler = { .valid_hooks = MANGLE_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, .me = THIS_MODULE, + .af = AF_INET, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index 47449ba83eb..fdb9e9c81e8 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -83,7 +83,8 @@ static struct ipt_table packet_raw = { .name = "raw", .valid_hooks = RAW_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, - .me = THIS_MODULE + .me = THIS_MODULE, + .af = AF_INET, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 0c56c52a383..167619f638c 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -575,7 +575,7 @@ MODULE_LICENSE("GPL"); static int __init init(void) { - need_nf_conntrack(); + need_conntrack(); return init_or_cleanup(1); } @@ -587,9 +587,4 @@ static void __exit fini(void) module_init(init); module_exit(fini); -void need_ip_conntrack(void) -{ -} - -EXPORT_SYMBOL(need_ip_conntrack); EXPORT_SYMBOL(nf_ct_ipv4_gather_frags); diff --git a/net/ipv4/xfrm4_state.c b/net/ipv4/xfrm4_state.c index d23e07fc81f..dbabf81a9b7 100644 --- a/net/ipv4/xfrm4_state.c +++ b/net/ipv4/xfrm4_state.c @@ -42,6 +42,21 @@ __xfrm4_init_tempsel(struct xfrm_state *x, struct flowi *fl, x->props.saddr = tmpl->saddr; if (x->props.saddr.a4 == 0) x->props.saddr.a4 = saddr->a4; + if (tmpl->mode && x->props.saddr.a4 == 0) { + struct rtable *rt; + struct flowi fl_tunnel = { + .nl_u = { + .ip4_u = { + .daddr = x->id.daddr.a4, + } + } + }; + if (!xfrm_dst_lookup((struct xfrm_dst **)&rt, + &fl_tunnel, AF_INET)) { + x->props.saddr.a4 = rt->rt_src; + dst_release(&rt->u.dst); + } + } x->props.mode = tmpl->mode; x->props.reqid = tmpl->reqid; x->props.family = AF_INET; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 7129d423975..dfb4f145a13 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2644,7 +2644,7 @@ static int if6_seq_show(struct seq_file *seq, void *v) { struct inet6_ifaddr *ifp = (struct inet6_ifaddr *)v; seq_printf(seq, - "%04x%04x%04x%04x%04x%04x%04x%04x %02x %02x %02x %02x %8s\n", + NIP6_FMT " %02x %02x %02x %02x %8s\n", NIP6(ifp->addr), ifp->idev->dev->ifindex, ifp->prefix_len, diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index 13cc7f89558..c7932cb420a 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -332,8 +332,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!x) return; - NETDEBUG(KERN_DEBUG "pmtu discovery on SA AH/%08x/" - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NETDEBUG(KERN_DEBUG "pmtu discovery on SA AH/%08x/" NIP6_FMT "\n", ntohl(ah->spi), NIP6(iph->daddr)); xfrm_state_put(x); diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 65e73ac0d6d..72bd08af2df 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -532,9 +532,7 @@ static int ac6_seq_show(struct seq_file *seq, void *v) struct ac6_iter_state *state = ac6_seq_private(seq); seq_printf(seq, - "%-4d %-15s " - "%04x%04x%04x%04x%04x%04x%04x%04x " - "%5d\n", + "%-4d %-15s " NIP6_FMT " %5d\n", state->dev->ifindex, state->dev->name, NIP6(im->aca_addr), im->aca_users); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 6de8ee1a5ad..7b5b94f1390 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -266,8 +266,7 @@ static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, x = xfrm_state_lookup((xfrm_address_t *)&iph->daddr, esph->spi, IPPROTO_ESP, AF_INET6); if (!x) return; - printk(KERN_DEBUG "pmtu discovery on SA ESP/%08x/" - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + printk(KERN_DEBUG "pmtu discovery on SA ESP/%08x/" NIP6_FMT "\n", ntohl(esph->spi), NIP6(iph->daddr)); xfrm_state_put(x); } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 53c81fcd20b..fcf883183ce 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -607,7 +607,7 @@ static int icmpv6_rcv(struct sk_buff **pskb) skb->csum = ~csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, 0); if (__skb_checksum_complete(skb)) { - LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x > %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x]\n", + LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [" NIP6_FMT " > " NIP6_FMT "]\n", NIP6(*saddr), NIP6(*daddr)); goto discard_it; } diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 964ad9d1276..4183c8dac7f 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -629,9 +629,7 @@ static void ip6fl_fl_seq_show(struct seq_file *seq, struct ip6_flowlabel *fl) { while(fl) { seq_printf(seq, - "%05X %-1d %-6d %-6d %-6ld %-8ld " - "%02x%02x%02x%02x%02x%02x%02x%02x " - "%-4d\n", + "%05X %-1d %-6d %-6d %-6ld %-8ld " NIP6_FMT " %-4d\n", (unsigned)ntohl(fl->label), fl->share, (unsigned)fl->owner, @@ -647,8 +645,8 @@ static void ip6fl_fl_seq_show(struct seq_file *seq, struct ip6_flowlabel *fl) static int ip6fl_seq_show(struct seq_file *seq, void *v) { if (v == SEQ_START_TOKEN) - seq_puts(seq, "Label S Owner Users Linger Expires " - "Dst Opt\n"); + seq_printf(seq, "%-5s %-1s %-6s %-6s %-6s %-8s %-39s %s\n", + "Label", "S", "Owner", "Users", "Linger", "Expires", "Dst", "Opt"); else ip6fl_fl_seq_show(seq, v); return 0; diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 626dd39685f..d511a884dad 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -212,8 +212,7 @@ static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (!x) return; - printk(KERN_DEBUG "pmtu discovery on SA IPCOMP/%08x/" - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + printk(KERN_DEBUG "pmtu discovery on SA IPCOMP/%08x/" NIP6_FMT "\n", spi, NIP6(iph->daddr)); xfrm_state_put(x); } diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index cc3e9f56086..0e03eabfb9d 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2373,7 +2373,7 @@ static int igmp6_mc_seq_show(struct seq_file *seq, void *v) struct igmp6_mc_iter_state *state = igmp6_mc_seq_private(seq); seq_printf(seq, - "%-4d %-15s %04x%04x%04x%04x%04x%04x%04x%04x %5d %08X %ld\n", + "%-4d %-15s " NIP6_FMT " %5d %08X %ld\n", state->dev->ifindex, state->dev->name, NIP6(im->mca_addr), im->mca_users, im->mca_flags, @@ -2542,15 +2542,12 @@ static int igmp6_mcf_seq_show(struct seq_file *seq, void *v) if (v == SEQ_START_TOKEN) { seq_printf(seq, "%3s %6s " - "%32s %32s %6s %6s\n", "Idx", + "%39s %39s %6s %6s\n", "Idx", "Device", "Multicast Address", "Source Address", "INC", "EXC"); } else { seq_printf(seq, - "%3d %6.6s " - "%04x%04x%04x%04x%04x%04x%04x%04x " - "%04x%04x%04x%04x%04x%04x%04x%04x " - "%6lu %6lu\n", + "%3d %6.6s " NIP6_FMT " " NIP6_FMT " %6lu %6lu\n", state->dev->ifindex, state->dev->name, NIP6(state->im->mca_addr), NIP6(psf->sf_addr), diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 305d9ee6d7d..cb8856b1d95 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -692,7 +692,7 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb) if (!(neigh->nud_state & NUD_VALID)) { ND_PRINTK1(KERN_DEBUG "%s(): trying to ucast probe in NUD_INVALID: " - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NIP6_FMT "\n", __FUNCTION__, NIP6(*target)); } diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 105dd69ee9f..2d6f8ecbc27 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -41,6 +41,7 @@ config IP6_NF_QUEUE config IP6_NF_IPTABLES tristate "IP6 tables support (required for filtering/masq/NAT)" + depends on NETFILTER_XTABLES help ip6tables is a general, extensible packet identification framework. Currently only the packet filtering and packet mangling subsystem @@ -50,25 +51,6 @@ config IP6_NF_IPTABLES To compile it as a module, choose M here. If unsure, say N. # The simple matches. -config IP6_NF_MATCH_LIMIT - tristate "limit match support" - depends on IP6_NF_IPTABLES - help - limit matching allows you to control the rate at which a rule can be - matched: mainly useful in combination with the LOG target ("LOG - target support", below) and to avoid some Denial of Service attacks. - - To compile it as a module, choose M here. If unsure, say N. - -config IP6_NF_MATCH_MAC - tristate "MAC address match support" - depends on IP6_NF_IPTABLES - help - mac matching allows you to match packets based on the source - Ethernet address of the packet. - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_MATCH_RT tristate "Routing header match support" depends on IP6_NF_IPTABLES @@ -124,16 +106,6 @@ config IP6_NF_MATCH_OWNER To compile it as a module, choose M here. If unsure, say N. -config IP6_NF_MATCH_MARK - tristate "netfilter MARK match support" - depends on IP6_NF_IPTABLES - help - Netfilter mark matching allows you to match packets based on the - `nfmark' value in the packet. This can be set by the MARK target - (see below). - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_MATCH_IPV6HEADER tristate "IPv6 Extension Headers Match" depends on IP6_NF_IPTABLES @@ -151,15 +123,6 @@ config IP6_NF_MATCH_AHESP To compile it as a module, choose M here. If unsure, say N. -config IP6_NF_MATCH_LENGTH - tristate "Packet Length match support" - depends on IP6_NF_IPTABLES - help - This option allows you to match the length of a packet against a - specific value or range of values. - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_MATCH_EUI64 tristate "EUI64 address check" depends on IP6_NF_IPTABLES @@ -170,15 +133,6 @@ config IP6_NF_MATCH_EUI64 To compile it as a module, choose M here. If unsure, say N. -config IP6_NF_MATCH_PHYSDEV - tristate "Physdev match support" - depends on IP6_NF_IPTABLES && BRIDGE_NETFILTER - help - Physdev packet matching matches against the physical bridge ports - the IP packet arrived on or will leave by. - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_MATCH_POLICY tristate "IPsec policy match support" depends on IP6_NF_IPTABLES && XFRM @@ -219,17 +173,6 @@ config IP6_NF_TARGET_REJECT To compile it as a module, choose M here. If unsure, say N. -config IP6_NF_TARGET_NFQUEUE - tristate "NFQUEUE Target Support" - depends on IP6_NF_IPTABLES - help - This Target replaced the old obsolete QUEUE target. - - As opposed to QUEUE, it supports 65535 different queues, - not just one. - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_MANGLE tristate "Packet mangling" depends on IP6_NF_IPTABLES @@ -240,19 +183,6 @@ config IP6_NF_MANGLE To compile it as a module, choose M here. If unsure, say N. -config IP6_NF_TARGET_MARK - tristate "MARK target support" - depends on IP6_NF_MANGLE - help - This option adds a `MARK' target, which allows you to create rules - in the `mangle' table which alter the netfilter mark (nfmark) field - associated with the packet packet prior to routing. This can change - the routing method (see `Use netfilter MARK value as routing - key') and can also be used by other subsystems to change their - behavior. - - To compile it as a module, choose M here. If unsure, say N. - config IP6_NF_TARGET_HL tristate 'HL (hoplimit) target support' depends on IP6_NF_MANGLE diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index c0c809b426e..663b4749820 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -4,10 +4,7 @@ # Link order matters here. obj-$(CONFIG_IP6_NF_IPTABLES) += ip6_tables.o -obj-$(CONFIG_IP6_NF_MATCH_LIMIT) += ip6t_limit.o -obj-$(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-$(CONFIG_IP6_NF_MATCH_RT) += ip6t_rt.o obj-$(CONFIG_IP6_NF_MATCH_OPTS) += ip6t_hbh.o ip6t_dst.o obj-$(CONFIG_IP6_NF_MATCH_IPV6HEADER) += ip6t_ipv6header.o @@ -17,12 +14,9 @@ obj-$(CONFIG_IP6_NF_MATCH_POLICY) += ip6t_policy.o obj-$(CONFIG_IP6_NF_MATCH_EUI64) += ip6t_eui64.o obj-$(CONFIG_IP6_NF_MATCH_MULTIPORT) += ip6t_multiport.o obj-$(CONFIG_IP6_NF_MATCH_OWNER) += ip6t_owner.o -obj-$(CONFIG_IP6_NF_MATCH_PHYSDEV) += ip6t_physdev.o 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_HL) += ip6t_HL.o -obj-$(CONFIG_IP6_NF_TARGET_NFQUEUE) += ip6t_NFQUEUE.o obj-$(CONFIG_IP6_NF_QUEUE) += ip6_queue.o obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 1390370186d..847068fd336 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -13,6 +13,9 @@ * a table * 06 Jun 2002 Andras Kis-Szabo <kisza@sch.bme.hu> * - new extension header parser code + * 15 Oct 2005 Harald Welte <laforge@netfilter.org> + * - Unification of {ip,ip6}_tables into x_tables + * - Removed tcp and udp code, since it's not ipv6 specific */ #include <linux/capability.h> @@ -23,8 +26,6 @@ #include <linux/vmalloc.h> #include <linux/netdevice.h> #include <linux/module.h> -#include <linux/tcp.h> -#include <linux/udp.h> #include <linux/icmpv6.h> #include <net/ipv6.h> #include <asm/uaccess.h> @@ -33,6 +34,7 @@ #include <linux/cpumask.h> #include <linux/netfilter_ipv6/ip6_tables.h> +#include <linux/netfilter/x_tables.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); @@ -67,13 +69,8 @@ do { \ #else #define IP_NF_ASSERT(x) #endif -#define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) -static DECLARE_MUTEX(ip6t_mutex); -/* Must have mutex */ -#define ASSERT_READ_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0) -#define ASSERT_WRITE_LOCK(x) IP_NF_ASSERT(down_trylock(&ip6t_mutex) != 0) #include <linux/netfilter_ipv4/listhelp.h> #if 0 @@ -91,30 +88,6 @@ static DECLARE_MUTEX(ip6t_mutex); Hence the start of any table is given by get_table() below. */ -/* The table itself */ -struct ip6t_table_info -{ - /* Size per table */ - unsigned int size; - /* Number of entries: FIXME. --RR */ - unsigned int number; - /* Initial number of entries. Needed for module usage count */ - unsigned int initial_entries; - - /* Entry points and underflows */ - unsigned int hook_entry[NF_IP6_NUMHOOKS]; - unsigned int underflow[NF_IP6_NUMHOOKS]; - - /* ip6t_entry tables: one per CPU */ - void *entries[NR_CPUS]; -}; - -static LIST_HEAD(ip6t_target); -static LIST_HEAD(ip6t_match); -static LIST_HEAD(ip6t_tables); -#define SET_COUNTER(c,b,p) do { (c).bcnt = (b); (c).pcnt = (p); } while(0) -#define ADD_COUNTER(c,b,p) do { (c).bcnt += (b); (c).pcnt += (p); } while(0) - #if 0 #define down(x) do { printk("DOWN:%u:" #x "\n", __LINE__); down(x); } while(0) #define down_interruptible(x) ({ int __r; printk("DOWNi:%u:" #x "\n", __LINE__); __r = down_interruptible(x); if (__r != 0) printk("ABORT-DOWNi:%u\n", __LINE__); __r; }) @@ -297,7 +270,7 @@ ip6t_do_table(struct sk_buff **pskb, unsigned int hook, const struct net_device *in, const struct net_device *out, - struct ip6t_table *table, + struct xt_table *table, void *userdata) { static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); @@ -309,6 +282,7 @@ ip6t_do_table(struct sk_buff **pskb, const char *indev, *outdev; void *table_base; struct ip6t_entry *e, *back; + struct xt_table_info *private; /* Initialization */ indev = in ? in->name : nulldevname; @@ -321,9 +295,10 @@ ip6t_do_table(struct sk_buff **pskb, * match it. */ read_lock_bh(&table->lock); + private = table->private; IP_NF_ASSERT(table->valid_hooks & (1 << hook)); - table_base = (void *)table->private->entries[smp_processor_id()]; - e = get_entry(table_base, table->private->hook_entry[hook]); + table_base = (void *)private->entries[smp_processor_id()]; + e = get_entry(table_base, private->hook_entry[hook]); #ifdef CONFIG_NETFILTER_DEBUG /* Check noone else using our table */ @@ -339,7 +314,7 @@ ip6t_do_table(struct sk_buff **pskb, #endif /* For return from builtin chain */ - back = get_entry(table_base, table->private->underflow[hook]); + back = get_entry(table_base, private->underflow[hook]); do { IP_NF_ASSERT(e); @@ -439,145 +414,6 @@ ip6t_do_table(struct sk_buff **pskb, #endif } -/* - * These are weird, but module loading must not be done with mutex - * held (since they will register), and we have to have a single - * function to use try_then_request_module(). - */ - -/* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ -static inline struct ip6t_table *find_table_lock(const char *name) -{ - struct ip6t_table *t; - - if (down_interruptible(&ip6t_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &ip6t_tables, list) - if (strcmp(t->name, name) == 0 && try_module_get(t->me)) - return t; - up(&ip6t_mutex); - return NULL; -} - -/* Find match, grabs ref. Returns ERR_PTR() on error. */ -static inline struct ip6t_match *find_match(const char *name, u8 revision) -{ - struct ip6t_match *m; - int err = 0; - - if (down_interruptible(&ip6t_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(m, &ip6t_match, list) { - if (strcmp(m->name, name) == 0) { - if (m->revision == revision) { - if (try_module_get(m->me)) { - up(&ip6t_mutex); - return m; - } - } else - err = -EPROTOTYPE; /* Found something. */ - } - } - up(&ip6t_mutex); - return ERR_PTR(err); -} - -/* Find target, grabs ref. Returns ERR_PTR() on error. */ -static inline struct ip6t_target *find_target(const char *name, u8 revision) -{ - struct ip6t_target *t; - int err = 0; - - if (down_interruptible(&ip6t_mutex) != 0) - return ERR_PTR(-EINTR); - - list_for_each_entry(t, &ip6t_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision == revision) { - if (try_module_get(t->me)) { - up(&ip6t_mutex); - return t; - } - } else - err = -EPROTOTYPE; /* Found something. */ - } - } - up(&ip6t_mutex); - return ERR_PTR(err); -} - -struct ip6t_target *ip6t_find_target(const char *name, u8 revision) -{ - struct ip6t_target *target; - - target = try_then_request_module(find_target(name, revision), - "ip6t_%s", name); - if (IS_ERR(target) || !target) - return NULL; - return target; -} - -static int match_revfn(const char *name, u8 revision, int *bestp) -{ - struct ip6t_match *m; - int have_rev = 0; - - list_for_each_entry(m, &ip6t_match, list) { - if (strcmp(m->name, name) == 0) { - if (m->revision > *bestp) - *bestp = m->revision; - if (m->revision == revision) - have_rev = 1; - } - } - return have_rev; -} - -static int target_revfn(const char *name, u8 revision, int *bestp) -{ - struct ip6t_target *t; - int have_rev = 0; - - list_for_each_entry(t, &ip6t_target, list) { - if (strcmp(t->name, name) == 0) { - if (t->revision > *bestp) - *bestp = t->revision; - if (t->revision == revision) - have_rev = 1; - } - } - return have_rev; -} - -/* Returns true or fals (if no such extension at all) */ -static inline int find_revision(const char *name, u8 revision, - int (*revfn)(const char *, u8, int *), - int *err) -{ - int have_rev, best = -1; - - if (down_interruptible(&ip6t_mutex) != 0) { - *err = -EINTR; - return 1; - } - have_rev = revfn(name, revision, &best); - up(&ip6t_mutex); - - /* Nothing at all? Return 0 to try loading module. */ - if (best == -1) { - *err = -ENOENT; - return 0; - } - - *err = best; - if (!have_rev) - *err = -EPROTONOSUPPORT; - return 1; -} - - /* All zeroes == unconditional rule. */ static inline int unconditional(const struct ip6t_ip6 *ipv6) @@ -594,7 +430,7 @@ unconditional(const struct ip6t_ip6 *ipv6) /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int -mark_source_chains(struct ip6t_table_info *newinfo, +mark_source_chains(struct xt_table_info *newinfo, unsigned int valid_hooks, void *entry0) { unsigned int hook; @@ -740,11 +576,11 @@ check_match(struct ip6t_entry_match *m, { struct ip6t_match *match; - match = try_then_request_module(find_match(m->u.user.name, - m->u.user.revision), + match = try_then_request_module(xt_find_match(AF_INET6, m->u.user.name, + m->u.user.revision), "ip6t_%s", m->u.user.name); if (IS_ERR(match) || !match) { - duprintf("check_match: `%s' not found\n", m->u.user.name); + duprintf("check_match: `%s' not found\n", m->u.user.name); return match ? PTR_ERR(match) : -ENOENT; } m->u.kernel.match = match; @@ -785,8 +621,9 @@ check_entry(struct ip6t_entry *e, const char *name, unsigned int size, goto cleanup_matches; t = ip6t_get_target(e); - target = try_then_request_module(find_target(t->u.user.name, - t->u.user.revision), + target = try_then_request_module(xt_find_target(AF_INET6, + t->u.user.name, + t->u.user.revision), "ip6t_%s", t->u.user.name); if (IS_ERR(target) || !target) { duprintf("check_entry: `%s' not found\n", t->u.user.name); @@ -822,7 +659,7 @@ check_entry(struct ip6t_entry *e, const char *name, unsigned int size, static inline int check_entry_size_and_hooks(struct ip6t_entry *e, - struct ip6t_table_info *newinfo, + struct xt_table_info *newinfo, unsigned char *base, unsigned char *limit, const unsigned int *hook_entries, @@ -856,7 +693,7 @@ check_entry_size_and_hooks(struct ip6t_entry *e, < 0 (not IP6T_RETURN). --RR */ /* Clear counters and comefrom */ - e->counters = ((struct ip6t_counters) { 0, 0 }); + e->counters = ((struct xt_counters) { 0, 0 }); e->comefrom = 0; (*i)++; @@ -886,7 +723,7 @@ cleanup_entry(struct ip6t_entry *e, unsigned int *i) static int translate_table(const char *name, unsigned int valid_hooks, - struct ip6t_table_info *newinfo, + struct xt_table_info *newinfo, void *entry0, unsigned int size, unsigned int number, @@ -963,48 +800,10 @@ translate_table(const char *name, return ret; } -static struct ip6t_table_info * -replace_table(struct ip6t_table *table, - unsigned int num_counters, - struct ip6t_table_info *newinfo, - int *error) -{ - struct ip6t_table_info *oldinfo; - -#ifdef CONFIG_NETFILTER_DEBUG - { - int cpu; - - for_each_cpu(cpu) { - struct ip6t_entry *table_base = newinfo->entries[cpu]; - if (table_base) - table_base->comefrom = 0xdead57ac; - } - } -#endif - - /* Do the substitution. */ - write_lock_bh(&table->lock); - /* Check inside lock: is the old number correct? */ - if (num_counters != table->private->number) { - duprintf("num_counters != table->private->number (%u/%u)\n", - num_counters, table->private->number); - write_unlock_bh(&table->lock); - *error = -EAGAIN; - return NULL; - } - oldinfo = table->private; - table->private = newinfo; - newinfo->initial_entries = oldinfo->initial_entries; - write_unlock_bh(&table->lock); - - return oldinfo; -} - /* Gets counters. */ static inline int add_entry_to_counter(const struct ip6t_entry *e, - struct ip6t_counters total[], + struct xt_counters total[], unsigned int *i) { ADD_COUNTER(total[*i], e->counters.bcnt, e->counters.pcnt); @@ -1025,8 +824,8 @@ set_entry_to_counter(const struct ip6t_entry *e, } static void -get_counters(const struct ip6t_table_info *t, - struct ip6t_counters counters[]) +get_counters(const struct xt_table_info *t, + struct xt_counters counters[]) { unsigned int cpu; unsigned int i; @@ -1060,19 +859,20 @@ get_counters(const struct ip6t_table_info *t, static int copy_entries_to_user(unsigned int total_size, - struct ip6t_table *table, + struct xt_table *table, void __user *userptr) { unsigned int off, num, countersize; struct ip6t_entry *e; - struct ip6t_counters *counters; + struct xt_counters *counters; + struct xt_table_info *private = table->private; int ret = 0; void *loc_cpu_entry; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care about). */ - countersize = sizeof(struct ip6t_counters) * table->private->number; + countersize = sizeof(struct xt_counters) * private->number; counters = vmalloc(countersize); if (counters == NULL) @@ -1080,11 +880,11 @@ copy_entries_to_user(unsigned int total_size, /* First, sum counters... */ write_lock_bh(&table->lock); - get_counters(table->private, counters); + get_counters(private, counters); write_unlock_bh(&table->lock); /* choose the copy that is on ourc node/cpu */ - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; if (copy_to_user(userptr, loc_cpu_entry, total_size) != 0) { ret = -EFAULT; goto free_counters; @@ -1143,87 +943,42 @@ get_entries(const struct ip6t_get_entries *entries, struct ip6t_get_entries __user *uptr) { int ret; - struct ip6t_table *t; + struct xt_table *t; - t = find_table_lock(entries->name); + t = xt_find_table_lock(AF_INET6, entries->name); if (t && !IS_ERR(t)) { - duprintf("t->private->number = %u\n", - t->private->number); - if (entries->size == t->private->size) - ret = copy_entries_to_user(t->private->size, + struct xt_table_info *private = t->private; + duprintf("t->private->number = %u\n", private->number); + if (entries->size == private->size) + ret = copy_entries_to_user(private->size, t, uptr->entrytable); else { duprintf("get_entries: I've got %u not %u!\n", - t->private->size, - entries->size); + private->size, entries->size); ret = -EINVAL; } module_put(t->me); - up(&ip6t_mutex); + xt_table_unlock(t); } else ret = t ? PTR_ERR(t) : -ENOENT; return ret; } -static void free_table_info(struct ip6t_table_info *info) -{ - int cpu; - for_each_cpu(cpu) { - if (info->size <= PAGE_SIZE) - kfree(info->entries[cpu]); - else - vfree(info->entries[cpu]); - } - kfree(info); -} - -static struct ip6t_table_info *alloc_table_info(unsigned int size) -{ - struct ip6t_table_info *newinfo; - int cpu; - - newinfo = kzalloc(sizeof(struct ip6t_table_info), GFP_KERNEL); - if (!newinfo) - return NULL; - - newinfo->size = size; - - for_each_cpu(cpu) { - if (size <= PAGE_SIZE) - newinfo->entries[cpu] = kmalloc_node(size, - GFP_KERNEL, - cpu_to_node(cpu)); - else - newinfo->entries[cpu] = vmalloc_node(size, - cpu_to_node(cpu)); - if (newinfo->entries[cpu] == NULL) { - free_table_info(newinfo); - return NULL; - } - } - - return newinfo; -} - static int do_replace(void __user *user, unsigned int len) { int ret; struct ip6t_replace tmp; - struct ip6t_table *t; - struct ip6t_table_info *newinfo, *oldinfo; - struct ip6t_counters *counters; + struct xt_table *t; + struct xt_table_info *newinfo, *oldinfo; + struct xt_counters *counters; void *loc_cpu_entry, *loc_cpu_old_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; - /* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */ - if ((SMP_ALIGN(tmp.size) >> PAGE_SHIFT) + 2 > num_physpages) - return -ENOMEM; - - newinfo = alloc_table_info(tmp.size); + newinfo = xt_alloc_table_info(tmp.size); if (!newinfo) return -ENOMEM; @@ -1235,7 +990,7 @@ do_replace(void __user *user, unsigned int len) goto free_newinfo; } - counters = vmalloc(tmp.num_counters * sizeof(struct ip6t_counters)); + counters = vmalloc(tmp.num_counters * sizeof(struct xt_counters)); if (!counters) { ret = -ENOMEM; goto free_newinfo; @@ -1249,7 +1004,7 @@ do_replace(void __user *user, unsigned int len) duprintf("ip_tables: Translated table\n"); - t = try_then_request_module(find_table_lock(tmp.name), + t = try_then_request_module(xt_find_table_lock(AF_INET6, tmp.name), "ip6table_%s", tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; @@ -1264,7 +1019,7 @@ do_replace(void __user *user, unsigned int len) goto put_module; } - oldinfo = replace_table(t, tmp.num_counters, newinfo, &ret); + oldinfo = xt_replace_table(t, tmp.num_counters, newinfo, &ret); if (!oldinfo) goto put_module; @@ -1283,23 +1038,23 @@ do_replace(void __user *user, unsigned int len) /* Decrease module usage counts and free resource */ loc_cpu_old_entry = oldinfo->entries[raw_smp_processor_id()]; IP6T_ENTRY_ITERATE(loc_cpu_old_entry, oldinfo->size, cleanup_entry,NULL); - free_table_info(oldinfo); + xt_free_table_info(oldinfo); if (copy_to_user(tmp.counters, counters, - sizeof(struct ip6t_counters) * tmp.num_counters) != 0) + sizeof(struct xt_counters) * tmp.num_counters) != 0) ret = -EFAULT; vfree(counters); - up(&ip6t_mutex); + xt_table_unlock(t); return ret; put_module: module_put(t->me); - up(&ip6t_mutex); + xt_table_unlock(t); free_newinfo_counters_untrans: IP6T_ENTRY_ITERATE(loc_cpu_entry, newinfo->size, cleanup_entry,NULL); free_newinfo_counters: vfree(counters); free_newinfo: - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } @@ -1307,7 +1062,7 @@ do_replace(void __user *user, unsigned int len) * and everything is OK. */ static inline int add_counter_to_entry(struct ip6t_entry *e, - const struct ip6t_counters addme[], + const struct xt_counters addme[], unsigned int *i) { #if 0 @@ -1329,15 +1084,16 @@ static int do_add_counters(void __user *user, unsigned int len) { unsigned int i; - struct ip6t_counters_info tmp, *paddc; - struct ip6t_table *t; + struct xt_counters_info tmp, *paddc; + struct xt_table_info *private; + struct xt_table *t; int ret = 0; void *loc_cpu_entry; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; - if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct ip6t_counters)) + if (len != sizeof(tmp) + tmp.num_counters*sizeof(struct xt_counters)) return -EINVAL; paddc = vmalloc(len); @@ -1349,29 +1105,30 @@ do_add_counters(void __user *user, unsigned int len) goto free; } - t = find_table_lock(tmp.name); + t = xt_find_table_lock(AF_INET6, tmp.name); if (!t || IS_ERR(t)) { ret = t ? PTR_ERR(t) : -ENOENT; goto free; } write_lock_bh(&t->lock); - if (t->private->number != paddc->num_counters) { + private = t->private; + if (private->number != paddc->num_counters) { ret = -EINVAL; goto unlock_up_free; } i = 0; /* Choose the copy that is on our node */ - loc_cpu_entry = t->private->entries[smp_processor_id()]; + loc_cpu_entry = private->entries[smp_processor_id()]; IP6T_ENTRY_ITERATE(loc_cpu_entry, - t->private->size, + private->size, add_counter_to_entry, paddc->counters, &i); unlock_up_free: write_unlock_bh(&t->lock); - up(&ip6t_mutex); + xt_table_unlock(t); module_put(t->me); free: vfree(paddc); @@ -1415,7 +1172,7 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IP6T_SO_GET_INFO: { char name[IP6T_TABLE_MAXNAMELEN]; - struct ip6t_table *t; + struct xt_table *t; if (*len != sizeof(struct ip6t_getinfo)) { duprintf("length %u != %u\n", *len, @@ -1430,25 +1187,26 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) } name[IP6T_TABLE_MAXNAMELEN-1] = '\0'; - t = try_then_request_module(find_table_lock(name), + t = try_then_request_module(xt_find_table_lock(AF_INET6, name), "ip6table_%s", name); if (t && !IS_ERR(t)) { struct ip6t_getinfo info; + struct xt_table_info *private = t->private; info.valid_hooks = t->valid_hooks; - memcpy(info.hook_entry, t->private->hook_entry, + memcpy(info.hook_entry, private->hook_entry, sizeof(info.hook_entry)); - memcpy(info.underflow, t->private->underflow, + memcpy(info.underflow, private->underflow, sizeof(info.underflow)); - info.num_entries = t->private->number; - info.size = t->private->size; + info.num_entries = private->number; + info.size = private->size; memcpy(info.name, name, sizeof(info.name)); if (copy_to_user(user, &info, *len) != 0) ret = -EFAULT; else ret = 0; - up(&ip6t_mutex); + xt_table_unlock(t); module_put(t->me); } else ret = t ? PTR_ERR(t) : -ENOENT; @@ -1475,7 +1233,7 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) case IP6T_SO_GET_REVISION_MATCH: case IP6T_SO_GET_REVISION_TARGET: { struct ip6t_get_revision rev; - int (*revfn)(const char *, u8, int *); + int target; if (*len != sizeof(rev)) { ret = -EINVAL; @@ -1487,12 +1245,13 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) } if (cmd == IP6T_SO_GET_REVISION_TARGET) - revfn = target_revfn; + target = 1; else - revfn = match_revfn; + target = 0; - try_then_request_module(find_revision(rev.name, rev.revision, - revfn, &ret), + try_then_request_module(xt_find_revision(AF_INET6, rev.name, + rev.revision, + target, &ret), "ip6t_%s", rev.name); break; } @@ -1505,61 +1264,16 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) return ret; } -/* Registration hooks for targets. */ -int -ip6t_register_target(struct ip6t_target *target) -{ - int ret; - - ret = down_interruptible(&ip6t_mutex); - if (ret != 0) - return ret; - list_add(&target->list, &ip6t_target); - up(&ip6t_mutex); - return ret; -} - -void -ip6t_unregister_target(struct ip6t_target *target) -{ - down(&ip6t_mutex); - LIST_DELETE(&ip6t_target, target); - up(&ip6t_mutex); -} - -int -ip6t_register_match(struct ip6t_match *match) -{ - int ret; - - ret = down_interruptible(&ip6t_mutex); - if (ret != 0) - return ret; - - list_add(&match->list, &ip6t_match); - up(&ip6t_mutex); - - return ret; -} - -void -ip6t_unregister_match(struct ip6t_match *match) -{ - down(&ip6t_mutex); - LIST_DELETE(&ip6t_match, match); - up(&ip6t_mutex); -} - -int ip6t_register_table(struct ip6t_table *table, +int ip6t_register_table(struct xt_table *table, const struct ip6t_replace *repl) { int ret; - struct ip6t_table_info *newinfo; - static struct ip6t_table_info bootstrap + struct xt_table_info *newinfo; + static struct xt_table_info bootstrap = { 0, 0, 0, { 0 }, { 0 }, { } }; void *loc_cpu_entry; - newinfo = alloc_table_info(repl->size); + newinfo = xt_alloc_table_info(repl->size); if (!newinfo) return -ENOMEM; @@ -1573,244 +1287,29 @@ int ip6t_register_table(struct ip6t_table *table, repl->hook_entry, repl->underflow); if (ret != 0) { - free_table_info(newinfo); + xt_free_table_info(newinfo); return ret; } - ret = down_interruptible(&ip6t_mutex); - if (ret != 0) { - free_table_info(newinfo); + if (xt_register_table(table, &bootstrap, newinfo) != 0) { + xt_free_table_info(newinfo); return ret; } - /* Don't autoload: we'd eat our tail... */ - if (list_named_find(&ip6t_tables, table->name)) { - ret = -EEXIST; - goto free_unlock; - } - - /* Simplifies replace_table code. */ - table->private = &bootstrap; - if (!replace_table(table, 0, newinfo, &ret)) - goto free_unlock; - - duprintf("table->private->number = %u\n", - table->private->number); - - /* save number of initial entries */ - table->private->initial_entries = table->private->number; - - rwlock_init(&table->lock); - list_prepend(&ip6t_tables, table); - - unlock: - up(&ip6t_mutex); - return ret; - - free_unlock: - free_table_info(newinfo); - goto unlock; + return 0; } -void ip6t_unregister_table(struct ip6t_table *table) +void ip6t_unregister_table(struct xt_table *table) { + struct xt_table_info *private; void *loc_cpu_entry; - down(&ip6t_mutex); - LIST_DELETE(&ip6t_tables, table); - up(&ip6t_mutex); + private = xt_unregister_table(table); /* Decrease module usage counts and free resources */ - loc_cpu_entry = table->private->entries[raw_smp_processor_id()]; - IP6T_ENTRY_ITERATE(loc_cpu_entry, table->private->size, - cleanup_entry, NULL); - free_table_info(table->private); -} - -/* Returns 1 if the port is matched by the range, 0 otherwise */ -static inline int -port_match(u_int16_t min, u_int16_t max, u_int16_t port, int invert) -{ - int ret; - - ret = (port >= min && port <= max) ^ invert; - return ret; -} - -static int -tcp_find_option(u_int8_t option, - const struct sk_buff *skb, - unsigned int tcpoff, - unsigned int optlen, - int invert, - int *hotdrop) -{ - /* tcp.doff is only 4 bits, ie. max 15 * 4 bytes */ - u_int8_t _opt[60 - sizeof(struct tcphdr)], *op; - unsigned int i; - - duprintf("tcp_match: finding option\n"); - if (!optlen) - return invert; - /* If we don't have the whole header, drop packet. */ - op = skb_header_pointer(skb, tcpoff + sizeof(struct tcphdr), optlen, - _opt); - if (op == NULL) { - *hotdrop = 1; - return 0; - } - - for (i = 0; i < optlen; ) { - if (op[i] == option) return !invert; - if (op[i] < 2) i++; - else i += op[i+1]?:1; - } - - return invert; -} - -static int -tcp_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - struct tcphdr _tcph, *th; - const struct ip6t_tcp *tcpinfo = matchinfo; - - if (offset) { - /* To quote Alan: - - Don't allow a fragment of TCP 8 bytes in. Nobody normal - causes this. Its a cracker trying to break in by doing a - flag overwrite to pass the direction checks. - */ - if (offset == 1) { - duprintf("Dropping evil TCP offset=1 frag.\n"); - *hotdrop = 1; - } - /* Must not be a fragment. */ - return 0; - } - -#define FWINVTCP(bool,invflg) ((bool) ^ !!(tcpinfo->invflags & invflg)) - - th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); - if (th == NULL) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - duprintf("Dropping evil TCP offset=0 tinygram.\n"); - *hotdrop = 1; - return 0; - } - - if (!port_match(tcpinfo->spts[0], tcpinfo->spts[1], - ntohs(th->source), - !!(tcpinfo->invflags & IP6T_TCP_INV_SRCPT))) - return 0; - if (!port_match(tcpinfo->dpts[0], tcpinfo->dpts[1], - ntohs(th->dest), - !!(tcpinfo->invflags & IP6T_TCP_INV_DSTPT))) - return 0; - if (!FWINVTCP((((unsigned char *)th)[13] & tcpinfo->flg_mask) - == tcpinfo->flg_cmp, - IP6T_TCP_INV_FLAGS)) - return 0; - if (tcpinfo->option) { - if (th->doff * 4 < sizeof(_tcph)) { - *hotdrop = 1; - return 0; - } - if (!tcp_find_option(tcpinfo->option, skb, protoff, - th->doff*4 - sizeof(*th), - tcpinfo->invflags & IP6T_TCP_INV_OPTION, - hotdrop)) - return 0; - } - return 1; -} - -/* Called when user tries to insert an entry of this type. */ -static int -tcp_checkentry(const char *tablename, - const struct ip6t_ip6 *ipv6, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - const struct ip6t_tcp *tcpinfo = matchinfo; - - /* Must specify proto == TCP, and no unknown invflags */ - return ipv6->proto == IPPROTO_TCP - && !(ipv6->invflags & IP6T_INV_PROTO) - && matchsize == IP6T_ALIGN(sizeof(struct ip6t_tcp)) - && !(tcpinfo->invflags & ~IP6T_TCP_INV_MASK); -} - -static int -udp_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - struct udphdr _udph, *uh; - const struct ip6t_udp *udpinfo = matchinfo; - - /* Must not be a fragment. */ - if (offset) - return 0; - - uh = skb_header_pointer(skb, protoff, sizeof(_udph), &_udph); - if (uh == NULL) { - /* We've been asked to examine this packet, and we - can't. Hence, no choice but to drop. */ - duprintf("Dropping evil UDP tinygram.\n"); - *hotdrop = 1; - return 0; - } - - return port_match(udpinfo->spts[0], udpinfo->spts[1], - ntohs(uh->source), - !!(udpinfo->invflags & IP6T_UDP_INV_SRCPT)) - && port_match(udpinfo->dpts[0], udpinfo->dpts[1], - ntohs(uh->dest), - !!(udpinfo->invflags & IP6T_UDP_INV_DSTPT)); -} - -/* Called when user tries to insert an entry of this type. */ -static int -udp_checkentry(const char *tablename, - const struct ip6t_ip6 *ipv6, - void *matchinfo, - unsigned int matchinfosize, - unsigned int hook_mask) -{ - const struct ip6t_udp *udpinfo = matchinfo; - - /* Must specify proto == UDP, and no unknown invflags */ - if (ipv6->proto != IPPROTO_UDP || (ipv6->invflags & IP6T_INV_PROTO)) { - duprintf("ip6t_udp: Protocol %u != %u\n", ipv6->proto, - IPPROTO_UDP); - return 0; - } - if (matchinfosize != IP6T_ALIGN(sizeof(struct ip6t_udp))) { - duprintf("ip6t_udp: matchsize %u != %u\n", - matchinfosize, IP6T_ALIGN(sizeof(struct ip6t_udp))); - return 0; - } - if (udpinfo->invflags & ~IP6T_UDP_INV_MASK) { - duprintf("ip6t_udp: unknown flags %X\n", - udpinfo->invflags); - return 0; - } - - return 1; + loc_cpu_entry = private->entries[raw_smp_processor_id()]; + IP6T_ENTRY_ITERATE(loc_cpu_entry, private->size, cleanup_entry, NULL); + xt_free_table_info(private); } /* Returns 1 if the type and code is matched by the range, 0 otherwise */ @@ -1858,11 +1357,12 @@ icmp6_match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int icmp6_checkentry(const char *tablename, - const struct ip6t_ip6 *ipv6, + const void *entry, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { + const struct ip6t_ip6 *ipv6 = entry; const struct ip6t_icmp *icmpinfo = matchinfo; /* Must specify proto == ICMP, and no unknown invflags */ @@ -1892,164 +1392,42 @@ static struct nf_sockopt_ops ip6t_sockopts = { .get = do_ip6t_get_ctl, }; -static struct ip6t_match tcp_matchstruct = { - .name = "tcp", - .match = &tcp_match, - .checkentry = &tcp_checkentry, -}; - -static struct ip6t_match udp_matchstruct = { - .name = "udp", - .match = &udp_match, - .checkentry = &udp_checkentry, -}; - static struct ip6t_match icmp6_matchstruct = { .name = "icmp6", .match = &icmp6_match, .checkentry = &icmp6_checkentry, }; -#ifdef CONFIG_PROC_FS -static inline int print_name(const char *i, - off_t start_offset, char *buffer, int length, - off_t *pos, unsigned int *count) -{ - if ((*count)++ >= start_offset) { - unsigned int namelen; - - namelen = sprintf(buffer + *pos, "%s\n", - i + sizeof(struct list_head)); - if (*pos + namelen > length) { - /* Stop iterating */ - return 1; - } - *pos += namelen; - } - return 0; -} - -static inline int print_target(const struct ip6t_target *t, - off_t start_offset, char *buffer, int length, - off_t *pos, unsigned int *count) -{ - if (t == &ip6t_standard_target || t == &ip6t_error_target) - return 0; - return print_name((char *)t, start_offset, buffer, length, pos, count); -} - -static int ip6t_get_tables(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ip6t_mutex) != 0) - return 0; - - LIST_FIND(&ip6t_tables, print_name, char *, - offset, buffer, length, &pos, &count); - - up(&ip6t_mutex); - - /* `start' hack - see fs/proc/generic.c line ~105 */ - *start=(char *)((unsigned long)count-offset); - return pos; -} - -static int ip6t_get_targets(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ip6t_mutex) != 0) - return 0; - - LIST_FIND(&ip6t_target, print_target, struct ip6t_target *, - offset, buffer, length, &pos, &count); - - up(&ip6t_mutex); - - *start = (char *)((unsigned long)count - offset); - return pos; -} - -static int ip6t_get_matches(char *buffer, char **start, off_t offset, int length) -{ - off_t pos = 0; - unsigned int count = 0; - - if (down_interruptible(&ip6t_mutex) != 0) - return 0; - - LIST_FIND(&ip6t_match, print_name, char *, - offset, buffer, length, &pos, &count); - - up(&ip6t_mutex); - - *start = (char *)((unsigned long)count - offset); - return pos; -} - -static const struct { char *name; get_info_t *get_info; } ip6t_proc_entry[] = -{ { "ip6_tables_names", ip6t_get_tables }, - { "ip6_tables_targets", ip6t_get_targets }, - { "ip6_tables_matches", ip6t_get_matches }, - { NULL, NULL} }; -#endif /*CONFIG_PROC_FS*/ - static int __init init(void) { int ret; + xt_proto_init(AF_INET6); + /* Noone else will be downing sem now, so we won't sleep */ - down(&ip6t_mutex); - list_append(&ip6t_target, &ip6t_standard_target); - list_append(&ip6t_target, &ip6t_error_target); - list_append(&ip6t_match, &tcp_matchstruct); - list_append(&ip6t_match, &udp_matchstruct); - list_append(&ip6t_match, &icmp6_matchstruct); - up(&ip6t_mutex); + xt_register_target(AF_INET6, &ip6t_standard_target); + xt_register_target(AF_INET6, &ip6t_error_target); + xt_register_match(AF_INET6, &icmp6_matchstruct); /* Register setsockopt */ ret = nf_register_sockopt(&ip6t_sockopts); if (ret < 0) { duprintf("Unable to register sockopts.\n"); + xt_proto_fini(AF_INET6); return ret; } -#ifdef CONFIG_PROC_FS - { - struct proc_dir_entry *proc; - int i; - - for (i = 0; ip6t_proc_entry[i].name; i++) { - proc = proc_net_create(ip6t_proc_entry[i].name, 0, - ip6t_proc_entry[i].get_info); - if (!proc) { - while (--i >= 0) - proc_net_remove(ip6t_proc_entry[i].name); - nf_unregister_sockopt(&ip6t_sockopts); - return -ENOMEM; - } - proc->owner = THIS_MODULE; - } - } -#endif - - printk("ip6_tables: (C) 2000-2002 Netfilter core team\n"); + printk("ip6_tables: (C) 2000-2006 Netfilter Core Team\n"); return 0; } static void __exit fini(void) { nf_unregister_sockopt(&ip6t_sockopts); -#ifdef CONFIG_PROC_FS - { - int i; - for (i = 0; ip6t_proc_entry[i].name; i++) - proc_net_remove(ip6t_proc_entry[i].name); - } -#endif + xt_unregister_match(AF_INET6, &icmp6_matchstruct); + xt_unregister_target(AF_INET6, &ip6t_error_target); + xt_unregister_target(AF_INET6, &ip6t_standard_target); + xt_proto_fini(AF_INET6); } /* @@ -2128,10 +1506,6 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset, EXPORT_SYMBOL(ip6t_register_table); EXPORT_SYMBOL(ip6t_unregister_table); EXPORT_SYMBOL(ip6t_do_table); -EXPORT_SYMBOL(ip6t_register_match); -EXPORT_SYMBOL(ip6t_unregister_match); -EXPORT_SYMBOL(ip6t_register_target); -EXPORT_SYMBOL(ip6t_unregister_target); EXPORT_SYMBOL(ip6t_ext_hdr); EXPORT_SYMBOL(ipv6_find_hdr); EXPORT_SYMBOL(ip6_masked_addrcmp); diff --git a/net/ipv6/netfilter/ip6t_HL.c b/net/ipv6/netfilter/ip6t_HL.c index 8f5549b7272..306200c3505 100644 --- a/net/ipv6/netfilter/ip6t_HL.c +++ b/net/ipv6/netfilter/ip6t_HL.c @@ -62,7 +62,7 @@ static unsigned int ip6t_hl_target(struct sk_buff **pskb, } static int ip6t_hl_checkentry(const char *tablename, - const struct ip6t_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c index ae4653bfd65..77c725832de 100644 --- a/net/ipv6/netfilter/ip6t_LOG.c +++ b/net/ipv6/netfilter/ip6t_LOG.c @@ -63,9 +63,8 @@ static void dump_packet(const struct nf_loginfo *info, return; } - /* Max length: 88 "SRC=0000.0000.0000.0000.0000.0000.0000.0000 DST=0000.0000.0000.0000.0000.0000.0000.0000" */ - printk("SRC=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(ih->saddr)); - printk("DST=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", NIP6(ih->daddr)); + /* Max length: 88 "SRC=0000.0000.0000.0000.0000.0000.0000.0000 DST=0000.0000.0000.0000.0000.0000.0000.0000 " */ + printk("SRC=" NIP6_FMT " DST=" NIP6_FMT " ", NIP6(ih->saddr), NIP6(ih->daddr)); /* Max length: 44 "LEN=65535 TC=255 HOPLIMIT=255 FLOWLBL=FFFFF " */ printk("LEN=%Zu TC=%u HOPLIMIT=%u FLOWLBL=%u ", @@ -444,7 +443,7 @@ ip6t_log_target(struct sk_buff **pskb, static int ip6t_log_checkentry(const char *tablename, - const struct ip6t_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_MARK.c b/net/ipv6/netfilter/ip6t_MARK.c deleted file mode 100644 index eab8fb864ee..00000000000 --- a/net/ipv6/netfilter/ip6t_MARK.c +++ /dev/null @@ -1,81 +0,0 @@ -/* This is a module which is used for setting the NFMARK field of an skb. */ - -/* (C) 1999-2001 Marc Boucher <marc@mbsi.ca> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> -#include <linux/ip.h> -#include <net/checksum.h> - -#include <linux/netfilter_ipv6/ip6_tables.h> -#include <linux/netfilter_ipv6/ip6t_MARK.h> - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); - -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) -{ - const struct ip6t_mark_target_info *markinfo = targinfo; - - if((*pskb)->nfmark != markinfo->mark) - (*pskb)->nfmark = markinfo->mark; - - return IP6T_CONTINUE; -} - -static int -checkentry(const char *tablename, - const struct ip6t_entry *e, - void *targinfo, - unsigned int targinfosize, - unsigned int hook_mask) -{ - if (targinfosize != IP6T_ALIGN(sizeof(struct ip6t_mark_target_info))) { - printk(KERN_WARNING "MARK: targinfosize %u != %Zu\n", - targinfosize, - IP6T_ALIGN(sizeof(struct ip6t_mark_target_info))); - return 0; - } - - if (strcmp(tablename, "mangle") != 0) { - printk(KERN_WARNING "MARK: can only be called from \"mangle\" table, not \"%s\"\n", tablename); - return 0; - } - - return 1; -} - -static struct ip6t_target ip6t_mark_reg = { - .name = "MARK", - .target = target, - .checkentry = checkentry, - .me = THIS_MODULE -}; - -static int __init init(void) -{ - printk(KERN_DEBUG "registering ipv6 mark target\n"); - if (ip6t_register_target(&ip6t_mark_reg)) - return -EINVAL; - - return 0; -} - -static void __exit fini(void) -{ - ip6t_unregister_target(&ip6t_mark_reg); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_NFQUEUE.c b/net/ipv6/netfilter/ip6t_NFQUEUE.c deleted file mode 100644 index c6e3730e740..00000000000 --- a/net/ipv6/netfilter/ip6t_NFQUEUE.c +++ /dev/null @@ -1,70 +0,0 @@ -/* ip6tables module for using new netfilter netlink queue - * - * (C) 2005 by Harald Welte <laforge@netfilter.org> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> - -#include <linux/netfilter.h> -#include <linux/netfilter_ipv6/ip6_tables.h> -#include <linux/netfilter_ipv4/ipt_NFQUEUE.h> - -MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); -MODULE_DESCRIPTION("ip6tables NFQUEUE target"); -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) -{ - const struct ipt_NFQ_info *tinfo = targinfo; - - return NF_QUEUE_NR(tinfo->queuenum); -} - -static int -checkentry(const char *tablename, - const struct ip6t_entry *e, - void *targinfo, - unsigned int targinfosize, - unsigned int hook_mask) -{ - if (targinfosize != IP6T_ALIGN(sizeof(struct ipt_NFQ_info))) { - printk(KERN_WARNING "NFQUEUE: targinfosize %u != %Zu\n", - targinfosize, - IP6T_ALIGN(sizeof(struct ipt_NFQ_info))); - return 0; - } - - return 1; -} - -static struct ip6t_target ipt_NFQ_reg = { - .name = "NFQUEUE", - .target = target, - .checkentry = checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ip6t_register_target(&ipt_NFQ_reg); -} - -static void __exit fini(void) -{ - ip6t_unregister_target(&ipt_NFQ_reg); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index b03e87adca9..c745717b4ce 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -218,12 +218,13 @@ static unsigned int reject6_target(struct sk_buff **pskb, } static int check(const char *tablename, - const struct ip6t_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { const struct ip6t_reject_info *rejinfo = targinfo; + const struct ip6t_entry *e = entry; if (targinfosize != IP6T_ALIGN(sizeof(struct ip6t_reject_info))) { DEBUGP("ip6t_REJECT: targinfosize %u != 0\n", targinfosize); diff --git a/net/ipv6/netfilter/ip6t_ah.c b/net/ipv6/netfilter/ip6t_ah.c index f5c1a7ff4a1..219a30365df 100644 --- a/net/ipv6/netfilter/ip6t_ah.c +++ b/net/ipv6/netfilter/ip6t_ah.c @@ -98,7 +98,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *entry, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_dst.c b/net/ipv6/netfilter/ip6t_dst.c index 48cf5f9efc9..80fe82669ce 100644 --- a/net/ipv6/netfilter/ip6t_dst.c +++ b/net/ipv6/netfilter/ip6t_dst.c @@ -178,7 +178,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *info, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_esp.c b/net/ipv6/netfilter/ip6t_esp.c index e1828f6d0a4..724285df871 100644 --- a/net/ipv6/netfilter/ip6t_esp.c +++ b/net/ipv6/netfilter/ip6t_esp.c @@ -76,7 +76,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_eui64.c b/net/ipv6/netfilter/ip6t_eui64.c index 616c2cbcd54..ddf5f571909 100644 --- a/net/ipv6/netfilter/ip6t_eui64.c +++ b/net/ipv6/netfilter/ip6t_eui64.c @@ -62,7 +62,7 @@ match(const struct sk_buff *skb, static int ip6t_eui64_checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_frag.c b/net/ipv6/netfilter/ip6t_frag.c index d1549b26866..a9964b946ed 100644 --- a/net/ipv6/netfilter/ip6t_frag.c +++ b/net/ipv6/netfilter/ip6t_frag.c @@ -115,7 +115,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_hbh.c b/net/ipv6/netfilter/ip6t_hbh.c index e3bc8e2700e..ed8ded18bbd 100644 --- a/net/ipv6/netfilter/ip6t_hbh.c +++ b/net/ipv6/netfilter/ip6t_hbh.c @@ -178,7 +178,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *entry, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_hl.c b/net/ipv6/netfilter/ip6t_hl.c index 0beaff5471d..c5d9079f2d9 100644 --- a/net/ipv6/netfilter/ip6t_hl.c +++ b/net/ipv6/netfilter/ip6t_hl.c @@ -48,7 +48,7 @@ static int match(const struct sk_buff *skb, const struct net_device *in, return 0; } -static int checkentry(const char *tablename, const struct ip6t_ip6 *ip, +static int checkentry(const char *tablename, const void *entry, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { diff --git a/net/ipv6/netfilter/ip6t_ipv6header.c b/net/ipv6/netfilter/ip6t_ipv6header.c index 32e67f05845..fda1ceaf5a2 100644 --- a/net/ipv6/netfilter/ip6t_ipv6header.c +++ b/net/ipv6/netfilter/ip6t_ipv6header.c @@ -124,7 +124,7 @@ ipv6header_match(const struct sk_buff *skb, static int ipv6header_checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_length.c b/net/ipv6/netfilter/ip6t_length.c deleted file mode 100644 index e0537d3811d..00000000000 --- a/net/ipv6/netfilter/ip6t_length.c +++ /dev/null @@ -1,66 +0,0 @@ -/* Length Match - IPv6 Port */ - -/* (C) 1999-2001 James Morris <jmorros@intercode.com.au> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> -#include <linux/netfilter_ipv6/ip6t_length.h> -#include <linux/netfilter_ipv6/ip6_tables.h> - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>"); -MODULE_DESCRIPTION("IPv6 packet length match"); - -static int -match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - const struct ip6t_length_info *info = matchinfo; - u_int16_t pktlen = ntohs(skb->nh.ipv6h->payload_len) + sizeof(struct ipv6hdr); - - return (pktlen >= info->min && pktlen <= info->max) ^ info->invert; -} - -static int -checkentry(const char *tablename, - const struct ip6t_ip6 *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_length_info))) - return 0; - - return 1; -} - -static struct ip6t_match length_match = { - .name = "length", - .match = &match, - .checkentry = &checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ip6t_register_match(&length_match); -} - -static void __exit fini(void) -{ - ip6t_unregister_match(&length_match); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_limit.c b/net/ipv6/netfilter/ip6t_limit.c deleted file mode 100644 index fb782f610be..00000000000 --- a/net/ipv6/netfilter/ip6t_limit.c +++ /dev/null @@ -1,147 +0,0 @@ -/* Kernel module to control the rate - * - * 2 September 1999: Changed from the target RATE to the match - * `limit', removed logging. Did I mention that - * Alexey is a fucking genius? - * Rusty Russell (rusty@rustcorp.com.au). */ - -/* (C) 1999 Jérôme de Vivie <devivie@info.enserb.u-bordeaux.fr> - * (C) 1999 Hervé Eychenne <eychenne@info.enserb.u-bordeaux.fr> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> -#include <linux/spinlock.h> -#include <linux/interrupt.h> - -#include <linux/netfilter_ipv6/ip6_tables.h> -#include <linux/netfilter_ipv6/ip6t_limit.h> - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Herve Eychenne <rv@wallfire.org>"); -MODULE_DESCRIPTION("rate limiting within ip6tables"); - -/* The algorithm used is the Simple Token Bucket Filter (TBF) - * see net/sched/sch_tbf.c in the linux source tree - */ - -static DEFINE_SPINLOCK(limit_lock); - -/* 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 avoid underflow, we multiply by 128 (ie. you get 128 credits per - jiffy). Hence a cost of 2^32-1, means one pass per 32768 seconds - at 1024HZ (or one every 9 hours). A cost of 1 means 12800 passes - per second at 100HZ. */ - -#define CREDITS_PER_JIFFY 128 - -static int -ip6t_limit_match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - struct ip6t_rateinfo *r = ((struct ip6t_rateinfo *)matchinfo)->master; - unsigned long now = jiffies; - - spin_lock_bh(&limit_lock); - r->credit += (now - xchg(&r->prev, now)) * CREDITS_PER_JIFFY; - if (r->credit > r->credit_cap) - r->credit = r->credit_cap; - - if (r->credit >= r->cost) { - /* We're not limited. */ - r->credit -= r->cost; - spin_unlock_bh(&limit_lock); - return 1; - } - - spin_unlock_bh(&limit_lock); - return 0; -} - -/* Precision saver. */ -static u_int32_t -user2credits(u_int32_t user) -{ - /* If multiplying would overflow... */ - if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) - /* Divide first. */ - return (user / IP6T_LIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; - - return (user * HZ * CREDITS_PER_JIFFY) / IP6T_LIMIT_SCALE; -} - -static int -ip6t_limit_checkentry(const char *tablename, - const struct ip6t_ip6 *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - struct ip6t_rateinfo *r = matchinfo; - - if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_rateinfo))) - return 0; - - /* Check for overflow. */ - if (r->burst == 0 - || user2credits(r->avg * r->burst) < user2credits(r->avg)) { - printk("Call rusty: overflow in ip6t_limit: %u/%u\n", - r->avg, r->burst); - return 0; - } - - /* User avg in seconds * IP6T_LIMIT_SCALE: convert to jiffies * - 128. */ - r->prev = jiffies; - r->credit = user2credits(r->avg * r->burst); /* Credits full. */ - r->credit_cap = user2credits(r->avg * r->burst); /* Credits full. */ - r->cost = user2credits(r->avg); - - /* For SMP, we only want to use one set of counters. */ - r->master = r; - - return 1; -} - -static struct ip6t_match ip6t_limit_reg = { - .name = "limit", - .match = ip6t_limit_match, - .checkentry = ip6t_limit_checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - if (ip6t_register_match(&ip6t_limit_reg)) - return -EINVAL; - return 0; -} - -static void __exit fini(void) -{ - ip6t_unregister_match(&ip6t_limit_reg); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_mac.c b/net/ipv6/netfilter/ip6t_mac.c deleted file mode 100644 index c848152315b..00000000000 --- a/net/ipv6/netfilter/ip6t_mac.c +++ /dev/null @@ -1,81 +0,0 @@ -/* Kernel module to match MAC address parameters. */ - -/* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> -#include <linux/if_ether.h> -#include <linux/etherdevice.h> - -#include <linux/netfilter_ipv6/ip6t_mac.h> -#include <linux/netfilter_ipv6/ip6_tables.h> - -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("MAC address matching module for IPv6"); -MODULE_AUTHOR("Netfilter Core Teaam <coreteam@netfilter.org>"); - -static int -match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - const struct ip6t_mac_info *info = matchinfo; - - /* Is mac pointer valid? */ - return (skb->mac.raw >= skb->head - && (skb->mac.raw + ETH_HLEN) <= skb->data - /* If so, compare... */ - && ((!compare_ether_addr(eth_hdr(skb)->h_source, info->srcaddr)) - ^ info->invert)); -} - -static int -ip6t_mac_checkentry(const char *tablename, - const struct ip6t_ip6 *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - if (hook_mask - & ~((1 << NF_IP6_PRE_ROUTING) | (1 << NF_IP6_LOCAL_IN) - | (1 << NF_IP6_FORWARD))) { - printk("ip6t_mac: only valid for PRE_ROUTING, LOCAL_IN or" - " FORWARD\n"); - return 0; - } - - if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_mac_info))) - return 0; - - return 1; -} - -static struct ip6t_match mac_match = { - .name = "mac", - .match = &match, - .checkentry = &ip6t_mac_checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ip6t_register_match(&mac_match); -} - -static void __exit fini(void) -{ - ip6t_unregister_match(&mac_match); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_mark.c b/net/ipv6/netfilter/ip6t_mark.c deleted file mode 100644 index affc3de364f..00000000000 --- a/net/ipv6/netfilter/ip6t_mark.c +++ /dev/null @@ -1,66 +0,0 @@ -/* Kernel module to match NFMARK values. */ - -/* (C) 1999-2001 Marc Boucher <marc@mbsi.ca> - * - * 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 <linux/module.h> -#include <linux/skbuff.h> - -#include <linux/netfilter_ipv6/ip6t_mark.h> -#include <linux/netfilter_ipv6/ip6_tables.h> - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); -MODULE_DESCRIPTION("ip6tables mark match"); - -static int -match(const struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - const void *matchinfo, - int offset, - unsigned int protoff, - int *hotdrop) -{ - const struct ip6t_mark_info *info = matchinfo; - - return ((skb->nfmark & info->mask) == info->mark) ^ info->invert; -} - -static int -checkentry(const char *tablename, - const struct ip6t_ip6 *ip, - void *matchinfo, - unsigned int matchsize, - unsigned int hook_mask) -{ - if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_mark_info))) - return 0; - - return 1; -} - -static struct ip6t_match mark_match = { - .name = "mark", - .match = &match, - .checkentry = &checkentry, - .me = THIS_MODULE, -}; - -static int __init init(void) -{ - return ip6t_register_match(&mark_match); -} - -static void __exit fini(void) -{ - ip6t_unregister_match(&mark_match); -} - -module_init(init); -module_exit(fini); diff --git a/net/ipv6/netfilter/ip6t_multiport.c b/net/ipv6/netfilter/ip6t_multiport.c index 6e3246153fa..49f7829dfbc 100644 --- a/net/ipv6/netfilter/ip6t_multiport.c +++ b/net/ipv6/netfilter/ip6t_multiport.c @@ -84,11 +84,12 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *info, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { + const struct ip6t_ip6 *ip = info; const struct ip6t_multiport *multiinfo = matchinfo; if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_multiport))) diff --git a/net/ipv6/netfilter/ip6t_owner.c b/net/ipv6/netfilter/ip6t_owner.c index 4de4cdad4b7..5409b375b51 100644 --- a/net/ipv6/netfilter/ip6t_owner.c +++ b/net/ipv6/netfilter/ip6t_owner.c @@ -53,7 +53,7 @@ match(const struct sk_buff *skb, static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6t_rt.c b/net/ipv6/netfilter/ip6t_rt.c index c1e770e4554..8465b437585 100644 --- a/net/ipv6/netfilter/ip6t_rt.c +++ b/net/ipv6/netfilter/ip6t_rt.c @@ -183,7 +183,7 @@ match(const struct sk_buff *skb, /* Called when user tries to insert an entry of this type. */ static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *entry, void *matchinfo, unsigned int matchinfosize, unsigned int hook_mask) diff --git a/net/ipv6/netfilter/ip6table_filter.c b/net/ipv6/netfilter/ip6table_filter.c index 4c0028671c2..ce4a968e1f7 100644 --- a/net/ipv6/netfilter/ip6table_filter.c +++ b/net/ipv6/netfilter/ip6table_filter.c @@ -97,6 +97,7 @@ static struct ip6t_table packet_filter = { .valid_hooks = FILTER_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, .me = THIS_MODULE, + .af = AF_INET6, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv6/netfilter/ip6table_mangle.c b/net/ipv6/netfilter/ip6table_mangle.c index 85c1e6eada1..30a4627e000 100644 --- a/net/ipv6/netfilter/ip6table_mangle.c +++ b/net/ipv6/netfilter/ip6table_mangle.c @@ -127,6 +127,7 @@ static struct ip6t_table packet_mangler = { .valid_hooks = MANGLE_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, .me = THIS_MODULE, + .af = AF_INET6, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv6/netfilter/ip6table_raw.c b/net/ipv6/netfilter/ip6table_raw.c index c2982efd14a..db28ba3855e 100644 --- a/net/ipv6/netfilter/ip6table_raw.c +++ b/net/ipv6/netfilter/ip6table_raw.c @@ -106,11 +106,12 @@ static struct } }; -static struct ip6t_table packet_raw = { +static struct xt_table packet_raw = { .name = "raw", .valid_hooks = RAW_VALID_HOOKS, .lock = RW_LOCK_UNLOCKED, - .me = THIS_MODULE + .me = THIS_MODULE, + .af = AF_INET6, }; /* The work comes in here from netfilter.c. */ diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index e57d6fc9957..ac702a29dd1 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -74,7 +74,7 @@ static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, static int ipv6_print_tuple(struct seq_file *s, const struct nf_conntrack_tuple *tuple) { - return seq_printf(s, "src=%x:%x:%x:%x:%x:%x:%x:%x dst=%x:%x:%x:%x:%x:%x:%x:%x ", + return seq_printf(s, "src=" NIP6_FMT " dst=" NIP6_FMT " ", NIP6(*((struct in6_addr *)tuple->src.u3.ip6)), NIP6(*((struct in6_addr *)tuple->dst.u3.ip6))); } @@ -584,7 +584,7 @@ MODULE_AUTHOR("Yasuyuki KOZAKAI @USAGI <yasuyuki.kozakai@toshiba.co.jp>"); static int __init init(void) { - need_nf_conntrack(); + need_conntrack(); return init_or_cleanup(1); } @@ -595,9 +595,3 @@ static void __exit fini(void) module_init(init); module_exit(fini); - -void need_ip6_conntrack(void) -{ -} - -EXPORT_SYMBOL(need_ip6_conntrack); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index f3e5ffbd592..84ef9a13108 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -70,8 +70,8 @@ struct nf_ct_frag6_skb_cb struct nf_ct_frag6_queue { - struct nf_ct_frag6_queue *next; - struct list_head lru_list; /* lru list member */ + struct hlist_node list; + struct list_head lru_list; /* lru list member */ __u32 id; /* fragment id */ struct in6_addr saddr; @@ -90,14 +90,13 @@ struct nf_ct_frag6_queue #define FIRST_IN 2 #define LAST_IN 1 __u16 nhoffset; - struct nf_ct_frag6_queue **pprev; }; /* Hash table. */ #define FRAG6Q_HASHSZ 64 -static struct nf_ct_frag6_queue *nf_ct_frag6_hash[FRAG6Q_HASHSZ]; +static struct hlist_head nf_ct_frag6_hash[FRAG6Q_HASHSZ]; static DEFINE_RWLOCK(nf_ct_frag6_lock); static u32 nf_ct_frag6_hash_rnd; static LIST_HEAD(nf_ct_frag6_lru_list); @@ -105,9 +104,7 @@ int nf_ct_frag6_nqueues = 0; static __inline__ void __fq_unlink(struct nf_ct_frag6_queue *fq) { - if (fq->next) - fq->next->pprev = fq->pprev; - *fq->pprev = fq->next; + hlist_del(&fq->list); list_del(&fq->lru_list); nf_ct_frag6_nqueues--; } @@ -158,28 +155,18 @@ static void nf_ct_frag6_secret_rebuild(unsigned long dummy) get_random_bytes(&nf_ct_frag6_hash_rnd, sizeof(u32)); for (i = 0; i < FRAG6Q_HASHSZ; i++) { struct nf_ct_frag6_queue *q; + struct hlist_node *p, *n; - q = nf_ct_frag6_hash[i]; - while (q) { - struct nf_ct_frag6_queue *next = q->next; + hlist_for_each_entry_safe(q, p, n, &nf_ct_frag6_hash[i], list) { unsigned int hval = ip6qhashfn(q->id, &q->saddr, &q->daddr); - if (hval != i) { - /* Unlink. */ - if (q->next) - q->next->pprev = q->pprev; - *q->pprev = q->next; - + hlist_del(&q->list); /* Relink to new hash chain. */ - if ((q->next = nf_ct_frag6_hash[hval]) != NULL) - q->next->pprev = &q->next; - nf_ct_frag6_hash[hval] = q; - q->pprev = &nf_ct_frag6_hash[hval]; + hlist_add_head(&q->list, + &nf_ct_frag6_hash[hval]); } - - q = next; } } write_unlock(&nf_ct_frag6_lock); @@ -314,15 +301,17 @@ out: /* Creation primitives. */ - static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, struct nf_ct_frag6_queue *fq_in) { struct nf_ct_frag6_queue *fq; +#ifdef CONFIG_SMP + struct hlist_node *n; +#endif write_lock(&nf_ct_frag6_lock); #ifdef CONFIG_SMP - for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + hlist_for_each_entry(fq, n, &nf_ct_frag6_hash[hash], list) { if (fq->id == fq_in->id && !ipv6_addr_cmp(&fq_in->saddr, &fq->saddr) && !ipv6_addr_cmp(&fq_in->daddr, &fq->daddr)) { @@ -340,10 +329,7 @@ static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, atomic_inc(&fq->refcnt); atomic_inc(&fq->refcnt); - if ((fq->next = nf_ct_frag6_hash[hash]) != NULL) - fq->next->pprev = &fq->next; - nf_ct_frag6_hash[hash] = fq; - fq->pprev = &nf_ct_frag6_hash[hash]; + hlist_add_head(&fq->list, &nf_ct_frag6_hash[hash]); INIT_LIST_HEAD(&fq->lru_list); list_add_tail(&fq->lru_list, &nf_ct_frag6_lru_list); nf_ct_frag6_nqueues++; @@ -384,10 +370,11 @@ static __inline__ struct nf_ct_frag6_queue * fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) { struct nf_ct_frag6_queue *fq; + struct hlist_node *n; unsigned int hash = ip6qhashfn(id, src, dst); read_lock(&nf_ct_frag6_lock); - for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + hlist_for_each_entry(fq, n, &nf_ct_frag6_hash[hash], list) { if (fq->id == id && !ipv6_addr_cmp(src, &fq->saddr) && !ipv6_addr_cmp(dst, &fq->daddr)) { diff --git a/net/ipv6/xfrm6_state.c b/net/ipv6/xfrm6_state.c index bf0d0abc387..a5723024d3b 100644 --- a/net/ipv6/xfrm6_state.c +++ b/net/ipv6/xfrm6_state.c @@ -15,6 +15,7 @@ #include <linux/pfkeyv2.h> #include <linux/ipsec.h> #include <net/ipv6.h> +#include <net/addrconf.h> static struct xfrm_state_afinfo xfrm6_state_afinfo; @@ -41,6 +42,22 @@ __xfrm6_init_tempsel(struct xfrm_state *x, struct flowi *fl, memcpy(&x->props.saddr, &tmpl->saddr, sizeof(x->props.saddr)); if (ipv6_addr_any((struct in6_addr*)&x->props.saddr)) memcpy(&x->props.saddr, saddr, sizeof(x->props.saddr)); + if (tmpl->mode && ipv6_addr_any((struct in6_addr*)&x->props.saddr)) { + struct rt6_info *rt; + struct flowi fl_tunnel = { + .nl_u = { + .ip6_u = { + .daddr = *(struct in6_addr *)daddr, + } + } + }; + if (!xfrm_dst_lookup((struct xfrm_dst **)&rt, + &fl_tunnel, AF_INET6)) { + ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)daddr, + (struct in6_addr *)&x->props.saddr); + dst_release(&rt->u.dst); + } + } x->props.mode = tmpl->mode; x->props.reqid = tmpl->reqid; x->props.family = AF_INET6; diff --git a/net/ipv6/xfrm6_tunnel.c b/net/ipv6/xfrm6_tunnel.c index da09ff25864..8cfc58b96fc 100644 --- a/net/ipv6/xfrm6_tunnel.c +++ b/net/ipv6/xfrm6_tunnel.c @@ -259,8 +259,7 @@ try_next_2:; spi = 0; goto out; alloc_spi: - X6TPRINTK3(KERN_DEBUG "%s(): allocate new spi for " - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + X6TPRINTK3(KERN_DEBUG "%s(): allocate new spi for " NIP6_FMT "\n", __FUNCTION__, NIP6(*(struct in6_addr *)saddr)); x6spi = kmem_cache_alloc(xfrm6_tunnel_spi_kmem, SLAB_ATOMIC); @@ -323,9 +322,8 @@ void xfrm6_tunnel_free_spi(xfrm_address_t *saddr) list_byaddr) { if (memcmp(&x6spi->addr, saddr, sizeof(x6spi->addr)) == 0) { - X6TPRINTK3(KERN_DEBUG "%s(): x6spi object " - "for %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x " - "found at %p\n", + X6TPRINTK3(KERN_DEBUG "%s(): x6spi object for " NIP6_FMT + " found at %p\n", __FUNCTION__, NIP6(*(struct in6_addr *)saddr), x6spi); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 7d55f9cbd85..99c0a0fa4a9 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -103,3 +103,261 @@ config NF_CT_NETLINK This option enables support for a netlink-based userspace interface endmenu + +config NETFILTER_XTABLES + tristate "Netfilter Xtables support (required for ip_tables)" + help + This is required if you intend to use any of ip_tables, + ip6_tables or arp_tables. + +# alphabetically ordered list of targets + +config NETFILTER_XT_TARGET_CLASSIFY + tristate '"CLASSIFY" target support' + depends on NETFILTER_XTABLES + help + This option adds a `CLASSIFY' target, which enables the user to set + the priority of a packet. Some qdiscs can use this value for + classification, among these are: + + atm, cbq, dsmark, pfifo_fast, htb, prio + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_TARGET_CONNMARK + tristate '"CONNMARK" target support' + depends on NETFILTER_XTABLES + depends on IP_NF_MANGLE || IP6_NF_MANGLE + depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4) + help + This option adds a `CONNMARK' target, which allows one to manipulate + the connection mark value. Similar to the MARK target, but + affects the connection mark value rather than the packet mark value. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. The module will be called + ipt_CONNMARK.o. If unsure, say `N'. + +config NETFILTER_XT_TARGET_MARK + tristate '"MARK" target support' + depends on NETFILTER_XTABLES + help + This option adds a `MARK' target, which allows you to create rules + in the `mangle' table which alter the netfilter mark (nfmark) field + associated with the packet prior to routing. This can change + the routing method (see `Use netfilter MARK value as routing + key') and can also be used by other subsystems to change their + behavior. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_TARGET_NFQUEUE + tristate '"NFQUEUE" target Support' + depends on NETFILTER_XTABLES + help + This Target replaced the old obsolete QUEUE target. + + As opposed to QUEUE, it supports 65535 different queues, + not just one. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_TARGET_NOTRACK + tristate '"NOTRACK" target support' + depends on NETFILTER_XTABLES + depends on IP_NF_RAW || IP6_NF_RAW + depends on IP_NF_CONNTRACK || NF_CONNTRACK + help + The NOTRACK target allows a select rule to specify + which packets *not* to enter the conntrack/NAT + subsystem with all the consequences (no ICMP error tracking, + no protocol helpers for the selected packets). + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_COMMENT + tristate '"comment" match support' + depends on NETFILTER_XTABLES + help + This option adds a `comment' dummy-match, which allows you to put + comments in your iptables ruleset. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_CONNBYTES + tristate '"connbytes" per-connection counter match support' + depends on NETFILTER_XTABLES + depends on (IP_NF_CONNTRACK && IP_NF_CT_ACCT) || NF_CT_ACCT + help + This option adds a `connbytes' match, which allows you to match the + number of bytes and/or packets for each direction within a connection. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_CONNMARK + tristate '"connmark" connection mark match support' + depends on NETFILTER_XTABLES + depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || NF_CONNTRACK_MARK + help + This option adds a `connmark' match, which allows you to match the + connection mark value previously set for the session by `CONNMARK'. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. The module will be called + ipt_connmark.o. If unsure, say `N'. + +config NETFILTER_XT_MATCH_CONNTRACK + tristate '"conntrack" connection tracking match support' + depends on NETFILTER_XTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK + help + This is a general conntrack match module, a superset of the state match. + + It allows matching on additional conntrack information, which is + useful in complex configurations, such as NAT gateways with multiple + internet links or tunnels. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_DCCP + tristate '"DCCP" protocol match support' + depends on NETFILTER_XTABLES + help + With this option enabled, you will be able to use the iptables + `dccp' match in order to match on DCCP source/destination ports + and DCCP flags. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_HELPER + tristate '"helper" match support' + depends on NETFILTER_XTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK + help + Helper matching allows you to match packets in dynamic connections + tracked by a conntrack-helper, ie. ip_conntrack_ftp + + To compile it as a module, choose M here. If unsure, say Y. + +config NETFILTER_XT_MATCH_LENGTH + tristate '"length" match support' + depends on NETFILTER_XTABLES + help + This option allows you to match the length of a packet against a + specific value or range of values. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_LIMIT + tristate '"limit" match support' + depends on NETFILTER_XTABLES + help + limit matching allows you to control the rate at which a rule can be + matched: mainly useful in combination with the LOG target ("LOG + target support", below) and to avoid some Denial of Service attacks. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_MAC + tristate '"mac" address match support' + depends on NETFILTER_XTABLES + help + MAC matching allows you to match packets based on the source + Ethernet address of the packet. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_MARK + tristate '"mark" match support' + depends on NETFILTER_XTABLES + help + Netfilter mark matching allows you to match packets based on the + `nfmark' value in the packet. This can be set by the MARK target + (see below). + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_PHYSDEV + tristate '"physdev" match support' + depends on NETFILTER_XTABLES && BRIDGE_NETFILTER + help + Physdev packet matching matches against the physical bridge ports + the IP packet arrived on or will leave by. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_PKTTYPE + tristate '"pkttype" packet type match support' + depends on NETFILTER_XTABLES + help + Packet type matching allows you to match a packet by + its "class", eg. BROADCAST, MULTICAST, ... + + Typical usage: + iptables -A INPUT -m pkttype --pkt-type broadcast -j LOG + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_REALM + tristate '"realm" match support' + depends on NETFILTER_XTABLES + select NET_CLS_ROUTE + help + This option adds a `realm' match, which allows you to use the realm + key from the routing subsystem inside iptables. + + This match pretty much resembles the CONFIG_NET_CLS_ROUTE4 option + in tc world. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_SCTP + tristate '"sctp" protocol match support' + depends on NETFILTER_XTABLES + help + With this option enabled, you will be able to use the + `sctp' match in order to match on SCTP source/destination ports + and SCTP chunk types. + + If you want to compile it as a module, say M here and read + <file:Documentation/modules.txt>. If unsure, say `N'. + +config NETFILTER_XT_MATCH_STATE + tristate '"state" match support' + depends on NETFILTER_XTABLES + depends on IP_NF_CONNTRACK || NF_CONNTRACK + help + Connection state matching allows you to match packets based on their + relationship to a tracked connection (ie. previous packets). This + is a powerful tool for packet classification. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_STRING + tristate '"string" match support' + depends on NETFILTER_XTABLES + select TEXTSEARCH + select TEXTSEARCH_KMP + select TEXTSEARCH_BM + select TEXTSEARCH_FSM + help + This option adds a `string' match, which allows you to look for + pattern matchings in packets. + + To compile it as a module, choose M here. If unsure, say N. + +config NETFILTER_XT_MATCH_TCPMSS + tristate '"tcpmss" match support' + depends on NETFILTER_XTABLES + help + This option adds a `tcpmss' match, which allows you to examine the + MSS value of TCP SYN packets, which control the maximum packet size + for that connection. + + To compile it as a module, choose M here. If unsure, say N. + diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index cb2183145c3..746172ebc91 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -1,4 +1,5 @@ netfilter-objs := core.o nf_log.o nf_queue.o nf_sockopt.o +nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o obj-$(CONFIG_NETFILTER) = netfilter.o @@ -6,13 +7,43 @@ obj-$(CONFIG_NETFILTER_NETLINK) += nfnetlink.o obj-$(CONFIG_NETFILTER_NETLINK_QUEUE) += nfnetlink_queue.o obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o -nf_conntrack-objs := nf_conntrack_core.o nf_conntrack_standalone.o nf_conntrack_l3proto_generic.o nf_conntrack_proto_generic.o nf_conntrack_proto_tcp.o nf_conntrack_proto_udp.o - +# connection tracking obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o -obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o # SCTP protocol connection tracking obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o # netlink interface for nf_conntrack obj-$(CONFIG_NF_CT_NETLINK) += nf_conntrack_netlink.o + +# connection tracking helpers +obj-$(CONFIG_NF_CONNTRACK_FTP) += nf_conntrack_ftp.o + +# generic X tables +obj-$(CONFIG_NETFILTER_XTABLES) += x_tables.o xt_tcpudp.o + +# targets +obj-$(CONFIG_NETFILTER_XT_TARGET_CLASSIFY) += xt_CLASSIFY.o +obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o +obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o +obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o +obj-$(CONFIG_NETFILTER_XT_TARGET_NOTRACK) += xt_NOTRACK.o + +# matches +obj-$(CONFIG_NETFILTER_XT_MATCH_COMMENT) += xt_comment.o +obj-$(CONFIG_NETFILTER_XT_MATCH_CONNBYTES) += xt_connbytes.o +obj-$(CONFIG_NETFILTER_XT_MATCH_CONNMARK) += xt_connmark.o +obj-$(CONFIG_NETFILTER_XT_MATCH_CONNTRACK) += xt_conntrack.o +obj-$(CONFIG_NETFILTER_XT_MATCH_DCCP) += xt_dccp.o +obj-$(CONFIG_NETFILTER_XT_MATCH_HELPER) += xt_helper.o +obj-$(CONFIG_NETFILTER_XT_MATCH_LENGTH) += xt_length.o +obj-$(CONFIG_NETFILTER_XT_MATCH_LIMIT) += xt_limit.o +obj-$(CONFIG_NETFILTER_XT_MATCH_MAC) += xt_mac.o +obj-$(CONFIG_NETFILTER_XT_MATCH_MARK) += xt_mark.o +obj-$(CONFIG_NETFILTER_XT_MATCH_PKTTYPE) += xt_pkttype.o +obj-$(CONFIG_NETFILTER_XT_MATCH_REALM) += xt_realm.o +obj-$(CONFIG_NETFILTER_XT_MATCH_SCTP) += xt_sctp.o +obj-$(CONFIG_NETFILTER_XT_MATCH_STATE) += xt_state.o +obj-$(CONFIG_NETFILTER_XT_MATCH_STRING) += xt_string.o +obj-$(CONFIG_NETFILTER_XT_MATCH_TCPMSS) += xt_tcpmss.o +obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index d5a6eaf4a1d..ab0c920f0d3 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -545,11 +545,11 @@ static int help(struct sk_buff **pskb, different IP address. Simply don't record it for NAT. */ if (cmd.l3num == PF_INET) { - DEBUGP("conntrack_ftp: NOT RECORDING: %u,%u,%u,%u != %u.%u.%u.%u\n", + DEBUGP("conntrack_ftp: NOT RECORDING: " NIPQUAD_FMT " != " NIPQUAD_FMT "\n", NIPQUAD(cmd.u3.ip), NIPQUAD(ct->tuplehash[dir].tuple.src.u3.ip)); } else { - DEBUGP("conntrack_ftp: NOT RECORDING: %x:%x:%x:%x:%x:%x:%x:%x != %x:%x:%x:%x:%x:%x:%x:%x\n", + DEBUGP("conntrack_ftp: NOT RECORDING: " NIP6_FMT " != " NIP6_FMT "\n", NIP6(*((struct in6_addr *)cmd.u3.ip6)), NIP6(*((struct in6_addr *)ct->tuplehash[dir] .tuple.src.u3.ip6))); diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 3531d142f69..617599aeeea 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -821,7 +821,7 @@ module_exit(fini); /* Some modules need us, but don't depend directly on any symbol. They should call this. */ -void need_nf_conntrack(void) +void need_conntrack(void) { } @@ -841,7 +841,7 @@ EXPORT_SYMBOL(nf_conntrack_protocol_unregister); EXPORT_SYMBOL(nf_ct_invert_tuplepr); EXPORT_SYMBOL(nf_conntrack_alter_reply); EXPORT_SYMBOL(nf_conntrack_destroyed); -EXPORT_SYMBOL(need_nf_conntrack); +EXPORT_SYMBOL(need_conntrack); EXPORT_SYMBOL(nf_conntrack_helper_register); EXPORT_SYMBOL(nf_conntrack_helper_unregister); EXPORT_SYMBOL(nf_ct_iterate_cleanup); diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c new file mode 100644 index 00000000000..d7817afc6b9 --- /dev/null +++ b/net/netfilter/x_tables.c @@ -0,0 +1,624 @@ +/* + * x_tables core - Backend for {ip,ip6,arp}_tables + * + * Copyright (C) 2006-2006 Harald Welte <laforge@netfilter.org> + * + * Based on existing ip_tables code which is + * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling + * Copyright (C) 2000-2005 Netfilter Core Team <coreteam@netfilter.org> + * + * 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 <linux/config.h> +#include <linux/kernel.h> +#include <linux/socket.h> +#include <linux/net.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/string.h> +#include <linux/vmalloc.h> + +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter_arp.h> + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); +MODULE_DESCRIPTION("[ip,ip6,arp]_tables backend module"); + +#define SMP_ALIGN(x) (((x) + SMP_CACHE_BYTES-1) & ~(SMP_CACHE_BYTES-1)) + +struct xt_af { + struct semaphore mutex; + struct list_head match; + struct list_head target; + struct list_head tables; +}; + +static struct xt_af *xt; + +#ifdef DEBUG_IP_FIREWALL_USER +#define duprintf(format, args...) printk(format , ## args) +#else +#define duprintf(format, args...) +#endif + +enum { + TABLE, + TARGET, + MATCH, +}; + +/* Registration hooks for targets. */ +int +xt_register_target(int af, struct xt_target *target) +{ + int ret; + + ret = down_interruptible(&xt[af].mutex); + if (ret != 0) + return ret; + list_add(&target->list, &xt[af].target); + up(&xt[af].mutex); + return ret; +} +EXPORT_SYMBOL(xt_register_target); + +void +xt_unregister_target(int af, struct xt_target *target) +{ + down(&xt[af].mutex); + LIST_DELETE(&xt[af].target, target); + up(&xt[af].mutex); +} +EXPORT_SYMBOL(xt_unregister_target); + +int +xt_register_match(int af, struct xt_match *match) +{ + int ret; + + ret = down_interruptible(&xt[af].mutex); + if (ret != 0) + return ret; + + list_add(&match->list, &xt[af].match); + up(&xt[af].mutex); + + return ret; +} +EXPORT_SYMBOL(xt_register_match); + +void +xt_unregister_match(int af, struct xt_match *match) +{ + down(&xt[af].mutex); + LIST_DELETE(&xt[af].match, match); + up(&xt[af].mutex); +} +EXPORT_SYMBOL(xt_unregister_match); + + +/* + * These are weird, but module loading must not be done with mutex + * held (since they will register), and we have to have a single + * function to use try_then_request_module(). + */ + +/* Find match, grabs ref. Returns ERR_PTR() on error. */ +struct xt_match *xt_find_match(int af, const char *name, u8 revision) +{ + struct xt_match *m; + int err = 0; + + if (down_interruptible(&xt[af].mutex) != 0) + return ERR_PTR(-EINTR); + + list_for_each_entry(m, &xt[af].match, list) { + if (strcmp(m->name, name) == 0) { + if (m->revision == revision) { + if (try_module_get(m->me)) { + up(&xt[af].mutex); + return m; + } + } else + err = -EPROTOTYPE; /* Found something. */ + } + } + up(&xt[af].mutex); + return ERR_PTR(err); +} +EXPORT_SYMBOL(xt_find_match); + +/* Find target, grabs ref. Returns ERR_PTR() on error. */ +struct xt_target *xt_find_target(int af, const char *name, u8 revision) +{ + struct xt_target *t; + int err = 0; + + if (down_interruptible(&xt[af].mutex) != 0) + return ERR_PTR(-EINTR); + + list_for_each_entry(t, &xt[af].target, list) { + if (strcmp(t->name, name) == 0) { + if (t->revision == revision) { + if (try_module_get(t->me)) { + up(&xt[af].mutex); + return t; + } + } else + err = -EPROTOTYPE; /* Found something. */ + } + } + up(&xt[af].mutex); + return ERR_PTR(err); +} +EXPORT_SYMBOL(xt_find_target); + +static const char *xt_prefix[NPROTO] = { + [AF_INET] = "ipt_%s", + [AF_INET6] = "ip6t_%s", + [NF_ARP] = "arpt_%s", +}; + +struct xt_target *xt_request_find_target(int af, const char *name, u8 revision) +{ + struct xt_target *target; + + target = try_then_request_module(xt_find_target(af, name, revision), + xt_prefix[af], name); + if (IS_ERR(target) || !target) + return NULL; + return target; +} +EXPORT_SYMBOL_GPL(xt_request_find_target); + +static int match_revfn(int af, const char *name, u8 revision, int *bestp) +{ + struct xt_match *m; + int have_rev = 0; + + list_for_each_entry(m, &xt[af].match, list) { + if (strcmp(m->name, name) == 0) { + if (m->revision > *bestp) + *bestp = m->revision; + if (m->revision == revision) + have_rev = 1; + } + } + return have_rev; +} + +static int target_revfn(int af, const char *name, u8 revision, int *bestp) +{ + struct xt_target *t; + int have_rev = 0; + + list_for_each_entry(t, &xt[af].target, list) { + if (strcmp(t->name, name) == 0) { + if (t->revision > *bestp) + *bestp = t->revision; + if (t->revision == revision) + have_rev = 1; + } + } + return have_rev; +} + +/* Returns true or false (if no such extension at all) */ +int xt_find_revision(int af, const char *name, u8 revision, int target, + int *err) +{ + int have_rev, best = -1; + + if (down_interruptible(&xt[af].mutex) != 0) { + *err = -EINTR; + return 1; + } + if (target == 1) + have_rev = target_revfn(af, name, revision, &best); + else + have_rev = match_revfn(af, name, revision, &best); + up(&xt[af].mutex); + + /* Nothing at all? Return 0 to try loading module. */ + if (best == -1) { + *err = -ENOENT; + return 0; + } + + *err = best; + if (!have_rev) + *err = -EPROTONOSUPPORT; + return 1; +} +EXPORT_SYMBOL_GPL(xt_find_revision); + +struct xt_table_info *xt_alloc_table_info(unsigned int size) +{ + struct xt_table_info *newinfo; + int cpu; + + /* Pedantry: prevent them from hitting BUG() in vmalloc.c --RR */ + if ((SMP_ALIGN(size) >> PAGE_SHIFT) + 2 > num_physpages) + return NULL; + + newinfo = kzalloc(sizeof(struct xt_table_info), GFP_KERNEL); + if (!newinfo) + return NULL; + + newinfo->size = size; + + for_each_cpu(cpu) { + if (size <= PAGE_SIZE) + newinfo->entries[cpu] = kmalloc_node(size, + GFP_KERNEL, + cpu_to_node(cpu)); + else + newinfo->entries[cpu] = vmalloc_node(size, + cpu_to_node(cpu)); + + if (newinfo->entries[cpu] == NULL) { + xt_free_table_info(newinfo); + return NULL; + } + } + + return newinfo; +} +EXPORT_SYMBOL(xt_alloc_table_info); + +void xt_free_table_info(struct xt_table_info *info) +{ + int cpu; + + for_each_cpu(cpu) { + if (info->size <= PAGE_SIZE) + kfree(info->entries[cpu]); + else + vfree(info->entries[cpu]); + } + kfree(info); +} +EXPORT_SYMBOL(xt_free_table_info); + +/* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ +struct xt_table *xt_find_table_lock(int af, const char *name) +{ + struct xt_table *t; + + if (down_interruptible(&xt[af].mutex) != 0) + return ERR_PTR(-EINTR); + + list_for_each_entry(t, &xt[af].tables, list) + if (strcmp(t->name, name) == 0 && try_module_get(t->me)) + return t; + up(&xt[af].mutex); + return NULL; +} +EXPORT_SYMBOL_GPL(xt_find_table_lock); + +void xt_table_unlock(struct xt_table *table) +{ + up(&xt[table->af].mutex); +} +EXPORT_SYMBOL_GPL(xt_table_unlock); + + +struct xt_table_info * +xt_replace_table(struct xt_table *table, + unsigned int num_counters, + struct xt_table_info *newinfo, + int *error) +{ + struct xt_table_info *oldinfo, *private; + + /* Do the substitution. */ + write_lock_bh(&table->lock); + private = table->private; + /* Check inside lock: is the old number correct? */ + if (num_counters != private->number) { + duprintf("num_counters != table->private->number (%u/%u)\n", + num_counters, private->number); + write_unlock_bh(&table->lock); + *error = -EAGAIN; + return NULL; + } + oldinfo = private; + table->private = newinfo; + newinfo->initial_entries = oldinfo->initial_entries; + write_unlock_bh(&table->lock); + + return oldinfo; +} +EXPORT_SYMBOL_GPL(xt_replace_table); + +int xt_register_table(struct xt_table *table, + struct xt_table_info *bootstrap, + struct xt_table_info *newinfo) +{ + int ret; + struct xt_table_info *private; + + ret = down_interruptible(&xt[table->af].mutex); + if (ret != 0) + return ret; + + /* Don't autoload: we'd eat our tail... */ + if (list_named_find(&xt[table->af].tables, table->name)) { + ret = -EEXIST; + goto unlock; + } + + /* Simplifies replace_table code. */ + table->private = bootstrap; + if (!xt_replace_table(table, 0, newinfo, &ret)) + goto unlock; + + private = table->private; + duprintf("table->private->number = %u\n", private->number); + + /* save number of initial entries */ + private->initial_entries = private->number; + + rwlock_init(&table->lock); + list_prepend(&xt[table->af].tables, table); + + ret = 0; + unlock: + up(&xt[table->af].mutex); + return ret; +} +EXPORT_SYMBOL_GPL(xt_register_table); + +void *xt_unregister_table(struct xt_table *table) +{ + struct xt_table_info *private; + + down(&xt[table->af].mutex); + private = table->private; + LIST_DELETE(&xt[table->af].tables, table); + up(&xt[table->af].mutex); + + return private; +} +EXPORT_SYMBOL_GPL(xt_unregister_table); + +#ifdef CONFIG_PROC_FS +static char *xt_proto_prefix[NPROTO] = { + [AF_INET] = "ip", + [AF_INET6] = "ip6", + [NF_ARP] = "arp", +}; + +static struct list_head *xt_get_idx(struct list_head *list, struct seq_file *seq, loff_t pos) +{ + struct list_head *head = list->next; + + if (!head || list_empty(list)) + return NULL; + + while (pos && (head = head->next)) { + if (head == list) + return NULL; + pos--; + } + return pos ? NULL : head; +} + +static struct list_head *type2list(u_int16_t af, u_int16_t type) +{ + struct list_head *list; + + switch (type) { + case TARGET: + list = &xt[af].target; + break; + case MATCH: + list = &xt[af].match; + break; + case TABLE: + list = &xt[af].tables; + break; + default: + list = NULL; + break; + } + + return list; +} + +static void *xt_tgt_seq_start(struct seq_file *seq, loff_t *pos) +{ + struct proc_dir_entry *pde = (struct proc_dir_entry *) seq->private; + u_int16_t af = (unsigned long)pde->data & 0xffff; + u_int16_t type = (unsigned long)pde->data >> 16; + struct list_head *list; + + if (af >= NPROTO) + return NULL; + + list = type2list(af, type); + if (!list) + return NULL; + + if (down_interruptible(&xt[af].mutex) != 0) + return NULL; + + return xt_get_idx(list, seq, *pos); +} + +static void *xt_tgt_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct proc_dir_entry *pde = seq->private; + u_int16_t af = (unsigned long)pde->data & 0xffff; + u_int16_t type = (unsigned long)pde->data >> 16; + struct list_head *list; + + if (af >= NPROTO) + return NULL; + + list = type2list(af, type); + if (!list) + return NULL; + + (*pos)++; + return xt_get_idx(list, seq, *pos); +} + +static void xt_tgt_seq_stop(struct seq_file *seq, void *v) +{ + struct proc_dir_entry *pde = seq->private; + u_int16_t af = (unsigned long)pde->data & 0xffff; + + up(&xt[af].mutex); +} + +static int xt_name_seq_show(struct seq_file *seq, void *v) +{ + char *name = (char *)v + sizeof(struct list_head); + + if (strlen(name)) + return seq_printf(seq, "%s\n", name); + else + return 0; +} + +static struct seq_operations xt_tgt_seq_ops = { + .start = xt_tgt_seq_start, + .next = xt_tgt_seq_next, + .stop = xt_tgt_seq_stop, + .show = xt_name_seq_show, +}; + +static int xt_tgt_open(struct inode *inode, struct file *file) +{ + int ret; + + ret = seq_open(file, &xt_tgt_seq_ops); + if (!ret) { + struct seq_file *seq = file->private_data; + struct proc_dir_entry *pde = PDE(inode); + + seq->private = pde; + } + + return ret; +} + +static struct file_operations xt_file_ops = { + .owner = THIS_MODULE, + .open = xt_tgt_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +#define FORMAT_TABLES "_tables_names" +#define FORMAT_MATCHES "_tables_matches" +#define FORMAT_TARGETS "_tables_targets" + +#endif /* CONFIG_PROC_FS */ + +int xt_proto_init(int af) +{ +#ifdef CONFIG_PROC_FS + char buf[XT_FUNCTION_MAXNAMELEN]; + struct proc_dir_entry *proc; +#endif + + if (af >= NPROTO) + return -EINVAL; + + +#ifdef CONFIG_PROC_FS + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_TABLES, sizeof(buf)); + proc = proc_net_fops_create(buf, 0440, &xt_file_ops); + if (!proc) + goto out; + proc->data = (void *) ((unsigned long) af | (TABLE << 16)); + + + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_MATCHES, sizeof(buf)); + proc = proc_net_fops_create(buf, 0440, &xt_file_ops); + if (!proc) + goto out_remove_tables; + proc->data = (void *) ((unsigned long) af | (MATCH << 16)); + + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_TARGETS, sizeof(buf)); + proc = proc_net_fops_create(buf, 0440, &xt_file_ops); + if (!proc) + goto out_remove_matches; + proc->data = (void *) ((unsigned long) af | (TARGET << 16)); +#endif + + return 0; + +#ifdef CONFIG_PROC_FS +out_remove_matches: + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_MATCHES, sizeof(buf)); + proc_net_remove(buf); + +out_remove_tables: + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_TABLES, sizeof(buf)); + proc_net_remove(buf); +out: + return -1; +#endif +} +EXPORT_SYMBOL_GPL(xt_proto_init); + +void xt_proto_fini(int af) +{ +#ifdef CONFIG_PROC_FS + char buf[XT_FUNCTION_MAXNAMELEN]; + + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_TABLES, sizeof(buf)); + proc_net_remove(buf); + + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_TARGETS, sizeof(buf)); + proc_net_remove(buf); + + strlcpy(buf, xt_proto_prefix[af], sizeof(buf)); + strlcat(buf, FORMAT_MATCHES, sizeof(buf)); + proc_net_remove(buf); +#endif /*CONFIG_PROC_FS*/ +} +EXPORT_SYMBOL_GPL(xt_proto_fini); + + +static int __init xt_init(void) +{ + int i; + + xt = kmalloc(sizeof(struct xt_af) * NPROTO, GFP_KERNEL); + if (!xt) + return -ENOMEM; + + for (i = 0; i < NPROTO; i++) { + init_MUTEX(&xt[i].mutex); + INIT_LIST_HEAD(&xt[i].target); + INIT_LIST_HEAD(&xt[i].match); + INIT_LIST_HEAD(&xt[i].tables); + } + return 0; +} + +static void __exit xt_fini(void) +{ + kfree(xt); +} + +module_init(xt_init); +module_exit(xt_fini); + diff --git a/net/ipv4/netfilter/ipt_CLASSIFY.c b/net/netfilter/xt_CLASSIFY.c index dab78d8bd49..78ee266a12e 100644 --- a/net/ipv4/netfilter/ipt_CLASSIFY.c +++ b/net/netfilter/xt_CLASSIFY.c @@ -15,12 +15,13 @@ #include <linux/ip.h> #include <net/checksum.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_CLASSIFY.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_CLASSIFY.h> MODULE_AUTHOR("Patrick McHardy <kaber@trash.net>"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("iptables qdisc classification target module"); +MODULE_ALIAS("ipt_CLASSIFY"); static unsigned int target(struct sk_buff **pskb, @@ -30,25 +31,25 @@ target(struct sk_buff **pskb, const void *targinfo, void *userinfo) { - const struct ipt_classify_target_info *clinfo = targinfo; + const struct xt_classify_target_info *clinfo = targinfo; - if((*pskb)->priority != clinfo->priority) + if ((*pskb)->priority != clinfo->priority) (*pskb)->priority = clinfo->priority; - return IPT_CONTINUE; + return XT_CONTINUE; } static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *e, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { - if (targinfosize != IPT_ALIGN(sizeof(struct ipt_classify_target_info))){ + if (targinfosize != XT_ALIGN(sizeof(struct xt_classify_target_info))){ printk(KERN_ERR "CLASSIFY: invalid size (%u != %Zu).\n", targinfosize, - IPT_ALIGN(sizeof(struct ipt_classify_target_info))); + XT_ALIGN(sizeof(struct xt_classify_target_info))); return 0; } @@ -69,21 +70,39 @@ checkentry(const char *tablename, return 1; } -static struct ipt_target ipt_classify_reg = { +static struct xt_target classify_reg = { + .name = "CLASSIFY", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE, +}; +static struct xt_target classify6_reg = { .name = "CLASSIFY", .target = target, .checkentry = checkentry, .me = THIS_MODULE, }; + static int __init init(void) { - return ipt_register_target(&ipt_classify_reg); + int ret; + + ret = xt_register_target(AF_INET, &classify_reg); + if (ret) + return ret; + + ret = xt_register_target(AF_INET6, &classify6_reg); + if (ret) + xt_unregister_target(AF_INET, &classify_reg); + + return ret; } static void __exit fini(void) { - ipt_unregister_target(&ipt_classify_reg); + xt_unregister_target(AF_INET, &classify_reg); + xt_unregister_target(AF_INET6, &classify6_reg); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_CONNMARK.c b/net/netfilter/xt_CONNMARK.c index 8acac5a40a9..22506e376be 100644 --- a/net/ipv4/netfilter/ipt_CONNMARK.c +++ b/net/netfilter/xt_CONNMARK.c @@ -26,9 +26,10 @@ MODULE_AUTHOR("Henrik Nordstrom <hno@marasytems.com>"); MODULE_DESCRIPTION("IP tables CONNMARK matching module"); MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_CONNMARK"); -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_CONNMARK.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_CONNMARK.h> #include <net/netfilter/nf_conntrack_compat.h> static unsigned int @@ -39,7 +40,7 @@ target(struct sk_buff **pskb, const void *targinfo, void *userinfo) { - const struct ipt_connmark_target_info *markinfo = targinfo; + const struct xt_connmark_target_info *markinfo = targinfo; u_int32_t diff; u_int32_t nfmark; u_int32_t newmark; @@ -48,17 +49,17 @@ target(struct sk_buff **pskb, if (ctmark) { switch(markinfo->mode) { - case IPT_CONNMARK_SET: + case XT_CONNMARK_SET: newmark = (*ctmark & ~markinfo->mask) | markinfo->mark; if (newmark != *ctmark) *ctmark = newmark; break; - case IPT_CONNMARK_SAVE: + case XT_CONNMARK_SAVE: newmark = (*ctmark & ~markinfo->mask) | ((*pskb)->nfmark & markinfo->mask); if (*ctmark != newmark) *ctmark = newmark; break; - case IPT_CONNMARK_RESTORE: + case XT_CONNMARK_RESTORE: nfmark = (*pskb)->nfmark; diff = (*ctmark ^ nfmark) & markinfo->mask; if (diff != 0) @@ -67,25 +68,25 @@ target(struct sk_buff **pskb, } } - return IPT_CONTINUE; + return XT_CONTINUE; } static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { - struct ipt_connmark_target_info *matchinfo = targinfo; - if (targinfosize != IPT_ALIGN(sizeof(struct ipt_connmark_target_info))) { + struct xt_connmark_target_info *matchinfo = targinfo; + if (targinfosize != XT_ALIGN(sizeof(struct xt_connmark_target_info))) { printk(KERN_WARNING "CONNMARK: targinfosize %u != %Zu\n", targinfosize, - IPT_ALIGN(sizeof(struct ipt_connmark_target_info))); + XT_ALIGN(sizeof(struct xt_connmark_target_info))); return 0; } - if (matchinfo->mode == IPT_CONNMARK_RESTORE) { + if (matchinfo->mode == XT_CONNMARK_RESTORE) { if (strcmp(tablename, "mangle") != 0) { printk(KERN_WARNING "CONNMARK: restore can only be called from \"mangle\" table, not \"%s\"\n", tablename); return 0; @@ -100,7 +101,13 @@ checkentry(const char *tablename, return 1; } -static struct ipt_target ipt_connmark_reg = { +static struct xt_target connmark_reg = { + .name = "CONNMARK", + .target = &target, + .checkentry = &checkentry, + .me = THIS_MODULE +}; +static struct xt_target connmark6_reg = { .name = "CONNMARK", .target = &target, .checkentry = &checkentry, @@ -109,13 +116,25 @@ static struct ipt_target ipt_connmark_reg = { static int __init init(void) { - need_ip_conntrack(); - return ipt_register_target(&ipt_connmark_reg); + int ret; + + need_conntrack(); + + ret = xt_register_target(AF_INET, &connmark_reg); + if (ret) + return ret; + + ret = xt_register_target(AF_INET6, &connmark6_reg); + if (ret) + xt_unregister_target(AF_INET, &connmark_reg); + + return ret; } static void __exit fini(void) { - ipt_unregister_target(&ipt_connmark_reg); + xt_unregister_target(AF_INET, &connmark_reg); + xt_unregister_target(AF_INET6, &connmark6_reg); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_MARK.c b/net/netfilter/xt_MARK.c index 52b4f2c296b..0c11ee9550f 100644 --- a/net/ipv4/netfilter/ipt_MARK.c +++ b/net/netfilter/xt_MARK.c @@ -12,12 +12,14 @@ #include <linux/ip.h> #include <net/checksum.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_MARK.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_MARK.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>"); -MODULE_DESCRIPTION("iptables MARK modification module"); +MODULE_DESCRIPTION("ip[6]tables MARK modification module"); +MODULE_ALIAS("ipt_MARK"); +MODULE_ALIAS("ip6t_MARK"); static unsigned int target_v0(struct sk_buff **pskb, @@ -27,12 +29,12 @@ target_v0(struct sk_buff **pskb, const void *targinfo, void *userinfo) { - const struct ipt_mark_target_info *markinfo = targinfo; + const struct xt_mark_target_info *markinfo = targinfo; if((*pskb)->nfmark != markinfo->mark) (*pskb)->nfmark = markinfo->mark; - return IPT_CONTINUE; + return XT_CONTINUE; } static unsigned int @@ -43,19 +45,19 @@ target_v1(struct sk_buff **pskb, const void *targinfo, void *userinfo) { - const struct ipt_mark_target_info_v1 *markinfo = targinfo; + const struct xt_mark_target_info_v1 *markinfo = targinfo; int mark = 0; switch (markinfo->mode) { - case IPT_MARK_SET: + case XT_MARK_SET: mark = markinfo->mark; break; - case IPT_MARK_AND: + case XT_MARK_AND: mark = (*pskb)->nfmark & markinfo->mark; break; - case IPT_MARK_OR: + case XT_MARK_OR: mark = (*pskb)->nfmark | markinfo->mark; break; } @@ -63,23 +65,23 @@ target_v1(struct sk_buff **pskb, if((*pskb)->nfmark != mark) (*pskb)->nfmark = mark; - return IPT_CONTINUE; + return XT_CONTINUE; } static int checkentry_v0(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { - struct ipt_mark_target_info *markinfo = targinfo; + struct xt_mark_target_info *markinfo = targinfo; - if (targinfosize != IPT_ALIGN(sizeof(struct ipt_mark_target_info))) { + if (targinfosize != XT_ALIGN(sizeof(struct xt_mark_target_info))) { printk(KERN_WARNING "MARK: targinfosize %u != %Zu\n", targinfosize, - IPT_ALIGN(sizeof(struct ipt_mark_target_info))); + XT_ALIGN(sizeof(struct xt_mark_target_info))); return 0; } @@ -98,17 +100,17 @@ checkentry_v0(const char *tablename, static int checkentry_v1(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) { - struct ipt_mark_target_info_v1 *markinfo = targinfo; + struct xt_mark_target_info_v1 *markinfo = targinfo; - if (targinfosize != IPT_ALIGN(sizeof(struct ipt_mark_target_info_v1))){ + if (targinfosize != XT_ALIGN(sizeof(struct xt_mark_target_info_v1))){ printk(KERN_WARNING "MARK: targinfosize %u != %Zu\n", targinfosize, - IPT_ALIGN(sizeof(struct ipt_mark_target_info_v1))); + XT_ALIGN(sizeof(struct xt_mark_target_info_v1))); return 0; } @@ -117,9 +119,9 @@ checkentry_v1(const char *tablename, return 0; } - if (markinfo->mode != IPT_MARK_SET - && markinfo->mode != IPT_MARK_AND - && markinfo->mode != IPT_MARK_OR) { + if (markinfo->mode != XT_MARK_SET + && markinfo->mode != XT_MARK_AND + && markinfo->mode != XT_MARK_OR) { printk(KERN_WARNING "MARK: unknown mode %u\n", markinfo->mode); return 0; @@ -133,7 +135,7 @@ checkentry_v1(const char *tablename, return 1; } -static struct ipt_target ipt_mark_reg_v0 = { +static struct xt_target ipt_mark_reg_v0 = { .name = "MARK", .target = target_v0, .checkentry = checkentry_v0, @@ -141,7 +143,7 @@ static struct ipt_target ipt_mark_reg_v0 = { .revision = 0, }; -static struct ipt_target ipt_mark_reg_v1 = { +static struct xt_target ipt_mark_reg_v1 = { .name = "MARK", .target = target_v1, .checkentry = checkentry_v1, @@ -149,23 +151,40 @@ static struct ipt_target ipt_mark_reg_v1 = { .revision = 1, }; +static struct xt_target ip6t_mark_reg_v0 = { + .name = "MARK", + .target = target_v0, + .checkentry = checkentry_v0, + .me = THIS_MODULE, + .revision = 0, +}; + static int __init init(void) { int err; - err = ipt_register_target(&ipt_mark_reg_v0); - if (!err) { - err = ipt_register_target(&ipt_mark_reg_v1); - if (err) - ipt_unregister_target(&ipt_mark_reg_v0); + err = xt_register_target(AF_INET, &ipt_mark_reg_v0); + if (err) + return err; + + err = xt_register_target(AF_INET, &ipt_mark_reg_v1); + if (err) + xt_unregister_target(AF_INET, &ipt_mark_reg_v0); + + err = xt_register_target(AF_INET6, &ip6t_mark_reg_v0); + if (err) { + xt_unregister_target(AF_INET, &ipt_mark_reg_v0); + xt_unregister_target(AF_INET, &ipt_mark_reg_v1); } + return err; } static void __exit fini(void) { - ipt_unregister_target(&ipt_mark_reg_v0); - ipt_unregister_target(&ipt_mark_reg_v1); + xt_unregister_target(AF_INET, &ipt_mark_reg_v0); + xt_unregister_target(AF_INET, &ipt_mark_reg_v1); + xt_unregister_target(AF_INET6, &ip6t_mark_reg_v0); } module_init(init); diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c new file mode 100644 index 00000000000..8b76b6f8d1e --- /dev/null +++ b/net/netfilter/xt_NFQUEUE.c @@ -0,0 +1,107 @@ +/* iptables module for using new netfilter netlink queue + * + * (C) 2005 by Harald Welte <laforge@netfilter.org> + * + * 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 <linux/module.h> +#include <linux/skbuff.h> + +#include <linux/netfilter.h> +#include <linux/netfilter_arp.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_NFQUEUE.h> + +MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); +MODULE_DESCRIPTION("[ip,ip6,arp]_tables NFQUEUE target"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_NFQUEUE"); +MODULE_ALIAS("ip6t_NFQUEUE"); +MODULE_ALIAS("arpt_NFQUEUE"); + +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) +{ + const struct xt_NFQ_info *tinfo = targinfo; + + return NF_QUEUE_NR(tinfo->queuenum); +} + +static int +checkentry(const char *tablename, + const void *entry, + void *targinfo, + unsigned int targinfosize, + unsigned int hook_mask) +{ + if (targinfosize != XT_ALIGN(sizeof(struct xt_NFQ_info))) { + printk(KERN_WARNING "NFQUEUE: targinfosize %u != %Zu\n", + targinfosize, + XT_ALIGN(sizeof(struct xt_NFQ_info))); + return 0; + } + + return 1; +} + +static struct xt_target ipt_NFQ_reg = { + .name = "NFQUEUE", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE, +}; + +static struct xt_target ip6t_NFQ_reg = { + .name = "NFQUEUE", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE, +}; + +static struct xt_target arpt_NFQ_reg = { + .name = "NFQUEUE", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE, +}; + +static int __init init(void) +{ + int ret; + ret = xt_register_target(AF_INET, &ipt_NFQ_reg); + if (ret) + return ret; + ret = xt_register_target(AF_INET6, &ip6t_NFQ_reg); + if (ret) + goto out_ip; + ret = xt_register_target(NF_ARP, &arpt_NFQ_reg); + if (ret) + goto out_ip6; + + return ret; +out_ip6: + xt_unregister_target(AF_INET6, &ip6t_NFQ_reg); +out_ip: + xt_unregister_target(AF_INET, &ipt_NFQ_reg); + + return ret; +} + +static void __exit fini(void) +{ + xt_unregister_target(NF_ARP, &arpt_NFQ_reg); + xt_unregister_target(AF_INET6, &ip6t_NFQ_reg); + xt_unregister_target(AF_INET, &ipt_NFQ_reg); +} + +module_init(init); +module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_NOTRACK.c b/net/netfilter/xt_NOTRACK.c index e3c69d072c6..24d477afa93 100644 --- a/net/ipv4/netfilter/ipt_NOTRACK.c +++ b/net/netfilter/xt_NOTRACK.c @@ -4,9 +4,12 @@ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter/x_tables.h> #include <net/netfilter/nf_conntrack_compat.h> +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_NOTRACK"); + static unsigned int target(struct sk_buff **pskb, const struct net_device *in, @@ -17,7 +20,7 @@ target(struct sk_buff **pskb, { /* Previously seen (loopback)? Ignore. */ if ((*pskb)->nfct != NULL) - return IPT_CONTINUE; + return XT_CONTINUE; /* Attach fake conntrack entry. If there is a real ct entry correspondig to this packet, @@ -27,12 +30,12 @@ target(struct sk_buff **pskb, (*pskb)->nfctinfo = IP_CT_NEW; nf_conntrack_get((*pskb)->nfct); - return IPT_CONTINUE; + return XT_CONTINUE; } static int checkentry(const char *tablename, - const struct ipt_entry *e, + const void *entry, void *targinfo, unsigned int targinfosize, unsigned int hook_mask) @@ -51,26 +54,39 @@ checkentry(const char *tablename, return 1; } -static struct ipt_target ipt_notrack_reg = { +static struct xt_target notrack_reg = { .name = "NOTRACK", .target = target, .checkentry = checkentry, - .me = THIS_MODULE + .me = THIS_MODULE, +}; +static struct xt_target notrack6_reg = { + .name = "NOTRACK", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE, }; static int __init init(void) { - if (ipt_register_target(&ipt_notrack_reg)) - return -EINVAL; + int ret; + + ret = xt_register_target(AF_INET, ¬rack_reg); + if (ret) + return ret; - return 0; + ret = xt_register_target(AF_INET6, ¬rack6_reg); + if (ret) + xt_unregister_target(AF_INET, ¬rack_reg); + + return ret; } static void __exit fini(void) { - ipt_unregister_target(&ipt_notrack_reg); + xt_unregister_target(AF_INET6, ¬rack6_reg); + xt_unregister_target(AF_INET, ¬rack_reg); } module_init(init); module_exit(fini); -MODULE_LICENSE("GPL"); diff --git a/net/ipv4/netfilter/ipt_comment.c b/net/netfilter/xt_comment.c index 6b76a1ea524..4ba6fd65c6e 100644 --- a/net/ipv4/netfilter/ipt_comment.c +++ b/net/netfilter/xt_comment.c @@ -6,12 +6,14 @@ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_comment.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_comment.h> MODULE_AUTHOR("Brad Fisher <brad@info-link.net>"); MODULE_DESCRIPTION("iptables comment match module"); MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_comment"); +MODULE_ALIAS("ip6t_comment"); static int match(const struct sk_buff *skb, @@ -19,6 +21,7 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protooff, int *hotdrop) { /* We always match */ @@ -27,18 +30,25 @@ match(const struct sk_buff *skb, static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { /* Check the size */ - if (matchsize != IPT_ALIGN(sizeof(struct ipt_comment_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_comment_info))) return 0; return 1; } -static struct ipt_match comment_match = { +static struct xt_match comment_match = { + .name = "comment", + .match = match, + .checkentry = checkentry, + .me = THIS_MODULE +}; + +static struct xt_match comment6_match = { .name = "comment", .match = match, .checkentry = checkentry, @@ -47,12 +57,23 @@ static struct ipt_match comment_match = { static int __init init(void) { - return ipt_register_match(&comment_match); + int ret; + + ret = xt_register_match(AF_INET, &comment_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &comment6_match); + if (ret) + xt_unregister_match(AF_INET, &comment_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&comment_match); + xt_unregister_match(AF_INET, &comment_match); + xt_unregister_match(AF_INET6, &comment6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_connbytes.c b/net/netfilter/xt_connbytes.c index d68a048b717..150d2a4b0f7 100644 --- a/net/ipv4/netfilter/ipt_connbytes.c +++ b/net/netfilter/xt_connbytes.c @@ -6,13 +6,15 @@ * - add functionality to match number of packets * - add functionality to match average packet size * - add support to match directions seperately + * 2005-10-16 Harald Welte <laforge@netfilter.org> + * - Port to x_tables * */ #include <linux/module.h> #include <linux/skbuff.h> #include <net/netfilter/nf_conntrack_compat.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_connbytes.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_connbytes.h> #include <asm/div64.h> #include <asm/bitops.h> @@ -20,6 +22,7 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); MODULE_DESCRIPTION("iptables match for matching number of pkts/bytes per connection"); +MODULE_ALIAS("ipt_connbytes"); /* 64bit divisor, dividend and result. dynamic precision */ static u_int64_t div64_64(u_int64_t dividend, u_int64_t divisor) @@ -43,9 +46,10 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_connbytes_info *sinfo = matchinfo; + const struct xt_connbytes_info *sinfo = matchinfo; u_int64_t what = 0; /* initialize to make gcc happy */ const struct ip_conntrack_counter *counters; @@ -53,45 +57,45 @@ match(const struct sk_buff *skb, return 0; /* no match */ switch (sinfo->what) { - case IPT_CONNBYTES_PKTS: + case XT_CONNBYTES_PKTS: switch (sinfo->direction) { - case IPT_CONNBYTES_DIR_ORIGINAL: + case XT_CONNBYTES_DIR_ORIGINAL: what = counters[IP_CT_DIR_ORIGINAL].packets; break; - case IPT_CONNBYTES_DIR_REPLY: + case XT_CONNBYTES_DIR_REPLY: what = counters[IP_CT_DIR_REPLY].packets; break; - case IPT_CONNBYTES_DIR_BOTH: + case XT_CONNBYTES_DIR_BOTH: what = counters[IP_CT_DIR_ORIGINAL].packets; what += counters[IP_CT_DIR_REPLY].packets; break; } break; - case IPT_CONNBYTES_BYTES: + case XT_CONNBYTES_BYTES: switch (sinfo->direction) { - case IPT_CONNBYTES_DIR_ORIGINAL: + case XT_CONNBYTES_DIR_ORIGINAL: what = counters[IP_CT_DIR_ORIGINAL].bytes; break; - case IPT_CONNBYTES_DIR_REPLY: + case XT_CONNBYTES_DIR_REPLY: what = counters[IP_CT_DIR_REPLY].bytes; break; - case IPT_CONNBYTES_DIR_BOTH: + case XT_CONNBYTES_DIR_BOTH: what = counters[IP_CT_DIR_ORIGINAL].bytes; what += counters[IP_CT_DIR_REPLY].bytes; break; } break; - case IPT_CONNBYTES_AVGPKT: + case XT_CONNBYTES_AVGPKT: switch (sinfo->direction) { - case IPT_CONNBYTES_DIR_ORIGINAL: + case XT_CONNBYTES_DIR_ORIGINAL: what = div64_64(counters[IP_CT_DIR_ORIGINAL].bytes, counters[IP_CT_DIR_ORIGINAL].packets); break; - case IPT_CONNBYTES_DIR_REPLY: + case XT_CONNBYTES_DIR_REPLY: what = div64_64(counters[IP_CT_DIR_REPLY].bytes, counters[IP_CT_DIR_REPLY].packets); break; - case IPT_CONNBYTES_DIR_BOTH: + case XT_CONNBYTES_DIR_BOTH: { u_int64_t bytes; u_int64_t pkts; @@ -117,30 +121,36 @@ match(const struct sk_buff *skb, } static int check(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - const struct ipt_connbytes_info *sinfo = matchinfo; + const struct xt_connbytes_info *sinfo = matchinfo; - if (matchsize != IPT_ALIGN(sizeof(struct ipt_connbytes_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_connbytes_info))) return 0; - if (sinfo->what != IPT_CONNBYTES_PKTS && - sinfo->what != IPT_CONNBYTES_BYTES && - sinfo->what != IPT_CONNBYTES_AVGPKT) + if (sinfo->what != XT_CONNBYTES_PKTS && + sinfo->what != XT_CONNBYTES_BYTES && + sinfo->what != XT_CONNBYTES_AVGPKT) return 0; - if (sinfo->direction != IPT_CONNBYTES_DIR_ORIGINAL && - sinfo->direction != IPT_CONNBYTES_DIR_REPLY && - sinfo->direction != IPT_CONNBYTES_DIR_BOTH) + if (sinfo->direction != XT_CONNBYTES_DIR_ORIGINAL && + sinfo->direction != XT_CONNBYTES_DIR_REPLY && + sinfo->direction != XT_CONNBYTES_DIR_BOTH) return 0; return 1; } -static struct ipt_match state_match = { +static struct xt_match connbytes_match = { + .name = "connbytes", + .match = &match, + .checkentry = &check, + .me = THIS_MODULE +}; +static struct xt_match connbytes6_match = { .name = "connbytes", .match = &match, .checkentry = &check, @@ -149,12 +159,21 @@ static struct ipt_match state_match = { static int __init init(void) { - return ipt_register_match(&state_match); + int ret; + ret = xt_register_match(AF_INET, &connbytes_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &connbytes6_match); + if (ret) + xt_unregister_match(AF_INET, &connbytes_match); + return ret; } static void __exit fini(void) { - ipt_unregister_match(&state_match); + xt_unregister_match(AF_INET, &connbytes_match); + xt_unregister_match(AF_INET6, &connbytes6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_connmark.c b/net/netfilter/xt_connmark.c index 5306ef293b9..d06e925032d 100644 --- a/net/ipv4/netfilter/ipt_connmark.c +++ b/net/netfilter/xt_connmark.c @@ -25,9 +25,10 @@ MODULE_AUTHOR("Henrik Nordstrom <hno@marasytems.com>"); MODULE_DESCRIPTION("IP tables connmark match module"); MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_connmark"); -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_connmark.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_connmark.h> #include <net/netfilter/nf_conntrack_compat.h> static int @@ -36,9 +37,10 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_connmark_info *info = matchinfo; + const struct xt_connmark_info *info = matchinfo; u_int32_t ctinfo; const u_int32_t *ctmark = nf_ct_get_mark(skb, &ctinfo); if (!ctmark) @@ -49,14 +51,14 @@ match(const struct sk_buff *skb, static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - struct ipt_connmark_info *cm = - (struct ipt_connmark_info *)matchinfo; - if (matchsize != IPT_ALIGN(sizeof(struct ipt_connmark_info))) + struct xt_connmark_info *cm = + (struct xt_connmark_info *)matchinfo; + if (matchsize != XT_ALIGN(sizeof(struct xt_connmark_info))) return 0; if (cm->mark > 0xffffffff || cm->mask > 0xffffffff) { @@ -67,21 +69,40 @@ checkentry(const char *tablename, return 1; } -static struct ipt_match connmark_match = { +static struct xt_match connmark_match = { + .name = "connmark", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE +}; +static struct xt_match connmark6_match = { .name = "connmark", .match = &match, .checkentry = &checkentry, .me = THIS_MODULE }; + static int __init init(void) { - return ipt_register_match(&connmark_match); + int ret; + + need_conntrack(); + + ret = xt_register_match(AF_INET, &connmark_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &connmark6_match); + if (ret) + xt_unregister_match(AF_INET, &connmark_match); + return ret; } static void __exit fini(void) { - ipt_unregister_match(&connmark_match); + xt_unregister_match(AF_INET6, &connmark6_match); + xt_unregister_match(AF_INET, &connmark_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_conntrack.c b/net/netfilter/xt_conntrack.c index c8d18705469..ffdebc95eb9 100644 --- a/net/ipv4/netfilter/ipt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -18,12 +18,13 @@ #include <net/netfilter/nf_conntrack.h> #endif -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_conntrack.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_conntrack.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>"); MODULE_DESCRIPTION("iptables connection tracking match module"); +MODULE_ALIAS("ipt_conntrack"); #if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) @@ -33,9 +34,10 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_conntrack_info *sinfo = matchinfo; + const struct xt_conntrack_info *sinfo = matchinfo; struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; unsigned int statebit; @@ -45,58 +47,58 @@ match(const struct sk_buff *skb, #define FWINV(bool,invflg) ((bool) ^ !!(sinfo->invflags & invflg)) if (ct == &ip_conntrack_untracked) - statebit = IPT_CONNTRACK_STATE_UNTRACKED; + statebit = XT_CONNTRACK_STATE_UNTRACKED; else if (ct) - statebit = IPT_CONNTRACK_STATE_BIT(ctinfo); + statebit = XT_CONNTRACK_STATE_BIT(ctinfo); else - statebit = IPT_CONNTRACK_STATE_INVALID; + statebit = XT_CONNTRACK_STATE_INVALID; - if(sinfo->flags & IPT_CONNTRACK_STATE) { + if(sinfo->flags & XT_CONNTRACK_STATE) { if (ct) { if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip != ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip) - statebit |= IPT_CONNTRACK_STATE_SNAT; + statebit |= XT_CONNTRACK_STATE_SNAT; if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip != ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip) - statebit |= IPT_CONNTRACK_STATE_DNAT; + statebit |= XT_CONNTRACK_STATE_DNAT; } - if (FWINV((statebit & sinfo->statemask) == 0, IPT_CONNTRACK_STATE)) + if (FWINV((statebit & sinfo->statemask) == 0, XT_CONNTRACK_STATE)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_PROTO) { - if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, IPT_CONNTRACK_PROTO)) + if(sinfo->flags & XT_CONNTRACK_PROTO) { + if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, XT_CONNTRACK_PROTO)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_ORIGSRC) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, IPT_CONNTRACK_ORIGSRC)) + if(sinfo->flags & XT_CONNTRACK_ORIGSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, XT_CONNTRACK_ORIGSRC)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_ORIGDST) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, IPT_CONNTRACK_ORIGDST)) + if(sinfo->flags & XT_CONNTRACK_ORIGDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, XT_CONNTRACK_ORIGDST)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_REPLSRC) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, IPT_CONNTRACK_REPLSRC)) + if(sinfo->flags & XT_CONNTRACK_REPLSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, XT_CONNTRACK_REPLSRC)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_REPLDST) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, IPT_CONNTRACK_REPLDST)) + if(sinfo->flags & XT_CONNTRACK_REPLDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, XT_CONNTRACK_REPLDST)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_STATUS) { - if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, IPT_CONNTRACK_STATUS)) + if(sinfo->flags & XT_CONNTRACK_STATUS) { + if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, XT_CONNTRACK_STATUS)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_EXPIRES) { + if(sinfo->flags & XT_CONNTRACK_EXPIRES) { unsigned long expires; if(!ct) @@ -104,7 +106,7 @@ match(const struct sk_buff *skb, expires = timer_pending(&ct->timeout) ? (ct->timeout.expires - jiffies)/HZ : 0; - if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), IPT_CONNTRACK_EXPIRES)) + if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), XT_CONNTRACK_EXPIRES)) return 0; } @@ -118,9 +120,10 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_conntrack_info *sinfo = matchinfo; + const struct xt_conntrack_info *sinfo = matchinfo; struct nf_conn *ct; enum ip_conntrack_info ctinfo; unsigned int statebit; @@ -130,58 +133,58 @@ match(const struct sk_buff *skb, #define FWINV(bool,invflg) ((bool) ^ !!(sinfo->invflags & invflg)) if (ct == &nf_conntrack_untracked) - statebit = IPT_CONNTRACK_STATE_UNTRACKED; + statebit = XT_CONNTRACK_STATE_UNTRACKED; else if (ct) - statebit = IPT_CONNTRACK_STATE_BIT(ctinfo); + statebit = XT_CONNTRACK_STATE_BIT(ctinfo); else - statebit = IPT_CONNTRACK_STATE_INVALID; + statebit = XT_CONNTRACK_STATE_INVALID; - if(sinfo->flags & IPT_CONNTRACK_STATE) { + if(sinfo->flags & XT_CONNTRACK_STATE) { if (ct) { if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip != ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip) - statebit |= IPT_CONNTRACK_STATE_SNAT; + statebit |= XT_CONNTRACK_STATE_SNAT; if(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip != ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip) - statebit |= IPT_CONNTRACK_STATE_DNAT; + statebit |= XT_CONNTRACK_STATE_DNAT; } - if (FWINV((statebit & sinfo->statemask) == 0, IPT_CONNTRACK_STATE)) + if (FWINV((statebit & sinfo->statemask) == 0, XT_CONNTRACK_STATE)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_PROTO) { - if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, IPT_CONNTRACK_PROTO)) + if(sinfo->flags & XT_CONNTRACK_PROTO) { + if (!ct || FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, XT_CONNTRACK_PROTO)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_ORIGSRC) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, IPT_CONNTRACK_ORIGSRC)) + if(sinfo->flags & XT_CONNTRACK_ORIGSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].src.ip, XT_CONNTRACK_ORIGSRC)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_ORIGDST) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, IPT_CONNTRACK_ORIGDST)) + if(sinfo->flags & XT_CONNTRACK_ORIGDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_ORIGINAL].s_addr) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.ip, XT_CONNTRACK_ORIGDST)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_REPLSRC) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, IPT_CONNTRACK_REPLSRC)) + if(sinfo->flags & XT_CONNTRACK_REPLSRC) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip&sinfo->sipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].src.ip, XT_CONNTRACK_REPLSRC)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_REPLDST) { - if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, IPT_CONNTRACK_REPLDST)) + if(sinfo->flags & XT_CONNTRACK_REPLDST) { + if (!ct || FWINV((ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip&sinfo->dipmsk[IP_CT_DIR_REPLY].s_addr) != sinfo->tuple[IP_CT_DIR_REPLY].dst.ip, XT_CONNTRACK_REPLDST)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_STATUS) { - if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, IPT_CONNTRACK_STATUS)) + if(sinfo->flags & XT_CONNTRACK_STATUS) { + if (!ct || FWINV((ct->status & sinfo->statusmask) == 0, XT_CONNTRACK_STATUS)) return 0; } - if(sinfo->flags & IPT_CONNTRACK_EXPIRES) { + if(sinfo->flags & XT_CONNTRACK_EXPIRES) { unsigned long expires; if(!ct) @@ -189,7 +192,7 @@ match(const struct sk_buff *skb, expires = timer_pending(&ct->timeout) ? (ct->timeout.expires - jiffies)/HZ : 0; - if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), IPT_CONNTRACK_EXPIRES)) + if (FWINV(!(expires >= sinfo->expires_min && expires <= sinfo->expires_max), XT_CONNTRACK_EXPIRES)) return 0; } @@ -199,18 +202,18 @@ match(const struct sk_buff *skb, #endif /* CONFIG_NF_IP_CONNTRACK */ static int check(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - if (matchsize != IPT_ALIGN(sizeof(struct ipt_conntrack_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_conntrack_info))) return 0; return 1; } -static struct ipt_match conntrack_match = { +static struct xt_match conntrack_match = { .name = "conntrack", .match = &match, .checkentry = &check, @@ -219,13 +222,16 @@ static struct ipt_match conntrack_match = { static int __init init(void) { - need_ip_conntrack(); - return ipt_register_match(&conntrack_match); + int ret; + need_conntrack(); + ret = xt_register_match(AF_INET, &conntrack_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&conntrack_match); + xt_unregister_match(AF_INET, &conntrack_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_dccp.c b/net/netfilter/xt_dccp.c index ad3278bba6c..779f42fc952 100644 --- a/net/ipv4/netfilter/ipt_dccp.c +++ b/net/netfilter/xt_dccp.c @@ -14,8 +14,16 @@ #include <net/ip.h> #include <linux/dccp.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_dccp.h> + #include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_dccp.h> +#include <linux/netfilter_ipv6/ip6_tables.h> + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); +MODULE_DESCRIPTION("Match for DCCP protocol packets"); +MODULE_ALIAS("ipt_dccp"); #define DCCHECK(cond, option, flag, invflag) (!((flag) & (option)) \ || (!!((invflag) & (option)) ^ (cond))) @@ -26,6 +34,7 @@ static DEFINE_SPINLOCK(dccp_buflock); static inline int dccp_find_option(u_int8_t option, const struct sk_buff *skb, + unsigned int protoff, const struct dccp_hdr *dh, int *hotdrop) { @@ -44,9 +53,7 @@ dccp_find_option(u_int8_t option, return 0; spin_lock_bh(&dccp_buflock); - op = skb_header_pointer(skb, - skb->nh.iph->ihl*4 + optoff, - optlen, dccp_optbuf); + op = skb_header_pointer(skb, protoff + optoff, optlen, dccp_optbuf); if (op == NULL) { /* If we don't have the whole header, drop packet. */ spin_unlock_bh(&dccp_buflock); @@ -78,10 +85,10 @@ match_types(const struct dccp_hdr *dh, u_int16_t typemask) } static inline int -match_option(u_int8_t option, const struct sk_buff *skb, +match_option(u_int8_t option, const struct sk_buff *skb, unsigned int protoff, const struct dccp_hdr *dh, int *hotdrop) { - return dccp_find_option(option, skb, dh, hotdrop); + return dccp_find_option(option, skb, protoff, dh, hotdrop); } static int @@ -90,16 +97,17 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_dccp_info *info = - (const struct ipt_dccp_info *)matchinfo; + const struct xt_dccp_info *info = + (const struct xt_dccp_info *)matchinfo; struct dccp_hdr _dh, *dh; if (offset) return 0; - dh = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_dh), &_dh); + dh = skb_header_pointer(skb, protoff, sizeof(_dh), &_dh); if (dh == NULL) { *hotdrop = 1; return 0; @@ -107,42 +115,73 @@ match(const struct sk_buff *skb, return DCCHECK(((ntohs(dh->dccph_sport) >= info->spts[0]) && (ntohs(dh->dccph_sport) <= info->spts[1])), - IPT_DCCP_SRC_PORTS, info->flags, info->invflags) + XT_DCCP_SRC_PORTS, info->flags, info->invflags) && DCCHECK(((ntohs(dh->dccph_dport) >= info->dpts[0]) && (ntohs(dh->dccph_dport) <= info->dpts[1])), - IPT_DCCP_DEST_PORTS, info->flags, info->invflags) + XT_DCCP_DEST_PORTS, info->flags, info->invflags) && DCCHECK(match_types(dh, info->typemask), - IPT_DCCP_TYPE, info->flags, info->invflags) - && DCCHECK(match_option(info->option, skb, dh, hotdrop), - IPT_DCCP_OPTION, info->flags, info->invflags); + XT_DCCP_TYPE, info->flags, info->invflags) + && DCCHECK(match_option(info->option, skb, protoff, dh, + hotdrop), + XT_DCCP_OPTION, info->flags, info->invflags); } static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - const struct ipt_dccp_info *info; + const struct ipt_ip *ip = inf; + const struct xt_dccp_info *info; - info = (const struct ipt_dccp_info *)matchinfo; + info = (const struct xt_dccp_info *)matchinfo; return ip->proto == IPPROTO_DCCP - && !(ip->invflags & IPT_INV_PROTO) - && matchsize == IPT_ALIGN(sizeof(struct ipt_dccp_info)) - && !(info->flags & ~IPT_DCCP_VALID_FLAGS) - && !(info->invflags & ~IPT_DCCP_VALID_FLAGS) + && !(ip->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_dccp_info)) + && !(info->flags & ~XT_DCCP_VALID_FLAGS) + && !(info->invflags & ~XT_DCCP_VALID_FLAGS) && !(info->invflags & ~info->flags); } -static struct ipt_match dccp_match = +static int +checkentry6(const char *tablename, + const void *inf, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ip6t_ip6 *ip = inf; + const struct xt_dccp_info *info; + + info = (const struct xt_dccp_info *)matchinfo; + + return ip->proto == IPPROTO_DCCP + && !(ip->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_dccp_info)) + && !(info->flags & ~XT_DCCP_VALID_FLAGS) + && !(info->invflags & ~XT_DCCP_VALID_FLAGS) + && !(info->invflags & ~info->flags); +} + + +static struct xt_match dccp_match = { .name = "dccp", .match = &match, .checkentry = &checkentry, .me = THIS_MODULE, }; +static struct xt_match dccp6_match = +{ + .name = "dccp", + .match = &match, + .checkentry = &checkentry6, + .me = THIS_MODULE, +}; + static int __init init(void) { @@ -154,23 +193,29 @@ static int __init init(void) dccp_optbuf = kmalloc(256 * 4, GFP_KERNEL); if (!dccp_optbuf) return -ENOMEM; - ret = ipt_register_match(&dccp_match); + ret = xt_register_match(AF_INET, &dccp_match); if (ret) - kfree(dccp_optbuf); + goto out_kfree; + ret = xt_register_match(AF_INET6, &dccp6_match); + if (ret) + goto out_unreg; + + return ret; + +out_unreg: + xt_unregister_match(AF_INET, &dccp_match); +out_kfree: + kfree(dccp_optbuf); return ret; } static void __exit fini(void) { - ipt_unregister_match(&dccp_match); + xt_unregister_match(AF_INET6, &dccp6_match); + xt_unregister_match(AF_INET, &dccp_match); kfree(dccp_optbuf); } module_init(init); module_exit(fini); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>"); -MODULE_DESCRIPTION("Match for DCCP protocol packets"); - diff --git a/net/ipv4/netfilter/ipt_helper.c b/net/netfilter/xt_helper.c index aef649e393a..38b6715e1db 100644 --- a/net/ipv4/netfilter/ipt_helper.c +++ b/net/netfilter/xt_helper.c @@ -13,7 +13,6 @@ #include <linux/module.h> #include <linux/skbuff.h> #include <linux/netfilter.h> -#include <linux/interrupt.h> #if defined(CONFIG_IP_NF_CONNTRACK) || defined(CONFIG_IP_NF_CONNTRACK_MODULE) #include <linux/netfilter_ipv4/ip_conntrack.h> #include <linux/netfilter_ipv4/ip_conntrack_core.h> @@ -23,12 +22,14 @@ #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_helper.h> #endif -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_helper.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_helper.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Martin Josefsson <gandalf@netfilter.org>"); MODULE_DESCRIPTION("iptables helper match module"); +MODULE_ALIAS("ipt_helper"); +MODULE_ALIAS("ip6t_helper"); #if 0 #define DEBUGP printk @@ -43,27 +44,28 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_helper_info *info = matchinfo; + const struct xt_helper_info *info = matchinfo; struct ip_conntrack *ct; enum ip_conntrack_info ctinfo; int ret = info->invert; ct = ip_conntrack_get((struct sk_buff *)skb, &ctinfo); if (!ct) { - DEBUGP("ipt_helper: Eek! invalid conntrack?\n"); + DEBUGP("xt_helper: Eek! invalid conntrack?\n"); return ret; } if (!ct->master) { - DEBUGP("ipt_helper: conntrack %p has no master\n", ct); + DEBUGP("xt_helper: conntrack %p has no master\n", ct); return ret; } read_lock_bh(&ip_conntrack_lock); if (!ct->master->helper) { - DEBUGP("ipt_helper: master ct %p has no helper\n", + DEBUGP("xt_helper: master ct %p has no helper\n", exp->expectant); goto out_unlock; } @@ -89,27 +91,28 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_helper_info *info = matchinfo; + const struct xt_helper_info *info = matchinfo; struct nf_conn *ct; enum ip_conntrack_info ctinfo; int ret = info->invert; ct = nf_ct_get((struct sk_buff *)skb, &ctinfo); if (!ct) { - DEBUGP("ipt_helper: Eek! invalid conntrack?\n"); + DEBUGP("xt_helper: Eek! invalid conntrack?\n"); return ret; } if (!ct->master) { - DEBUGP("ipt_helper: conntrack %p has no master\n", ct); + DEBUGP("xt_helper: conntrack %p has no master\n", ct); return ret; } read_lock_bh(&nf_conntrack_lock); if (!ct->master->helper) { - DEBUGP("ipt_helper: master ct %p has no helper\n", + DEBUGP("xt_helper: master ct %p has no helper\n", exp->expectant); goto out_unlock; } @@ -129,23 +132,29 @@ out_unlock: #endif static int check(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - struct ipt_helper_info *info = matchinfo; + struct xt_helper_info *info = matchinfo; info->name[29] = '\0'; /* verify size */ - if (matchsize != IPT_ALIGN(sizeof(struct ipt_helper_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_helper_info))) return 0; return 1; } -static struct ipt_match helper_match = { +static struct xt_match helper_match = { + .name = "helper", + .match = &match, + .checkentry = &check, + .me = THIS_MODULE, +}; +static struct xt_match helper6_match = { .name = "helper", .match = &match, .checkentry = &check, @@ -154,13 +163,24 @@ static struct ipt_match helper_match = { static int __init init(void) { - need_ip_conntrack(); - return ipt_register_match(&helper_match); + int ret; + need_conntrack(); + + ret = xt_register_match(AF_INET, &helper_match); + if (ret < 0) + return ret; + + ret = xt_register_match(AF_INET6, &helper6_match); + if (ret < 0) + xt_unregister_match(AF_INET, &helper_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&helper_match); + xt_unregister_match(AF_INET, &helper_match); + xt_unregister_match(AF_INET6, &helper6_match); } module_init(init); diff --git a/net/netfilter/xt_length.c b/net/netfilter/xt_length.c new file mode 100644 index 00000000000..ab6c710cf88 --- /dev/null +++ b/net/netfilter/xt_length.c @@ -0,0 +1,98 @@ +/* Kernel module to match packet length. */ +/* (C) 1999-2001 James Morris <jmorros@intercode.com.au> + * + * 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 <linux/module.h> +#include <linux/skbuff.h> +#include <net/ip.h> + +#include <linux/netfilter/xt_length.h> +#include <linux/netfilter/x_tables.h> + +MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>"); +MODULE_DESCRIPTION("IP tables packet length matching module"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_length"); +MODULE_ALIAS("ip6t_length"); + +static int +match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + const struct xt_length_info *info = matchinfo; + u_int16_t pktlen = ntohs(skb->nh.iph->tot_len); + + return (pktlen >= info->min && pktlen <= info->max) ^ info->invert; +} + +static int +match6(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + const struct xt_length_info *info = matchinfo; + u_int16_t pktlen = ntohs(skb->nh.ipv6h->payload_len) + sizeof(struct ipv6hdr); + + return (pktlen >= info->min && pktlen <= info->max) ^ info->invert; +} + +static int +checkentry(const char *tablename, + const void *ip, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + if (matchsize != XT_ALIGN(sizeof(struct xt_length_info))) + return 0; + + return 1; +} + +static struct xt_match length_match = { + .name = "length", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE, +}; +static struct xt_match length6_match = { + .name = "length", + .match = &match6, + .checkentry = &checkentry, + .me = THIS_MODULE, +}; + +static int __init init(void) +{ + int ret; + ret = xt_register_match(AF_INET, &length_match); + if (ret) + return ret; + ret = xt_register_match(AF_INET6, &length6_match); + if (ret) + xt_unregister_match(AF_INET, &length_match); + + return ret; +} + +static void __exit fini(void) +{ + xt_unregister_match(AF_INET, &length_match); + xt_unregister_match(AF_INET6, &length6_match); +} + +module_init(init); +module_exit(fini); diff --git a/net/ipv4/netfilter/ipt_limit.c b/net/netfilter/xt_limit.c index 0c24dcc703a..15e40506bc3 100644 --- a/net/ipv4/netfilter/ipt_limit.c +++ b/net/netfilter/xt_limit.c @@ -18,12 +18,14 @@ #include <linux/spinlock.h> #include <linux/interrupt.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_limit.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_limit.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Herve Eychenne <rv@wallfire.org>"); MODULE_DESCRIPTION("iptables rate limit match"); +MODULE_ALIAS("ipt_limit"); +MODULE_ALIAS("ip6t_limit"); /* The algorithm used is the Simple Token Bucket Filter (TBF) * see net/sched/sch_tbf.c in the linux source tree @@ -68,9 +70,10 @@ ipt_limit_match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - struct ipt_rateinfo *r = ((struct ipt_rateinfo *)matchinfo)->master; + struct xt_rateinfo *r = ((struct xt_rateinfo *)matchinfo)->master; unsigned long now = jiffies; spin_lock_bh(&limit_lock); @@ -96,32 +99,32 @@ user2credits(u_int32_t user) /* If multiplying would overflow... */ if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY)) /* Divide first. */ - return (user / IPT_LIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; + return (user / XT_LIMIT_SCALE) * HZ * CREDITS_PER_JIFFY; - return (user * HZ * CREDITS_PER_JIFFY) / IPT_LIMIT_SCALE; + return (user * HZ * CREDITS_PER_JIFFY) / XT_LIMIT_SCALE; } static int ipt_limit_checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - struct ipt_rateinfo *r = matchinfo; + struct xt_rateinfo *r = matchinfo; - if (matchsize != IPT_ALIGN(sizeof(struct ipt_rateinfo))) + if (matchsize != XT_ALIGN(sizeof(struct xt_rateinfo))) return 0; /* Check for overflow. */ if (r->burst == 0 || user2credits(r->avg * r->burst) < user2credits(r->avg)) { - printk("Overflow in ipt_limit, try lower: %u/%u\n", + printk("Overflow in xt_limit, try lower: %u/%u\n", r->avg, r->burst); return 0; } - /* User avg in seconds * IPT_LIMIT_SCALE: convert to jiffies * + /* User avg in seconds * XT_LIMIT_SCALE: convert to jiffies * 128. */ r->prev = jiffies; r->credit = user2credits(r->avg * r->burst); /* Credits full. */ @@ -134,7 +137,13 @@ ipt_limit_checkentry(const char *tablename, return 1; } -static struct ipt_match ipt_limit_reg = { +static struct xt_match ipt_limit_reg = { + .name = "limit", + .match = ipt_limit_match, + .checkentry = ipt_limit_checkentry, + .me = THIS_MODULE, +}; +static struct xt_match limit6_reg = { .name = "limit", .match = ipt_limit_match, .checkentry = ipt_limit_checkentry, @@ -143,14 +152,23 @@ static struct ipt_match ipt_limit_reg = { static int __init init(void) { - if (ipt_register_match(&ipt_limit_reg)) - return -EINVAL; - return 0; + int ret; + + ret = xt_register_match(AF_INET, &ipt_limit_reg); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &limit6_reg); + if (ret) + xt_unregister_match(AF_INET, &ipt_limit_reg); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&ipt_limit_reg); + xt_unregister_match(AF_INET, &ipt_limit_reg); + xt_unregister_match(AF_INET6, &limit6_reg); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_mac.c b/net/netfilter/xt_mac.c index 1b9bb4559f8..0461dcb5fc7 100644 --- a/net/ipv4/netfilter/ipt_mac.c +++ b/net/netfilter/xt_mac.c @@ -13,12 +13,15 @@ #include <linux/if_ether.h> #include <linux/etherdevice.h> -#include <linux/netfilter_ipv4/ipt_mac.h> -#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv4.h> +#include <linux/netfilter/xt_mac.h> +#include <linux/netfilter/x_tables.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Netfilter Core Team <coreteam@netfilter.org>"); MODULE_DESCRIPTION("iptables mac matching module"); +MODULE_ALIAS("ipt_mac"); +MODULE_ALIAS("ip6t_mac"); static int match(const struct sk_buff *skb, @@ -26,9 +29,10 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_mac_info *info = matchinfo; + const struct xt_mac_info *info = matchinfo; /* Is mac pointer valid? */ return (skb->mac.raw >= skb->head @@ -40,7 +44,7 @@ match(const struct sk_buff *skb, static int ipt_mac_checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) @@ -49,17 +53,23 @@ ipt_mac_checkentry(const char *tablename, if (hook_mask & ~((1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_IN) | (1 << NF_IP_FORWARD))) { - printk("ipt_mac: only valid for PRE_ROUTING, LOCAL_IN or FORWARD.\n"); + printk("xt_mac: only valid for PRE_ROUTING, LOCAL_IN or FORWARD.\n"); return 0; } - if (matchsize != IPT_ALIGN(sizeof(struct ipt_mac_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_mac_info))) return 0; return 1; } -static struct ipt_match mac_match = { +static struct xt_match mac_match = { + .name = "mac", + .match = &match, + .checkentry = &ipt_mac_checkentry, + .me = THIS_MODULE, +}; +static struct xt_match mac6_match = { .name = "mac", .match = &match, .checkentry = &ipt_mac_checkentry, @@ -68,12 +78,22 @@ static struct ipt_match mac_match = { static int __init init(void) { - return ipt_register_match(&mac_match); + int ret; + ret = xt_register_match(AF_INET, &mac_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &mac6_match); + if (ret) + xt_unregister_match(AF_INET, &mac_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&mac_match); + xt_unregister_match(AF_INET, &mac_match); + xt_unregister_match(AF_INET6, &mac6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_mark.c b/net/netfilter/xt_mark.c index 00bef6cdd3f..2a0ac62b72c 100644 --- a/net/ipv4/netfilter/ipt_mark.c +++ b/net/netfilter/xt_mark.c @@ -10,12 +10,14 @@ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ipt_mark.h> -#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter/xt_mark.h> +#include <linux/netfilter/x_tables.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>"); MODULE_DESCRIPTION("iptables mark matching module"); +MODULE_ALIAS("ipt_mark"); +MODULE_ALIAS("ip6t_mark"); static int match(const struct sk_buff *skb, @@ -23,23 +25,24 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_mark_info *info = matchinfo; + const struct xt_mark_info *info = matchinfo; return ((skb->nfmark & info->mask) == info->mark) ^ info->invert; } static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *entry, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - struct ipt_mark_info *minfo = (struct ipt_mark_info *) matchinfo; + struct xt_mark_info *minfo = (struct xt_mark_info *) matchinfo; - if (matchsize != IPT_ALIGN(sizeof(struct ipt_mark_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_mark_info))) return 0; if (minfo->mark > 0xffffffff || minfo->mask > 0xffffffff) { @@ -50,7 +53,14 @@ checkentry(const char *tablename, return 1; } -static struct ipt_match mark_match = { +static struct xt_match mark_match = { + .name = "mark", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE, +}; + +static struct xt_match mark6_match = { .name = "mark", .match = &match, .checkentry = &checkentry, @@ -59,12 +69,22 @@ static struct ipt_match mark_match = { static int __init init(void) { - return ipt_register_match(&mark_match); + int ret; + ret = xt_register_match(AF_INET, &mark_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &mark6_match); + if (ret) + xt_unregister_match(AF_INET, &mark_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&mark_match); + xt_unregister_match(AF_INET, &mark_match); + xt_unregister_match(AF_INET6, &mark6_match); } module_init(init); diff --git a/net/ipv6/netfilter/ip6t_physdev.c b/net/netfilter/xt_physdev.c index 71515c86ece..19bb57c14df 100644 --- a/net/ipv6/netfilter/ip6t_physdev.c +++ b/net/netfilter/xt_physdev.c @@ -10,8 +10,8 @@ #include <linux/module.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv6/ip6t_physdev.h> -#include <linux/netfilter_ipv6/ip6_tables.h> +#include <linux/netfilter/xt_physdev.h> +#include <linux/netfilter/x_tables.h> #include <linux/netfilter_bridge.h> #define MATCH 1 #define NOMATCH 0 @@ -19,6 +19,8 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Bart De Schuymer <bdschuym@pandora.be>"); MODULE_DESCRIPTION("iptables bridge physical device match module"); +MODULE_ALIAS("ipt_physdev"); +MODULE_ALIAS("ip6t_physdev"); static int match(const struct sk_buff *skb, @@ -31,7 +33,7 @@ match(const struct sk_buff *skb, { int i; static const char nulldevname[IFNAMSIZ]; - const struct ip6t_physdev_info *info = matchinfo; + const struct xt_physdev_info *info = matchinfo; unsigned int ret; const char *indev, *outdev; struct nf_bridge_info *nf_bridge; @@ -41,37 +43,37 @@ match(const struct sk_buff *skb, * the destination device will be a bridge. */ if (!(nf_bridge = skb->nf_bridge)) { /* Return MATCH if the invert flags of the used options are on */ - if ((info->bitmask & IP6T_PHYSDEV_OP_BRIDGED) && - !(info->invert & IP6T_PHYSDEV_OP_BRIDGED)) + if ((info->bitmask & XT_PHYSDEV_OP_BRIDGED) && + !(info->invert & XT_PHYSDEV_OP_BRIDGED)) return NOMATCH; - if ((info->bitmask & IP6T_PHYSDEV_OP_ISIN) && - !(info->invert & IP6T_PHYSDEV_OP_ISIN)) + if ((info->bitmask & XT_PHYSDEV_OP_ISIN) && + !(info->invert & XT_PHYSDEV_OP_ISIN)) return NOMATCH; - if ((info->bitmask & IP6T_PHYSDEV_OP_ISOUT) && - !(info->invert & IP6T_PHYSDEV_OP_ISOUT)) + if ((info->bitmask & XT_PHYSDEV_OP_ISOUT) && + !(info->invert & XT_PHYSDEV_OP_ISOUT)) return NOMATCH; - if ((info->bitmask & IP6T_PHYSDEV_OP_IN) && - !(info->invert & IP6T_PHYSDEV_OP_IN)) + if ((info->bitmask & XT_PHYSDEV_OP_IN) && + !(info->invert & XT_PHYSDEV_OP_IN)) return NOMATCH; - if ((info->bitmask & IP6T_PHYSDEV_OP_OUT) && - !(info->invert & IP6T_PHYSDEV_OP_OUT)) + if ((info->bitmask & XT_PHYSDEV_OP_OUT) && + !(info->invert & XT_PHYSDEV_OP_OUT)) return NOMATCH; return MATCH; } /* This only makes sense in the FORWARD and POSTROUTING chains */ - if ((info->bitmask & IP6T_PHYSDEV_OP_BRIDGED) && + if ((info->bitmask & XT_PHYSDEV_OP_BRIDGED) && (!!(nf_bridge->mask & BRNF_BRIDGED) ^ - !(info->invert & IP6T_PHYSDEV_OP_BRIDGED))) + !(info->invert & XT_PHYSDEV_OP_BRIDGED))) return NOMATCH; - if ((info->bitmask & IP6T_PHYSDEV_OP_ISIN && - (!nf_bridge->physindev ^ !!(info->invert & IP6T_PHYSDEV_OP_ISIN))) || - (info->bitmask & IP6T_PHYSDEV_OP_ISOUT && - (!nf_bridge->physoutdev ^ !!(info->invert & IP6T_PHYSDEV_OP_ISOUT)))) + if ((info->bitmask & XT_PHYSDEV_OP_ISIN && + (!nf_bridge->physindev ^ !!(info->invert & XT_PHYSDEV_OP_ISIN))) || + (info->bitmask & XT_PHYSDEV_OP_ISOUT && + (!nf_bridge->physoutdev ^ !!(info->invert & XT_PHYSDEV_OP_ISOUT)))) return NOMATCH; - if (!(info->bitmask & IP6T_PHYSDEV_OP_IN)) + if (!(info->bitmask & XT_PHYSDEV_OP_IN)) goto match_outdev; indev = nf_bridge->physindev ? nf_bridge->physindev->name : nulldevname; for (i = 0, ret = 0; i < IFNAMSIZ/sizeof(unsigned int); i++) { @@ -80,11 +82,11 @@ match(const struct sk_buff *skb, & ((const unsigned int *)info->in_mask)[i]; } - if ((ret == 0) ^ !(info->invert & IP6T_PHYSDEV_OP_IN)) + if ((ret == 0) ^ !(info->invert & XT_PHYSDEV_OP_IN)) return NOMATCH; match_outdev: - if (!(info->bitmask & IP6T_PHYSDEV_OP_OUT)) + if (!(info->bitmask & XT_PHYSDEV_OP_OUT)) return MATCH; outdev = nf_bridge->physoutdev ? nf_bridge->physoutdev->name : nulldevname; @@ -94,27 +96,34 @@ match_outdev: & ((const unsigned int *)info->out_mask)[i]; } - return (ret != 0) ^ !(info->invert & IP6T_PHYSDEV_OP_OUT); + return (ret != 0) ^ !(info->invert & XT_PHYSDEV_OP_OUT); } static int checkentry(const char *tablename, - const struct ip6t_ip6 *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - const struct ip6t_physdev_info *info = matchinfo; + const struct xt_physdev_info *info = matchinfo; - if (matchsize != IP6T_ALIGN(sizeof(struct ip6t_physdev_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_physdev_info))) return 0; - if (!(info->bitmask & IP6T_PHYSDEV_OP_MASK) || - info->bitmask & ~IP6T_PHYSDEV_OP_MASK) + if (!(info->bitmask & XT_PHYSDEV_OP_MASK) || + info->bitmask & ~XT_PHYSDEV_OP_MASK) return 0; return 1; } -static struct ip6t_match physdev_match = { +static struct xt_match physdev_match = { + .name = "physdev", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE, +}; + +static struct xt_match physdev6_match = { .name = "physdev", .match = &match, .checkentry = &checkentry, @@ -123,12 +132,23 @@ static struct ip6t_match physdev_match = { static int __init init(void) { - return ip6t_register_match(&physdev_match); + int ret; + + ret = xt_register_match(AF_INET, &physdev_match); + if (ret < 0) + return ret; + + ret = xt_register_match(AF_INET6, &physdev6_match); + if (ret < 0) + xt_unregister_match(AF_INET, &physdev_match); + + return ret; } static void __exit fini(void) { - ip6t_unregister_match(&physdev_match); + xt_unregister_match(AF_INET, &physdev_match); + xt_unregister_match(AF_INET6, &physdev6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_pkttype.c b/net/netfilter/xt_pkttype.c index 8ddb1dc5e5a..ab1b2630f97 100644 --- a/net/ipv4/netfilter/ipt_pkttype.c +++ b/net/netfilter/xt_pkttype.c @@ -10,60 +10,72 @@ #include <linux/if_ether.h> #include <linux/if_packet.h> -#include <linux/netfilter_ipv4/ipt_pkttype.h> -#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter/xt_pkttype.h> +#include <linux/netfilter/x_tables.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Michal Ludvig <michal@logix.cz>"); MODULE_DESCRIPTION("IP tables match to match on linklayer packet type"); +MODULE_ALIAS("ipt_pkttype"); +MODULE_ALIAS("ip6t_pkttype"); static int match(const struct sk_buff *skb, const struct net_device *in, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_pkttype_info *info = matchinfo; + const struct xt_pkttype_info *info = matchinfo; - return (skb->pkt_type == info->pkttype) ^ info->invert; + return (skb->pkt_type == info->pkttype) ^ info->invert; } static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { -/* - if (hook_mask - & ~((1 << NF_IP_PRE_ROUTING) | (1 << NF_IP_LOCAL_IN) - | (1 << NF_IP_FORWARD))) { - printk("ipt_pkttype: only valid for PRE_ROUTING, LOCAL_IN or FORWARD.\n"); - return 0; - } -*/ - if (matchsize != IPT_ALIGN(sizeof(struct ipt_pkttype_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_pkttype_info))) return 0; return 1; } -static struct ipt_match pkttype_match = { +static struct xt_match pkttype_match = { .name = "pkttype", .match = &match, .checkentry = &checkentry, .me = THIS_MODULE, }; +static struct xt_match pkttype6_match = { + .name = "pkttype", + .match = &match, + .checkentry = &checkentry, + .me = THIS_MODULE, +}; + static int __init init(void) { - return ipt_register_match(&pkttype_match); + int ret; + ret = xt_register_match(AF_INET, &pkttype_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &pkttype6_match); + if (ret) + xt_unregister_match(AF_INET, &pkttype_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&pkttype_match); + xt_unregister_match(AF_INET, &pkttype_match); + xt_unregister_match(AF_INET6, &pkttype6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_realm.c b/net/netfilter/xt_realm.c index 54a6897ebaa..2b7e1781d34 100644 --- a/net/ipv4/netfilter/ipt_realm.c +++ b/net/netfilter/xt_realm.c @@ -14,12 +14,14 @@ #include <linux/netdevice.h> #include <net/route.h> -#include <linux/netfilter_ipv4/ipt_realm.h> -#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv4.h> +#include <linux/netfilter/xt_realm.h> +#include <linux/netfilter/x_tables.h> MODULE_AUTHOR("Sampsa Ranta <sampsa@netsonic.fi>"); MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("iptables realm match"); +MODULE_DESCRIPTION("X_tables realm match"); +MODULE_ALIAS("ipt_realm"); static int match(const struct sk_buff *skb, @@ -27,16 +29,17 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_realm_info *info = matchinfo; + const struct xt_realm_info *info = matchinfo; struct dst_entry *dst = skb->dst; return (info->id == (dst->tclassid & info->mask)) ^ info->invert; } static int check(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) @@ -44,18 +47,18 @@ static int check(const char *tablename, 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, " + printk("xt_realm: only valid for POST_ROUTING, LOCAL_OUT, " "LOCAL_IN or FORWARD.\n"); return 0; } - if (matchsize != IPT_ALIGN(sizeof(struct ipt_realm_info))) { - printk("ipt_realm: invalid matchsize.\n"); + if (matchsize != XT_ALIGN(sizeof(struct xt_realm_info))) { + printk("xt_realm: invalid matchsize.\n"); return 0; } return 1; } -static struct ipt_match realm_match = { +static struct xt_match realm_match = { .name = "realm", .match = match, .checkentry = check, @@ -64,12 +67,12 @@ static struct ipt_match realm_match = { static int __init init(void) { - return ipt_register_match(&realm_match); + return xt_register_match(AF_INET, &realm_match); } static void __exit fini(void) { - ipt_unregister_match(&realm_match); + xt_unregister_match(AF_INET, &realm_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_sctp.c b/net/netfilter/xt_sctp.c index fe2b327bcaa..10fbfc5ba75 100644 --- a/net/ipv4/netfilter/ipt_sctp.c +++ b/net/netfilter/xt_sctp.c @@ -1,10 +1,18 @@ #include <linux/module.h> #include <linux/skbuff.h> #include <net/ip.h> +#include <net/ipv6.h> #include <linux/sctp.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_sctp.h> #include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_sctp.h> +#include <linux/netfilter_ipv6/ip6_tables.h> + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Kiran Kumar Immidi"); +MODULE_DESCRIPTION("Match for SCTP protocol packets"); +MODULE_ALIAS("ipt_sctp"); #ifdef DEBUG_SCTP #define duprintf(format, args...) printk(format , ## args) @@ -16,7 +24,7 @@ || (!!((invflag) & (option)) ^ (cond))) static int -match_flags(const struct ipt_sctp_flag_info *flag_info, +match_flags(const struct xt_sctp_flag_info *flag_info, const int flag_count, u_int8_t chunktype, u_int8_t chunkflags) @@ -32,15 +40,15 @@ match_flags(const struct ipt_sctp_flag_info *flag_info, return 1; } -static int +static inline int match_packet(const struct sk_buff *skb, + unsigned int offset, const u_int32_t *chunkmap, int chunk_match_type, - const struct ipt_sctp_flag_info *flag_info, + const struct xt_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, *sch; @@ -52,7 +60,6 @@ match_packet(const struct sk_buff *skb, SCTP_CHUNKMAP_COPY(chunkmapcopy, chunkmap); } - offset = skb->nh.iph->ihl * 4 + sizeof (sctp_sctphdr_t); do { sch = skb_header_pointer(skb, offset, sizeof(_sch), &_sch); if (sch == NULL) { @@ -118,19 +125,20 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_sctp_info *info; + const struct xt_sctp_info *info; sctp_sctphdr_t _sh, *sh; - info = (const struct ipt_sctp_info *)matchinfo; + info = (const struct xt_sctp_info *)matchinfo; if (offset) { duprintf("Dropping non-first fragment.. FIXME\n"); return 0; } - sh = skb_header_pointer(skb, skb->nh.iph->ihl*4, sizeof(_sh), &_sh); + sh = skb_header_pointer(skb, protoff, sizeof(_sh), &_sh); if (sh == NULL) { duprintf("Dropping evil TCP offset=0 tinygram.\n"); *hotdrop = 1; @@ -140,64 +148,103 @@ match(const struct sk_buff *skb, return SCCHECK(((ntohs(sh->source) >= info->spts[0]) && (ntohs(sh->source) <= info->spts[1])), - IPT_SCTP_SRC_PORTS, info->flags, info->invflags) + XT_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, + XT_SCTP_DEST_PORTS, info->flags, info->invflags) + && SCCHECK(match_packet(skb, protoff, + info->chunkmap, info->chunk_match_type, info->flag_info, info->flag_count, hotdrop), - IPT_SCTP_CHUNK_TYPES, info->flags, info->invflags); + XT_SCTP_CHUNK_TYPES, info->flags, info->invflags); } static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *inf, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct xt_sctp_info *info; + const struct ipt_ip *ip = inf; + + info = (const struct xt_sctp_info *)matchinfo; + + return ip->proto == IPPROTO_SCTP + && !(ip->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_sctp_info)) + && !(info->flags & ~XT_SCTP_VALID_FLAGS) + && !(info->invflags & ~XT_SCTP_VALID_FLAGS) + && !(info->invflags & ~info->flags) + && ((!(info->flags & XT_SCTP_CHUNK_TYPES)) || + (info->chunk_match_type & + (SCTP_CHUNK_MATCH_ALL + | SCTP_CHUNK_MATCH_ANY + | SCTP_CHUNK_MATCH_ONLY))); +} + +static int +checkentry6(const char *tablename, + const void *inf, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - const struct ipt_sctp_info *info; + const struct xt_sctp_info *info; + const struct ip6t_ip6 *ip = inf; - info = (const struct ipt_sctp_info *)matchinfo; + info = (const struct xt_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) + && !(ip->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_sctp_info)) + && !(info->flags & ~XT_SCTP_VALID_FLAGS) + && !(info->invflags & ~XT_SCTP_VALID_FLAGS) && !(info->invflags & ~info->flags) - && ((!(info->flags & IPT_SCTP_CHUNK_TYPES)) || + && ((!(info->flags & XT_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 = + +static struct xt_match sctp_match = { - .list = { NULL, NULL}, .name = "sctp", .match = &match, .checkentry = &checkentry, - .destroy = NULL, + .me = THIS_MODULE +}; +static struct xt_match sctp6_match = +{ + .name = "sctp", + .match = &match, + .checkentry = &checkentry6, .me = THIS_MODULE }; + static int __init init(void) { - return ipt_register_match(&sctp_match); + int ret; + ret = xt_register_match(AF_INET, &sctp_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &sctp6_match); + if (ret) + xt_unregister_match(AF_INET, &sctp_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&sctp_match); + xt_unregister_match(AF_INET6, &sctp6_match); + xt_unregister_match(AF_INET, &sctp_match); } module_init(init); module_exit(fini); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Kiran Kumar Immidi"); -MODULE_DESCRIPTION("Match for SCTP protocol packets"); - diff --git a/net/ipv4/netfilter/ipt_state.c b/net/netfilter/xt_state.c index 4d7f16b70ce..39ce808d40e 100644 --- a/net/ipv4/netfilter/ipt_state.c +++ b/net/netfilter/xt_state.c @@ -1,7 +1,7 @@ /* Kernel module to match connection tracking information. */ /* (C) 1999-2001 Paul `Rusty' Russell - * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * (C) 2002-2005 Netfilter Core Team <coreteam@netfilter.org> * * 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 @@ -11,12 +11,14 @@ #include <linux/module.h> #include <linux/skbuff.h> #include <net/netfilter/nf_conntrack_compat.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_state.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_state.h> MODULE_LICENSE("GPL"); MODULE_AUTHOR("Rusty Russell <rusty@rustcorp.com.au>"); -MODULE_DESCRIPTION("iptables connection tracking state match module"); +MODULE_DESCRIPTION("ip[6]_tables connection tracking state match module"); +MODULE_ALIAS("ipt_state"); +MODULE_ALIAS("ip6t_state"); static int match(const struct sk_buff *skb, @@ -24,35 +26,43 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_state_info *sinfo = matchinfo; + const struct xt_state_info *sinfo = matchinfo; enum ip_conntrack_info ctinfo; unsigned int statebit; if (nf_ct_is_untracked(skb)) - statebit = IPT_STATE_UNTRACKED; + statebit = XT_STATE_UNTRACKED; else if (!nf_ct_get_ctinfo(skb, &ctinfo)) - statebit = IPT_STATE_INVALID; + statebit = XT_STATE_INVALID; else - statebit = IPT_STATE_BIT(ctinfo); + statebit = XT_STATE_BIT(ctinfo); return (sinfo->statemask & statebit); } static int check(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - if (matchsize != IPT_ALIGN(sizeof(struct ipt_state_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_state_info))) return 0; return 1; } -static struct ipt_match state_match = { +static struct xt_match state_match = { + .name = "state", + .match = &match, + .checkentry = &check, + .me = THIS_MODULE, +}; + +static struct xt_match state6_match = { .name = "state", .match = &match, .checkentry = &check, @@ -61,13 +71,25 @@ static struct ipt_match state_match = { static int __init init(void) { - need_ip_conntrack(); - return ipt_register_match(&state_match); + int ret; + + need_conntrack(); + + ret = xt_register_match(AF_INET, &state_match); + if (ret < 0) + return ret; + + ret = xt_register_match(AF_INET6, &state6_match); + if (ret < 0) + xt_unregister_match(AF_INET,&state_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&state_match); + xt_unregister_match(AF_INET, &state_match); + xt_unregister_match(AF_INET6, &state6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_string.c b/net/netfilter/xt_string.c index b5def204d79..7c7d5c8807d 100644 --- a/net/ipv4/netfilter/ipt_string.c +++ b/net/netfilter/xt_string.c @@ -11,23 +11,26 @@ #include <linux/module.h> #include <linux/kernel.h> #include <linux/skbuff.h> -#include <linux/netfilter_ipv4/ip_tables.h> -#include <linux/netfilter_ipv4/ipt_string.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_string.h> #include <linux/textsearch.h> MODULE_AUTHOR("Pablo Neira Ayuso <pablo@eurodev.net>"); MODULE_DESCRIPTION("IP tables string match module"); MODULE_LICENSE("GPL"); +MODULE_ALIAS("ipt_string"); +MODULE_ALIAS("ip6t_string"); static int match(const struct sk_buff *skb, const struct net_device *in, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { struct ts_state state; - struct ipt_string_info *conf = (struct ipt_string_info *) matchinfo; + struct xt_string_info *conf = (struct xt_string_info *) matchinfo; memset(&state, 0, sizeof(struct ts_state)); @@ -36,18 +39,18 @@ static int match(const struct sk_buff *skb, != UINT_MAX) && !conf->invert; } -#define STRING_TEXT_PRIV(m) ((struct ipt_string_info *) m) +#define STRING_TEXT_PRIV(m) ((struct xt_string_info *) m) static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ip, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - struct ipt_string_info *conf = matchinfo; + struct xt_string_info *conf = matchinfo; struct ts_config *ts_conf; - if (matchsize != IPT_ALIGN(sizeof(struct ipt_string_info))) + if (matchsize != XT_ALIGN(sizeof(struct xt_string_info))) return 0; /* Damn, can't handle this case properly with iptables... */ @@ -69,7 +72,14 @@ static void destroy(void *matchinfo, unsigned int matchsize) textsearch_destroy(STRING_TEXT_PRIV(matchinfo)->config); } -static struct ipt_match string_match = { +static struct xt_match string_match = { + .name = "string", + .match = match, + .checkentry = checkentry, + .destroy = destroy, + .me = THIS_MODULE +}; +static struct xt_match string6_match = { .name = "string", .match = match, .checkentry = checkentry, @@ -79,12 +89,22 @@ static struct ipt_match string_match = { static int __init init(void) { - return ipt_register_match(&string_match); + int ret; + + ret = xt_register_match(AF_INET, &string_match); + if (ret) + return ret; + ret = xt_register_match(AF_INET6, &string6_match); + if (ret) + xt_unregister_match(AF_INET, &string_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&string_match); + xt_unregister_match(AF_INET, &string_match); + xt_unregister_match(AF_INET6, &string6_match); } module_init(init); diff --git a/net/ipv4/netfilter/ipt_tcpmss.c b/net/netfilter/xt_tcpmss.c index 4dc9b16ab4a..acf7f533e9f 100644 --- a/net/ipv4/netfilter/ipt_tcpmss.c +++ b/net/netfilter/xt_tcpmss.c @@ -1,6 +1,7 @@ /* Kernel module to match TCP MSS values. */ /* Copyright (C) 2000 Marc Boucher <marc@mbsi.ca> + * Portions (C) 2005 by Harald Welte <laforge@netfilter.org> * * 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 @@ -11,19 +12,24 @@ #include <linux/skbuff.h> #include <net/tcp.h> -#include <linux/netfilter_ipv4/ipt_tcpmss.h> +#include <linux/netfilter/xt_tcpmss.h> +#include <linux/netfilter/x_tables.h> + #include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv6/ip6_tables.h> #define TH_SYN 0x02 MODULE_LICENSE("GPL"); MODULE_AUTHOR("Marc Boucher <marc@mbsi.ca>"); MODULE_DESCRIPTION("iptables TCP MSS match module"); +MODULE_ALIAS("ipt_tcpmss"); /* Returns 1 if the mss option is set and matched by the range, 0 otherwise */ static inline int mssoption_match(u_int16_t min, u_int16_t max, const struct sk_buff *skb, + unsigned int protoff, int invert, int *hotdrop) { @@ -33,8 +39,7 @@ mssoption_match(u_int16_t min, u_int16_t max, unsigned int i, optlen; /* If we don't have the whole header, drop packet. */ - th = skb_header_pointer(skb, skb->nh.iph->ihl * 4, - sizeof(_tcph), &_tcph); + th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); if (th == NULL) goto dropit; @@ -47,8 +52,7 @@ mssoption_match(u_int16_t min, u_int16_t max, goto out; /* Truncated options. */ - op = skb_header_pointer(skb, skb->nh.iph->ihl * 4 + sizeof(*th), - optlen, _opt); + op = skb_header_pointer(skb, protoff + sizeof(*th), optlen, _opt); if (op == NULL) goto dropit; @@ -79,22 +83,24 @@ match(const struct sk_buff *skb, const struct net_device *out, const void *matchinfo, int offset, + unsigned int protoff, int *hotdrop) { - const struct ipt_tcpmss_match_info *info = matchinfo; + const struct xt_tcpmss_match_info *info = matchinfo; - return mssoption_match(info->mss_min, info->mss_max, skb, + return mssoption_match(info->mss_min, info->mss_max, skb, protoff, info->invert, hotdrop); } static int checkentry(const char *tablename, - const struct ipt_ip *ip, + const void *ipinfo, void *matchinfo, unsigned int matchsize, unsigned int hook_mask) { - if (matchsize != IPT_ALIGN(sizeof(struct ipt_tcpmss_match_info))) + const struct ipt_ip *ip = ipinfo; + if (matchsize != XT_ALIGN(sizeof(struct xt_tcpmss_match_info))) return 0; /* Must specify -p tcp */ @@ -106,21 +112,60 @@ checkentry(const char *tablename, return 1; } -static struct ipt_match tcpmss_match = { +static int +checkentry6(const char *tablename, + const void *ipinfo, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ip6t_ip6 *ip = ipinfo; + + if (matchsize != XT_ALIGN(sizeof(struct xt_tcpmss_match_info))) + return 0; + + /* Must specify -p tcp */ + if (ip->proto != IPPROTO_TCP || (ip->invflags & XT_INV_PROTO)) { + printk("tcpmss: Only works on TCP packets\n"); + return 0; + } + + return 1; +} + +static struct xt_match tcpmss_match = { .name = "tcpmss", .match = &match, .checkentry = &checkentry, .me = THIS_MODULE, }; +static struct xt_match tcpmss6_match = { + .name = "tcpmss", + .match = &match, + .checkentry = &checkentry6, + .me = THIS_MODULE, +}; + + static int __init init(void) { - return ipt_register_match(&tcpmss_match); + int ret; + ret = xt_register_match(AF_INET, &tcpmss_match); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &tcpmss6_match); + if (ret) + xt_unregister_match(AF_INET, &tcpmss_match); + + return ret; } static void __exit fini(void) { - ipt_unregister_match(&tcpmss_match); + xt_unregister_match(AF_INET6, &tcpmss6_match); + xt_unregister_match(AF_INET, &tcpmss_match); } module_init(init); diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c new file mode 100644 index 00000000000..33f86fd6f3e --- /dev/null +++ b/net/netfilter/xt_tcpudp.c @@ -0,0 +1,333 @@ +#include <linux/types.h> +#include <linux/module.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/tcp.h> +#include <net/udp.h> +#include <linux/netfilter/x_tables.h> +#include <linux/netfilter/xt_tcpudp.h> +#include <linux/netfilter_ipv4/ip_tables.h> +#include <linux/netfilter_ipv6/ip6_tables.h> + +MODULE_DESCRIPTION("x_tables match for TCP and UDP, supports IPv4 and IPv6"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("xt_tcp"); +MODULE_ALIAS("xt_udp"); +MODULE_ALIAS("ipt_udp"); +MODULE_ALIAS("ipt_tcp"); +MODULE_ALIAS("ip6t_udp"); +MODULE_ALIAS("ip6t_tcp"); + +#ifdef DEBUG_IP_FIREWALL_USER +#define duprintf(format, args...) printk(format , ## args) +#else +#define duprintf(format, args...) +#endif + + +/* Returns 1 if the port is matched by the range, 0 otherwise */ +static inline int +port_match(u_int16_t min, u_int16_t max, u_int16_t port, int invert) +{ + int ret; + + ret = (port >= min && port <= max) ^ invert; + return ret; +} + +static int +tcp_find_option(u_int8_t option, + const struct sk_buff *skb, + unsigned int protoff, + unsigned int optlen, + int invert, + int *hotdrop) +{ + /* tcp.doff is only 4 bits, ie. max 15 * 4 bytes */ + u_int8_t _opt[60 - sizeof(struct tcphdr)], *op; + unsigned int i; + + duprintf("tcp_match: finding option\n"); + + if (!optlen) + return invert; + + /* If we don't have the whole header, drop packet. */ + op = skb_header_pointer(skb, protoff + sizeof(struct tcphdr), + optlen, _opt); + if (op == NULL) { + *hotdrop = 1; + return 0; + } + + for (i = 0; i < optlen; ) { + if (op[i] == option) return !invert; + if (op[i] < 2) i++; + else i += op[i+1]?:1; + } + + return invert; +} + +static int +tcp_match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + struct tcphdr _tcph, *th; + const struct xt_tcp *tcpinfo = matchinfo; + + if (offset) { + /* To quote Alan: + + Don't allow a fragment of TCP 8 bytes in. Nobody normal + causes this. Its a cracker trying to break in by doing a + flag overwrite to pass the direction checks. + */ + if (offset == 1) { + duprintf("Dropping evil TCP offset=1 frag.\n"); + *hotdrop = 1; + } + /* Must not be a fragment. */ + return 0; + } + +#define FWINVTCP(bool,invflg) ((bool) ^ !!(tcpinfo->invflags & invflg)) + + th = skb_header_pointer(skb, protoff, sizeof(_tcph), &_tcph); + if (th == NULL) { + /* We've been asked to examine this packet, and we + can't. Hence, no choice but to drop. */ + duprintf("Dropping evil TCP offset=0 tinygram.\n"); + *hotdrop = 1; + return 0; + } + + if (!port_match(tcpinfo->spts[0], tcpinfo->spts[1], + ntohs(th->source), + !!(tcpinfo->invflags & XT_TCP_INV_SRCPT))) + return 0; + if (!port_match(tcpinfo->dpts[0], tcpinfo->dpts[1], + ntohs(th->dest), + !!(tcpinfo->invflags & XT_TCP_INV_DSTPT))) + return 0; + if (!FWINVTCP((((unsigned char *)th)[13] & tcpinfo->flg_mask) + == tcpinfo->flg_cmp, + XT_TCP_INV_FLAGS)) + return 0; + if (tcpinfo->option) { + if (th->doff * 4 < sizeof(_tcph)) { + *hotdrop = 1; + return 0; + } + if (!tcp_find_option(tcpinfo->option, skb, protoff, + th->doff*4 - sizeof(_tcph), + tcpinfo->invflags & XT_TCP_INV_OPTION, + hotdrop)) + return 0; + } + return 1; +} + +/* Called when user tries to insert an entry of this type. */ +static int +tcp_checkentry(const char *tablename, + const void *info, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ipt_ip *ip = info; + const struct xt_tcp *tcpinfo = matchinfo; + + /* Must specify proto == TCP, and no unknown invflags */ + return ip->proto == IPPROTO_TCP + && !(ip->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_tcp)) + && !(tcpinfo->invflags & ~XT_TCP_INV_MASK); +} + +/* Called when user tries to insert an entry of this type. */ +static int +tcp6_checkentry(const char *tablename, + const void *entry, + void *matchinfo, + unsigned int matchsize, + unsigned int hook_mask) +{ + const struct ip6t_ip6 *ipv6 = entry; + const struct xt_tcp *tcpinfo = matchinfo; + + /* Must specify proto == TCP, and no unknown invflags */ + return ipv6->proto == IPPROTO_TCP + && !(ipv6->invflags & XT_INV_PROTO) + && matchsize == XT_ALIGN(sizeof(struct xt_tcp)) + && !(tcpinfo->invflags & ~XT_TCP_INV_MASK); +} + + +static int +udp_match(const struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + const void *matchinfo, + int offset, + unsigned int protoff, + int *hotdrop) +{ + struct udphdr _udph, *uh; + const struct xt_udp *udpinfo = matchinfo; + + /* Must not be a fragment. */ + if (offset) + return 0; + + uh = skb_header_pointer(skb, protoff, sizeof(_udph), &_udph); + if (uh == NULL) { + /* We've been asked to examine this packet, and we + can't. Hence, no choice but to drop. */ + duprintf("Dropping evil UDP tinygram.\n"); + *hotdrop = 1; + return 0; + } + + return port_match(udpinfo->spts[0], udpinfo->spts[1], + ntohs(uh->source), + !!(udpinfo->invflags & XT_UDP_INV_SRCPT)) + && port_match(udpinfo->dpts[0], udpinfo->dpts[1], + ntohs(uh->dest), + !!(udpinfo->invflags & XT_UDP_INV_DSTPT)); +} + +/* Called when user tries to insert an entry of this type. */ +static int +udp_checkentry(const char *tablename, + const void *info, + void *matchinfo, + unsigned int matchinfosize, + unsigned int hook_mask) +{ + const struct ipt_ip *ip = info; + const struct xt_udp *udpinfo = matchinfo; + + /* Must specify proto == UDP, and no unknown invflags */ + if (ip->proto != IPPROTO_UDP || (ip->invflags & XT_INV_PROTO)) { + duprintf("ipt_udp: Protocol %u != %u\n", ip->proto, + IPPROTO_UDP); + return 0; + } + if (matchinfosize != XT_ALIGN(sizeof(struct xt_udp))) { + duprintf("ipt_udp: matchsize %u != %u\n", + matchinfosize, XT_ALIGN(sizeof(struct xt_udp))); + return 0; + } + if (udpinfo->invflags & ~XT_UDP_INV_MASK) { + duprintf("ipt_udp: unknown flags %X\n", + udpinfo->invflags); + return 0; + } + + return 1; +} + +/* Called when user tries to insert an entry of this type. */ +static int +udp6_checkentry(const char *tablename, + const void *entry, + void *matchinfo, + unsigned int matchinfosize, + unsigned int hook_mask) +{ + const struct ip6t_ip6 *ipv6 = entry; + const struct xt_udp *udpinfo = matchinfo; + + /* Must specify proto == UDP, and no unknown invflags */ + if (ipv6->proto != IPPROTO_UDP || (ipv6->invflags & XT_INV_PROTO)) { + duprintf("ip6t_udp: Protocol %u != %u\n", ipv6->proto, + IPPROTO_UDP); + return 0; + } + if (matchinfosize != XT_ALIGN(sizeof(struct xt_udp))) { + duprintf("ip6t_udp: matchsize %u != %u\n", + matchinfosize, XT_ALIGN(sizeof(struct xt_udp))); + return 0; + } + if (udpinfo->invflags & ~XT_UDP_INV_MASK) { + duprintf("ip6t_udp: unknown flags %X\n", + udpinfo->invflags); + return 0; + } + + return 1; +} + +static struct xt_match tcp_matchstruct = { + .name = "tcp", + .match = &tcp_match, + .checkentry = &tcp_checkentry, + .me = THIS_MODULE, +}; +static struct xt_match tcp6_matchstruct = { + .name = "tcp", + .match = &tcp_match, + .checkentry = &tcp6_checkentry, + .me = THIS_MODULE, +}; + +static struct xt_match udp_matchstruct = { + .name = "udp", + .match = &udp_match, + .checkentry = &udp_checkentry, + .me = THIS_MODULE, +}; +static struct xt_match udp6_matchstruct = { + .name = "udp", + .match = &udp_match, + .checkentry = &udp6_checkentry, + .me = THIS_MODULE, +}; + +static int __init init(void) +{ + int ret; + ret = xt_register_match(AF_INET, &tcp_matchstruct); + if (ret) + return ret; + + ret = xt_register_match(AF_INET6, &tcp6_matchstruct); + if (ret) + goto out_unreg_tcp; + + ret = xt_register_match(AF_INET, &udp_matchstruct); + if (ret) + goto out_unreg_tcp6; + + ret = xt_register_match(AF_INET6, &udp6_matchstruct); + if (ret) + goto out_unreg_udp; + + return ret; + +out_unreg_udp: + xt_unregister_match(AF_INET, &tcp_matchstruct); +out_unreg_tcp6: + xt_unregister_match(AF_INET6, &tcp6_matchstruct); +out_unreg_tcp: + xt_unregister_match(AF_INET, &tcp_matchstruct); + return ret; +} + +static void __exit fini(void) +{ + xt_unregister_match(AF_INET6, &udp6_matchstruct); + xt_unregister_match(AF_INET, &udp_matchstruct); + xt_unregister_match(AF_INET6, &tcp6_matchstruct); + xt_unregister_match(AF_INET, &tcp_matchstruct); +} + +module_init(init); +module_exit(fini); diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 3b1378498d5..4ae1538c54a 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -222,11 +222,6 @@ int genl_register_family(struct genl_family *family) goto errout_locked; } - if (!try_module_get(family->owner)) { - err = -EBUSY; - goto errout_locked; - } - if (family->id == GENL_ID_GENERATE) { u16 newid = genl_generate_id(); @@ -283,7 +278,6 @@ int genl_unregister_family(struct genl_family *family) INIT_LIST_HEAD(&family->ops_list); genl_unlock(); - module_put(family->owner); kfree(family->attrbuf); genl_ctrl_event(CTRL_CMD_DELFAMILY, family); return 0; @@ -535,7 +529,6 @@ static struct genl_family genl_ctrl = { .name = "nlctrl", .version = 0x1, .maxattr = CTRL_ATTR_MAX, - .owner = THIS_MODULE, }; static int __init genl_init(void) diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 8a260d43cee..778b1e5a4b5 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -44,7 +44,7 @@ if NET_SCHED choice prompt "Packet scheduler clock source" - default NET_SCH_CLK_JIFFIES + default NET_SCH_CLK_GETTIMEOFDAY ---help--- Packet schedulers need a monotonic clock that increments at a static rate. The kernel provides several suitable interfaces, each with diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c index b5001939b74..39a22a3ffe7 100644 --- a/net/sched/act_ipt.c +++ b/net/sched/act_ipt.c @@ -62,7 +62,7 @@ ipt_init_target(struct ipt_entry_target *t, char *table, unsigned int hook) struct ipt_target *target; int ret = 0; - target = ipt_find_target(t->u.user.name, t->u.user.revision); + target = xt_find_target(AF_INET, t->u.user.name, t->u.user.revision); if (!target) return -ENOENT; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 04c7fab4edc..2e266129a76 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -180,8 +180,7 @@ static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport, } SCTP_DEBUG_PRINTK("%s: skb:%p, len:%d, " - "src:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x " - "dst:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + "src:" NIP6_FMT " dst:" NIP6_FMT "\n", __FUNCTION__, skb, skb->len, NIP6(fl.fl6_src), NIP6(fl.fl6_dst)); @@ -206,13 +205,13 @@ static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc, fl.oif = daddr->v6.sin6_scope_id; - SCTP_DEBUG_PRINTK("%s: DST=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", + SCTP_DEBUG_PRINTK("%s: DST=" NIP6_FMT " ", __FUNCTION__, NIP6(fl.fl6_dst)); if (saddr) { ipv6_addr_copy(&fl.fl6_src, &saddr->v6.sin6_addr); SCTP_DEBUG_PRINTK( - "SRC=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x - ", + "SRC=" NIP6_FMT " - ", NIP6(fl.fl6_src)); } @@ -221,8 +220,7 @@ static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc, struct rt6_info *rt; rt = (struct rt6_info *)dst; SCTP_DEBUG_PRINTK( - "rt6_dst:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x " - "rt6_src:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + "rt6_dst:" NIP6_FMT " rt6_src:" NIP6_FMT "\n", NIP6(rt->rt6i_dst.addr), NIP6(rt->rt6i_src.addr)); } else { SCTP_DEBUG_PRINTK("NO ROUTE\n"); @@ -271,13 +269,12 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc, __u8 bmatchlen; SCTP_DEBUG_PRINTK("%s: asoc:%p dst:%p " - "daddr:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", + "daddr:" NIP6_FMT " ", __FUNCTION__, asoc, dst, NIP6(daddr->v6.sin6_addr)); if (!asoc) { ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr); - SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n", NIP6(saddr->v6.sin6_addr)); return; } @@ -305,13 +302,11 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc, if (baddr) { memcpy(saddr, baddr, sizeof(union sctp_addr)); - SCTP_DEBUG_PRINTK("saddr: " - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + SCTP_DEBUG_PRINTK("saddr: " NIP6_FMT "\n", NIP6(saddr->v6.sin6_addr)); } else { printk(KERN_ERR "%s: asoc:%p Could not find a valid source " - "address for the " - "dest:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + "address for the dest:" NIP6_FMT "\n", __FUNCTION__, asoc, NIP6(daddr->v6.sin6_addr)); } @@ -675,8 +670,7 @@ static int sctp_v6_is_ce(const struct sk_buff *skb) /* Dump the v6 addr to the seq file. */ static void sctp_v6_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr) { - seq_printf(seq, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", - NIP6(addr->v6.sin6_addr)); + seq_printf(seq, NIP6_FMT " ", NIP6(addr->v6.sin6_addr)); } /* Initialize a PF_INET6 socket msg_name. */ diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 557a7d90b92..477d7f80dba 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1036,14 +1036,14 @@ sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep, if (from_addr.sa.sa_family == AF_INET6) { printk(KERN_WARNING "%s association %p could not find address " - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NIP6_FMT "\n", __FUNCTION__, asoc, NIP6(from_addr.v6.sin6_addr)); } else { printk(KERN_WARNING "%s association %p could not find address " - "%u.%u.%u.%u\n", + NIPQUAD_FMT "\n", __FUNCTION__, asoc, NIPQUAD(from_addr.v4.sin_addr.s_addr)); diff --git a/net/tipc/Kconfig b/net/tipc/Kconfig new file mode 100644 index 00000000000..05ab18e62de --- /dev/null +++ b/net/tipc/Kconfig @@ -0,0 +1,112 @@ +# +# TIPC configuration +# + +menu "TIPC Configuration (EXPERIMENTAL)" + depends on INET && EXPERIMENTAL + +config TIPC + tristate "The TIPC Protocol (EXPERIMENTAL)" + ---help--- + TBD. + + This protocol support is also available as a module ( = code which + can be inserted in and removed from the running kernel whenever you + want). The module will be called tipc. If you want to compile it + as a module, say M here and read <file:Documentation/modules.txt>. + + If in doubt, say N. + +config TIPC_ADVANCED + bool "TIPC: Advanced configuration" + depends on TIPC + default n + help + Saying Y here will open some advanced configuration + for TIPC. Most users do not need to bother, so if + unsure, just say N. + +config TIPC_ZONES + int "Maximum number of zones in network" + depends on TIPC && TIPC_ADVANCED + default "3" + help + Max number of zones inside TIPC network. Max supported value + is 255 zones, minimum is 1 + + Default is 3 zones in a network; setting this to higher + allows more zones but might use more memory. + +config TIPC_CLUSTERS + int "Maximum number of clusters in a zone" + depends on TIPC && TIPC_ADVANCED + default "1" + help + ***Only 1 (one cluster in a zone) is supported by current code. + Any value set here will be overridden.*** + + (Max number of clusters inside TIPC zone. Max supported + value is 4095 clusters, minimum is 1. + + Default is 1; setting this to smaller value might save + some memory, setting it to higher + allows more clusters and might consume more memory.) + +config TIPC_NODES + int "Maximum number of nodes in cluster" + depends on TIPC && TIPC_ADVANCED + default "255" + help + Maximum number of nodes inside a TIPC cluster. Maximum + supported value is 2047 nodes, minimum is 8. + + Setting this to a smaller value saves some memory, + setting it to higher allows more nodes. + +config TIPC_SLAVE_NODES + int "Maximum number of slave nodes in cluster" + depends on TIPC && TIPC_ADVANCED + default "0" + help + ***This capability is not supported by current code.*** + + Maximum number of slave nodes inside a TIPC cluster. Maximum + supported value is 2047 nodes, minimum is 0. + + Setting this to a smaller value saves some memory, + setting it to higher allows more nodes. + +config TIPC_PORTS + int "Maximum number of ports in a node" + depends on TIPC && TIPC_ADVANCED + default "8191" + help + Maximum number of ports within a node. Maximum + supported value is 64535 nodes, minimum is 127. + + Setting this to a smaller value saves some memory, + setting it to higher allows more ports. + +config TIPC_LOG + int "Size of log buffer" + depends on TIPC && TIPC_ADVANCED + default 0 + help + Size (in bytes) of TIPC's internal log buffer, which records the + occurrence of significant events. Maximum supported value + is 32768 bytes, minimum is 0. + + There is no need to enable the log buffer unless the node will be + managed remotely via TIPC. + +config TIPC_DEBUG + bool "Enable debugging support" + depends on TIPC + default n + help + This will enable debugging of TIPC. + + Only say Y here if you are having trouble with TIPC. It will + enable the display of detailed information about what is going on. + +endmenu diff --git a/net/tipc/Makefile b/net/tipc/Makefile new file mode 100644 index 00000000000..dceb7027946 --- /dev/null +++ b/net/tipc/Makefile @@ -0,0 +1,13 @@ +# +# Makefile for the Linux TIPC layer +# + +obj-$(CONFIG_TIPC) := tipc.o + +tipc-y += addr.o bcast.o bearer.o config.o cluster.o \ + core.o handler.o link.o discover.o msg.o \ + name_distr.o subscr.o name_table.o net.o \ + netlink.o node.o node_subscr.o port.o ref.o \ + socket.o user_reg.o zone.o dbg.o eth_media.o + +# End of file diff --git a/net/tipc/addr.c b/net/tipc/addr.c new file mode 100644 index 00000000000..eca22260c98 --- /dev/null +++ b/net/tipc/addr.c @@ -0,0 +1,94 @@ +/* + * net/tipc/addr.c: TIPC address utility routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "addr.h" +#include "zone.h" +#include "cluster.h" +#include "net.h" + +u32 tipc_get_addr(void) +{ + return tipc_own_addr; +} + +/** + * addr_domain_valid - validates a network domain address + * + * Accepts <Z.C.N>, <Z.C.0>, <Z.0.0>, and <0.0.0>, + * where Z, C, and N are non-zero and do not exceed the configured limits. + * + * Returns 1 if domain address is valid, otherwise 0 + */ + +int addr_domain_valid(u32 addr) +{ + u32 n = tipc_node(addr); + u32 c = tipc_cluster(addr); + u32 z = tipc_zone(addr); + u32 max_nodes = tipc_max_nodes; + + if (is_slave(addr)) + max_nodes = LOWEST_SLAVE + tipc_max_slaves; + if (n > max_nodes) + return 0; + if (c > tipc_max_clusters) + return 0; + if (z > tipc_max_zones) + return 0; + + if (n && (!z || !c)) + return 0; + if (c && !z) + return 0; + return 1; +} + +/** + * addr_node_valid - validates a proposed network address for this node + * + * Accepts <Z.C.N>, where Z, C, and N are non-zero and do not exceed + * the configured limits. + * + * Returns 1 if address can be used, otherwise 0 + */ + +int addr_node_valid(u32 addr) +{ + return (addr_domain_valid(addr) && tipc_node(addr)); +} + diff --git a/net/tipc/addr.h b/net/tipc/addr.h new file mode 100644 index 00000000000..02ca71783e2 --- /dev/null +++ b/net/tipc/addr.h @@ -0,0 +1,128 @@ +/* + * net/tipc/addr.h: Include file for TIPC address utility routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_ADDR_H +#define _TIPC_ADDR_H + +static inline u32 own_node(void) +{ + return tipc_node(tipc_own_addr); +} + +static inline u32 own_cluster(void) +{ + return tipc_cluster(tipc_own_addr); +} + +static inline u32 own_zone(void) +{ + return tipc_zone(tipc_own_addr); +} + +static inline int in_own_cluster(u32 addr) +{ + return !((addr ^ tipc_own_addr) >> 12); +} + +static inline int in_own_zone(u32 addr) +{ + return !((addr ^ tipc_own_addr) >> 24); +} + +static inline int is_slave(u32 addr) +{ + return addr & 0x800; +} + +static inline int may_route(u32 addr) +{ + return(addr ^ tipc_own_addr) >> 11; +} + +static inline int in_scope(u32 domain, u32 addr) +{ + if (!domain || (domain == addr)) + return 1; + if (domain == (addr & 0xfffff000u)) /* domain <Z.C.0> */ + return 1; + if (domain == (addr & 0xff000000u)) /* domain <Z.0.0> */ + return 1; + return 0; +} + +/** + * addr_scope - convert message lookup domain to equivalent 2-bit scope value + */ + +static inline int addr_scope(u32 domain) +{ + if (likely(!domain)) + return TIPC_ZONE_SCOPE; + if (tipc_node(domain)) + return TIPC_NODE_SCOPE; + if (tipc_cluster(domain)) + return TIPC_CLUSTER_SCOPE; + return TIPC_ZONE_SCOPE; +} + +/** + * addr_domain - convert 2-bit scope value to equivalent message lookup domain + * + * Needed when address of a named message must be looked up a second time + * after a network hop. + */ + +static inline int addr_domain(int sc) +{ + if (likely(sc == TIPC_NODE_SCOPE)) + return tipc_own_addr; + if (sc == TIPC_CLUSTER_SCOPE) + return tipc_addr(tipc_zone(tipc_own_addr), + tipc_cluster(tipc_own_addr), 0); + return tipc_addr(tipc_zone(tipc_own_addr), 0, 0); +} + +static inline char *addr_string_fill(char *string, u32 addr) +{ + snprintf(string, 16, "<%u.%u.%u>", + tipc_zone(addr), tipc_cluster(addr), tipc_node(addr)); + return string; +} + +int addr_domain_valid(u32); +int addr_node_valid(u32 addr); + +#endif diff --git a/net/tipc/bcast.c b/net/tipc/bcast.c new file mode 100644 index 00000000000..9713d622efb --- /dev/null +++ b/net/tipc/bcast.c @@ -0,0 +1,806 @@ +/* + * net/tipc/bcast.c: TIPC broadcast code + * + * Copyright (c) 2004-2006, Ericsson AB + * Copyright (c) 2004, Intel Corporation. + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "msg.h" +#include "dbg.h" +#include "link.h" +#include "net.h" +#include "node.h" +#include "port.h" +#include "addr.h" +#include "node_subscr.h" +#include "name_distr.h" +#include "bearer.h" +#include "name_table.h" +#include "bcast.h" + + +#define MAX_PKT_DEFAULT_MCAST 1500 /* bcast link max packet size (fixed) */ + +#define BCLINK_WIN_DEFAULT 20 /* bcast link window size (default) */ + +#define BCLINK_LOG_BUF_SIZE 0 + +/** + * struct bcbearer_pair - a pair of bearers used by broadcast link + * @primary: pointer to primary bearer + * @secondary: pointer to secondary bearer + * + * Bearers must have same priority and same set of reachable destinations + * to be paired. + */ + +struct bcbearer_pair { + struct bearer *primary; + struct bearer *secondary; +}; + +/** + * struct bcbearer - bearer used by broadcast link + * @bearer: (non-standard) broadcast bearer structure + * @media: (non-standard) broadcast media structure + * @bpairs: array of bearer pairs + * @bpairs_temp: array of bearer pairs used during creation of "bpairs" + */ + +struct bcbearer { + struct bearer bearer; + struct media media; + struct bcbearer_pair bpairs[MAX_BEARERS]; + struct bcbearer_pair bpairs_temp[TIPC_NUM_LINK_PRI]; +}; + +/** + * struct bclink - link used for broadcast messages + * @link: (non-standard) broadcast link structure + * @node: (non-standard) node structure representing b'cast link's peer node + * + * Handles sequence numbering, fragmentation, bundling, etc. + */ + +struct bclink { + struct link link; + struct node node; +}; + + +static struct bcbearer *bcbearer = NULL; +static struct bclink *bclink = NULL; +static struct link *bcl = NULL; +static spinlock_t bc_lock = SPIN_LOCK_UNLOCKED; + +char bc_link_name[] = "multicast-link"; + + +static inline u32 buf_seqno(struct sk_buff *buf) +{ + return msg_seqno(buf_msg(buf)); +} + +static inline u32 bcbuf_acks(struct sk_buff *buf) +{ + return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle; +} + +static inline void bcbuf_set_acks(struct sk_buff *buf, u32 acks) +{ + TIPC_SKB_CB(buf)->handle = (void *)(unsigned long)acks; +} + +static inline void bcbuf_decr_acks(struct sk_buff *buf) +{ + bcbuf_set_acks(buf, bcbuf_acks(buf) - 1); +} + + +/** + * bclink_set_gap - set gap according to contents of current deferred pkt queue + * + * Called with 'node' locked, bc_lock unlocked + */ + +static inline void bclink_set_gap(struct node *n_ptr) +{ + struct sk_buff *buf = n_ptr->bclink.deferred_head; + + n_ptr->bclink.gap_after = n_ptr->bclink.gap_to = + mod(n_ptr->bclink.last_in); + if (unlikely(buf != NULL)) + n_ptr->bclink.gap_to = mod(buf_seqno(buf) - 1); +} + +/** + * bclink_ack_allowed - test if ACK or NACK message can be sent at this moment + * + * This mechanism endeavours to prevent all nodes in network from trying + * to ACK or NACK at the same time. + * + * Note: TIPC uses a different trigger to distribute ACKs than it does to + * distribute NACKs, but tries to use the same spacing (divide by 16). + */ + +static inline int bclink_ack_allowed(u32 n) +{ + return((n % TIPC_MIN_LINK_WIN) == tipc_own_tag); +} + + +/** + * bclink_retransmit_pkt - retransmit broadcast packets + * @after: sequence number of last packet to *not* retransmit + * @to: sequence number of last packet to retransmit + * + * Called with 'node' locked, bc_lock unlocked + */ + +static void bclink_retransmit_pkt(u32 after, u32 to) +{ + struct sk_buff *buf; + + spin_lock_bh(&bc_lock); + buf = bcl->first_out; + while (buf && less_eq(buf_seqno(buf), after)) { + buf = buf->next; + } + if (buf != NULL) + link_retransmit(bcl, buf, mod(to - after)); + spin_unlock_bh(&bc_lock); +} + +/** + * bclink_acknowledge - handle acknowledgement of broadcast packets + * @n_ptr: node that sent acknowledgement info + * @acked: broadcast sequence # that has been acknowledged + * + * Node is locked, bc_lock unlocked. + */ + +void bclink_acknowledge(struct node *n_ptr, u32 acked) +{ + struct sk_buff *crs; + struct sk_buff *next; + unsigned int released = 0; + + if (less_eq(acked, n_ptr->bclink.acked)) + return; + + spin_lock_bh(&bc_lock); + + /* Skip over packets that node has previously acknowledged */ + + crs = bcl->first_out; + while (crs && less_eq(buf_seqno(crs), n_ptr->bclink.acked)) { + crs = crs->next; + } + + /* Update packets that node is now acknowledging */ + + while (crs && less_eq(buf_seqno(crs), acked)) { + next = crs->next; + bcbuf_decr_acks(crs); + if (bcbuf_acks(crs) == 0) { + bcl->first_out = next; + bcl->out_queue_size--; + buf_discard(crs); + released = 1; + } + crs = next; + } + n_ptr->bclink.acked = acked; + + /* Try resolving broadcast link congestion, if necessary */ + + if (unlikely(bcl->next_out)) + link_push_queue(bcl); + if (unlikely(released && !list_empty(&bcl->waiting_ports))) + link_wakeup_ports(bcl, 0); + spin_unlock_bh(&bc_lock); +} + +/** + * bclink_send_ack - unicast an ACK msg + * + * net_lock and node lock set + */ + +static void bclink_send_ack(struct node *n_ptr) +{ + struct link *l_ptr = n_ptr->active_links[n_ptr->addr & 1]; + + if (l_ptr != NULL) + link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); +} + +/** + * bclink_send_nack- broadcast a NACK msg + * + * net_lock and node lock set + */ + +static void bclink_send_nack(struct node *n_ptr) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + + if (!less(n_ptr->bclink.gap_after, n_ptr->bclink.gap_to)) + return; + + buf = buf_acquire(INT_H_SIZE); + if (buf) { + msg = buf_msg(buf); + msg_init(msg, BCAST_PROTOCOL, STATE_MSG, + TIPC_OK, INT_H_SIZE, n_ptr->addr); + msg_set_mc_netid(msg, tipc_net_id); + msg_set_bcast_ack(msg, mod(n_ptr->bclink.last_in)); + msg_set_bcgap_after(msg, n_ptr->bclink.gap_after); + msg_set_bcgap_to(msg, n_ptr->bclink.gap_to); + msg_set_bcast_tag(msg, tipc_own_tag); + + if (bearer_send(&bcbearer->bearer, buf, 0)) { + bcl->stats.sent_nacks++; + buf_discard(buf); + } else { + bearer_schedule(bcl->b_ptr, bcl); + bcl->proto_msg_queue = buf; + bcl->stats.bearer_congs++; + } + + /* + * Ensure we doesn't send another NACK msg to the node + * until 16 more deferred messages arrive from it + * (i.e. helps prevent all nodes from NACK'ing at same time) + */ + + n_ptr->bclink.nack_sync = tipc_own_tag; + } +} + +/** + * bclink_check_gap - send a NACK if a sequence gap exists + * + * net_lock and node lock set + */ + +void bclink_check_gap(struct node *n_ptr, u32 last_sent) +{ + if (!n_ptr->bclink.supported || + less_eq(last_sent, mod(n_ptr->bclink.last_in))) + return; + + bclink_set_gap(n_ptr); + if (n_ptr->bclink.gap_after == n_ptr->bclink.gap_to) + n_ptr->bclink.gap_to = last_sent; + bclink_send_nack(n_ptr); +} + +/** + * bclink_peek_nack - process a NACK msg meant for another node + * + * Only net_lock set. + */ + +void bclink_peek_nack(u32 dest, u32 sender_tag, u32 gap_after, u32 gap_to) +{ + struct node *n_ptr = node_find(dest); + u32 my_after, my_to; + + if (unlikely(!n_ptr || !node_is_up(n_ptr))) + return; + node_lock(n_ptr); + /* + * Modify gap to suppress unnecessary NACKs from this node + */ + my_after = n_ptr->bclink.gap_after; + my_to = n_ptr->bclink.gap_to; + + if (less_eq(gap_after, my_after)) { + if (less(my_after, gap_to) && less(gap_to, my_to)) + n_ptr->bclink.gap_after = gap_to; + else if (less_eq(my_to, gap_to)) + n_ptr->bclink.gap_to = n_ptr->bclink.gap_after; + } else if (less_eq(gap_after, my_to)) { + if (less_eq(my_to, gap_to)) + n_ptr->bclink.gap_to = gap_after; + } else { + /* + * Expand gap if missing bufs not in deferred queue: + */ + struct sk_buff *buf = n_ptr->bclink.deferred_head; + u32 prev = n_ptr->bclink.gap_to; + + for (; buf; buf = buf->next) { + u32 seqno = buf_seqno(buf); + + if (mod(seqno - prev) != 1) + buf = NULL; + if (seqno == gap_after) + break; + prev = seqno; + } + if (buf == NULL) + n_ptr->bclink.gap_to = gap_after; + } + /* + * Some nodes may send a complementary NACK now: + */ + if (bclink_ack_allowed(sender_tag + 1)) { + if (n_ptr->bclink.gap_to != n_ptr->bclink.gap_after) { + bclink_send_nack(n_ptr); + bclink_set_gap(n_ptr); + } + } + node_unlock(n_ptr); +} + +/** + * bclink_send_msg - broadcast a packet to all nodes in cluster + */ + +int bclink_send_msg(struct sk_buff *buf) +{ + int res; + + spin_lock_bh(&bc_lock); + + res = link_send_buf(bcl, buf); + if (unlikely(res == -ELINKCONG)) + buf_discard(buf); + else + bcl->stats.sent_info++; + + if (bcl->out_queue_size > bcl->stats.max_queue_sz) + bcl->stats.max_queue_sz = bcl->out_queue_size; + bcl->stats.queue_sz_counts++; + bcl->stats.accu_queue_sz += bcl->out_queue_size; + + spin_unlock_bh(&bc_lock); + return res; +} + +/** + * bclink_recv_pkt - receive a broadcast packet, and deliver upwards + * + * net_lock is read_locked, no other locks set + */ + +void bclink_recv_pkt(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct node* node = node_find(msg_prevnode(msg)); + u32 next_in; + u32 seqno; + struct sk_buff *deferred; + + msg_dbg(msg, "<BC<<<"); + + if (unlikely(!node || !node_is_up(node) || !node->bclink.supported || + (msg_mc_netid(msg) != tipc_net_id))) { + buf_discard(buf); + return; + } + + if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) { + msg_dbg(msg, "<BCNACK<<<"); + if (msg_destnode(msg) == tipc_own_addr) { + node_lock(node); + bclink_acknowledge(node, msg_bcast_ack(msg)); + node_unlock(node); + bcl->stats.recv_nacks++; + bclink_retransmit_pkt(msg_bcgap_after(msg), + msg_bcgap_to(msg)); + } else { + bclink_peek_nack(msg_destnode(msg), + msg_bcast_tag(msg), + msg_bcgap_after(msg), + msg_bcgap_to(msg)); + } + buf_discard(buf); + return; + } + + node_lock(node); +receive: + deferred = node->bclink.deferred_head; + next_in = mod(node->bclink.last_in + 1); + seqno = msg_seqno(msg); + + if (likely(seqno == next_in)) { + bcl->stats.recv_info++; + node->bclink.last_in++; + bclink_set_gap(node); + if (unlikely(bclink_ack_allowed(seqno))) { + bclink_send_ack(node); + bcl->stats.sent_acks++; + } + if (likely(msg_isdata(msg))) { + node_unlock(node); + port_recv_mcast(buf, NULL); + } else if (msg_user(msg) == MSG_BUNDLER) { + bcl->stats.recv_bundles++; + bcl->stats.recv_bundled += msg_msgcnt(msg); + node_unlock(node); + link_recv_bundle(buf); + } else if (msg_user(msg) == MSG_FRAGMENTER) { + bcl->stats.recv_fragments++; + if (link_recv_fragment(&node->bclink.defragm, + &buf, &msg)) + bcl->stats.recv_fragmented++; + node_unlock(node); + net_route_msg(buf); + } else { + node_unlock(node); + net_route_msg(buf); + } + if (deferred && (buf_seqno(deferred) == mod(next_in + 1))) { + node_lock(node); + buf = deferred; + msg = buf_msg(buf); + node->bclink.deferred_head = deferred->next; + goto receive; + } + return; + } else if (less(next_in, seqno)) { + u32 gap_after = node->bclink.gap_after; + u32 gap_to = node->bclink.gap_to; + + if (link_defer_pkt(&node->bclink.deferred_head, + &node->bclink.deferred_tail, + buf)) { + node->bclink.nack_sync++; + bcl->stats.deferred_recv++; + if (seqno == mod(gap_after + 1)) + node->bclink.gap_after = seqno; + else if (less(gap_after, seqno) && less(seqno, gap_to)) + node->bclink.gap_to = seqno; + } + if (bclink_ack_allowed(node->bclink.nack_sync)) { + if (gap_to != gap_after) + bclink_send_nack(node); + bclink_set_gap(node); + } + } else { + bcl->stats.duplicates++; + buf_discard(buf); + } + node_unlock(node); +} + +u32 bclink_get_last_sent(void) +{ + u32 last_sent = mod(bcl->next_out_no - 1); + + if (bcl->next_out) + last_sent = mod(buf_seqno(bcl->next_out) - 1); + return last_sent; +} + +u32 bclink_acks_missing(struct node *n_ptr) +{ + return (n_ptr->bclink.supported && + (bclink_get_last_sent() != n_ptr->bclink.acked)); +} + + +/** + * bcbearer_send - send a packet through the broadcast pseudo-bearer + * + * Send through as many bearers as necessary to reach all nodes + * that support TIPC multicasting. + * + * Returns 0 if packet sent successfully, non-zero if not + */ + +int bcbearer_send(struct sk_buff *buf, + struct tipc_bearer *unused1, + struct tipc_media_addr *unused2) +{ + static int send_count = 0; + + struct node_map remains; + struct node_map remains_new; + int bp_index; + int swap_time; + + /* Prepare buffer for broadcasting (if first time trying to send it) */ + + if (likely(!msg_non_seq(buf_msg(buf)))) { + struct tipc_msg *msg; + + assert(cluster_bcast_nodes.count != 0); + bcbuf_set_acks(buf, cluster_bcast_nodes.count); + msg = buf_msg(buf); + msg_set_non_seq(msg); + msg_set_mc_netid(msg, tipc_net_id); + } + + /* Determine if bearer pairs should be swapped following this attempt */ + + if ((swap_time = (++send_count >= 10))) + send_count = 0; + + /* Send buffer over bearers until all targets reached */ + + remains = cluster_bcast_nodes; + + for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) { + struct bearer *p = bcbearer->bpairs[bp_index].primary; + struct bearer *s = bcbearer->bpairs[bp_index].secondary; + + if (!p) + break; /* no more bearers to try */ + + nmap_diff(&remains, &p->nodes, &remains_new); + if (remains_new.count == remains.count) + continue; /* bearer pair doesn't add anything */ + + if (!p->publ.blocked && + !p->media->send_msg(buf, &p->publ, &p->media->bcast_addr)) { + if (swap_time && s && !s->publ.blocked) + goto swap; + else + goto update; + } + + if (!s || s->publ.blocked || + s->media->send_msg(buf, &s->publ, &s->media->bcast_addr)) + continue; /* unable to send using bearer pair */ +swap: + bcbearer->bpairs[bp_index].primary = s; + bcbearer->bpairs[bp_index].secondary = p; +update: + if (remains_new.count == 0) + return TIPC_OK; + + remains = remains_new; + } + + /* Unable to reach all targets */ + + bcbearer->bearer.publ.blocked = 1; + bcl->stats.bearer_congs++; + return ~TIPC_OK; +} + +/** + * bcbearer_sort - create sets of bearer pairs used by broadcast bearer + */ + +void bcbearer_sort(void) +{ + struct bcbearer_pair *bp_temp = bcbearer->bpairs_temp; + struct bcbearer_pair *bp_curr; + int b_index; + int pri; + + spin_lock_bh(&bc_lock); + + /* Group bearers by priority (can assume max of two per priority) */ + + memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp)); + + for (b_index = 0; b_index < MAX_BEARERS; b_index++) { + struct bearer *b = &bearers[b_index]; + + if (!b->active || !b->nodes.count) + continue; + + if (!bp_temp[b->priority].primary) + bp_temp[b->priority].primary = b; + else + bp_temp[b->priority].secondary = b; + } + + /* Create array of bearer pairs for broadcasting */ + + bp_curr = bcbearer->bpairs; + memset(bcbearer->bpairs, 0, sizeof(bcbearer->bpairs)); + + for (pri = (TIPC_NUM_LINK_PRI - 1); pri >= 0; pri--) { + + if (!bp_temp[pri].primary) + continue; + + bp_curr->primary = bp_temp[pri].primary; + + if (bp_temp[pri].secondary) { + if (nmap_equal(&bp_temp[pri].primary->nodes, + &bp_temp[pri].secondary->nodes)) { + bp_curr->secondary = bp_temp[pri].secondary; + } else { + bp_curr++; + bp_curr->primary = bp_temp[pri].secondary; + } + } + + bp_curr++; + } + + spin_unlock_bh(&bc_lock); +} + +/** + * bcbearer_push - resolve bearer congestion + * + * Forces bclink to push out any unsent packets, until all packets are gone + * or congestion reoccurs. + * No locks set when function called + */ + +void bcbearer_push(void) +{ + struct bearer *b_ptr; + + spin_lock_bh(&bc_lock); + b_ptr = &bcbearer->bearer; + if (b_ptr->publ.blocked) { + b_ptr->publ.blocked = 0; + bearer_lock_push(b_ptr); + } + spin_unlock_bh(&bc_lock); +} + + +int bclink_stats(char *buf, const u32 buf_size) +{ + struct print_buf pb; + + if (!bcl) + return 0; + + printbuf_init(&pb, buf, buf_size); + + spin_lock_bh(&bc_lock); + + tipc_printf(&pb, "Link <%s>\n" + " Window:%u packets\n", + bcl->name, bcl->queue_limit[0]); + tipc_printf(&pb, " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + bcl->stats.recv_info, + bcl->stats.recv_fragments, + bcl->stats.recv_fragmented, + bcl->stats.recv_bundles, + bcl->stats.recv_bundled); + tipc_printf(&pb, " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + bcl->stats.sent_info, + bcl->stats.sent_fragments, + bcl->stats.sent_fragmented, + bcl->stats.sent_bundles, + bcl->stats.sent_bundled); + tipc_printf(&pb, " RX naks:%u defs:%u dups:%u\n", + bcl->stats.recv_nacks, + bcl->stats.deferred_recv, + bcl->stats.duplicates); + tipc_printf(&pb, " TX naks:%u acks:%u dups:%u\n", + bcl->stats.sent_nacks, + bcl->stats.sent_acks, + bcl->stats.retransmitted); + tipc_printf(&pb, " Congestion bearer:%u link:%u Send queue max:%u avg:%u\n", + bcl->stats.bearer_congs, + bcl->stats.link_congs, + bcl->stats.max_queue_sz, + bcl->stats.queue_sz_counts + ? (bcl->stats.accu_queue_sz / bcl->stats.queue_sz_counts) + : 0); + + spin_unlock_bh(&bc_lock); + return printbuf_validate(&pb); +} + +int bclink_reset_stats(void) +{ + if (!bcl) + return -ENOPROTOOPT; + + spin_lock_bh(&bc_lock); + memset(&bcl->stats, 0, sizeof(bcl->stats)); + spin_unlock_bh(&bc_lock); + return TIPC_OK; +} + +int bclink_set_queue_limits(u32 limit) +{ + if (!bcl) + return -ENOPROTOOPT; + if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN)) + return -EINVAL; + + spin_lock_bh(&bc_lock); + link_set_queue_limits(bcl, limit); + spin_unlock_bh(&bc_lock); + return TIPC_OK; +} + +int bclink_init(void) +{ + bcbearer = kmalloc(sizeof(*bcbearer), GFP_ATOMIC); + bclink = kmalloc(sizeof(*bclink), GFP_ATOMIC); + if (!bcbearer || !bclink) { + nomem: + warn("Memory squeeze; Failed to create multicast link\n"); + kfree(bcbearer); + bcbearer = NULL; + kfree(bclink); + bclink = NULL; + return -ENOMEM; + } + + memset(bcbearer, 0, sizeof(struct bcbearer)); + INIT_LIST_HEAD(&bcbearer->bearer.cong_links); + bcbearer->bearer.media = &bcbearer->media; + bcbearer->media.send_msg = bcbearer_send; + sprintf(bcbearer->media.name, "tipc-multicast"); + + bcl = &bclink->link; + memset(bclink, 0, sizeof(struct bclink)); + INIT_LIST_HEAD(&bcl->waiting_ports); + bcl->next_out_no = 1; + bclink->node.lock = SPIN_LOCK_UNLOCKED; + bcl->owner = &bclink->node; + bcl->max_pkt = MAX_PKT_DEFAULT_MCAST; + link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT); + bcl->b_ptr = &bcbearer->bearer; + bcl->state = WORKING_WORKING; + sprintf(bcl->name, bc_link_name); + + if (BCLINK_LOG_BUF_SIZE) { + char *pb = kmalloc(BCLINK_LOG_BUF_SIZE, GFP_ATOMIC); + + if (!pb) + goto nomem; + printbuf_init(&bcl->print_buf, pb, BCLINK_LOG_BUF_SIZE); + } + + return TIPC_OK; +} + +void bclink_stop(void) +{ + spin_lock_bh(&bc_lock); + if (bcbearer) { + link_stop(bcl); + if (BCLINK_LOG_BUF_SIZE) + kfree(bcl->print_buf.buf); + bcl = NULL; + kfree(bclink); + bclink = NULL; + kfree(bcbearer); + bcbearer = NULL; + } + spin_unlock_bh(&bc_lock); +} + diff --git a/net/tipc/bcast.h b/net/tipc/bcast.h new file mode 100644 index 00000000000..5430e524b4f --- /dev/null +++ b/net/tipc/bcast.h @@ -0,0 +1,223 @@ +/* + * net/tipc/bcast.h: Include file for TIPC broadcast code + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_BCAST_H +#define _TIPC_BCAST_H + +#define MAX_NODES 4096 +#define WSIZE 32 + +/** + * struct node_map - set of node identifiers + * @count: # of nodes in set + * @map: bitmap of node identifiers that are in the set + */ + +struct node_map { + u32 count; + u32 map[MAX_NODES / WSIZE]; +}; + + +#define PLSIZE 32 + +/** + * struct port_list - set of node local destination ports + * @count: # of ports in set (only valid for first entry in list) + * @next: pointer to next entry in list + * @ports: array of port references + */ + +struct port_list { + int count; + struct port_list *next; + u32 ports[PLSIZE]; +}; + + +struct node; + +extern char bc_link_name[]; + + +/** + * nmap_get - determine if node exists in a node map + */ + +static inline int nmap_get(struct node_map *nm_ptr, u32 node) +{ + int n = tipc_node(node); + int w = n / WSIZE; + int b = n % WSIZE; + + return nm_ptr->map[w] & (1 << b); +} + +/** + * nmap_add - add a node to a node map + */ + +static inline void nmap_add(struct node_map *nm_ptr, u32 node) +{ + int n = tipc_node(node); + int w = n / WSIZE; + u32 mask = (1 << (n % WSIZE)); + + if ((nm_ptr->map[w] & mask) == 0) { + nm_ptr->count++; + nm_ptr->map[w] |= mask; + } +} + +/** + * nmap_remove - remove a node from a node map + */ + +static inline void nmap_remove(struct node_map *nm_ptr, u32 node) +{ + int n = tipc_node(node); + int w = n / WSIZE; + u32 mask = (1 << (n % WSIZE)); + + if ((nm_ptr->map[w] & mask) != 0) { + nm_ptr->map[w] &= ~mask; + nm_ptr->count--; + } +} + +/** + * nmap_equal - test for equality of node maps + */ + +static inline int nmap_equal(struct node_map *nm_a, struct node_map *nm_b) +{ + return !memcmp(nm_a, nm_b, sizeof(*nm_a)); +} + +/** + * nmap_diff - find differences between node maps + * @nm_a: input node map A + * @nm_b: input node map B + * @nm_diff: output node map A-B (i.e. nodes of A that are not in B) + */ + +static inline void nmap_diff(struct node_map *nm_a, struct node_map *nm_b, + struct node_map *nm_diff) +{ + int stop = sizeof(nm_a->map) / sizeof(u32); + int w; + int b; + u32 map; + + memset(nm_diff, 0, sizeof(*nm_diff)); + for (w = 0; w < stop; w++) { + map = nm_a->map[w] ^ (nm_a->map[w] & nm_b->map[w]); + nm_diff->map[w] = map; + if (map != 0) { + for (b = 0 ; b < WSIZE; b++) { + if (map & (1 << b)) + nm_diff->count++; + } + } + } +} + +/** + * port_list_add - add a port to a port list, ensuring no duplicates + */ + +static inline void port_list_add(struct port_list *pl_ptr, u32 port) +{ + struct port_list *item = pl_ptr; + int i; + int item_sz = PLSIZE; + int cnt = pl_ptr->count; + + for (; ; cnt -= item_sz, item = item->next) { + if (cnt < PLSIZE) + item_sz = cnt; + for (i = 0; i < item_sz; i++) + if (item->ports[i] == port) + return; + if (i < PLSIZE) { + item->ports[i] = port; + pl_ptr->count++; + return; + } + if (!item->next) { + item->next = kmalloc(sizeof(*item), GFP_ATOMIC); + if (!item->next) { + warn("Memory squeeze: multicast destination port list is incomplete\n"); + return; + } + item->next->next = NULL; + } + } +} + +/** + * port_list_free - free dynamically created entries in port_list chain + * + * Note: First item is on stack, so it doesn't need to be released + */ + +static inline void port_list_free(struct port_list *pl_ptr) +{ + struct port_list *item; + struct port_list *next; + + for (item = pl_ptr->next; item; item = next) { + next = item->next; + kfree(item); + } +} + + +int bclink_init(void); +void bclink_stop(void); +void bclink_acknowledge(struct node *n_ptr, u32 acked); +int bclink_send_msg(struct sk_buff *buf); +void bclink_recv_pkt(struct sk_buff *buf); +u32 bclink_get_last_sent(void); +u32 bclink_acks_missing(struct node *n_ptr); +void bclink_check_gap(struct node *n_ptr, u32 seqno); +int bclink_stats(char *stats_buf, const u32 buf_size); +int bclink_reset_stats(void); +int bclink_set_queue_limits(u32 limit); +void bcbearer_sort(void); +void bcbearer_push(void); + +#endif diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c new file mode 100644 index 00000000000..3dd19fdc5a2 --- /dev/null +++ b/net/tipc/bearer.c @@ -0,0 +1,692 @@ +/* + * net/tipc/bearer.c: TIPC bearer code + * + * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "dbg.h" +#include "bearer.h" +#include "link.h" +#include "port.h" +#include "discover.h" +#include "bcast.h" + +#define MAX_ADDR_STR 32 + +static struct media *media_list = 0; +static u32 media_count = 0; + +struct bearer *bearers = 0; + +/** + * media_name_valid - validate media name + * + * Returns 1 if media name is valid, otherwise 0. + */ + +static int media_name_valid(const char *name) +{ + u32 len; + + len = strlen(name); + if ((len + 1) > TIPC_MAX_MEDIA_NAME) + return 0; + return (strspn(name, tipc_alphabet) == len); +} + +/** + * media_find - locates specified media object by name + */ + +static struct media *media_find(const char *name) +{ + struct media *m_ptr; + u32 i; + + for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { + if (!strcmp(m_ptr->name, name)) + return m_ptr; + } + return 0; +} + +/** + * tipc_register_media - register a media type + * + * Bearers for this media type must be activated separately at a later stage. + */ + +int tipc_register_media(u32 media_type, + char *name, + int (*enable)(struct tipc_bearer *), + void (*disable)(struct tipc_bearer *), + int (*send_msg)(struct sk_buff *, + struct tipc_bearer *, + struct tipc_media_addr *), + char *(*addr2str)(struct tipc_media_addr *a, + char *str_buf, int str_size), + struct tipc_media_addr *bcast_addr, + const u32 bearer_priority, + const u32 link_tolerance, /* [ms] */ + const u32 send_window_limit) +{ + struct media *m_ptr; + u32 media_id; + u32 i; + int res = -EINVAL; + + write_lock_bh(&net_lock); + if (!media_list) + goto exit; + + if (!media_name_valid(name)) { + warn("Media registration error: illegal name <%s>\n", name); + goto exit; + } + if (!bcast_addr) { + warn("Media registration error: no broadcast address supplied\n"); + goto exit; + } + if (bearer_priority >= TIPC_NUM_LINK_PRI) { + warn("Media registration error: priority %u\n", bearer_priority); + goto exit; + } + if ((link_tolerance < TIPC_MIN_LINK_TOL) || + (link_tolerance > TIPC_MAX_LINK_TOL)) { + warn("Media registration error: tolerance %u\n", link_tolerance); + goto exit; + } + + media_id = media_count++; + if (media_id >= MAX_MEDIA) { + warn("Attempt to register more than %u media\n", MAX_MEDIA); + media_count--; + goto exit; + } + for (i = 0; i < media_id; i++) { + if (media_list[i].type_id == media_type) { + warn("Attempt to register second media with type %u\n", + media_type); + media_count--; + goto exit; + } + if (!strcmp(name, media_list[i].name)) { + warn("Attempt to re-register media name <%s>\n", name); + media_count--; + goto exit; + } + } + + m_ptr = &media_list[media_id]; + m_ptr->type_id = media_type; + m_ptr->send_msg = send_msg; + m_ptr->enable_bearer = enable; + m_ptr->disable_bearer = disable; + m_ptr->addr2str = addr2str; + memcpy(&m_ptr->bcast_addr, bcast_addr, sizeof(*bcast_addr)); + m_ptr->bcast = 1; + strcpy(m_ptr->name, name); + m_ptr->priority = bearer_priority; + m_ptr->tolerance = link_tolerance; + m_ptr->window = send_window_limit; + dbg("Media <%s> registered\n", name); + res = 0; +exit: + write_unlock_bh(&net_lock); + return res; +} + +/** + * media_addr_printf - record media address in print buffer + */ + +void media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a) +{ + struct media *m_ptr; + u32 media_type; + u32 i; + + media_type = ntohl(a->type); + for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { + if (m_ptr->type_id == media_type) + break; + } + + if ((i < media_count) && (m_ptr->addr2str != NULL)) { + char addr_str[MAX_ADDR_STR]; + + tipc_printf(pb, "%s(%s) ", m_ptr->name, + m_ptr->addr2str(a, addr_str, sizeof(addr_str))); + } else { + unchar *addr = (unchar *)&a->dev_addr; + + tipc_printf(pb, "UNKNOWN(%u):", media_type); + for (i = 0; i < (sizeof(*a) - sizeof(a->type)); i++) { + tipc_printf(pb, "%02x ", addr[i]); + } + } +} + +/** + * media_get_names - record names of registered media in buffer + */ + +struct sk_buff *media_get_names(void) +{ + struct sk_buff *buf; + struct media *m_ptr; + int i; + + buf = cfg_reply_alloc(MAX_MEDIA * TLV_SPACE(TIPC_MAX_MEDIA_NAME)); + if (!buf) + return NULL; + + read_lock_bh(&net_lock); + for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { + cfg_append_tlv(buf, TIPC_TLV_MEDIA_NAME, m_ptr->name, + strlen(m_ptr->name) + 1); + } + read_unlock_bh(&net_lock); + return buf; +} + +/** + * bearer_name_validate - validate & (optionally) deconstruct bearer name + * @name - ptr to bearer name string + * @name_parts - ptr to area for bearer name components (or NULL if not needed) + * + * Returns 1 if bearer name is valid, otherwise 0. + */ + +static int bearer_name_validate(const char *name, + struct bearer_name *name_parts) +{ + char name_copy[TIPC_MAX_BEARER_NAME]; + char *media_name; + char *if_name; + u32 media_len; + u32 if_len; + + /* copy bearer name & ensure length is OK */ + + name_copy[TIPC_MAX_BEARER_NAME - 1] = 0; + /* need above in case non-Posix strncpy() doesn't pad with nulls */ + strncpy(name_copy, name, TIPC_MAX_BEARER_NAME); + if (name_copy[TIPC_MAX_BEARER_NAME - 1] != 0) + return 0; + + /* ensure all component parts of bearer name are present */ + + media_name = name_copy; + if ((if_name = strchr(media_name, ':')) == NULL) + return 0; + *(if_name++) = 0; + media_len = if_name - media_name; + if_len = strlen(if_name) + 1; + + /* validate component parts of bearer name */ + + if ((media_len <= 1) || (media_len > TIPC_MAX_MEDIA_NAME) || + (if_len <= 1) || (if_len > TIPC_MAX_IF_NAME) || + (strspn(media_name, tipc_alphabet) != (media_len - 1)) || + (strspn(if_name, tipc_alphabet) != (if_len - 1))) + return 0; + + /* return bearer name components, if necessary */ + + if (name_parts) { + strcpy(name_parts->media_name, media_name); + strcpy(name_parts->if_name, if_name); + } + return 1; +} + +/** + * bearer_find - locates bearer object with matching bearer name + */ + +static struct bearer *bearer_find(const char *name) +{ + struct bearer *b_ptr; + u32 i; + + for (i = 0, b_ptr = bearers; i < MAX_BEARERS; i++, b_ptr++) { + if (b_ptr->active && (!strcmp(b_ptr->publ.name, name))) + return b_ptr; + } + return 0; +} + +/** + * bearer_find - locates bearer object with matching interface name + */ + +struct bearer *bearer_find_interface(const char *if_name) +{ + struct bearer *b_ptr; + char *b_if_name; + u32 i; + + for (i = 0, b_ptr = bearers; i < MAX_BEARERS; i++, b_ptr++) { + if (!b_ptr->active) + continue; + b_if_name = strchr(b_ptr->publ.name, ':') + 1; + if (!strcmp(b_if_name, if_name)) + return b_ptr; + } + return 0; +} + +/** + * bearer_get_names - record names of bearers in buffer + */ + +struct sk_buff *bearer_get_names(void) +{ + struct sk_buff *buf; + struct media *m_ptr; + struct bearer *b_ptr; + int i, j; + + buf = cfg_reply_alloc(MAX_BEARERS * TLV_SPACE(TIPC_MAX_BEARER_NAME)); + if (!buf) + return NULL; + + read_lock_bh(&net_lock); + for (i = 0, m_ptr = media_list; i < media_count; i++, m_ptr++) { + for (j = 0; j < MAX_BEARERS; j++) { + b_ptr = &bearers[j]; + if (b_ptr->active && (b_ptr->media == m_ptr)) { + cfg_append_tlv(buf, TIPC_TLV_BEARER_NAME, + b_ptr->publ.name, + strlen(b_ptr->publ.name) + 1); + } + } + } + read_unlock_bh(&net_lock); + return buf; +} + +void bearer_add_dest(struct bearer *b_ptr, u32 dest) +{ + nmap_add(&b_ptr->nodes, dest); + disc_update_link_req(b_ptr->link_req); + bcbearer_sort(); +} + +void bearer_remove_dest(struct bearer *b_ptr, u32 dest) +{ + nmap_remove(&b_ptr->nodes, dest); + disc_update_link_req(b_ptr->link_req); + bcbearer_sort(); +} + +/* + * bearer_push(): Resolve bearer congestion. Force the waiting + * links to push out their unsent packets, one packet per link + * per iteration, until all packets are gone or congestion reoccurs. + * 'net_lock' is read_locked when this function is called + * bearer.lock must be taken before calling + * Returns binary true(1) ore false(0) + */ +static int bearer_push(struct bearer *b_ptr) +{ + u32 res = TIPC_OK; + struct link *ln, *tln; + + if (b_ptr->publ.blocked) + return 0; + + while (!list_empty(&b_ptr->cong_links) && (res != PUSH_FAILED)) { + list_for_each_entry_safe(ln, tln, &b_ptr->cong_links, link_list) { + res = link_push_packet(ln); + if (res == PUSH_FAILED) + break; + if (res == PUSH_FINISHED) + list_move_tail(&ln->link_list, &b_ptr->links); + } + } + return list_empty(&b_ptr->cong_links); +} + +void bearer_lock_push(struct bearer *b_ptr) +{ + int res; + + spin_lock_bh(&b_ptr->publ.lock); + res = bearer_push(b_ptr); + spin_unlock_bh(&b_ptr->publ.lock); + if (res) + bcbearer_push(); +} + + +/* + * Interrupt enabling new requests after bearer congestion or blocking: + * See bearer_send(). + */ +void tipc_continue(struct tipc_bearer *tb_ptr) +{ + struct bearer *b_ptr = (struct bearer *)tb_ptr; + + spin_lock_bh(&b_ptr->publ.lock); + b_ptr->continue_count++; + if (!list_empty(&b_ptr->cong_links)) + k_signal((Handler)bearer_lock_push, (unsigned long)b_ptr); + b_ptr->publ.blocked = 0; + spin_unlock_bh(&b_ptr->publ.lock); +} + +/* + * Schedule link for sending of messages after the bearer + * has been deblocked by 'continue()'. This method is called + * when somebody tries to send a message via this link while + * the bearer is congested. 'net_lock' is in read_lock here + * bearer.lock is busy + */ + +static void bearer_schedule_unlocked(struct bearer *b_ptr, struct link *l_ptr) +{ + list_move_tail(&l_ptr->link_list, &b_ptr->cong_links); +} + +/* + * Schedule link for sending of messages after the bearer + * has been deblocked by 'continue()'. This method is called + * when somebody tries to send a message via this link while + * the bearer is congested. 'net_lock' is in read_lock here, + * bearer.lock is free + */ + +void bearer_schedule(struct bearer *b_ptr, struct link *l_ptr) +{ + spin_lock_bh(&b_ptr->publ.lock); + bearer_schedule_unlocked(b_ptr, l_ptr); + spin_unlock_bh(&b_ptr->publ.lock); +} + + +/* + * bearer_resolve_congestion(): Check if there is bearer congestion, + * and if there is, try to resolve it before returning. + * 'net_lock' is read_locked when this function is called + */ +int bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr) +{ + int res = 1; + + if (list_empty(&b_ptr->cong_links)) + return 1; + spin_lock_bh(&b_ptr->publ.lock); + if (!bearer_push(b_ptr)) { + bearer_schedule_unlocked(b_ptr, l_ptr); + res = 0; + } + spin_unlock_bh(&b_ptr->publ.lock); + return res; +} + + +/** + * tipc_enable_bearer - enable bearer with the given name + */ + +int tipc_enable_bearer(const char *name, u32 bcast_scope, u32 priority) +{ + struct bearer *b_ptr; + struct media *m_ptr; + struct bearer_name b_name; + char addr_string[16]; + u32 bearer_id; + u32 with_this_prio; + u32 i; + int res = -EINVAL; + + if (tipc_mode != TIPC_NET_MODE) + return -ENOPROTOOPT; + if (!bearer_name_validate(name, &b_name) || + !addr_domain_valid(bcast_scope) || + !in_scope(bcast_scope, tipc_own_addr) || + (priority > TIPC_NUM_LINK_PRI)) + return -EINVAL; + + write_lock_bh(&net_lock); + if (!bearers) + goto failed; + + m_ptr = media_find(b_name.media_name); + if (!m_ptr) { + warn("No media <%s>\n", b_name.media_name); + goto failed; + } + if (priority == TIPC_NUM_LINK_PRI) + priority = m_ptr->priority; + +restart: + bearer_id = MAX_BEARERS; + with_this_prio = 1; + for (i = MAX_BEARERS; i-- != 0; ) { + if (!bearers[i].active) { + bearer_id = i; + continue; + } + if (!strcmp(name, bearers[i].publ.name)) { + warn("Bearer <%s> already enabled\n", name); + goto failed; + } + if ((bearers[i].priority == priority) && + (++with_this_prio > 2)) { + if (priority-- == 0) { + warn("Third bearer <%s> with priority %u, unable to lower to %u\n", + name, priority + 1, priority); + goto failed; + } + warn("Third bearer <%s> with priority %u, lowering to %u\n", + name, priority + 1, priority); + goto restart; + } + } + if (bearer_id >= MAX_BEARERS) { + warn("Attempt to enable more than %d bearers\n", MAX_BEARERS); + goto failed; + } + + b_ptr = &bearers[bearer_id]; + memset(b_ptr, 0, sizeof(struct bearer)); + + strcpy(b_ptr->publ.name, name); + res = m_ptr->enable_bearer(&b_ptr->publ); + if (res) { + warn("Failed to enable bearer <%s>\n", name); + goto failed; + } + + b_ptr->identity = bearer_id; + b_ptr->media = m_ptr; + b_ptr->net_plane = bearer_id + 'A'; + b_ptr->active = 1; + b_ptr->detect_scope = bcast_scope; + b_ptr->priority = priority; + INIT_LIST_HEAD(&b_ptr->cong_links); + INIT_LIST_HEAD(&b_ptr->links); + if (m_ptr->bcast) { + b_ptr->link_req = disc_init_link_req(b_ptr, &m_ptr->bcast_addr, + bcast_scope, 2); + } + b_ptr->publ.lock = SPIN_LOCK_UNLOCKED; + write_unlock_bh(&net_lock); + info("Enabled bearer <%s>, discovery domain %s\n", + name, addr_string_fill(addr_string, bcast_scope)); + return 0; +failed: + write_unlock_bh(&net_lock); + return res; +} + +/** + * tipc_block_bearer(): Block the bearer with the given name, + * and reset all its links + */ + +int tipc_block_bearer(const char *name) +{ + struct bearer *b_ptr = 0; + struct link *l_ptr; + struct link *temp_l_ptr; + + if (tipc_mode != TIPC_NET_MODE) + return -ENOPROTOOPT; + + read_lock_bh(&net_lock); + b_ptr = bearer_find(name); + if (!b_ptr) { + warn("Attempt to block unknown bearer <%s>\n", name); + read_unlock_bh(&net_lock); + return -EINVAL; + } + + spin_lock_bh(&b_ptr->publ.lock); + b_ptr->publ.blocked = 1; + list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { + struct node *n_ptr = l_ptr->owner; + + spin_lock_bh(&n_ptr->lock); + link_reset(l_ptr); + spin_unlock_bh(&n_ptr->lock); + } + spin_unlock_bh(&b_ptr->publ.lock); + read_unlock_bh(&net_lock); + info("Blocked bearer <%s>\n", name); + return TIPC_OK; +} + +/** + * bearer_disable - + * + * Note: This routine assumes caller holds net_lock. + */ + +static int bearer_disable(const char *name) +{ + struct bearer *b_ptr; + struct link *l_ptr; + struct link *temp_l_ptr; + + if (tipc_mode != TIPC_NET_MODE) + return -ENOPROTOOPT; + + b_ptr = bearer_find(name); + if (!b_ptr) { + warn("Attempt to disable unknown bearer <%s>\n", name); + return -EINVAL; + } + + disc_stop_link_req(b_ptr->link_req); + spin_lock_bh(&b_ptr->publ.lock); + b_ptr->link_req = NULL; + b_ptr->publ.blocked = 1; + if (b_ptr->media->disable_bearer) { + spin_unlock_bh(&b_ptr->publ.lock); + write_unlock_bh(&net_lock); + b_ptr->media->disable_bearer(&b_ptr->publ); + write_lock_bh(&net_lock); + spin_lock_bh(&b_ptr->publ.lock); + } + list_for_each_entry_safe(l_ptr, temp_l_ptr, &b_ptr->links, link_list) { + link_delete(l_ptr); + } + spin_unlock_bh(&b_ptr->publ.lock); + info("Disabled bearer <%s>\n", name); + memset(b_ptr, 0, sizeof(struct bearer)); + return TIPC_OK; +} + +int tipc_disable_bearer(const char *name) +{ + int res; + + write_lock_bh(&net_lock); + res = bearer_disable(name); + write_unlock_bh(&net_lock); + return res; +} + + + +int bearer_init(void) +{ + int res; + + write_lock_bh(&net_lock); + bearers = kmalloc(MAX_BEARERS * sizeof(struct bearer), GFP_ATOMIC); + media_list = kmalloc(MAX_MEDIA * sizeof(struct media), GFP_ATOMIC); + if (bearers && media_list) { + memset(bearers, 0, MAX_BEARERS * sizeof(struct bearer)); + memset(media_list, 0, MAX_MEDIA * sizeof(struct media)); + res = TIPC_OK; + } else { + kfree(bearers); + kfree(media_list); + bearers = 0; + media_list = 0; + res = -ENOMEM; + } + write_unlock_bh(&net_lock); + return res; +} + +void bearer_stop(void) +{ + u32 i; + + if (!bearers) + return; + + for (i = 0; i < MAX_BEARERS; i++) { + if (bearers[i].active) + bearers[i].publ.blocked = 1; + } + for (i = 0; i < MAX_BEARERS; i++) { + if (bearers[i].active) + bearer_disable(bearers[i].publ.name); + } + kfree(bearers); + kfree(media_list); + bearers = 0; + media_list = 0; + media_count = 0; +} + + diff --git a/net/tipc/bearer.h b/net/tipc/bearer.h new file mode 100644 index 00000000000..21e63d3f018 --- /dev/null +++ b/net/tipc/bearer.h @@ -0,0 +1,172 @@ +/* + * net/tipc/bearer.h: Include file for TIPC bearer code + * + * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_BEARER_H +#define _TIPC_BEARER_H + +#include <net/tipc/tipc_bearer.h> +#include "bcast.h" + +#define MAX_BEARERS 8 +#define MAX_MEDIA 4 + + +/** + * struct media - TIPC media information available to internal users + * @send_msg: routine which handles buffer transmission + * @enable_bearer: routine which enables a bearer + * @disable_bearer: routine which disables a bearer + * @addr2str: routine which converts bearer's address to string form + * @bcast_addr: media address used in broadcasting + * @bcast: non-zero if media supports broadcasting [currently mandatory] + * @priority: default link (and bearer) priority + * @tolerance: default time (in ms) before declaring link failure + * @window: default window (in packets) before declaring link congestion + * @type_id: TIPC media identifier [defined in tipc_bearer.h] + * @name: media name + */ + +struct media { + int (*send_msg)(struct sk_buff *buf, + struct tipc_bearer *b_ptr, + struct tipc_media_addr *dest); + int (*enable_bearer)(struct tipc_bearer *b_ptr); + void (*disable_bearer)(struct tipc_bearer *b_ptr); + char *(*addr2str)(struct tipc_media_addr *a, + char *str_buf, int str_size); + struct tipc_media_addr bcast_addr; + int bcast; + u32 priority; + u32 tolerance; + u32 window; + u32 type_id; + char name[TIPC_MAX_MEDIA_NAME]; +}; + +/** + * struct bearer - TIPC bearer information available to internal users + * @publ: bearer information available to privileged users + * @media: ptr to media structure associated with bearer + * @priority: default link priority for bearer + * @detect_scope: network address mask used during automatic link creation + * @identity: array index of this bearer within TIPC bearer array + * @link_req: ptr to (optional) structure making periodic link setup requests + * @links: list of non-congested links associated with bearer + * @cong_links: list of congested links associated with bearer + * @continue_count: # of times bearer has resumed after congestion or blocking + * @active: non-zero if bearer structure is represents a bearer + * @net_plane: network plane ('A' through 'H') currently associated with bearer + * @nodes: indicates which nodes in cluster can be reached through bearer + */ + +struct bearer { + struct tipc_bearer publ; + struct media *media; + u32 priority; + u32 detect_scope; + u32 identity; + struct link_req *link_req; + struct list_head links; + struct list_head cong_links; + u32 continue_count; + int active; + char net_plane; + struct node_map nodes; +}; + +struct bearer_name { + char media_name[TIPC_MAX_MEDIA_NAME]; + char if_name[TIPC_MAX_IF_NAME]; +}; + +struct link; + +extern struct bearer *bearers; + +void media_addr_printf(struct print_buf *pb, struct tipc_media_addr *a); +struct sk_buff *media_get_names(void); + +struct sk_buff *bearer_get_names(void); +void bearer_add_dest(struct bearer *b_ptr, u32 dest); +void bearer_remove_dest(struct bearer *b_ptr, u32 dest); +void bearer_schedule(struct bearer *b_ptr, struct link *l_ptr); +struct bearer *bearer_find_interface(const char *if_name); +int bearer_resolve_congestion(struct bearer *b_ptr, struct link *l_ptr); +int bearer_init(void); +void bearer_stop(void); +int bearer_broadcast(struct sk_buff *buf, struct tipc_bearer *b_ptr, + struct tipc_media_addr *dest); +void bearer_lock_push(struct bearer *b_ptr); + + +/** + * bearer_send- sends buffer to destination over bearer + * + * Returns true (1) if successful, or false (0) if unable to send + * + * IMPORTANT: + * The media send routine must not alter the buffer being passed in + * as it may be needed for later retransmission! + * + * If the media send routine returns a non-zero value (indicating that + * it was unable to send the buffer), it must: + * 1) mark the bearer as blocked, + * 2) call tipc_continue() once the bearer is able to send again. + * Media types that are unable to meet these two critera must ensure their + * send routine always returns success -- even if the buffer was not sent -- + * and let TIPC's link code deal with the undelivered message. + */ + +static inline int bearer_send(struct bearer *b_ptr, struct sk_buff *buf, + struct tipc_media_addr *dest) +{ + return !b_ptr->media->send_msg(buf, &b_ptr->publ, dest); +} + +/** + * bearer_congested - determines if bearer is currently congested + */ + +static inline int bearer_congested(struct bearer *b_ptr, struct link *l_ptr) +{ + if (unlikely(b_ptr->publ.blocked)) + return 1; + if (likely(list_empty(&b_ptr->cong_links))) + return 0; + return !bearer_resolve_congestion(b_ptr, l_ptr); +} + +#endif diff --git a/net/tipc/cluster.c b/net/tipc/cluster.c new file mode 100644 index 00000000000..f0f7bac51d4 --- /dev/null +++ b/net/tipc/cluster.c @@ -0,0 +1,576 @@ +/* + * net/tipc/cluster.c: TIPC cluster management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "cluster.h" +#include "addr.h" +#include "node_subscr.h" +#include "link.h" +#include "node.h" +#include "net.h" +#include "msg.h" +#include "bearer.h" + +void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, + u32 lower, u32 upper); +struct sk_buff *cluster_prepare_routing_msg(u32 data_size, u32 dest); + +struct node **local_nodes = 0; +struct node_map cluster_bcast_nodes = {0,{0,}}; +u32 highest_allowed_slave = 0; + +struct cluster *cluster_create(u32 addr) +{ + struct _zone *z_ptr; + struct cluster *c_ptr; + int max_nodes; + int alloc; + + c_ptr = (struct cluster *)kmalloc(sizeof(*c_ptr), GFP_ATOMIC); + if (c_ptr == NULL) + return 0; + memset(c_ptr, 0, sizeof(*c_ptr)); + + c_ptr->addr = tipc_addr(tipc_zone(addr), tipc_cluster(addr), 0); + if (in_own_cluster(addr)) + max_nodes = LOWEST_SLAVE + tipc_max_slaves; + else + max_nodes = tipc_max_nodes + 1; + alloc = sizeof(void *) * (max_nodes + 1); + c_ptr->nodes = (struct node **)kmalloc(alloc, GFP_ATOMIC); + if (c_ptr->nodes == NULL) { + kfree(c_ptr); + return 0; + } + memset(c_ptr->nodes, 0, alloc); + if (in_own_cluster(addr)) + local_nodes = c_ptr->nodes; + c_ptr->highest_slave = LOWEST_SLAVE - 1; + c_ptr->highest_node = 0; + + z_ptr = zone_find(tipc_zone(addr)); + if (z_ptr == NULL) { + z_ptr = zone_create(addr); + } + if (z_ptr != NULL) { + zone_attach_cluster(z_ptr, c_ptr); + c_ptr->owner = z_ptr; + } + else { + kfree(c_ptr); + c_ptr = 0; + } + + return c_ptr; +} + +void cluster_delete(struct cluster *c_ptr) +{ + u32 n_num; + + if (!c_ptr) + return; + for (n_num = 1; n_num <= c_ptr->highest_node; n_num++) { + node_delete(c_ptr->nodes[n_num]); + } + for (n_num = LOWEST_SLAVE; n_num <= c_ptr->highest_slave; n_num++) { + node_delete(c_ptr->nodes[n_num]); + } + kfree(c_ptr->nodes); + kfree(c_ptr); +} + +u32 cluster_next_node(struct cluster *c_ptr, u32 addr) +{ + struct node *n_ptr; + u32 n_num = tipc_node(addr) + 1; + + if (!c_ptr) + return addr; + for (; n_num <= c_ptr->highest_node; n_num++) { + n_ptr = c_ptr->nodes[n_num]; + if (n_ptr && node_has_active_links(n_ptr)) + return n_ptr->addr; + } + for (n_num = 1; n_num < tipc_node(addr); n_num++) { + n_ptr = c_ptr->nodes[n_num]; + if (n_ptr && node_has_active_links(n_ptr)) + return n_ptr->addr; + } + return 0; +} + +void cluster_attach_node(struct cluster *c_ptr, struct node *n_ptr) +{ + u32 n_num = tipc_node(n_ptr->addr); + u32 max_n_num = tipc_max_nodes; + + if (in_own_cluster(n_ptr->addr)) + max_n_num = highest_allowed_slave; + assert(n_num > 0); + assert(n_num <= max_n_num); + assert(c_ptr->nodes[n_num] == 0); + c_ptr->nodes[n_num] = n_ptr; + if (n_num > c_ptr->highest_node) + c_ptr->highest_node = n_num; +} + +/** + * cluster_select_router - select router to a cluster + * + * Uses deterministic and fair algorithm. + */ + +u32 cluster_select_router(struct cluster *c_ptr, u32 ref) +{ + u32 n_num; + u32 ulim = c_ptr->highest_node; + u32 mask; + u32 tstart; + + assert(!in_own_cluster(c_ptr->addr)); + if (!ulim) + return 0; + + /* Start entry must be random */ + mask = tipc_max_nodes; + while (mask > ulim) + mask >>= 1; + tstart = ref & mask; + n_num = tstart; + + /* Lookup upwards with wrap-around */ + do { + if (node_is_up(c_ptr->nodes[n_num])) + break; + } while (++n_num <= ulim); + if (n_num > ulim) { + n_num = 1; + do { + if (node_is_up(c_ptr->nodes[n_num])) + break; + } while (++n_num < tstart); + if (n_num == tstart) + return 0; + } + assert(n_num <= ulim); + return node_select_router(c_ptr->nodes[n_num], ref); +} + +/** + * cluster_select_node - select destination node within a remote cluster + * + * Uses deterministic and fair algorithm. + */ + +struct node *cluster_select_node(struct cluster *c_ptr, u32 selector) +{ + u32 n_num; + u32 mask = tipc_max_nodes; + u32 start_entry; + + assert(!in_own_cluster(c_ptr->addr)); + if (!c_ptr->highest_node) + return 0; + + /* Start entry must be random */ + while (mask > c_ptr->highest_node) { + mask >>= 1; + } + start_entry = (selector & mask) ? selector & mask : 1u; + assert(start_entry <= c_ptr->highest_node); + + /* Lookup upwards with wrap-around */ + for (n_num = start_entry; n_num <= c_ptr->highest_node; n_num++) { + if (node_has_active_links(c_ptr->nodes[n_num])) + return c_ptr->nodes[n_num]; + } + for (n_num = 1; n_num < start_entry; n_num++) { + if (node_has_active_links(c_ptr->nodes[n_num])) + return c_ptr->nodes[n_num]; + } + return 0; +} + +/* + * Routing table management: See description in node.c + */ + +struct sk_buff *cluster_prepare_routing_msg(u32 data_size, u32 dest) +{ + u32 size = INT_H_SIZE + data_size; + struct sk_buff *buf = buf_acquire(size); + struct tipc_msg *msg; + + if (buf) { + msg = buf_msg(buf); + memset((char *)msg, 0, size); + msg_init(msg, ROUTE_DISTRIBUTOR, 0, TIPC_OK, INT_H_SIZE, dest); + } + return buf; +} + +void cluster_bcast_new_route(struct cluster *c_ptr, u32 dest, + u32 lower, u32 upper) +{ + struct sk_buff *buf = cluster_prepare_routing_msg(0, c_ptr->addr); + struct tipc_msg *msg; + + if (buf) { + msg = buf_msg(buf); + msg_set_remote_node(msg, dest); + msg_set_type(msg, ROUTE_ADDITION); + cluster_multicast(c_ptr, buf, lower, upper); + } else { + warn("Memory squeeze: broadcast of new route failed\n"); + } +} + +void cluster_bcast_lost_route(struct cluster *c_ptr, u32 dest, + u32 lower, u32 upper) +{ + struct sk_buff *buf = cluster_prepare_routing_msg(0, c_ptr->addr); + struct tipc_msg *msg; + + if (buf) { + msg = buf_msg(buf); + msg_set_remote_node(msg, dest); + msg_set_type(msg, ROUTE_REMOVAL); + cluster_multicast(c_ptr, buf, lower, upper); + } else { + warn("Memory squeeze: broadcast of lost route failed\n"); + } +} + +void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + u32 highest = c_ptr->highest_slave; + u32 n_num; + int send = 0; + + assert(!is_slave(dest)); + assert(in_own_cluster(dest)); + assert(in_own_cluster(c_ptr->addr)); + if (highest <= LOWEST_SLAVE) + return; + buf = cluster_prepare_routing_msg(highest - LOWEST_SLAVE + 1, + c_ptr->addr); + if (buf) { + msg = buf_msg(buf); + msg_set_remote_node(msg, c_ptr->addr); + msg_set_type(msg, SLAVE_ROUTING_TABLE); + for (n_num = LOWEST_SLAVE; n_num <= highest; n_num++) { + if (c_ptr->nodes[n_num] && + node_has_active_links(c_ptr->nodes[n_num])) { + send = 1; + msg_set_dataoctet(msg, n_num); + } + } + if (send) + link_send(buf, dest, dest); + else + buf_discard(buf); + } else { + warn("Memory squeeze: broadcast of lost route failed\n"); + } +} + +void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + u32 highest = c_ptr->highest_node; + u32 n_num; + int send = 0; + + if (in_own_cluster(c_ptr->addr)) + return; + assert(!is_slave(dest)); + assert(in_own_cluster(dest)); + highest = c_ptr->highest_node; + buf = cluster_prepare_routing_msg(highest + 1, c_ptr->addr); + if (buf) { + msg = buf_msg(buf); + msg_set_remote_node(msg, c_ptr->addr); + msg_set_type(msg, EXT_ROUTING_TABLE); + for (n_num = 1; n_num <= highest; n_num++) { + if (c_ptr->nodes[n_num] && + node_has_active_links(c_ptr->nodes[n_num])) { + send = 1; + msg_set_dataoctet(msg, n_num); + } + } + if (send) + link_send(buf, dest, dest); + else + buf_discard(buf); + } else { + warn("Memory squeeze: broadcast of external route failed\n"); + } +} + +void cluster_send_local_routes(struct cluster *c_ptr, u32 dest) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + u32 highest = c_ptr->highest_node; + u32 n_num; + int send = 0; + + assert(is_slave(dest)); + assert(in_own_cluster(c_ptr->addr)); + buf = cluster_prepare_routing_msg(highest, c_ptr->addr); + if (buf) { + msg = buf_msg(buf); + msg_set_remote_node(msg, c_ptr->addr); + msg_set_type(msg, LOCAL_ROUTING_TABLE); + for (n_num = 1; n_num <= highest; n_num++) { + if (c_ptr->nodes[n_num] && + node_has_active_links(c_ptr->nodes[n_num])) { + send = 1; + msg_set_dataoctet(msg, n_num); + } + } + if (send) + link_send(buf, dest, dest); + else + buf_discard(buf); + } else { + warn("Memory squeeze: broadcast of local route failed\n"); + } +} + +void cluster_recv_routing_table(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct cluster *c_ptr; + struct node *n_ptr; + unchar *node_table; + u32 table_size; + u32 router; + u32 rem_node = msg_remote_node(msg); + u32 z_num; + u32 c_num; + u32 n_num; + + c_ptr = cluster_find(rem_node); + if (!c_ptr) { + c_ptr = cluster_create(rem_node); + if (!c_ptr) { + buf_discard(buf); + return; + } + } + + node_table = buf->data + msg_hdr_sz(msg); + table_size = msg_size(msg) - msg_hdr_sz(msg); + router = msg_prevnode(msg); + z_num = tipc_zone(rem_node); + c_num = tipc_cluster(rem_node); + + switch (msg_type(msg)) { + case LOCAL_ROUTING_TABLE: + assert(is_slave(tipc_own_addr)); + case EXT_ROUTING_TABLE: + for (n_num = 1; n_num < table_size; n_num++) { + if (node_table[n_num]) { + u32 addr = tipc_addr(z_num, c_num, n_num); + n_ptr = c_ptr->nodes[n_num]; + if (!n_ptr) { + n_ptr = node_create(addr); + } + if (n_ptr) + node_add_router(n_ptr, router); + } + } + break; + case SLAVE_ROUTING_TABLE: + assert(!is_slave(tipc_own_addr)); + assert(in_own_cluster(c_ptr->addr)); + for (n_num = 1; n_num < table_size; n_num++) { + if (node_table[n_num]) { + u32 slave_num = n_num + LOWEST_SLAVE; + u32 addr = tipc_addr(z_num, c_num, slave_num); + n_ptr = c_ptr->nodes[slave_num]; + if (!n_ptr) { + n_ptr = node_create(addr); + } + if (n_ptr) + node_add_router(n_ptr, router); + } + } + break; + case ROUTE_ADDITION: + if (!is_slave(tipc_own_addr)) { + assert(!in_own_cluster(c_ptr->addr) + || is_slave(rem_node)); + } else { + assert(in_own_cluster(c_ptr->addr) + && !is_slave(rem_node)); + } + n_ptr = c_ptr->nodes[tipc_node(rem_node)]; + if (!n_ptr) + n_ptr = node_create(rem_node); + if (n_ptr) + node_add_router(n_ptr, router); + break; + case ROUTE_REMOVAL: + if (!is_slave(tipc_own_addr)) { + assert(!in_own_cluster(c_ptr->addr) + || is_slave(rem_node)); + } else { + assert(in_own_cluster(c_ptr->addr) + && !is_slave(rem_node)); + } + n_ptr = c_ptr->nodes[tipc_node(rem_node)]; + if (n_ptr) + node_remove_router(n_ptr, router); + break; + default: + assert(!"Illegal routing manager message received\n"); + } + buf_discard(buf); +} + +void cluster_remove_as_router(struct cluster *c_ptr, u32 router) +{ + u32 start_entry; + u32 tstop; + u32 n_num; + + if (is_slave(router)) + return; /* Slave nodes can not be routers */ + + if (in_own_cluster(c_ptr->addr)) { + start_entry = LOWEST_SLAVE; + tstop = c_ptr->highest_slave; + } else { + start_entry = 1; + tstop = c_ptr->highest_node; + } + + for (n_num = start_entry; n_num <= tstop; n_num++) { + if (c_ptr->nodes[n_num]) { + node_remove_router(c_ptr->nodes[n_num], router); + } + } +} + +/** + * cluster_multicast - multicast message to local nodes + */ + +void cluster_multicast(struct cluster *c_ptr, struct sk_buff *buf, + u32 lower, u32 upper) +{ + struct sk_buff *buf_copy; + struct node *n_ptr; + u32 n_num; + u32 tstop; + + assert(lower <= upper); + assert(((lower >= 1) && (lower <= tipc_max_nodes)) || + ((lower >= LOWEST_SLAVE) && (lower <= highest_allowed_slave))); + assert(((upper >= 1) && (upper <= tipc_max_nodes)) || + ((upper >= LOWEST_SLAVE) && (upper <= highest_allowed_slave))); + assert(in_own_cluster(c_ptr->addr)); + + tstop = is_slave(upper) ? c_ptr->highest_slave : c_ptr->highest_node; + if (tstop > upper) + tstop = upper; + for (n_num = lower; n_num <= tstop; n_num++) { + n_ptr = c_ptr->nodes[n_num]; + if (n_ptr && node_has_active_links(n_ptr)) { + buf_copy = skb_copy(buf, GFP_ATOMIC); + if (buf_copy == NULL) + break; + msg_set_destnode(buf_msg(buf_copy), n_ptr->addr); + link_send(buf_copy, n_ptr->addr, n_ptr->addr); + } + } + buf_discard(buf); +} + +/** + * cluster_broadcast - broadcast message to all nodes within cluster + */ + +void cluster_broadcast(struct sk_buff *buf) +{ + struct sk_buff *buf_copy; + struct cluster *c_ptr; + struct node *n_ptr; + u32 n_num; + u32 tstart; + u32 tstop; + u32 node_type; + + if (tipc_mode == TIPC_NET_MODE) { + c_ptr = cluster_find(tipc_own_addr); + assert(in_own_cluster(c_ptr->addr)); /* For now */ + + /* Send to standard nodes, then repeat loop sending to slaves */ + tstart = 1; + tstop = c_ptr->highest_node; + for (node_type = 1; node_type <= 2; node_type++) { + for (n_num = tstart; n_num <= tstop; n_num++) { + n_ptr = c_ptr->nodes[n_num]; + if (n_ptr && node_has_active_links(n_ptr)) { + buf_copy = skb_copy(buf, GFP_ATOMIC); + if (buf_copy == NULL) + goto exit; + msg_set_destnode(buf_msg(buf_copy), + n_ptr->addr); + link_send(buf_copy, n_ptr->addr, + n_ptr->addr); + } + } + tstart = LOWEST_SLAVE; + tstop = c_ptr->highest_slave; + } + } +exit: + buf_discard(buf); +} + +int cluster_init(void) +{ + highest_allowed_slave = LOWEST_SLAVE + tipc_max_slaves; + return cluster_create(tipc_own_addr) ? TIPC_OK : -ENOMEM; +} + diff --git a/net/tipc/cluster.h b/net/tipc/cluster.h new file mode 100644 index 00000000000..1ffb095991d --- /dev/null +++ b/net/tipc/cluster.h @@ -0,0 +1,92 @@ +/* + * net/tipc/cluster.h: Include file for TIPC cluster management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_CLUSTER_H +#define _TIPC_CLUSTER_H + +#include "addr.h" +#include "zone.h" + +#define LOWEST_SLAVE 2048u + +/** + * struct cluster - TIPC cluster structure + * @addr: network address of cluster + * @owner: pointer to zone that cluster belongs to + * @nodes: array of pointers to all nodes within cluster + * @highest_node: id of highest numbered node within cluster + * @highest_slave: (used for secondary node support) + */ + +struct cluster { + u32 addr; + struct _zone *owner; + struct node **nodes; + u32 highest_node; + u32 highest_slave; +}; + + +extern struct node **local_nodes; +extern u32 highest_allowed_slave; +extern struct node_map cluster_bcast_nodes; + +void cluster_remove_as_router(struct cluster *c_ptr, u32 router); +void cluster_send_ext_routes(struct cluster *c_ptr, u32 dest); +struct node *cluster_select_node(struct cluster *c_ptr, u32 selector); +u32 cluster_select_router(struct cluster *c_ptr, u32 ref); +void cluster_recv_routing_table(struct sk_buff *buf); +struct cluster *cluster_create(u32 addr); +void cluster_delete(struct cluster *c_ptr); +void cluster_attach_node(struct cluster *c_ptr, struct node *n_ptr); +void cluster_send_slave_routes(struct cluster *c_ptr, u32 dest); +void cluster_broadcast(struct sk_buff *buf); +int cluster_init(void); +u32 cluster_next_node(struct cluster *c_ptr, u32 addr); +void cluster_bcast_new_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); +void cluster_send_local_routes(struct cluster *c_ptr, u32 dest); +void cluster_bcast_lost_route(struct cluster *c_ptr, u32 dest, u32 lo, u32 hi); + +static inline struct cluster *cluster_find(u32 addr) +{ + struct _zone *z_ptr = zone_find(addr); + + if (z_ptr) + return z_ptr->clusters[1]; + return 0; +} + +#endif diff --git a/net/tipc/config.c b/net/tipc/config.c new file mode 100644 index 00000000000..8ddef4fce2c --- /dev/null +++ b/net/tipc/config.c @@ -0,0 +1,718 @@ +/* + * net/tipc/config.c: TIPC configuration management code + * + * Copyright (c) 2002-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "bearer.h" +#include "port.h" +#include "link.h" +#include "zone.h" +#include "addr.h" +#include "name_table.h" +#include "node.h" +#include "config.h" +#include "discover.h" + +struct subscr_data { + char usr_handle[8]; + u32 domain; + u32 port_ref; + struct list_head subd_list; +}; + +struct manager { + u32 user_ref; + u32 port_ref; + u32 subscr_ref; + u32 link_subscriptions; + struct list_head link_subscribers; +}; + +static struct manager mng = { 0}; + +static spinlock_t config_lock = SPIN_LOCK_UNLOCKED; + +static const void *req_tlv_area; /* request message TLV area */ +static int req_tlv_space; /* request message TLV area size */ +static int rep_headroom; /* reply message headroom to use */ + + +void cfg_link_event(u32 addr, char *name, int up) +{ + /* TIPC DOESN'T HANDLE LINK EVENT SUBSCRIPTIONS AT THE MOMENT */ +} + + +struct sk_buff *cfg_reply_alloc(int payload_size) +{ + struct sk_buff *buf; + + buf = alloc_skb(rep_headroom + payload_size, GFP_ATOMIC); + if (buf) + skb_reserve(buf, rep_headroom); + return buf; +} + +int cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size) +{ + struct tlv_desc *tlv = (struct tlv_desc *)buf->tail; + int new_tlv_space = TLV_SPACE(tlv_data_size); + + if (skb_tailroom(buf) < new_tlv_space) { + dbg("cfg_append_tlv unable to append TLV\n"); + return 0; + } + skb_put(buf, new_tlv_space); + tlv->tlv_type = htons(tlv_type); + tlv->tlv_len = htons(TLV_LENGTH(tlv_data_size)); + if (tlv_data_size && tlv_data) + memcpy(TLV_DATA(tlv), tlv_data, tlv_data_size); + return 1; +} + +struct sk_buff *cfg_reply_unsigned_type(u16 tlv_type, u32 value) +{ + struct sk_buff *buf; + u32 value_net; + + buf = cfg_reply_alloc(TLV_SPACE(sizeof(value))); + if (buf) { + value_net = htonl(value); + cfg_append_tlv(buf, tlv_type, &value_net, + sizeof(value_net)); + } + return buf; +} + +struct sk_buff *cfg_reply_string_type(u16 tlv_type, char *string) +{ + struct sk_buff *buf; + int string_len = strlen(string) + 1; + + buf = cfg_reply_alloc(TLV_SPACE(string_len)); + if (buf) + cfg_append_tlv(buf, tlv_type, string, string_len); + return buf; +} + + + + +#if 0 + +/* Now obsolete code for handling commands not yet implemented the new way */ + +int tipc_cfg_cmd(const struct tipc_cmd_msg * msg, + char *data, + u32 sz, + u32 *ret_size, + struct tipc_portid *orig) +{ + int rv = -EINVAL; + u32 cmd = msg->cmd; + + *ret_size = 0; + switch (cmd) { + case TIPC_REMOVE_LINK: + case TIPC_CMD_BLOCK_LINK: + case TIPC_CMD_UNBLOCK_LINK: + if (!cfg_check_connection(orig)) + rv = link_control(msg->argv.link_name, msg->cmd, 0); + break; + case TIPC_ESTABLISH: + { + int connected; + + tipc_isconnected(mng.conn_port_ref, &connected); + if (connected || !orig) { + rv = TIPC_FAILURE; + break; + } + rv = tipc_connect2port(mng.conn_port_ref, orig); + if (rv == TIPC_OK) + orig = 0; + break; + } + case TIPC_GET_PEER_ADDRESS: + *ret_size = link_peer_addr(msg->argv.link_name, data, sz); + break; + case TIPC_GET_ROUTES: + rv = TIPC_OK; + break; + default: {} + } + if (*ret_size) + rv = TIPC_OK; + return rv; +} + +static void cfg_cmd_event(struct tipc_cmd_msg *msg, + char *data, + u32 sz, + struct tipc_portid const *orig) +{ + int rv = -EINVAL; + struct tipc_cmd_result_msg rmsg; + struct iovec msg_sect[2]; + int *arg; + + msg->cmd = ntohl(msg->cmd); + + cfg_prepare_res_msg(msg->cmd, msg->usr_handle, rv, &rmsg, msg_sect, + data, 0); + if (ntohl(msg->magic) != TIPC_MAGIC) + goto exit; + + switch (msg->cmd) { + case TIPC_CREATE_LINK: + if (!cfg_check_connection(orig)) + rv = disc_create_link(&msg->argv.create_link); + break; + case TIPC_LINK_SUBSCRIBE: + { + struct subscr_data *sub; + + if (mng.link_subscriptions > 64) + break; + sub = (struct subscr_data *)kmalloc(sizeof(*sub), + GFP_ATOMIC); + if (sub == NULL) { + warn("Memory squeeze; dropped remote link subscription\n"); + break; + } + INIT_LIST_HEAD(&sub->subd_list); + tipc_createport(mng.user_ref, + (void *)sub, + TIPC_HIGH_IMPORTANCE, + 0, + 0, + (tipc_conn_shutdown_event)cfg_linksubscr_cancel, + 0, + 0, + (tipc_conn_msg_event)cfg_linksubscr_cancel, + 0, + &sub->port_ref); + if (!sub->port_ref) { + kfree(sub); + break; + } + memcpy(sub->usr_handle,msg->usr_handle, + sizeof(sub->usr_handle)); + sub->domain = msg->argv.domain; + list_add_tail(&sub->subd_list, &mng.link_subscribers); + tipc_connect2port(sub->port_ref, orig); + rmsg.retval = TIPC_OK; + tipc_send(sub->port_ref, 2u, msg_sect); + mng.link_subscriptions++; + return; + } + default: + rv = tipc_cfg_cmd(msg, data, sz, (u32 *)&msg_sect[1].iov_len, orig); + } + exit: + rmsg.result_len = htonl(msg_sect[1].iov_len); + rmsg.retval = htonl(rv); + cfg_respond(msg_sect, 2u, orig); +} +#endif + +static struct sk_buff *cfg_enable_bearer(void) +{ + struct tipc_bearer_config *args; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_CONFIG)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + args = (struct tipc_bearer_config *)TLV_DATA(req_tlv_area); + if (tipc_enable_bearer(args->name, + ntohl(args->detect_scope), + ntohl(args->priority))) + return cfg_reply_error_string("unable to enable bearer"); + + return cfg_reply_none(); +} + +static struct sk_buff *cfg_disable_bearer(void) +{ + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_BEARER_NAME)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + if (tipc_disable_bearer((char *)TLV_DATA(req_tlv_area))) + return cfg_reply_error_string("unable to disable bearer"); + + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_own_addr(void) +{ + u32 addr; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + addr = *(u32 *)TLV_DATA(req_tlv_area); + addr = ntohl(addr); + if (addr == tipc_own_addr) + return cfg_reply_none(); + if (!addr_node_valid(addr)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (node address)"); + if (tipc_own_addr) + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change node address once assigned)"); + + spin_unlock_bh(&config_lock); + stop_net(); + tipc_own_addr = addr; + start_net(); + spin_lock_bh(&config_lock); + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_remote_mng(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + tipc_remote_management = (value != 0); + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_max_publications(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 1, 65535)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max publications must be 1-65535)"); + tipc_max_publications = value; + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_max_subscriptions(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 1, 65535)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max subscriptions must be 1-65535"); + tipc_max_subscriptions = value; + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_max_ports(void) +{ + int orig_mode; + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 127, 65535)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max ports must be 127-65535)"); + + if (value == tipc_max_ports) + return cfg_reply_none(); + + if (atomic_read(&tipc_user_count) > 2) + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change max ports while TIPC users exist)"); + + spin_unlock_bh(&config_lock); + orig_mode = tipc_get_mode(); + if (orig_mode == TIPC_NET_MODE) + stop_net(); + stop_core(); + tipc_max_ports = value; + start_core(); + if (orig_mode == TIPC_NET_MODE) + start_net(); + spin_lock_bh(&config_lock); + return cfg_reply_none(); +} + +static struct sk_buff *set_net_max(int value, int *parameter) +{ + int orig_mode; + + if (value != *parameter) { + orig_mode = tipc_get_mode(); + if (orig_mode == TIPC_NET_MODE) + stop_net(); + *parameter = value; + if (orig_mode == TIPC_NET_MODE) + start_net(); + } + + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_max_zones(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 1, 255)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max zones must be 1-255)"); + return set_net_max(value, &tipc_max_zones); +} + +static struct sk_buff *cfg_set_max_clusters(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != 1) + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (max clusters fixed at 1)"); + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_max_nodes(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 8, 2047)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (max nodes must be 8-2047)"); + return set_net_max(value, &tipc_max_nodes); +} + +static struct sk_buff *cfg_set_max_slaves(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != 0) + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (max secondary nodes fixed at 0)"); + return cfg_reply_none(); +} + +static struct sk_buff *cfg_set_netid(void) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 1, 9999)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network id must be 1-9999)"); + + if (tipc_own_addr) + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change network id once part of network)"); + + return set_net_max(value, &tipc_net_id); +} + +struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, const void *request_area, + int request_space, int reply_headroom) +{ + struct sk_buff *rep_tlv_buf; + + spin_lock_bh(&config_lock); + + /* Save request and reply details in a well-known location */ + + req_tlv_area = request_area; + req_tlv_space = request_space; + rep_headroom = reply_headroom; + + /* Check command authorization */ + + if (likely(orig_node == tipc_own_addr)) { + /* command is permitted */ + } else if (cmd >= 0x8000) { + rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot be done remotely)"); + goto exit; + } else if (!tipc_remote_management) { + rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NO_REMOTE); + goto exit; + } + else if (cmd >= 0x4000) { + u32 domain = 0; + + if ((nametbl_translate(TIPC_ZM_SRV, 0, &domain) == 0) || + (domain != orig_node)) { + rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_ZONE_MSTR); + goto exit; + } + } + + /* Call appropriate processing routine */ + + switch (cmd) { + case TIPC_CMD_NOOP: + rep_tlv_buf = cfg_reply_none(); + break; + case TIPC_CMD_GET_NODES: + rep_tlv_buf = node_get_nodes(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_GET_LINKS: + rep_tlv_buf = node_get_links(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_SHOW_LINK_STATS: + rep_tlv_buf = link_cmd_show_stats(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_RESET_LINK_STATS: + rep_tlv_buf = link_cmd_reset_stats(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_SHOW_NAME_TABLE: + rep_tlv_buf = nametbl_get(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_GET_BEARER_NAMES: + rep_tlv_buf = bearer_get_names(); + break; + case TIPC_CMD_GET_MEDIA_NAMES: + rep_tlv_buf = media_get_names(); + break; + case TIPC_CMD_SHOW_PORTS: + rep_tlv_buf = port_get_ports(); + break; +#if 0 + case TIPC_CMD_SHOW_PORT_STATS: + rep_tlv_buf = port_show_stats(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_RESET_PORT_STATS: + rep_tlv_buf = cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED); + break; +#endif + case TIPC_CMD_SET_LOG_SIZE: + rep_tlv_buf = log_resize(req_tlv_area, req_tlv_space); + break; + case TIPC_CMD_DUMP_LOG: + rep_tlv_buf = log_dump(); + break; + case TIPC_CMD_SET_LINK_TOL: + case TIPC_CMD_SET_LINK_PRI: + case TIPC_CMD_SET_LINK_WINDOW: + rep_tlv_buf = link_cmd_config(req_tlv_area, req_tlv_space, cmd); + break; + case TIPC_CMD_ENABLE_BEARER: + rep_tlv_buf = cfg_enable_bearer(); + break; + case TIPC_CMD_DISABLE_BEARER: + rep_tlv_buf = cfg_disable_bearer(); + break; + case TIPC_CMD_SET_NODE_ADDR: + rep_tlv_buf = cfg_set_own_addr(); + break; + case TIPC_CMD_SET_REMOTE_MNG: + rep_tlv_buf = cfg_set_remote_mng(); + break; + case TIPC_CMD_SET_MAX_PORTS: + rep_tlv_buf = cfg_set_max_ports(); + break; + case TIPC_CMD_SET_MAX_PUBL: + rep_tlv_buf = cfg_set_max_publications(); + break; + case TIPC_CMD_SET_MAX_SUBSCR: + rep_tlv_buf = cfg_set_max_subscriptions(); + break; + case TIPC_CMD_SET_MAX_ZONES: + rep_tlv_buf = cfg_set_max_zones(); + break; + case TIPC_CMD_SET_MAX_CLUSTERS: + rep_tlv_buf = cfg_set_max_clusters(); + break; + case TIPC_CMD_SET_MAX_NODES: + rep_tlv_buf = cfg_set_max_nodes(); + break; + case TIPC_CMD_SET_MAX_SLAVES: + rep_tlv_buf = cfg_set_max_slaves(); + break; + case TIPC_CMD_SET_NETID: + rep_tlv_buf = cfg_set_netid(); + break; + case TIPC_CMD_GET_REMOTE_MNG: + rep_tlv_buf = cfg_reply_unsigned(tipc_remote_management); + break; + case TIPC_CMD_GET_MAX_PORTS: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_ports); + break; + case TIPC_CMD_GET_MAX_PUBL: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_publications); + break; + case TIPC_CMD_GET_MAX_SUBSCR: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_subscriptions); + break; + case TIPC_CMD_GET_MAX_ZONES: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_zones); + break; + case TIPC_CMD_GET_MAX_CLUSTERS: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_clusters); + break; + case TIPC_CMD_GET_MAX_NODES: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_nodes); + break; + case TIPC_CMD_GET_MAX_SLAVES: + rep_tlv_buf = cfg_reply_unsigned(tipc_max_slaves); + break; + case TIPC_CMD_GET_NETID: + rep_tlv_buf = cfg_reply_unsigned(tipc_net_id); + break; + default: + rep_tlv_buf = NULL; + break; + } + + /* Return reply buffer */ +exit: + spin_unlock_bh(&config_lock); + return rep_tlv_buf; +} + +static void cfg_named_msg_event(void *userdata, + u32 port_ref, + struct sk_buff **buf, + const unchar *msg, + u32 size, + u32 importance, + struct tipc_portid const *orig, + struct tipc_name_seq const *dest) +{ + struct tipc_cfg_msg_hdr *req_hdr; + struct tipc_cfg_msg_hdr *rep_hdr; + struct sk_buff *rep_buf; + + /* Validate configuration message header (ignore invalid message) */ + + req_hdr = (struct tipc_cfg_msg_hdr *)msg; + if ((size < sizeof(*req_hdr)) || + (size != TCM_ALIGN(ntohl(req_hdr->tcm_len))) || + (ntohs(req_hdr->tcm_flags) != TCM_F_REQUEST)) { + warn("discarded invalid configuration message\n"); + return; + } + + /* Generate reply for request (if can't, return request) */ + + rep_buf = cfg_do_cmd(orig->node, + ntohs(req_hdr->tcm_type), + msg + sizeof(*req_hdr), + size - sizeof(*req_hdr), + BUF_HEADROOM + MAX_H_SIZE + sizeof(*rep_hdr)); + if (rep_buf) { + skb_push(rep_buf, sizeof(*rep_hdr)); + rep_hdr = (struct tipc_cfg_msg_hdr *)rep_buf->data; + memcpy(rep_hdr, req_hdr, sizeof(*rep_hdr)); + rep_hdr->tcm_len = htonl(rep_buf->len); + rep_hdr->tcm_flags &= htons(~TCM_F_REQUEST); + } else { + rep_buf = *buf; + *buf = NULL; + } + + /* NEED TO ADD CODE TO HANDLE FAILED SEND (SUCH AS CONGESTION) */ + tipc_send_buf2port(port_ref, orig, rep_buf, rep_buf->len); +} + +int cfg_init(void) +{ + struct tipc_name_seq seq; + int res; + + memset(&mng, 0, sizeof(mng)); + INIT_LIST_HEAD(&mng.link_subscribers); + + res = tipc_attach(&mng.user_ref, 0, 0); + if (res) + goto failed; + + res = tipc_createport(mng.user_ref, 0, TIPC_CRITICAL_IMPORTANCE, + NULL, NULL, NULL, + NULL, cfg_named_msg_event, NULL, + NULL, &mng.port_ref); + if (res) + goto failed; + + seq.type = TIPC_CFG_SRV; + seq.lower = seq.upper = tipc_own_addr; + res = nametbl_publish_rsv(mng.port_ref, TIPC_ZONE_SCOPE, &seq); + if (res) + goto failed; + + return 0; + +failed: + err("Unable to create configuration service\n"); + tipc_detach(mng.user_ref); + mng.user_ref = 0; + return res; +} + +void cfg_stop(void) +{ + if (mng.user_ref) { + tipc_detach(mng.user_ref); + mng.user_ref = 0; + } +} diff --git a/net/tipc/config.h b/net/tipc/config.h new file mode 100644 index 00000000000..646377d4045 --- /dev/null +++ b/net/tipc/config.h @@ -0,0 +1,80 @@ +/* + * net/tipc/config.h: Include file for TIPC configuration service code + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_CONFIG_H +#define _TIPC_CONFIG_H + +/* ---------------------------------------------------------------------- */ + +#include <linux/tipc.h> +#include <linux/tipc_config.h> +#include "link.h" + +struct sk_buff *cfg_reply_alloc(int payload_size); +int cfg_append_tlv(struct sk_buff *buf, int tlv_type, + void *tlv_data, int tlv_data_size); +struct sk_buff *cfg_reply_unsigned_type(u16 tlv_type, u32 value); +struct sk_buff *cfg_reply_string_type(u16 tlv_type, char *string); + +static inline struct sk_buff *cfg_reply_none(void) +{ + return cfg_reply_alloc(0); +} + +static inline struct sk_buff *cfg_reply_unsigned(u32 value) +{ + return cfg_reply_unsigned_type(TIPC_TLV_UNSIGNED, value); +} + +static inline struct sk_buff *cfg_reply_error_string(char *string) +{ + return cfg_reply_string_type(TIPC_TLV_ERROR_STRING, string); +} + +static inline struct sk_buff *cfg_reply_ultra_string(char *string) +{ + return cfg_reply_string_type(TIPC_TLV_ULTRA_STRING, string); +} + +struct sk_buff *cfg_do_cmd(u32 orig_node, u16 cmd, + const void *req_tlv_area, int req_tlv_space, + int headroom); + +void cfg_link_event(u32 addr, char *name, int up); +int cfg_init(void); +void cfg_stop(void); + +#endif diff --git a/net/tipc/core.c b/net/tipc/core.c new file mode 100644 index 00000000000..e83ac06e31b --- /dev/null +++ b/net/tipc/core.c @@ -0,0 +1,285 @@ +/* + * net/tipc/core.c: TIPC module code + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/init.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/version.h> +#include <linux/random.h> + +#include "core.h" +#include "dbg.h" +#include "ref.h" +#include "net.h" +#include "user_reg.h" +#include "name_table.h" +#include "subscr.h" +#include "config.h" + +int eth_media_start(void); +void eth_media_stop(void); +int handler_start(void); +void handler_stop(void); +int socket_init(void); +void socket_stop(void); +int netlink_start(void); +void netlink_stop(void); + +#define MOD_NAME "tipc_start: " + +#ifndef CONFIG_TIPC_ZONES +#define CONFIG_TIPC_ZONES 3 +#endif + +#ifndef CONFIG_TIPC_CLUSTERS +#define CONFIG_TIPC_CLUSTERS 1 +#endif + +#ifndef CONFIG_TIPC_NODES +#define CONFIG_TIPC_NODES 255 +#endif + +#ifndef CONFIG_TIPC_SLAVE_NODES +#define CONFIG_TIPC_SLAVE_NODES 0 +#endif + +#ifndef CONFIG_TIPC_PORTS +#define CONFIG_TIPC_PORTS 8191 +#endif + +#ifndef CONFIG_TIPC_LOG +#define CONFIG_TIPC_LOG 0 +#endif + +/* global variables used by multiple sub-systems within TIPC */ + +int tipc_mode = TIPC_NOT_RUNNING; +int tipc_random; +atomic_t tipc_user_count = ATOMIC_INIT(0); + +const char tipc_alphabet[] = + "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_"; + +/* configurable TIPC parameters */ + +u32 tipc_own_addr; +int tipc_max_zones; +int tipc_max_clusters; +int tipc_max_nodes; +int tipc_max_slaves; +int tipc_max_ports; +int tipc_max_subscriptions; +int tipc_max_publications; +int tipc_net_id; +int tipc_remote_management; + + +int tipc_get_mode(void) +{ + return tipc_mode; +} + +/** + * stop_net - shut down TIPC networking sub-systems + */ + +void stop_net(void) +{ + eth_media_stop(); + tipc_stop_net(); +} + +/** + * start_net - start TIPC networking sub-systems + */ + +int start_net(void) +{ + int res; + + if ((res = tipc_start_net()) || + (res = eth_media_start())) { + stop_net(); + } + return res; +} + +/** + * stop_core - switch TIPC from SINGLE NODE to NOT RUNNING mode + */ + +void stop_core(void) +{ + if (tipc_mode != TIPC_NODE_MODE) + return; + + tipc_mode = TIPC_NOT_RUNNING; + + netlink_stop(); + handler_stop(); + cfg_stop(); + subscr_stop(); + reg_stop(); + nametbl_stop(); + ref_table_stop(); + socket_stop(); +} + +/** + * start_core - switch TIPC from NOT RUNNING to SINGLE NODE mode + */ + +int start_core(void) +{ + int res; + + if (tipc_mode != TIPC_NOT_RUNNING) + return -ENOPROTOOPT; + + get_random_bytes(&tipc_random, sizeof(tipc_random)); + tipc_mode = TIPC_NODE_MODE; + + if ((res = handler_start()) || + (res = ref_table_init(tipc_max_ports + tipc_max_subscriptions, + tipc_random)) || + (res = reg_start()) || + (res = nametbl_init()) || + (res = k_signal((Handler)subscr_start, 0)) || + (res = k_signal((Handler)cfg_init, 0)) || + (res = netlink_start()) || + (res = socket_init())) { + stop_core(); + } + return res; +} + + +static int __init tipc_init(void) +{ + int res; + + log_reinit(CONFIG_TIPC_LOG); + info("Activated (compiled " __DATE__ " " __TIME__ ")\n"); + + tipc_own_addr = 0; + tipc_remote_management = 1; + tipc_max_publications = 10000; + tipc_max_subscriptions = 2000; + tipc_max_ports = delimit(CONFIG_TIPC_PORTS, 127, 65536); + tipc_max_zones = delimit(CONFIG_TIPC_ZONES, 1, 511); + tipc_max_clusters = delimit(CONFIG_TIPC_CLUSTERS, 1, 1); + tipc_max_nodes = delimit(CONFIG_TIPC_NODES, 8, 2047); + tipc_max_slaves = delimit(CONFIG_TIPC_SLAVE_NODES, 0, 2047); + tipc_net_id = 4711; + + if ((res = start_core())) + err("Unable to start in single node mode\n"); + else + info("Started in single node mode\n"); + return res; +} + +static void __exit tipc_exit(void) +{ + stop_net(); + stop_core(); + info("Deactivated\n"); + log_stop(); +} + +module_init(tipc_init); +module_exit(tipc_exit); + +MODULE_DESCRIPTION("TIPC: Transparent Inter Process Communication"); +MODULE_LICENSE("Dual BSD/GPL"); + +/* Native TIPC API for kernel-space applications (see tipc.h) */ + +EXPORT_SYMBOL(tipc_attach); +EXPORT_SYMBOL(tipc_detach); +EXPORT_SYMBOL(tipc_get_addr); +EXPORT_SYMBOL(tipc_get_mode); +EXPORT_SYMBOL(tipc_createport); +EXPORT_SYMBOL(tipc_deleteport); +EXPORT_SYMBOL(tipc_ownidentity); +EXPORT_SYMBOL(tipc_portimportance); +EXPORT_SYMBOL(tipc_set_portimportance); +EXPORT_SYMBOL(tipc_portunreliable); +EXPORT_SYMBOL(tipc_set_portunreliable); +EXPORT_SYMBOL(tipc_portunreturnable); +EXPORT_SYMBOL(tipc_set_portunreturnable); +EXPORT_SYMBOL(tipc_publish); +EXPORT_SYMBOL(tipc_withdraw); +EXPORT_SYMBOL(tipc_connect2port); +EXPORT_SYMBOL(tipc_disconnect); +EXPORT_SYMBOL(tipc_shutdown); +EXPORT_SYMBOL(tipc_isconnected); +EXPORT_SYMBOL(tipc_peer); +EXPORT_SYMBOL(tipc_ref_valid); +EXPORT_SYMBOL(tipc_send); +EXPORT_SYMBOL(tipc_send_buf); +EXPORT_SYMBOL(tipc_send2name); +EXPORT_SYMBOL(tipc_forward2name); +EXPORT_SYMBOL(tipc_send_buf2name); +EXPORT_SYMBOL(tipc_forward_buf2name); +EXPORT_SYMBOL(tipc_send2port); +EXPORT_SYMBOL(tipc_forward2port); +EXPORT_SYMBOL(tipc_send_buf2port); +EXPORT_SYMBOL(tipc_forward_buf2port); +EXPORT_SYMBOL(tipc_multicast); +/* EXPORT_SYMBOL(tipc_multicast_buf); not available yet */ +EXPORT_SYMBOL(tipc_ispublished); +EXPORT_SYMBOL(tipc_available_nodes); + +/* TIPC API for external bearers (see tipc_bearer.h) */ + +EXPORT_SYMBOL(tipc_block_bearer); +EXPORT_SYMBOL(tipc_continue); +EXPORT_SYMBOL(tipc_disable_bearer); +EXPORT_SYMBOL(tipc_enable_bearer); +EXPORT_SYMBOL(tipc_recv_msg); +EXPORT_SYMBOL(tipc_register_media); + +/* TIPC API for external APIs (see tipc_port.h) */ + +EXPORT_SYMBOL(tipc_createport_raw); +EXPORT_SYMBOL(tipc_set_msg_option); +EXPORT_SYMBOL(tipc_reject_msg); +EXPORT_SYMBOL(tipc_send_buf_fast); +EXPORT_SYMBOL(tipc_acknowledge); +EXPORT_SYMBOL(tipc_get_port); +EXPORT_SYMBOL(tipc_get_handle); + diff --git a/net/tipc/core.h b/net/tipc/core.h new file mode 100644 index 00000000000..b69b60b2cc8 --- /dev/null +++ b/net/tipc/core.h @@ -0,0 +1,316 @@ +/* + * net/tipc/core.h: Include file for TIPC global declarations + * + * Copyright (c) 2005-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_CORE_H +#define _TIPC_CORE_H + +#include <net/tipc/tipc.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/mm.h> +#include <linux/timer.h> +#include <linux/string.h> +#include <asm/uaccess.h> +#include <linux/interrupt.h> +#include <asm/atomic.h> +#include <asm/hardirq.h> +#include <linux/netdevice.h> +#include <linux/in.h> +#include <linux/list.h> +#include <linux/vmalloc.h> + +/* + * TIPC debugging code + */ + +#define assert(i) BUG_ON(!(i)) + +struct tipc_msg; +extern struct print_buf *CONS, *LOG; +extern struct print_buf *TEE(struct print_buf *, struct print_buf *); +void msg_print(struct print_buf*,struct tipc_msg *,const char*); +void tipc_printf(struct print_buf *, const char *fmt, ...); +void tipc_dump(struct print_buf*,const char *fmt, ...); + +#ifdef CONFIG_TIPC_DEBUG + +/* + * TIPC debug support included: + * - system messages are printed to TIPC_OUTPUT print buffer + * - debug messages are printed to DBG_OUTPUT print buffer + */ + +#define err(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_ERR "TIPC: " fmt, ## arg) +#define warn(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_WARNING "TIPC: " fmt, ## arg) +#define info(fmt, arg...) tipc_printf(TIPC_OUTPUT, KERN_NOTICE "TIPC: " fmt, ## arg) + +#define dbg(fmt, arg...) do {if (DBG_OUTPUT) tipc_printf(DBG_OUTPUT, fmt, ## arg);} while(0) +#define msg_dbg(msg, txt) do {if (DBG_OUTPUT) msg_print(DBG_OUTPUT, msg, txt);} while(0) +#define dump(fmt, arg...) do {if (DBG_OUTPUT) tipc_dump(DBG_OUTPUT, fmt, ##arg);} while(0) + + +/* + * By default, TIPC_OUTPUT is defined to be system console and TIPC log buffer, + * while DBG_OUTPUT is the null print buffer. These defaults can be changed + * here, or on a per .c file basis, by redefining these symbols. The following + * print buffer options are available: + * + * NULL : Output to null print buffer (i.e. print nowhere) + * CONS : Output to system console + * LOG : Output to TIPC log buffer + * &buf : Output to user-defined buffer (struct print_buf *) + * TEE(&buf_a,&buf_b) : Output to two print buffers (eg. TEE(CONS,LOG) ) + */ + +#ifndef TIPC_OUTPUT +#define TIPC_OUTPUT TEE(CONS,LOG) +#endif + +#ifndef DBG_OUTPUT +#define DBG_OUTPUT NULL +#endif + +#else + +#ifndef DBG_OUTPUT +#define DBG_OUTPUT NULL +#endif + +/* + * TIPC debug support not included: + * - system messages are printed to system console + * - debug messages are not printed + */ + +#define err(fmt, arg...) printk(KERN_ERR "TIPC: " fmt , ## arg) +#define info(fmt, arg...) printk(KERN_INFO "TIPC: " fmt , ## arg) +#define warn(fmt, arg...) printk(KERN_WARNING "TIPC: " fmt , ## arg) + +#define dbg(fmt, arg...) do {} while (0) +#define msg_dbg(msg,txt) do {} while (0) +#define dump(fmt,arg...) do {} while (0) + +#endif + + +/* + * TIPC-specific error codes + */ + +#define ELINKCONG EAGAIN /* link congestion <=> resource unavailable */ + +/* + * Global configuration variables + */ + +extern u32 tipc_own_addr; +extern int tipc_max_zones; +extern int tipc_max_clusters; +extern int tipc_max_nodes; +extern int tipc_max_slaves; +extern int tipc_max_ports; +extern int tipc_max_subscriptions; +extern int tipc_max_publications; +extern int tipc_net_id; +extern int tipc_remote_management; + +/* + * Other global variables + */ + +extern int tipc_mode; +extern int tipc_random; +extern const char tipc_alphabet[]; +extern atomic_t tipc_user_count; + + +/* + * Routines available to privileged subsystems + */ + +extern int start_core(void); +extern void stop_core(void); +extern int start_net(void); +extern void stop_net(void); + +static inline int delimit(int val, int min, int max) +{ + if (val > max) + return max; + if (val < min) + return min; + return val; +} + + +/* + * TIPC timer and signal code + */ + +typedef void (*Handler) (unsigned long); + +u32 k_signal(Handler routine, unsigned long argument); + +/** + * k_init_timer - initialize a timer + * @timer: pointer to timer structure + * @routine: pointer to routine to invoke when timer expires + * @argument: value to pass to routine when timer expires + * + * Timer must be initialized before use (and terminated when no longer needed). + */ + +static inline void k_init_timer(struct timer_list *timer, Handler routine, + unsigned long argument) +{ + dbg("initializing timer %p\n", timer); + init_timer(timer); + timer->function = routine; + timer->data = argument; +} + +/** + * k_start_timer - start a timer + * @timer: pointer to timer structure + * @msec: time to delay (in ms) + * + * Schedules a previously initialized timer for later execution. + * If timer is already running, the new timeout overrides the previous request. + * + * To ensure the timer doesn't expire before the specified delay elapses, + * the amount of delay is rounded up when converting to the jiffies + * then an additional jiffy is added to account for the fact that + * the starting time may be in the middle of the current jiffy. + */ + +static inline void k_start_timer(struct timer_list *timer, unsigned long msec) +{ + dbg("starting timer %p for %u\n", timer, msec); + mod_timer(timer, jiffies + msecs_to_jiffies(msec) + 1); +} + +/** + * k_cancel_timer - cancel a timer + * @timer: pointer to timer structure + * + * Cancels a previously initialized timer. + * Can be called safely even if the timer is already inactive. + * + * WARNING: Must not be called when holding locks required by the timer's + * timeout routine, otherwise deadlock can occur on SMP systems! + */ + +static inline void k_cancel_timer(struct timer_list *timer) +{ + dbg("cancelling timer %p\n", timer); + del_timer_sync(timer); +} + +/** + * k_term_timer - terminate a timer + * @timer: pointer to timer structure + * + * Prevents further use of a previously initialized timer. + * + * WARNING: Caller must ensure timer isn't currently running. + * + * (Do not "enhance" this routine to automatically cancel an active timer, + * otherwise deadlock can arise when a timeout routine calls k_term_timer.) + */ + +static inline void k_term_timer(struct timer_list *timer) +{ + dbg("terminating timer %p\n", timer); +} + + +/* + * TIPC message buffer code + * + * TIPC message buffer headroom leaves room for 14 byte Ethernet header, + * while ensuring TIPC header is word aligned for quicker access + */ + +#define BUF_HEADROOM 16u + +struct tipc_skb_cb { + void *handle; +}; + +#define TIPC_SKB_CB(__skb) ((struct tipc_skb_cb *)&((__skb)->cb[0])) + + +static inline struct tipc_msg *buf_msg(struct sk_buff *skb) +{ + return (struct tipc_msg *)skb->data; +} + +/** + * buf_acquire - creates a TIPC message buffer + * @size: message size (including TIPC header) + * + * Returns a new buffer. Space is reserved for a data link header. + */ + +static inline struct sk_buff *buf_acquire(u32 size) +{ + struct sk_buff *skb; + unsigned int buf_size = (BUF_HEADROOM + size + 3) & ~3u; + + skb = alloc_skb(buf_size, GFP_ATOMIC); + if (skb) { + skb_reserve(skb, BUF_HEADROOM); + skb_put(skb, size); + skb->next = NULL; + } + return skb; +} + +/** + * buf_discard - frees a TIPC message buffer + * @skb: message buffer + * + * Frees a new buffer. If passed NULL, just returns. + */ + +static inline void buf_discard(struct sk_buff *skb) +{ + if (likely(skb != NULL)) + kfree_skb(skb); +} + +#endif diff --git a/net/tipc/dbg.c b/net/tipc/dbg.c new file mode 100644 index 00000000000..7ed60a1cfbb --- /dev/null +++ b/net/tipc/dbg.c @@ -0,0 +1,395 @@ +/* + * net/tipc/dbg.c: TIPC print buffer routines for debuggign + * + * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "dbg.h" + +#define MAX_STRING 512 + +static char print_string[MAX_STRING]; +static spinlock_t print_lock = SPIN_LOCK_UNLOCKED; + +static struct print_buf cons_buf = { NULL, 0, NULL, NULL }; +struct print_buf *CONS = &cons_buf; + +static struct print_buf log_buf = { NULL, 0, NULL, NULL }; +struct print_buf *LOG = &log_buf; + + +#define FORMAT(PTR,LEN,FMT) \ +{\ + va_list args;\ + va_start(args, FMT);\ + LEN = vsprintf(PTR, FMT, args);\ + va_end(args);\ + *(PTR + LEN) = '\0';\ +} + +/* + * Locking policy when using print buffers. + * + * 1) Routines of the form printbuf_XXX() rely on the caller to prevent + * simultaneous use of the print buffer(s) being manipulated. + * 2) tipc_printf() uses 'print_lock' to prevent simultaneous use of + * 'print_string' and to protect its print buffer(s). + * 3) TEE() uses 'print_lock' to protect its print buffer(s). + * 4) Routines of the form log_XXX() uses 'print_lock' to protect LOG. + */ + +/** + * printbuf_init - initialize print buffer to empty + */ + +void printbuf_init(struct print_buf *pb, char *raw, u32 sz) +{ + if (!pb || !raw || (sz < (MAX_STRING + 1))) + return; + + pb->crs = pb->buf = raw; + pb->size = sz; + pb->next = 0; + pb->buf[0] = 0; + pb->buf[sz-1] = ~0; +} + +/** + * printbuf_reset - reinitialize print buffer to empty state + */ + +void printbuf_reset(struct print_buf *pb) +{ + if (pb && pb->buf) + printbuf_init(pb, pb->buf, pb->size); +} + +/** + * printbuf_empty - test if print buffer is in empty state + */ + +int printbuf_empty(struct print_buf *pb) +{ + return (!pb || !pb->buf || (pb->crs == pb->buf)); +} + +/** + * printbuf_validate - check for print buffer overflow + * + * Verifies that a print buffer has captured all data written to it. + * If data has been lost, linearize buffer and prepend an error message + * + * Returns length of print buffer data string (including trailing NULL) + */ + +int printbuf_validate(struct print_buf *pb) +{ + char *err = " *** PRINT BUFFER WRAPPED AROUND ***\n"; + char *cp_buf; + struct print_buf cb; + + if (!pb || !pb->buf) + return 0; + + if (pb->buf[pb->size - 1] == '\0') { + cp_buf = kmalloc(pb->size, GFP_ATOMIC); + if (cp_buf != NULL){ + printbuf_init(&cb, cp_buf, pb->size); + printbuf_move(&cb, pb); + printbuf_move(pb, &cb); + kfree(cp_buf); + memcpy(pb->buf, err, strlen(err)); + } else { + printbuf_reset(pb); + tipc_printf(pb, err); + } + } + return (pb->crs - pb->buf + 1); +} + +/** + * printbuf_move - move print buffer contents to another print buffer + * + * Current contents of destination print buffer (if any) are discarded. + * Source print buffer becomes empty if a successful move occurs. + */ + +void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from) +{ + int len; + + /* Handle the cases where contents can't be moved */ + + if (!pb_to || !pb_to->buf) + return; + + if (!pb_from || !pb_from->buf) { + printbuf_reset(pb_to); + return; + } + + if (pb_to->size < pb_from->size) { + printbuf_reset(pb_to); + tipc_printf(pb_to, "*** PRINT BUFFER OVERFLOW ***"); + return; + } + + /* Copy data from char after cursor to end (if used) */ + len = pb_from->buf + pb_from->size - pb_from->crs - 2; + if ((pb_from->buf[pb_from->size-1] == 0) && (len > 0)) { + strcpy(pb_to->buf, pb_from->crs + 1); + pb_to->crs = pb_to->buf + len; + } else + pb_to->crs = pb_to->buf; + + /* Copy data from start to cursor (always) */ + len = pb_from->crs - pb_from->buf; + strcpy(pb_to->crs, pb_from->buf); + pb_to->crs += len; + + printbuf_reset(pb_from); +} + +/** + * tipc_printf - append formatted output to print buffer chain + */ + +void tipc_printf(struct print_buf *pb, const char *fmt, ...) +{ + int chars_to_add; + int chars_left; + char save_char; + struct print_buf *pb_next; + + spin_lock_bh(&print_lock); + FORMAT(print_string, chars_to_add, fmt); + if (chars_to_add >= MAX_STRING) + strcpy(print_string, "*** STRING TOO LONG ***"); + + while (pb) { + if (pb == CONS) + printk(print_string); + else if (pb->buf) { + chars_left = pb->buf + pb->size - pb->crs - 1; + if (chars_to_add <= chars_left) { + strcpy(pb->crs, print_string); + pb->crs += chars_to_add; + } else { + strcpy(pb->buf, print_string + chars_left); + save_char = print_string[chars_left]; + print_string[chars_left] = 0; + strcpy(pb->crs, print_string); + print_string[chars_left] = save_char; + pb->crs = pb->buf + chars_to_add - chars_left; + } + } + pb_next = pb->next; + pb->next = 0; + pb = pb_next; + } + spin_unlock_bh(&print_lock); +} + +/** + * TEE - perform next output operation on both print buffers + */ + +struct print_buf *TEE(struct print_buf *b0, struct print_buf *b1) +{ + struct print_buf *pb = b0; + + if (!b0 || (b0 == b1)) + return b1; + if (!b1) + return b0; + + spin_lock_bh(&print_lock); + while (pb->next) { + if ((pb->next == b1) || (pb->next == b0)) + pb->next = pb->next->next; + else + pb = pb->next; + } + pb->next = b1; + spin_unlock_bh(&print_lock); + return b0; +} + +/** + * print_to_console - write string of bytes to console in multiple chunks + */ + +static void print_to_console(char *crs, int len) +{ + int rest = len; + + while (rest > 0) { + int sz = rest < MAX_STRING ? rest : MAX_STRING; + char c = crs[sz]; + + crs[sz] = 0; + printk((const char *)crs); + crs[sz] = c; + rest -= sz; + crs += sz; + } +} + +/** + * printbuf_dump - write print buffer contents to console + */ + +static void printbuf_dump(struct print_buf *pb) +{ + int len; + + /* Dump print buffer from char after cursor to end (if used) */ + len = pb->buf + pb->size - pb->crs - 2; + if ((pb->buf[pb->size - 1] == 0) && (len > 0)) + print_to_console(pb->crs + 1, len); + + /* Dump print buffer from start to cursor (always) */ + len = pb->crs - pb->buf; + print_to_console(pb->buf, len); +} + +/** + * tipc_dump - dump non-console print buffer(s) to console + */ + +void tipc_dump(struct print_buf *pb, const char *fmt, ...) +{ + int len; + + spin_lock_bh(&print_lock); + FORMAT(CONS->buf, len, fmt); + printk(CONS->buf); + + for (; pb; pb = pb->next) { + if (pb == CONS) + continue; + printk("\n---- Start of dump,%s log ----\n\n", + (pb == LOG) ? "global" : "local"); + printbuf_dump(pb); + printbuf_reset(pb); + printk("\n-------- End of dump --------\n"); + } + spin_unlock_bh(&print_lock); +} + +/** + * log_stop - free up TIPC log print buffer + */ + +void log_stop(void) +{ + spin_lock_bh(&print_lock); + if (LOG->buf) { + kfree(LOG->buf); + LOG->buf = NULL; + } + spin_unlock_bh(&print_lock); +} + +/** + * log_reinit - set TIPC log print buffer to specified size + */ + +void log_reinit(int log_size) +{ + log_stop(); + + if (log_size) { + if (log_size <= MAX_STRING) + log_size = MAX_STRING + 1; + spin_lock_bh(&print_lock); + printbuf_init(LOG, kmalloc(log_size, GFP_ATOMIC), log_size); + spin_unlock_bh(&print_lock); + } +} + +/** + * log_resize - reconfigure size of TIPC log buffer + */ + +struct sk_buff *log_resize(const void *req_tlv_area, int req_tlv_space) +{ + u32 value; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_UNSIGNED)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + value = *(u32 *)TLV_DATA(req_tlv_area); + value = ntohl(value); + if (value != delimit(value, 0, 32768)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (log size must be 0-32768)"); + log_reinit(value); + return cfg_reply_none(); +} + +/** + * log_dump - capture TIPC log buffer contents in configuration message + */ + +struct sk_buff *log_dump(void) +{ + struct sk_buff *reply; + + spin_lock_bh(&print_lock); + if (!LOG->buf) + reply = cfg_reply_ultra_string("log not activated\n"); + else if (printbuf_empty(LOG)) + reply = cfg_reply_ultra_string("log is empty\n"); + else { + struct tlv_desc *rep_tlv; + struct print_buf pb; + int str_len; + + str_len = min(LOG->size, 32768u); + reply = cfg_reply_alloc(TLV_SPACE(str_len)); + if (reply) { + rep_tlv = (struct tlv_desc *)reply->data; + printbuf_init(&pb, TLV_DATA(rep_tlv), str_len); + printbuf_move(&pb, LOG); + str_len = strlen(TLV_DATA(rep_tlv)) + 1; + skb_put(reply, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + } + } + spin_unlock_bh(&print_lock); + return reply; +} + diff --git a/net/tipc/dbg.h b/net/tipc/dbg.h new file mode 100644 index 00000000000..c6b2a64c224 --- /dev/null +++ b/net/tipc/dbg.h @@ -0,0 +1,59 @@ +/* + * net/tipc/dbg.h: Include file for TIPC print buffer routines + * + * Copyright (c) 1997-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_DBG_H +#define _TIPC_DBG_H + +struct print_buf { + char *buf; + u32 size; + char *crs; + struct print_buf *next; +}; + +void printbuf_init(struct print_buf *pb, char *buf, u32 sz); +void printbuf_reset(struct print_buf *pb); +int printbuf_empty(struct print_buf *pb); +int printbuf_validate(struct print_buf *pb); +void printbuf_move(struct print_buf *pb_to, struct print_buf *pb_from); + +void log_reinit(int log_size); +void log_stop(void); + +struct sk_buff *log_resize(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *log_dump(void); + +#endif diff --git a/net/tipc/discover.c b/net/tipc/discover.c new file mode 100644 index 00000000000..b106ef1621c --- /dev/null +++ b/net/tipc/discover.c @@ -0,0 +1,318 @@ +/* + * net/tipc/discover.c + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "link.h" +#include "zone.h" +#include "discover.h" +#include "port.h" +#include "name_table.h" + +#define TIPC_LINK_REQ_INIT 125 /* min delay during bearer start up */ +#define TIPC_LINK_REQ_FAST 2000 /* normal delay if bearer has no links */ +#define TIPC_LINK_REQ_SLOW 600000 /* normal delay if bearer has links */ + +#if 0 +#define GET_NODE_INFO 300 +#define GET_NODE_INFO_RESULT 301 +#define FORWARD_LINK_PROBE 302 +#define LINK_REQUEST_REJECTED 303 +#define LINK_REQUEST_ACCEPTED 304 +#define DROP_LINK_REQUEST 305 +#define CHECK_LINK_COUNT 306 +#endif + +/* + * TODO: Most of the inter-cluster setup stuff should be + * rewritten, and be made conformant with specification. + */ + + +/** + * struct link_req - information about an ongoing link setup request + * @bearer: bearer issuing requests + * @dest: destination address for request messages + * @buf: request message to be (repeatedly) sent + * @timer: timer governing period between requests + * @timer_intv: current interval between requests (in ms) + */ +struct link_req { + struct bearer *bearer; + struct tipc_media_addr dest; + struct sk_buff *buf; + struct timer_list timer; + unsigned int timer_intv; +}; + + +#if 0 +int disc_create_link(const struct tipc_link_create *argv) +{ + /* + * Code for inter cluster link setup here + */ + return TIPC_OK; +} +#endif + +/* + * disc_lost_link(): A link has lost contact + */ + +void disc_link_event(u32 addr, char *name, int up) +{ + if (in_own_cluster(addr)) + return; + /* + * Code for inter cluster link setup here + */ +} + +/** + * disc_init_msg - initialize a link setup message + * @type: message type (request or response) + * @req_links: number of links associated with message + * @dest_domain: network domain of node(s) which should respond to message + * @b_ptr: ptr to bearer issuing message + */ + +struct sk_buff *disc_init_msg(u32 type, + u32 req_links, + u32 dest_domain, + struct bearer *b_ptr) +{ + struct sk_buff *buf = buf_acquire(DSC_H_SIZE); + struct tipc_msg *msg; + + if (buf) { + msg = buf_msg(buf); + msg_init(msg, LINK_CONFIG, type, TIPC_OK, DSC_H_SIZE, + dest_domain); + msg_set_non_seq(msg); + msg_set_req_links(msg, req_links); + msg_set_dest_domain(msg, dest_domain); + msg_set_bc_netid(msg, tipc_net_id); + msg_set_media_addr(msg, &b_ptr->publ.addr); + } + return buf; +} + +/** + * disc_recv_msg - handle incoming link setup message (request or response) + * @buf: buffer containing message + */ + +void disc_recv_msg(struct sk_buff *buf) +{ + struct bearer *b_ptr = (struct bearer *)TIPC_SKB_CB(buf)->handle; + struct link *link; + struct tipc_media_addr media_addr; + struct tipc_msg *msg = buf_msg(buf); + u32 dest = msg_dest_domain(msg); + u32 orig = msg_prevnode(msg); + u32 net_id = msg_bc_netid(msg); + u32 type = msg_type(msg); + + msg_get_media_addr(msg,&media_addr); + msg_dbg(msg, "RECV:"); + buf_discard(buf); + + if (net_id != tipc_net_id) + return; + if (!addr_domain_valid(dest)) + return; + if (!addr_node_valid(orig)) + return; + if (orig == tipc_own_addr) + return; + if (!in_scope(dest, tipc_own_addr)) + return; + if (is_slave(tipc_own_addr) && is_slave(orig)) + return; + if (is_slave(orig) && !in_own_cluster(orig)) + return; + if (in_own_cluster(orig)) { + /* Always accept link here */ + struct sk_buff *rbuf; + struct tipc_media_addr *addr; + struct node *n_ptr = node_find(orig); + int link_up; + dbg(" in own cluster\n"); + if (n_ptr == NULL) { + n_ptr = node_create(orig); + } + if (n_ptr == NULL) { + warn("Memory squeeze; Failed to create node\n"); + return; + } + spin_lock_bh(&n_ptr->lock); + link = n_ptr->links[b_ptr->identity]; + if (!link) { + dbg("creating link\n"); + link = link_create(b_ptr, orig, &media_addr); + if (!link) { + spin_unlock_bh(&n_ptr->lock); + return; + } + } + addr = &link->media_addr; + if (memcmp(addr, &media_addr, sizeof(*addr))) { + char addr_string[16]; + + warn("New bearer address for %s\n", + addr_string_fill(addr_string, orig)); + memcpy(addr, &media_addr, sizeof(*addr)); + link_reset(link); + } + link_up = link_is_up(link); + spin_unlock_bh(&n_ptr->lock); + if ((type == DSC_RESP_MSG) || link_up) + return; + rbuf = disc_init_msg(DSC_RESP_MSG, 1, orig, b_ptr); + if (rbuf != NULL) { + msg_dbg(buf_msg(rbuf),"SEND:"); + b_ptr->media->send_msg(rbuf, &b_ptr->publ, &media_addr); + buf_discard(rbuf); + } + } +} + +/** + * disc_stop_link_req - stop sending periodic link setup requests + * @req: ptr to link request structure + */ + +void disc_stop_link_req(struct link_req *req) +{ + if (!req) + return; + + k_cancel_timer(&req->timer); + k_term_timer(&req->timer); + buf_discard(req->buf); + kfree(req); +} + +/** + * disc_update_link_req - update frequency of periodic link setup requests + * @req: ptr to link request structure + */ + +void disc_update_link_req(struct link_req *req) +{ + if (!req) + return; + + if (req->timer_intv == TIPC_LINK_REQ_SLOW) { + if (!req->bearer->nodes.count) { + req->timer_intv = TIPC_LINK_REQ_FAST; + k_start_timer(&req->timer, req->timer_intv); + } + } else if (req->timer_intv == TIPC_LINK_REQ_FAST) { + if (req->bearer->nodes.count) { + req->timer_intv = TIPC_LINK_REQ_SLOW; + k_start_timer(&req->timer, req->timer_intv); + } + } else { + /* leave timer "as is" if haven't yet reached a "normal" rate */ + } +} + +/** + * disc_timeout - send a periodic link setup request + * @req: ptr to link request structure + * + * Called whenever a link setup request timer associated with a bearer expires. + */ + +static void disc_timeout(struct link_req *req) +{ + spin_lock_bh(&req->bearer->publ.lock); + + req->bearer->media->send_msg(req->buf, &req->bearer->publ, &req->dest); + + if ((req->timer_intv == TIPC_LINK_REQ_SLOW) || + (req->timer_intv == TIPC_LINK_REQ_FAST)) { + /* leave timer interval "as is" if already at a "normal" rate */ + } else { + req->timer_intv *= 2; + if (req->timer_intv > TIPC_LINK_REQ_SLOW) + req->timer_intv = TIPC_LINK_REQ_SLOW; + if ((req->timer_intv == TIPC_LINK_REQ_FAST) && + (req->bearer->nodes.count)) + req->timer_intv = TIPC_LINK_REQ_SLOW; + } + k_start_timer(&req->timer, req->timer_intv); + + spin_unlock_bh(&req->bearer->publ.lock); +} + +/** + * disc_init_link_req - start sending periodic link setup requests + * @b_ptr: ptr to bearer issuing requests + * @dest: destination address for request messages + * @dest_domain: network domain of node(s) which should respond to message + * @req_links: max number of desired links + * + * Returns pointer to link request structure, or NULL if unable to create. + */ + +struct link_req *disc_init_link_req(struct bearer *b_ptr, + const struct tipc_media_addr *dest, + u32 dest_domain, + u32 req_links) +{ + struct link_req *req; + + req = (struct link_req *)kmalloc(sizeof(*req), GFP_ATOMIC); + if (!req) + return NULL; + + req->buf = disc_init_msg(DSC_REQ_MSG, req_links, dest_domain, b_ptr); + if (!req->buf) { + kfree(req); + return NULL; + } + + memcpy(&req->dest, dest, sizeof(*dest)); + req->bearer = b_ptr; + req->timer_intv = TIPC_LINK_REQ_INIT; + k_init_timer(&req->timer, (Handler)disc_timeout, (unsigned long)req); + k_start_timer(&req->timer, req->timer_intv); + return req; +} + diff --git a/net/tipc/discover.h b/net/tipc/discover.h new file mode 100644 index 00000000000..2a6114d9162 --- /dev/null +++ b/net/tipc/discover.h @@ -0,0 +1,58 @@ +/* + * net/tipc/discover.h + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_DISCOVER_H +#define _TIPC_DISCOVER_H + +#include <linux/tipc.h> + +struct link_req; + +struct link_req *disc_init_link_req(struct bearer *b_ptr, + const struct tipc_media_addr *dest, + u32 dest_domain, + u32 req_links); +void disc_update_link_req(struct link_req *req); +void disc_stop_link_req(struct link_req *req); + +void disc_recv_msg(struct sk_buff *buf); + +void disc_link_event(u32 addr, char *name, int up); +#if 0 +int disc_create_link(const struct tipc_link_create *argv); +#endif + +#endif diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c new file mode 100644 index 00000000000..34d0462db3a --- /dev/null +++ b/net/tipc/eth_media.c @@ -0,0 +1,299 @@ +/* + * net/tipc/eth_media.c: Ethernet bearer support for TIPC + * + * Copyright (c) 2001-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <net/tipc/tipc.h> +#include <net/tipc/tipc_bearer.h> +#include <net/tipc/tipc_msg.h> +#include <linux/netdevice.h> +#include <linux/version.h> + +#define MAX_ETH_BEARERS 2 +#define TIPC_PROTOCOL 0x88ca +#define ETH_LINK_PRIORITY 10 +#define ETH_LINK_TOLERANCE TIPC_DEF_LINK_TOL + + +/** + * struct eth_bearer - Ethernet bearer data structure + * @bearer: ptr to associated "generic" bearer structure + * @dev: ptr to associated Ethernet network device + * @tipc_packet_type: used in binding TIPC to Ethernet driver + */ + +struct eth_bearer { + struct tipc_bearer *bearer; + struct net_device *dev; + struct packet_type tipc_packet_type; +}; + +static struct eth_bearer eth_bearers[MAX_ETH_BEARERS]; +static int eth_started = 0; +static struct notifier_block notifier; + +/** + * send_msg - send a TIPC message out over an Ethernet interface + */ + +static int send_msg(struct sk_buff *buf, struct tipc_bearer *tb_ptr, + struct tipc_media_addr *dest) +{ + struct sk_buff *clone; + struct net_device *dev; + + clone = skb_clone(buf, GFP_ATOMIC); + if (clone) { + clone->nh.raw = clone->data; + dev = ((struct eth_bearer *)(tb_ptr->usr_handle))->dev; + clone->dev = dev; + dev->hard_header(clone, dev, TIPC_PROTOCOL, + &dest->dev_addr.eth_addr, + dev->dev_addr, clone->len); + dev_queue_xmit(clone); + } + return TIPC_OK; +} + +/** + * recv_msg - handle incoming TIPC message from an Ethernet interface + * + * Routine truncates any Ethernet padding/CRC appended to the message, + * and ensures message size matches actual length + */ + +static int recv_msg(struct sk_buff *buf, struct net_device *dev, + struct packet_type *pt, struct net_device *orig_dev) +{ + struct eth_bearer *eb_ptr = (struct eth_bearer *)pt->af_packet_priv; + u32 size; + + if (likely(eb_ptr->bearer)) { + size = msg_size((struct tipc_msg *)buf->data); + skb_trim(buf, size); + if (likely(buf->len == size)) { + buf->next = NULL; + tipc_recv_msg(buf, eb_ptr->bearer); + } else { + kfree_skb(buf); + } + } else { + kfree_skb(buf); + } + return TIPC_OK; +} + +/** + * enable_bearer - attach TIPC bearer to an Ethernet interface + */ + +static int enable_bearer(struct tipc_bearer *tb_ptr) +{ + struct net_device *dev = dev_base; + struct eth_bearer *eb_ptr = ð_bearers[0]; + struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; + char *driver_name = strchr((const char *)tb_ptr->name, ':') + 1; + + /* Find device with specified name */ + + while (dev && dev->name && + (memcmp(dev->name, driver_name, strlen(dev->name)))) { + dev = dev->next; + } + if (!dev) + return -ENODEV; + + /* Find Ethernet bearer for device (or create one) */ + + for (;(eb_ptr != stop) && eb_ptr->dev && (eb_ptr->dev != dev); eb_ptr++); + if (eb_ptr == stop) + return -EDQUOT; + if (!eb_ptr->dev) { + eb_ptr->dev = dev; + eb_ptr->tipc_packet_type.type = __constant_htons(TIPC_PROTOCOL); + eb_ptr->tipc_packet_type.dev = dev; + eb_ptr->tipc_packet_type.func = recv_msg; + eb_ptr->tipc_packet_type.af_packet_priv = eb_ptr; + INIT_LIST_HEAD(&(eb_ptr->tipc_packet_type.list)); + dev_hold(dev); + dev_add_pack(&eb_ptr->tipc_packet_type); + } + + /* Associate TIPC bearer with Ethernet bearer */ + + eb_ptr->bearer = tb_ptr; + tb_ptr->usr_handle = (void *)eb_ptr; + tb_ptr->mtu = dev->mtu; + tb_ptr->blocked = 0; + tb_ptr->addr.type = htonl(TIPC_MEDIA_TYPE_ETH); + memcpy(&tb_ptr->addr.dev_addr, &dev->dev_addr, ETH_ALEN); + return 0; +} + +/** + * disable_bearer - detach TIPC bearer from an Ethernet interface + * + * We really should do dev_remove_pack() here, but this function can not be + * called at tasklet level. => Use eth_bearer->bearer as a flag to throw away + * incoming buffers, & postpone dev_remove_pack() to eth_media_stop() on exit. + */ + +static void disable_bearer(struct tipc_bearer *tb_ptr) +{ + ((struct eth_bearer *)tb_ptr->usr_handle)->bearer = 0; +} + +/** + * recv_notification - handle device updates from OS + * + * Change the state of the Ethernet bearer (if any) associated with the + * specified device. + */ + +static int recv_notification(struct notifier_block *nb, unsigned long evt, + void *dv) +{ + struct net_device *dev = (struct net_device *)dv; + struct eth_bearer *eb_ptr = ð_bearers[0]; + struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; + + while ((eb_ptr->dev != dev)) { + if (++eb_ptr == stop) + return NOTIFY_DONE; /* couldn't find device */ + } + if (!eb_ptr->bearer) + return NOTIFY_DONE; /* bearer had been disabled */ + + eb_ptr->bearer->mtu = dev->mtu; + + switch (evt) { + case NETDEV_CHANGE: + if (netif_carrier_ok(dev)) + tipc_continue(eb_ptr->bearer); + else + tipc_block_bearer(eb_ptr->bearer->name); + break; + case NETDEV_UP: + tipc_continue(eb_ptr->bearer); + break; + case NETDEV_DOWN: + tipc_block_bearer(eb_ptr->bearer->name); + break; + case NETDEV_CHANGEMTU: + case NETDEV_CHANGEADDR: + tipc_block_bearer(eb_ptr->bearer->name); + tipc_continue(eb_ptr->bearer); + break; + case NETDEV_UNREGISTER: + case NETDEV_CHANGENAME: + tipc_disable_bearer(eb_ptr->bearer->name); + break; + } + return NOTIFY_OK; +} + +/** + * eth_addr2str - convert Ethernet address to string + */ + +static char *eth_addr2str(struct tipc_media_addr *a, char *str_buf, int str_size) +{ + unchar *addr = (unchar *)&a->dev_addr; + + if (str_size < 18) + *str_buf = '\0'; + else + sprintf(str_buf, "%02x:%02x:%02x:%02x:%02x:%02x", + addr[0], addr[1], addr[2], addr[3], addr[4], addr[5]); + return str_buf; +} + +/** + * eth_media_start - activate Ethernet bearer support + * + * Register Ethernet media type with TIPC bearer code. Also register + * with OS for notifications about device state changes. + */ + +int eth_media_start(void) +{ + struct tipc_media_addr bcast_addr; + int res; + + if (eth_started) + return -EINVAL; + + memset(&bcast_addr, 0xff, sizeof(bcast_addr)); + memset(eth_bearers, 0, sizeof(eth_bearers)); + + res = tipc_register_media(TIPC_MEDIA_TYPE_ETH, "eth", + enable_bearer, disable_bearer, send_msg, + eth_addr2str, &bcast_addr, ETH_LINK_PRIORITY, + ETH_LINK_TOLERANCE, TIPC_DEF_LINK_WIN); + if (res) + return res; + + notifier.notifier_call = &recv_notification; + notifier.priority = 0; + res = register_netdevice_notifier(¬ifier); + if (!res) + eth_started = 1; + return res; +} + +/** + * eth_media_stop - deactivate Ethernet bearer support + */ + +void eth_media_stop(void) +{ + int i; + + if (!eth_started) + return; + + unregister_netdevice_notifier(¬ifier); + for (i = 0; i < MAX_ETH_BEARERS ; i++) { + if (eth_bearers[i].bearer) { + eth_bearers[i].bearer->blocked = 1; + eth_bearers[i].bearer = 0; + } + if (eth_bearers[i].dev) { + dev_remove_pack(ð_bearers[i].tipc_packet_type); + dev_put(eth_bearers[i].dev); + } + } + memset(ð_bearers, 0, sizeof(eth_bearers)); + eth_started = 0; +} diff --git a/net/tipc/handler.c b/net/tipc/handler.c new file mode 100644 index 00000000000..f320010f8a6 --- /dev/null +++ b/net/tipc/handler.c @@ -0,0 +1,132 @@ +/* + * net/tipc/handler.c: TIPC signal handling + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" + +struct queue_item { + struct list_head next_signal; + void (*handler) (unsigned long); + unsigned long data; +}; + +static kmem_cache_t *tipc_queue_item_cache; +static struct list_head signal_queue_head; +static spinlock_t qitem_lock = SPIN_LOCK_UNLOCKED; +static int handler_enabled = 0; + +static void process_signal_queue(unsigned long dummy); + +static DECLARE_TASKLET_DISABLED(tipc_tasklet, process_signal_queue, 0); + + +unsigned int k_signal(Handler routine, unsigned long argument) +{ + struct queue_item *item; + + if (!handler_enabled) { + err("Signal request ignored by handler\n"); + return -ENOPROTOOPT; + } + + spin_lock_bh(&qitem_lock); + item = kmem_cache_alloc(tipc_queue_item_cache, GFP_ATOMIC); + if (!item) { + err("Signal queue out of memory\n"); + spin_unlock_bh(&qitem_lock); + return -ENOMEM; + } + item->handler = routine; + item->data = argument; + list_add_tail(&item->next_signal, &signal_queue_head); + spin_unlock_bh(&qitem_lock); + tasklet_schedule(&tipc_tasklet); + return 0; +} + +static void process_signal_queue(unsigned long dummy) +{ + struct queue_item *__volatile__ item; + struct list_head *l, *n; + + spin_lock_bh(&qitem_lock); + list_for_each_safe(l, n, &signal_queue_head) { + item = list_entry(l, struct queue_item, next_signal); + list_del(&item->next_signal); + spin_unlock_bh(&qitem_lock); + item->handler(item->data); + spin_lock_bh(&qitem_lock); + kmem_cache_free(tipc_queue_item_cache, item); + } + spin_unlock_bh(&qitem_lock); +} + +int handler_start(void) +{ + tipc_queue_item_cache = + kmem_cache_create("tipc_queue_items", sizeof(struct queue_item), + 0, SLAB_HWCACHE_ALIGN, NULL, NULL); + if (!tipc_queue_item_cache) + return -ENOMEM; + + INIT_LIST_HEAD(&signal_queue_head); + tasklet_enable(&tipc_tasklet); + handler_enabled = 1; + return 0; +} + +void handler_stop(void) +{ + struct list_head *l, *n; + struct queue_item *item; + + if (!handler_enabled) + return; + + handler_enabled = 0; + tasklet_disable(&tipc_tasklet); + tasklet_kill(&tipc_tasklet); + + spin_lock_bh(&qitem_lock); + list_for_each_safe(l, n, &signal_queue_head) { + item = list_entry(l, struct queue_item, next_signal); + list_del(&item->next_signal); + kmem_cache_free(tipc_queue_item_cache, item); + } + spin_unlock_bh(&qitem_lock); + + kmem_cache_destroy(tipc_queue_item_cache); +} + diff --git a/net/tipc/link.c b/net/tipc/link.c new file mode 100644 index 00000000000..7265f4be476 --- /dev/null +++ b/net/tipc/link.c @@ -0,0 +1,3167 @@ +/* + * net/tipc/link.c: TIPC link code + * + * Copyright (c) 1996-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "link.h" +#include "net.h" +#include "node.h" +#include "port.h" +#include "addr.h" +#include "node_subscr.h" +#include "name_distr.h" +#include "bearer.h" +#include "name_table.h" +#include "discover.h" +#include "config.h" +#include "bcast.h" + + +/* + * Limit for deferred reception queue: + */ + +#define DEF_QUEUE_LIMIT 256u + +/* + * Link state events: + */ + +#define STARTING_EVT 856384768 /* link processing trigger */ +#define TRAFFIC_MSG_EVT 560815u /* rx'd ??? */ +#define TIMEOUT_EVT 560817u /* link timer expired */ + +/* + * The following two 'message types' is really just implementation + * data conveniently stored in the message header. + * They must not be considered part of the protocol + */ +#define OPEN_MSG 0 +#define CLOSED_MSG 1 + +/* + * State value stored in 'exp_msg_count' + */ + +#define START_CHANGEOVER 100000u + +/** + * struct link_name - deconstructed link name + * @addr_local: network address of node at this end + * @if_local: name of interface at this end + * @addr_peer: network address of node at far end + * @if_peer: name of interface at far end + */ + +struct link_name { + u32 addr_local; + char if_local[TIPC_MAX_IF_NAME]; + u32 addr_peer; + char if_peer[TIPC_MAX_IF_NAME]; +}; + +#if 0 + +/* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ + +/** + * struct link_event - link up/down event notification + */ + +struct link_event { + u32 addr; + int up; + void (*fcn)(u32, char *, int); + char name[TIPC_MAX_LINK_NAME]; +}; + +#endif + +static void link_handle_out_of_seq_msg(struct link *l_ptr, + struct sk_buff *buf); +static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf); +static int link_recv_changeover_msg(struct link **l_ptr, struct sk_buff **buf); +static void link_set_supervision_props(struct link *l_ptr, u32 tolerance); +static int link_send_sections_long(struct port *sender, + struct iovec const *msg_sect, + u32 num_sect, u32 destnode); +static void link_check_defragm_bufs(struct link *l_ptr); +static void link_state_event(struct link *l_ptr, u32 event); +static void link_reset_statistics(struct link *l_ptr); +static void link_print(struct link *l_ptr, struct print_buf *buf, + const char *str); + +/* + * Debugging code used by link routines only + * + * When debugging link problems on a system that has multiple links, + * the standard TIPC debugging routines may not be useful since they + * allow the output from multiple links to be intermixed. For this reason + * routines of the form "dbg_link_XXX()" have been created that will capture + * debug info into a link's personal print buffer, which can then be dumped + * into the TIPC system log (LOG) upon request. + * + * To enable per-link debugging, use LINK_LOG_BUF_SIZE to specify the size + * of the print buffer used by each link. If LINK_LOG_BUF_SIZE is set to 0, + * the dbg_link_XXX() routines simply send their output to the standard + * debug print buffer (DBG_OUTPUT), if it has been defined; this can be useful + * when there is only a single link in the system being debugged. + * + * Notes: + * - When enabled, LINK_LOG_BUF_SIZE should be set to at least 1000 (bytes) + * - "l_ptr" must be valid when using dbg_link_XXX() macros + */ + +#define LINK_LOG_BUF_SIZE 0 + +#define dbg_link(fmt, arg...) do {if (LINK_LOG_BUF_SIZE) tipc_printf(&l_ptr->print_buf, fmt, ## arg); } while(0) +#define dbg_link_msg(msg, txt) do {if (LINK_LOG_BUF_SIZE) msg_print(&l_ptr->print_buf, msg, txt); } while(0) +#define dbg_link_state(txt) do {if (LINK_LOG_BUF_SIZE) link_print(l_ptr, &l_ptr->print_buf, txt); } while(0) +#define dbg_link_dump() do { \ + if (LINK_LOG_BUF_SIZE) { \ + tipc_printf(LOG, "\n\nDumping link <%s>:\n", l_ptr->name); \ + printbuf_move(LOG, &l_ptr->print_buf); \ + } \ +} while (0) + +static inline void dbg_print_link(struct link *l_ptr, const char *str) +{ + if (DBG_OUTPUT) + link_print(l_ptr, DBG_OUTPUT, str); +} + +static inline void dbg_print_buf_chain(struct sk_buff *root_buf) +{ + if (DBG_OUTPUT) { + struct sk_buff *buf = root_buf; + + while (buf) { + msg_dbg(buf_msg(buf), "In chain: "); + buf = buf->next; + } + } +} + +/* + * Simple inlined link routines + */ + +static inline unsigned int align(unsigned int i) +{ + return (i + 3) & ~3u; +} + +static inline int link_working_working(struct link *l_ptr) +{ + return (l_ptr->state == WORKING_WORKING); +} + +static inline int link_working_unknown(struct link *l_ptr) +{ + return (l_ptr->state == WORKING_UNKNOWN); +} + +static inline int link_reset_unknown(struct link *l_ptr) +{ + return (l_ptr->state == RESET_UNKNOWN); +} + +static inline int link_reset_reset(struct link *l_ptr) +{ + return (l_ptr->state == RESET_RESET); +} + +static inline int link_blocked(struct link *l_ptr) +{ + return (l_ptr->exp_msg_count || l_ptr->blocked); +} + +static inline int link_congested(struct link *l_ptr) +{ + return (l_ptr->out_queue_size >= l_ptr->queue_limit[0]); +} + +static inline u32 link_max_pkt(struct link *l_ptr) +{ + return l_ptr->max_pkt; +} + +static inline void link_init_max_pkt(struct link *l_ptr) +{ + u32 max_pkt; + + max_pkt = (l_ptr->b_ptr->publ.mtu & ~3); + if (max_pkt > MAX_MSG_SIZE) + max_pkt = MAX_MSG_SIZE; + + l_ptr->max_pkt_target = max_pkt; + if (l_ptr->max_pkt_target < MAX_PKT_DEFAULT) + l_ptr->max_pkt = l_ptr->max_pkt_target; + else + l_ptr->max_pkt = MAX_PKT_DEFAULT; + + l_ptr->max_pkt_probes = 0; +} + +static inline u32 link_next_sent(struct link *l_ptr) +{ + if (l_ptr->next_out) + return msg_seqno(buf_msg(l_ptr->next_out)); + return mod(l_ptr->next_out_no); +} + +static inline u32 link_last_sent(struct link *l_ptr) +{ + return mod(link_next_sent(l_ptr) - 1); +} + +/* + * Simple non-inlined link routines (i.e. referenced outside this file) + */ + +int link_is_up(struct link *l_ptr) +{ + if (!l_ptr) + return 0; + return (link_working_working(l_ptr) || link_working_unknown(l_ptr)); +} + +int link_is_active(struct link *l_ptr) +{ + return ((l_ptr->owner->active_links[0] == l_ptr) || + (l_ptr->owner->active_links[1] == l_ptr)); +} + +/** + * link_name_validate - validate & (optionally) deconstruct link name + * @name - ptr to link name string + * @name_parts - ptr to area for link name components (or NULL if not needed) + * + * Returns 1 if link name is valid, otherwise 0. + */ + +static int link_name_validate(const char *name, struct link_name *name_parts) +{ + char name_copy[TIPC_MAX_LINK_NAME]; + char *addr_local; + char *if_local; + char *addr_peer; + char *if_peer; + char dummy; + u32 z_local, c_local, n_local; + u32 z_peer, c_peer, n_peer; + u32 if_local_len; + u32 if_peer_len; + + /* copy link name & ensure length is OK */ + + name_copy[TIPC_MAX_LINK_NAME - 1] = 0; + /* need above in case non-Posix strncpy() doesn't pad with nulls */ + strncpy(name_copy, name, TIPC_MAX_LINK_NAME); + if (name_copy[TIPC_MAX_LINK_NAME - 1] != 0) + return 0; + + /* ensure all component parts of link name are present */ + + addr_local = name_copy; + if ((if_local = strchr(addr_local, ':')) == NULL) + return 0; + *(if_local++) = 0; + if ((addr_peer = strchr(if_local, '-')) == NULL) + return 0; + *(addr_peer++) = 0; + if_local_len = addr_peer - if_local; + if ((if_peer = strchr(addr_peer, ':')) == NULL) + return 0; + *(if_peer++) = 0; + if_peer_len = strlen(if_peer) + 1; + + /* validate component parts of link name */ + + if ((sscanf(addr_local, "%u.%u.%u%c", + &z_local, &c_local, &n_local, &dummy) != 3) || + (sscanf(addr_peer, "%u.%u.%u%c", + &z_peer, &c_peer, &n_peer, &dummy) != 3) || + (z_local > 255) || (c_local > 4095) || (n_local > 4095) || + (z_peer > 255) || (c_peer > 4095) || (n_peer > 4095) || + (if_local_len <= 1) || (if_local_len > TIPC_MAX_IF_NAME) || + (if_peer_len <= 1) || (if_peer_len > TIPC_MAX_IF_NAME) || + (strspn(if_local, tipc_alphabet) != (if_local_len - 1)) || + (strspn(if_peer, tipc_alphabet) != (if_peer_len - 1))) + return 0; + + /* return link name components, if necessary */ + + if (name_parts) { + name_parts->addr_local = tipc_addr(z_local, c_local, n_local); + strcpy(name_parts->if_local, if_local); + name_parts->addr_peer = tipc_addr(z_peer, c_peer, n_peer); + strcpy(name_parts->if_peer, if_peer); + } + return 1; +} + +/** + * link_timeout - handle expiration of link timer + * @l_ptr: pointer to link + * + * This routine must not grab "net_lock" to avoid a potential deadlock conflict + * with link_delete(). (There is no risk that the node will be deleted by + * another thread because link_delete() always cancels the link timer before + * node_delete() is called.) + */ + +static void link_timeout(struct link *l_ptr) +{ + node_lock(l_ptr->owner); + + /* update counters used in statistical profiling of send traffic */ + + l_ptr->stats.accu_queue_sz += l_ptr->out_queue_size; + l_ptr->stats.queue_sz_counts++; + + if (l_ptr->out_queue_size > l_ptr->stats.max_queue_sz) + l_ptr->stats.max_queue_sz = l_ptr->out_queue_size; + + if (l_ptr->first_out) { + struct tipc_msg *msg = buf_msg(l_ptr->first_out); + u32 length = msg_size(msg); + + if ((msg_user(msg) == MSG_FRAGMENTER) + && (msg_type(msg) == FIRST_FRAGMENT)) { + length = msg_size(msg_get_wrapped(msg)); + } + if (length) { + l_ptr->stats.msg_lengths_total += length; + l_ptr->stats.msg_length_counts++; + if (length <= 64) + l_ptr->stats.msg_length_profile[0]++; + else if (length <= 256) + l_ptr->stats.msg_length_profile[1]++; + else if (length <= 1024) + l_ptr->stats.msg_length_profile[2]++; + else if (length <= 4096) + l_ptr->stats.msg_length_profile[3]++; + else if (length <= 16384) + l_ptr->stats.msg_length_profile[4]++; + else if (length <= 32768) + l_ptr->stats.msg_length_profile[5]++; + else + l_ptr->stats.msg_length_profile[6]++; + } + } + + /* do all other link processing performed on a periodic basis */ + + link_check_defragm_bufs(l_ptr); + + link_state_event(l_ptr, TIMEOUT_EVT); + + if (l_ptr->next_out) + link_push_queue(l_ptr); + + node_unlock(l_ptr->owner); +} + +static inline void link_set_timer(struct link *l_ptr, u32 time) +{ + k_start_timer(&l_ptr->timer, time); +} + +/** + * link_create - create a new link + * @b_ptr: pointer to associated bearer + * @peer: network address of node at other end of link + * @media_addr: media address to use when sending messages over link + * + * Returns pointer to link. + */ + +struct link *link_create(struct bearer *b_ptr, const u32 peer, + const struct tipc_media_addr *media_addr) +{ + struct link *l_ptr; + struct tipc_msg *msg; + char *if_name; + + l_ptr = (struct link *)kmalloc(sizeof(*l_ptr), GFP_ATOMIC); + if (!l_ptr) { + warn("Memory squeeze; Failed to create link\n"); + return NULL; + } + memset(l_ptr, 0, sizeof(*l_ptr)); + + l_ptr->addr = peer; + if_name = strchr(b_ptr->publ.name, ':') + 1; + sprintf(l_ptr->name, "%u.%u.%u:%s-%u.%u.%u:", + tipc_zone(tipc_own_addr), tipc_cluster(tipc_own_addr), + tipc_node(tipc_own_addr), + if_name, + tipc_zone(peer), tipc_cluster(peer), tipc_node(peer)); + /* note: peer i/f is appended to link name by reset/activate */ + memcpy(&l_ptr->media_addr, media_addr, sizeof(*media_addr)); + k_init_timer(&l_ptr->timer, (Handler)link_timeout, (unsigned long)l_ptr); + list_add_tail(&l_ptr->link_list, &b_ptr->links); + l_ptr->checkpoint = 1; + l_ptr->b_ptr = b_ptr; + link_set_supervision_props(l_ptr, b_ptr->media->tolerance); + l_ptr->state = RESET_UNKNOWN; + + l_ptr->pmsg = (struct tipc_msg *)&l_ptr->proto_msg; + msg = l_ptr->pmsg; + msg_init(msg, LINK_PROTOCOL, RESET_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + msg_set_size(msg, sizeof(l_ptr->proto_msg)); + msg_set_session(msg, tipc_random); + msg_set_bearer_id(msg, b_ptr->identity); + strcpy((char *)msg_data(msg), if_name); + + l_ptr->priority = b_ptr->priority; + link_set_queue_limits(l_ptr, b_ptr->media->window); + + link_init_max_pkt(l_ptr); + + l_ptr->next_out_no = 1; + INIT_LIST_HEAD(&l_ptr->waiting_ports); + + link_reset_statistics(l_ptr); + + l_ptr->owner = node_attach_link(l_ptr); + if (!l_ptr->owner) { + kfree(l_ptr); + return NULL; + } + + if (LINK_LOG_BUF_SIZE) { + char *pb = kmalloc(LINK_LOG_BUF_SIZE, GFP_ATOMIC); + + if (!pb) { + kfree(l_ptr); + warn("Memory squeeze; Failed to create link\n"); + return NULL; + } + printbuf_init(&l_ptr->print_buf, pb, LINK_LOG_BUF_SIZE); + } + + k_signal((Handler)link_start, (unsigned long)l_ptr); + + dbg("link_create(): tolerance = %u,cont intv = %u, abort_limit = %u\n", + l_ptr->tolerance, l_ptr->continuity_interval, l_ptr->abort_limit); + + return l_ptr; +} + +/** + * link_delete - delete a link + * @l_ptr: pointer to link + * + * Note: 'net_lock' is write_locked, bearer is locked. + * This routine must not grab the node lock until after link timer cancellation + * to avoid a potential deadlock situation. + */ + +void link_delete(struct link *l_ptr) +{ + if (!l_ptr) { + err("Attempt to delete non-existent link\n"); + return; + } + + dbg("link_delete()\n"); + + k_cancel_timer(&l_ptr->timer); + + node_lock(l_ptr->owner); + link_reset(l_ptr); + node_detach_link(l_ptr->owner, l_ptr); + link_stop(l_ptr); + list_del_init(&l_ptr->link_list); + if (LINK_LOG_BUF_SIZE) + kfree(l_ptr->print_buf.buf); + node_unlock(l_ptr->owner); + k_term_timer(&l_ptr->timer); + kfree(l_ptr); +} + +void link_start(struct link *l_ptr) +{ + dbg("link_start %x\n", l_ptr); + link_state_event(l_ptr, STARTING_EVT); +} + +/** + * link_schedule_port - schedule port for deferred sending + * @l_ptr: pointer to link + * @origport: reference to sending port + * @sz: amount of data to be sent + * + * Schedules port for renewed sending of messages after link congestion + * has abated. + */ + +static int link_schedule_port(struct link *l_ptr, u32 origport, u32 sz) +{ + struct port *p_ptr; + + spin_lock_bh(&port_list_lock); + p_ptr = port_lock(origport); + if (p_ptr) { + if (!p_ptr->wakeup) + goto exit; + if (!list_empty(&p_ptr->wait_list)) + goto exit; + p_ptr->congested_link = l_ptr; + p_ptr->publ.congested = 1; + p_ptr->waiting_pkts = 1 + ((sz - 1) / link_max_pkt(l_ptr)); + list_add_tail(&p_ptr->wait_list, &l_ptr->waiting_ports); + l_ptr->stats.link_congs++; +exit: + port_unlock(p_ptr); + } + spin_unlock_bh(&port_list_lock); + return -ELINKCONG; +} + +void link_wakeup_ports(struct link *l_ptr, int all) +{ + struct port *p_ptr; + struct port *temp_p_ptr; + int win = l_ptr->queue_limit[0] - l_ptr->out_queue_size; + + if (all) + win = 100000; + if (win <= 0) + return; + if (!spin_trylock_bh(&port_list_lock)) + return; + if (link_congested(l_ptr)) + goto exit; + list_for_each_entry_safe(p_ptr, temp_p_ptr, &l_ptr->waiting_ports, + wait_list) { + if (win <= 0) + break; + list_del_init(&p_ptr->wait_list); + p_ptr->congested_link = 0; + assert(p_ptr->wakeup); + spin_lock_bh(p_ptr->publ.lock); + p_ptr->publ.congested = 0; + p_ptr->wakeup(&p_ptr->publ); + win -= p_ptr->waiting_pkts; + spin_unlock_bh(p_ptr->publ.lock); + } + +exit: + spin_unlock_bh(&port_list_lock); +} + +/** + * link_release_outqueue - purge link's outbound message queue + * @l_ptr: pointer to link + */ + +static void link_release_outqueue(struct link *l_ptr) +{ + struct sk_buff *buf = l_ptr->first_out; + struct sk_buff *next; + + while (buf) { + next = buf->next; + buf_discard(buf); + buf = next; + } + l_ptr->first_out = NULL; + l_ptr->out_queue_size = 0; +} + +/** + * link_reset_fragments - purge link's inbound message fragments queue + * @l_ptr: pointer to link + */ + +void link_reset_fragments(struct link *l_ptr) +{ + struct sk_buff *buf = l_ptr->defragm_buf; + struct sk_buff *next; + + while (buf) { + next = buf->next; + buf_discard(buf); + buf = next; + } + l_ptr->defragm_buf = NULL; +} + +/** + * link_stop - purge all inbound and outbound messages associated with link + * @l_ptr: pointer to link + */ + +void link_stop(struct link *l_ptr) +{ + struct sk_buff *buf; + struct sk_buff *next; + + buf = l_ptr->oldest_deferred_in; + while (buf) { + next = buf->next; + buf_discard(buf); + buf = next; + } + + buf = l_ptr->first_out; + while (buf) { + next = buf->next; + buf_discard(buf); + buf = next; + } + + link_reset_fragments(l_ptr); + + buf_discard(l_ptr->proto_msg_queue); + l_ptr->proto_msg_queue = NULL; +} + +#if 0 + +/* LINK EVENT CODE IS NOT SUPPORTED AT PRESENT */ + +static void link_recv_event(struct link_event *ev) +{ + ev->fcn(ev->addr, ev->name, ev->up); + kfree(ev); +} + +static void link_send_event(void (*fcn)(u32 a, char *n, int up), + struct link *l_ptr, int up) +{ + struct link_event *ev; + + ev = kmalloc(sizeof(*ev), GFP_ATOMIC); + if (!ev) { + warn("Link event allocation failure\n"); + return; + } + ev->addr = l_ptr->addr; + ev->up = up; + ev->fcn = fcn; + memcpy(ev->name, l_ptr->name, TIPC_MAX_LINK_NAME); + k_signal((Handler)link_recv_event, (unsigned long)ev); +} + +#else + +#define link_send_event(fcn, l_ptr, up) do { } while (0) + +#endif + +void link_reset(struct link *l_ptr) +{ + struct sk_buff *buf; + u32 prev_state = l_ptr->state; + u32 checkpoint = l_ptr->next_in_no; + + msg_set_session(l_ptr->pmsg, msg_session(l_ptr->pmsg) + 1); + + /* Link is down, accept any session: */ + l_ptr->peer_session = 0; + + /* Prepare for max packet size negotiation */ + link_init_max_pkt(l_ptr); + + l_ptr->state = RESET_UNKNOWN; + dbg_link_state("Resetting Link\n"); + + if ((prev_state == RESET_UNKNOWN) || (prev_state == RESET_RESET)) + return; + + node_link_down(l_ptr->owner, l_ptr); + bearer_remove_dest(l_ptr->b_ptr, l_ptr->addr); +#if 0 + tipc_printf(CONS, "\nReset link <%s>\n", l_ptr->name); + dbg_link_dump(); +#endif + if (node_has_active_links(l_ptr->owner) && + l_ptr->owner->permit_changeover) { + l_ptr->reset_checkpoint = checkpoint; + l_ptr->exp_msg_count = START_CHANGEOVER; + } + + /* Clean up all queues: */ + + link_release_outqueue(l_ptr); + buf_discard(l_ptr->proto_msg_queue); + l_ptr->proto_msg_queue = NULL; + buf = l_ptr->oldest_deferred_in; + while (buf) { + struct sk_buff *next = buf->next; + buf_discard(buf); + buf = next; + } + if (!list_empty(&l_ptr->waiting_ports)) + link_wakeup_ports(l_ptr, 1); + + l_ptr->retransm_queue_head = 0; + l_ptr->retransm_queue_size = 0; + l_ptr->last_out = NULL; + l_ptr->first_out = NULL; + l_ptr->next_out = NULL; + l_ptr->unacked_window = 0; + l_ptr->checkpoint = 1; + l_ptr->next_out_no = 1; + l_ptr->deferred_inqueue_sz = 0; + l_ptr->oldest_deferred_in = NULL; + l_ptr->newest_deferred_in = NULL; + l_ptr->fsm_msg_cnt = 0; + l_ptr->stale_count = 0; + link_reset_statistics(l_ptr); + + link_send_event(cfg_link_event, l_ptr, 0); + if (!in_own_cluster(l_ptr->addr)) + link_send_event(disc_link_event, l_ptr, 0); +} + + +static void link_activate(struct link *l_ptr) +{ + l_ptr->next_in_no = 1; + node_link_up(l_ptr->owner, l_ptr); + bearer_add_dest(l_ptr->b_ptr, l_ptr->addr); + link_send_event(cfg_link_event, l_ptr, 1); + if (!in_own_cluster(l_ptr->addr)) + link_send_event(disc_link_event, l_ptr, 1); +} + +/** + * link_state_event - link finite state machine + * @l_ptr: pointer to link + * @event: state machine event to process + */ + +static void link_state_event(struct link *l_ptr, unsigned event) +{ + struct link *other; + u32 cont_intv = l_ptr->continuity_interval; + + if (!l_ptr->started && (event != STARTING_EVT)) + return; /* Not yet. */ + + if (link_blocked(l_ptr)) { + if (event == TIMEOUT_EVT) { + link_set_timer(l_ptr, cont_intv); + } + return; /* Changeover going on */ + } + dbg_link("STATE_EV: <%s> ", l_ptr->name); + + switch (l_ptr->state) { + case WORKING_WORKING: + dbg_link("WW/"); + switch (event) { + case TRAFFIC_MSG_EVT: + dbg_link("TRF-"); + /* fall through */ + case ACTIVATE_MSG: + dbg_link("ACT\n"); + break; + case TIMEOUT_EVT: + dbg_link("TIM "); + if (l_ptr->next_in_no != l_ptr->checkpoint) { + l_ptr->checkpoint = l_ptr->next_in_no; + if (bclink_acks_missing(l_ptr->owner)) { + link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } else if (l_ptr->max_pkt < l_ptr->max_pkt_target) { + link_send_proto_msg(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } + link_set_timer(l_ptr, cont_intv); + break; + } + dbg_link(" -> WU\n"); + l_ptr->state = WORKING_UNKNOWN; + l_ptr->fsm_msg_cnt = 0; + link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv / 4); + break; + case RESET_MSG: + dbg_link("RES -> RR\n"); + link_reset(l_ptr); + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + default: + err("Unknown link event %u in WW state\n", event); + } + break; + case WORKING_UNKNOWN: + dbg_link("WU/"); + switch (event) { + case TRAFFIC_MSG_EVT: + dbg_link("TRF-"); + case ACTIVATE_MSG: + dbg_link("ACT -> WW\n"); + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + dbg_link("RES -> RR\n"); + link_reset(l_ptr); + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case TIMEOUT_EVT: + dbg_link("TIM "); + if (l_ptr->next_in_no != l_ptr->checkpoint) { + dbg_link("-> WW \n"); + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + l_ptr->checkpoint = l_ptr->next_in_no; + if (bclink_acks_missing(l_ptr->owner)) { + link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + } + link_set_timer(l_ptr, cont_intv); + } else if (l_ptr->fsm_msg_cnt < l_ptr->abort_limit) { + dbg_link("Probing %u/%u,timer = %u ms)\n", + l_ptr->fsm_msg_cnt, l_ptr->abort_limit, + cont_intv / 4); + link_send_proto_msg(l_ptr, STATE_MSG, + 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv / 4); + } else { /* Link has failed */ + dbg_link("-> RU (%u probes unanswered)\n", + l_ptr->fsm_msg_cnt); + link_reset(l_ptr); + l_ptr->state = RESET_UNKNOWN; + l_ptr->fsm_msg_cnt = 0; + link_send_proto_msg(l_ptr, RESET_MSG, + 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + } + break; + default: + err("Unknown link event %u in WU state\n", event); + } + break; + case RESET_UNKNOWN: + dbg_link("RU/"); + switch (event) { + case TRAFFIC_MSG_EVT: + dbg_link("TRF-\n"); + break; + case ACTIVATE_MSG: + other = l_ptr->owner->active_links[0]; + if (other && link_working_unknown(other)) { + dbg_link("ACT\n"); + break; + } + dbg_link("ACT -> WW\n"); + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_activate(l_ptr); + link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + dbg_link("RES \n"); + dbg_link(" -> RR\n"); + l_ptr->state = RESET_RESET; + l_ptr->fsm_msg_cnt = 0; + link_send_proto_msg(l_ptr, ACTIVATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case STARTING_EVT: + dbg_link("START-"); + l_ptr->started = 1; + /* fall through */ + case TIMEOUT_EVT: + dbg_link("TIM \n"); + link_send_proto_msg(l_ptr, RESET_MSG, 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + default: + err("Unknown link event %u in RU state\n", event); + } + break; + case RESET_RESET: + dbg_link("RR/ "); + switch (event) { + case TRAFFIC_MSG_EVT: + dbg_link("TRF-"); + /* fall through */ + case ACTIVATE_MSG: + other = l_ptr->owner->active_links[0]; + if (other && link_working_unknown(other)) { + dbg_link("ACT\n"); + break; + } + dbg_link("ACT -> WW\n"); + l_ptr->state = WORKING_WORKING; + l_ptr->fsm_msg_cnt = 0; + link_activate(l_ptr); + link_send_proto_msg(l_ptr, STATE_MSG, 1, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + break; + case RESET_MSG: + dbg_link("RES\n"); + break; + case TIMEOUT_EVT: + dbg_link("TIM\n"); + link_send_proto_msg(l_ptr, ACTIVATE_MSG, 0, 0, 0, 0, 0); + l_ptr->fsm_msg_cnt++; + link_set_timer(l_ptr, cont_intv); + dbg_link("fsm_msg_cnt %u\n", l_ptr->fsm_msg_cnt); + break; + default: + err("Unknown link event %u in RR state\n", event); + } + break; + default: + err("Unknown link state %u/%u\n", l_ptr->state, event); + } +} + +/* + * link_bundle_buf(): Append contents of a buffer to + * the tail of an existing one. + */ + +static int link_bundle_buf(struct link *l_ptr, + struct sk_buff *bundler, + struct sk_buff *buf) +{ + struct tipc_msg *bundler_msg = buf_msg(bundler); + struct tipc_msg *msg = buf_msg(buf); + u32 size = msg_size(msg); + u32 to_pos = align(msg_size(bundler_msg)); + u32 rest = link_max_pkt(l_ptr) - to_pos; + + if (msg_user(bundler_msg) != MSG_BUNDLER) + return 0; + if (msg_type(bundler_msg) != OPEN_MSG) + return 0; + if (rest < align(size)) + return 0; + + skb_put(bundler, (to_pos - msg_size(bundler_msg)) + size); + memcpy(bundler->data + to_pos, buf->data, size); + msg_set_size(bundler_msg, to_pos + size); + msg_set_msgcnt(bundler_msg, msg_msgcnt(bundler_msg) + 1); + dbg("Packed msg # %u(%u octets) into pos %u in buf(#%u)\n", + msg_msgcnt(bundler_msg), size, to_pos, msg_seqno(bundler_msg)); + msg_dbg(msg, "PACKD:"); + buf_discard(buf); + l_ptr->stats.sent_bundled++; + return 1; +} + +static inline void link_add_to_outqueue(struct link *l_ptr, + struct sk_buff *buf, + struct tipc_msg *msg) +{ + u32 ack = mod(l_ptr->next_in_no - 1); + u32 seqno = mod(l_ptr->next_out_no++); + + msg_set_word(msg, 2, ((ack << 16) | seqno)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + buf->next = NULL; + if (l_ptr->first_out) { + l_ptr->last_out->next = buf; + l_ptr->last_out = buf; + } else + l_ptr->first_out = l_ptr->last_out = buf; + l_ptr->out_queue_size++; +} + +/* + * link_send_buf() is the 'full path' for messages, called from + * inside TIPC when the 'fast path' in tipc_send_buf + * has failed, and from link_send() + */ + +int link_send_buf(struct link *l_ptr, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + u32 size = msg_size(msg); + u32 dsz = msg_data_sz(msg); + u32 queue_size = l_ptr->out_queue_size; + u32 imp = msg_tot_importance(msg); + u32 queue_limit = l_ptr->queue_limit[imp]; + u32 max_packet = link_max_pkt(l_ptr); + + msg_set_prevnode(msg, tipc_own_addr); /* If routed message */ + + /* Match msg importance against queue limits: */ + + if (unlikely(queue_size >= queue_limit)) { + if (imp <= TIPC_CRITICAL_IMPORTANCE) { + return link_schedule_port(l_ptr, msg_origport(msg), + size); + } + msg_dbg(msg, "TIPC: Congestion, throwing away\n"); + buf_discard(buf); + if (imp > CONN_MANAGER) { + warn("Resetting <%s>, send queue full", l_ptr->name); + link_reset(l_ptr); + } + return dsz; + } + + /* Fragmentation needed ? */ + + if (size > max_packet) + return link_send_long_buf(l_ptr, buf); + + /* Packet can be queued or sent: */ + + if (queue_size > l_ptr->stats.max_queue_sz) + l_ptr->stats.max_queue_sz = queue_size; + + if (likely(!bearer_congested(l_ptr->b_ptr, l_ptr) && + !link_congested(l_ptr))) { + link_add_to_outqueue(l_ptr, buf, msg); + + if (likely(bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr))) { + l_ptr->unacked_window = 0; + } else { + bearer_schedule(l_ptr->b_ptr, l_ptr); + l_ptr->stats.bearer_congs++; + l_ptr->next_out = buf; + } + return dsz; + } + /* Congestion: can message be bundled ?: */ + + if ((msg_user(msg) != CHANGEOVER_PROTOCOL) && + (msg_user(msg) != MSG_FRAGMENTER)) { + + /* Try adding message to an existing bundle */ + + if (l_ptr->next_out && + link_bundle_buf(l_ptr, l_ptr->last_out, buf)) { + bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); + return dsz; + } + + /* Try creating a new bundle */ + + if (size <= max_packet * 2 / 3) { + struct sk_buff *bundler = buf_acquire(max_packet); + struct tipc_msg bundler_hdr; + + if (bundler) { + msg_init(&bundler_hdr, MSG_BUNDLER, OPEN_MSG, + TIPC_OK, INT_H_SIZE, l_ptr->addr); + memcpy(bundler->data, (unchar *)&bundler_hdr, + INT_H_SIZE); + skb_trim(bundler, INT_H_SIZE); + link_bundle_buf(l_ptr, bundler, buf); + buf = bundler; + msg = buf_msg(buf); + l_ptr->stats.sent_bundles++; + } + } + } + if (!l_ptr->next_out) + l_ptr->next_out = buf; + link_add_to_outqueue(l_ptr, buf, msg); + bearer_resolve_congestion(l_ptr->b_ptr, l_ptr); + return dsz; +} + +/* + * link_send(): same as link_send_buf(), but the link to use has + * not been selected yet, and the the owner node is not locked + * Called by TIPC internal users, e.g. the name distributor + */ + +int link_send(struct sk_buff *buf, u32 dest, u32 selector) +{ + struct link *l_ptr; + struct node *n_ptr; + int res = -ELINKCONG; + + read_lock_bh(&net_lock); + n_ptr = node_select(dest, selector); + if (n_ptr) { + node_lock(n_ptr); + l_ptr = n_ptr->active_links[selector & 1]; + dbg("link_send: found link %x for dest %x\n", l_ptr, dest); + if (l_ptr) { + res = link_send_buf(l_ptr, buf); + } + node_unlock(n_ptr); + } else { + dbg("Attempt to send msg to unknown node:\n"); + msg_dbg(buf_msg(buf),">>>"); + buf_discard(buf); + } + read_unlock_bh(&net_lock); + return res; +} + +/* + * link_send_buf_fast: Entry for data messages where the + * destination link is known and the header is complete, + * inclusive total message length. Very time critical. + * Link is locked. Returns user data length. + */ + +static inline int link_send_buf_fast(struct link *l_ptr, struct sk_buff *buf, + u32 *used_max_pkt) +{ + struct tipc_msg *msg = buf_msg(buf); + int res = msg_data_sz(msg); + + if (likely(!link_congested(l_ptr))) { + if (likely(msg_size(msg) <= link_max_pkt(l_ptr))) { + if (likely(list_empty(&l_ptr->b_ptr->cong_links))) { + link_add_to_outqueue(l_ptr, buf, msg); + if (likely(bearer_send(l_ptr->b_ptr, buf, + &l_ptr->media_addr))) { + l_ptr->unacked_window = 0; + msg_dbg(msg,"SENT_FAST:"); + return res; + } + dbg("failed sent fast...\n"); + bearer_schedule(l_ptr->b_ptr, l_ptr); + l_ptr->stats.bearer_congs++; + l_ptr->next_out = buf; + return res; + } + } + else + *used_max_pkt = link_max_pkt(l_ptr); + } + return link_send_buf(l_ptr, buf); /* All other cases */ +} + +/* + * tipc_send_buf_fast: Entry for data messages where the + * destination node is known and the header is complete, + * inclusive total message length. + * Returns user data length. + */ +int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode) +{ + struct link *l_ptr; + struct node *n_ptr; + int res; + u32 selector = msg_origport(buf_msg(buf)) & 1; + u32 dummy; + + if (destnode == tipc_own_addr) + return port_recv_msg(buf); + + read_lock_bh(&net_lock); + n_ptr = node_select(destnode, selector); + if (likely(n_ptr)) { + node_lock(n_ptr); + l_ptr = n_ptr->active_links[selector]; + dbg("send_fast: buf %x selected %x, destnode = %x\n", + buf, l_ptr, destnode); + if (likely(l_ptr)) { + res = link_send_buf_fast(l_ptr, buf, &dummy); + node_unlock(n_ptr); + read_unlock_bh(&net_lock); + return res; + } + node_unlock(n_ptr); + } + read_unlock_bh(&net_lock); + res = msg_data_sz(buf_msg(buf)); + tipc_reject_msg(buf, TIPC_ERR_NO_NODE); + return res; +} + + +/* + * link_send_sections_fast: Entry for messages where the + * destination processor is known and the header is complete, + * except for total message length. + * Returns user data length or errno. + */ +int link_send_sections_fast(struct port *sender, + struct iovec const *msg_sect, + const u32 num_sect, + u32 destaddr) +{ + struct tipc_msg *hdr = &sender->publ.phdr; + struct link *l_ptr; + struct sk_buff *buf; + struct node *node; + int res; + u32 selector = msg_origport(hdr) & 1; + + assert(destaddr != tipc_own_addr); + +again: + /* + * Try building message using port's max_pkt hint. + * (Must not hold any locks while building message.) + */ + + res = msg_build(hdr, msg_sect, num_sect, sender->max_pkt, + !sender->user_port, &buf); + + read_lock_bh(&net_lock); + node = node_select(destaddr, selector); + if (likely(node)) { + node_lock(node); + l_ptr = node->active_links[selector]; + if (likely(l_ptr)) { + if (likely(buf)) { + res = link_send_buf_fast(l_ptr, buf, + &sender->max_pkt); + if (unlikely(res < 0)) + buf_discard(buf); +exit: + node_unlock(node); + read_unlock_bh(&net_lock); + return res; + } + + /* Exit if build request was invalid */ + + if (unlikely(res < 0)) + goto exit; + + /* Exit if link (or bearer) is congested */ + + if (link_congested(l_ptr) || + !list_empty(&l_ptr->b_ptr->cong_links)) { + res = link_schedule_port(l_ptr, + sender->publ.ref, res); + goto exit; + } + + /* + * Message size exceeds max_pkt hint; update hint, + * then re-try fast path or fragment the message + */ + + sender->max_pkt = link_max_pkt(l_ptr); + node_unlock(node); + read_unlock_bh(&net_lock); + + + if ((msg_hdr_sz(hdr) + res) <= sender->max_pkt) + goto again; + + return link_send_sections_long(sender, msg_sect, + num_sect, destaddr); + } + node_unlock(node); + } + read_unlock_bh(&net_lock); + + /* Couldn't find a link to the destination node */ + + if (buf) + return tipc_reject_msg(buf, TIPC_ERR_NO_NODE); + if (res >= 0) + return port_reject_sections(sender, hdr, msg_sect, num_sect, + TIPC_ERR_NO_NODE); + return res; +} + +/* + * link_send_sections_long(): Entry for long messages where the + * destination node is known and the header is complete, + * inclusive total message length. + * Link and bearer congestion status have been checked to be ok, + * and are ignored if they change. + * + * Note that fragments do not use the full link MTU so that they won't have + * to undergo refragmentation if link changeover causes them to be sent + * over another link with an additional tunnel header added as prefix. + * (Refragmentation will still occur if the other link has a smaller MTU.) + * + * Returns user data length or errno. + */ +static int link_send_sections_long(struct port *sender, + struct iovec const *msg_sect, + u32 num_sect, + u32 destaddr) +{ + struct link *l_ptr; + struct node *node; + struct tipc_msg *hdr = &sender->publ.phdr; + u32 dsz = msg_data_sz(hdr); + u32 max_pkt,fragm_sz,rest; + struct tipc_msg fragm_hdr; + struct sk_buff *buf,*buf_chain,*prev; + u32 fragm_crs,fragm_rest,hsz,sect_rest; + const unchar *sect_crs; + int curr_sect; + u32 fragm_no; + +again: + fragm_no = 1; + max_pkt = sender->max_pkt - INT_H_SIZE; + /* leave room for tunnel header in case of link changeover */ + fragm_sz = max_pkt - INT_H_SIZE; + /* leave room for fragmentation header in each fragment */ + rest = dsz; + fragm_crs = 0; + fragm_rest = 0; + sect_rest = 0; + sect_crs = 0; + curr_sect = -1; + + /* Prepare reusable fragment header: */ + + msg_dbg(hdr, ">FRAGMENTING>"); + msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, + TIPC_OK, INT_H_SIZE, msg_destnode(hdr)); + msg_set_link_selector(&fragm_hdr, sender->publ.ref); + msg_set_size(&fragm_hdr, max_pkt); + msg_set_fragm_no(&fragm_hdr, 1); + + /* Prepare header of first fragment: */ + + buf_chain = buf = buf_acquire(max_pkt); + if (!buf) + return -ENOMEM; + buf->next = NULL; + memcpy(buf->data, (unchar *)&fragm_hdr, INT_H_SIZE); + hsz = msg_hdr_sz(hdr); + memcpy(buf->data + INT_H_SIZE, (unchar *)hdr, hsz); + msg_dbg(buf_msg(buf), ">BUILD>"); + + /* Chop up message: */ + + fragm_crs = INT_H_SIZE + hsz; + fragm_rest = fragm_sz - hsz; + + do { /* For all sections */ + u32 sz; + + if (!sect_rest) { + sect_rest = msg_sect[++curr_sect].iov_len; + sect_crs = (const unchar *)msg_sect[curr_sect].iov_base; + } + + if (sect_rest < fragm_rest) + sz = sect_rest; + else + sz = fragm_rest; + + if (likely(!sender->user_port)) { + if (copy_from_user(buf->data + fragm_crs, sect_crs, sz)) { +error: + for (; buf_chain; buf_chain = buf) { + buf = buf_chain->next; + buf_discard(buf_chain); + } + return -EFAULT; + } + } else + memcpy(buf->data + fragm_crs, sect_crs, sz); + + sect_crs += sz; + sect_rest -= sz; + fragm_crs += sz; + fragm_rest -= sz; + rest -= sz; + + if (!fragm_rest && rest) { + + /* Initiate new fragment: */ + if (rest <= fragm_sz) { + fragm_sz = rest; + msg_set_type(&fragm_hdr,LAST_FRAGMENT); + } else { + msg_set_type(&fragm_hdr, FRAGMENT); + } + msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); + msg_set_fragm_no(&fragm_hdr, ++fragm_no); + prev = buf; + buf = buf_acquire(fragm_sz + INT_H_SIZE); + if (!buf) + goto error; + + buf->next = NULL; + prev->next = buf; + memcpy(buf->data, (unchar *)&fragm_hdr, INT_H_SIZE); + fragm_crs = INT_H_SIZE; + fragm_rest = fragm_sz; + msg_dbg(buf_msg(buf)," >BUILD>"); + } + } + while (rest > 0); + + /* + * Now we have a buffer chain. Select a link and check + * that packet size is still OK + */ + node = node_select(destaddr, sender->publ.ref & 1); + if (likely(node)) { + node_lock(node); + l_ptr = node->active_links[sender->publ.ref & 1]; + if (!l_ptr) { + node_unlock(node); + goto reject; + } + if (link_max_pkt(l_ptr) < max_pkt) { + sender->max_pkt = link_max_pkt(l_ptr); + node_unlock(node); + for (; buf_chain; buf_chain = buf) { + buf = buf_chain->next; + buf_discard(buf_chain); + } + goto again; + } + } else { +reject: + for (; buf_chain; buf_chain = buf) { + buf = buf_chain->next; + buf_discard(buf_chain); + } + return port_reject_sections(sender, hdr, msg_sect, num_sect, + TIPC_ERR_NO_NODE); + } + + /* Append whole chain to send queue: */ + + buf = buf_chain; + l_ptr->long_msg_seq_no = mod(l_ptr->long_msg_seq_no + 1); + if (!l_ptr->next_out) + l_ptr->next_out = buf_chain; + l_ptr->stats.sent_fragmented++; + while (buf) { + struct sk_buff *next = buf->next; + struct tipc_msg *msg = buf_msg(buf); + + l_ptr->stats.sent_fragments++; + msg_set_long_msgno(msg, l_ptr->long_msg_seq_no); + link_add_to_outqueue(l_ptr, buf, msg); + msg_dbg(msg, ">ADD>"); + buf = next; + } + + /* Send it, if possible: */ + + link_push_queue(l_ptr); + node_unlock(node); + return dsz; +} + +/* + * link_push_packet: Push one unsent packet to the media + */ +u32 link_push_packet(struct link *l_ptr) +{ + struct sk_buff *buf = l_ptr->first_out; + u32 r_q_size = l_ptr->retransm_queue_size; + u32 r_q_head = l_ptr->retransm_queue_head; + + /* Step to position where retransmission failed, if any, */ + /* consider that buffers may have been released in meantime */ + + if (r_q_size && buf) { + u32 last = lesser(mod(r_q_head + r_q_size), + link_last_sent(l_ptr)); + u32 first = msg_seqno(buf_msg(buf)); + + while (buf && less(first, r_q_head)) { + first = mod(first + 1); + buf = buf->next; + } + l_ptr->retransm_queue_head = r_q_head = first; + l_ptr->retransm_queue_size = r_q_size = mod(last - first); + } + + /* Continue retransmission now, if there is anything: */ + + if (r_q_size && buf && !skb_cloned(buf)) { + msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(buf_msg(buf), l_ptr->owner->bclink.last_in); + if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + msg_dbg(buf_msg(buf), ">DEF-RETR>"); + l_ptr->retransm_queue_head = mod(++r_q_head); + l_ptr->retransm_queue_size = --r_q_size; + l_ptr->stats.retransmitted++; + return TIPC_OK; + } else { + l_ptr->stats.bearer_congs++; + msg_dbg(buf_msg(buf), "|>DEF-RETR>"); + return PUSH_FAILED; + } + } + + /* Send deferred protocol message, if any: */ + + buf = l_ptr->proto_msg_queue; + if (buf) { + msg_set_ack(buf_msg(buf), mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(buf_msg(buf),l_ptr->owner->bclink.last_in); + if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + msg_dbg(buf_msg(buf), ">DEF-PROT>"); + l_ptr->unacked_window = 0; + buf_discard(buf); + l_ptr->proto_msg_queue = 0; + return TIPC_OK; + } else { + msg_dbg(buf_msg(buf), "|>DEF-PROT>"); + l_ptr->stats.bearer_congs++; + return PUSH_FAILED; + } + } + + /* Send one deferred data message, if send window not full: */ + + buf = l_ptr->next_out; + if (buf) { + struct tipc_msg *msg = buf_msg(buf); + u32 next = msg_seqno(msg); + u32 first = msg_seqno(buf_msg(l_ptr->first_out)); + + if (mod(next - first) < l_ptr->queue_limit[0]) { + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + if (msg_user(msg) == MSG_BUNDLER) + msg_set_type(msg, CLOSED_MSG); + msg_dbg(msg, ">PUSH-DATA>"); + l_ptr->next_out = buf->next; + return TIPC_OK; + } else { + msg_dbg(msg, "|PUSH-DATA|"); + l_ptr->stats.bearer_congs++; + return PUSH_FAILED; + } + } + } + return PUSH_FINISHED; +} + +/* + * push_queue(): push out the unsent messages of a link where + * congestion has abated. Node is locked + */ +void link_push_queue(struct link *l_ptr) +{ + u32 res; + + if (bearer_congested(l_ptr->b_ptr, l_ptr)) + return; + + do { + res = link_push_packet(l_ptr); + } + while (res == TIPC_OK); + if (res == PUSH_FAILED) + bearer_schedule(l_ptr->b_ptr, l_ptr); +} + +void link_retransmit(struct link *l_ptr, struct sk_buff *buf, + u32 retransmits) +{ + struct tipc_msg *msg; + + dbg("Retransmitting %u in link %x\n", retransmits, l_ptr); + + if (bearer_congested(l_ptr->b_ptr, l_ptr) && buf && !skb_cloned(buf)) { + msg_dbg(buf_msg(buf), ">NO_RETR->BCONG>"); + dbg_print_link(l_ptr, " "); + l_ptr->retransm_queue_head = msg_seqno(buf_msg(buf)); + l_ptr->retransm_queue_size = retransmits; + return; + } + while (retransmits && (buf != l_ptr->next_out) && buf && !skb_cloned(buf)) { + msg = buf_msg(buf); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + /* Catch if retransmissions fail repeatedly: */ + if (l_ptr->last_retransmitted == msg_seqno(msg)) { + if (++l_ptr->stale_count > 100) { + msg_print(CONS, buf_msg(buf), ">RETR>"); + info("...Retransmitted %u times\n", + l_ptr->stale_count); + link_print(l_ptr, CONS, "Resetting Link\n");; + link_reset(l_ptr); + break; + } + } else { + l_ptr->stale_count = 0; + } + l_ptr->last_retransmitted = msg_seqno(msg); + + msg_dbg(buf_msg(buf), ">RETR>"); + buf = buf->next; + retransmits--; + l_ptr->stats.retransmitted++; + } else { + bearer_schedule(l_ptr->b_ptr, l_ptr); + l_ptr->stats.bearer_congs++; + l_ptr->retransm_queue_head = msg_seqno(buf_msg(buf)); + l_ptr->retransm_queue_size = retransmits; + return; + } + } + l_ptr->retransm_queue_head = l_ptr->retransm_queue_size = 0; +} + +/* + * link_recv_non_seq: Receive packets which are outside + * the link sequence flow + */ + +static void link_recv_non_seq(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + + if (msg_user(msg) == LINK_CONFIG) + disc_recv_msg(buf); + else + bclink_recv_pkt(buf); +} + +/** + * link_insert_deferred_queue - insert deferred messages back into receive chain + */ + +static struct sk_buff *link_insert_deferred_queue(struct link *l_ptr, + struct sk_buff *buf) +{ + u32 seq_no; + + if (l_ptr->oldest_deferred_in == NULL) + return buf; + + seq_no = msg_seqno(buf_msg(l_ptr->oldest_deferred_in)); + if (seq_no == mod(l_ptr->next_in_no)) { + l_ptr->newest_deferred_in->next = buf; + buf = l_ptr->oldest_deferred_in; + l_ptr->oldest_deferred_in = NULL; + l_ptr->deferred_inqueue_sz = 0; + } + return buf; +} + +void tipc_recv_msg(struct sk_buff *head, struct tipc_bearer *tb_ptr) +{ + read_lock_bh(&net_lock); + while (head) { + struct bearer *b_ptr; + struct node *n_ptr; + struct link *l_ptr; + struct sk_buff *crs; + struct sk_buff *buf = head; + struct tipc_msg *msg = buf_msg(buf); + u32 seq_no = msg_seqno(msg); + u32 ackd = msg_ack(msg); + u32 released = 0; + int type; + + b_ptr = (struct bearer *)tb_ptr; + TIPC_SKB_CB(buf)->handle = b_ptr; + + head = head->next; + if (unlikely(msg_version(msg) != TIPC_VERSION)) + goto cont; +#if 0 + if (msg_user(msg) != LINK_PROTOCOL) +#endif + msg_dbg(msg,"<REC<"); + + if (unlikely(msg_non_seq(msg))) { + link_recv_non_seq(buf); + continue; + } + n_ptr = node_find(msg_prevnode(msg)); + if (unlikely(!n_ptr)) + goto cont; + + node_lock(n_ptr); + l_ptr = n_ptr->links[b_ptr->identity]; + if (unlikely(!l_ptr)) { + node_unlock(n_ptr); + goto cont; + } + /* + * Release acked messages + */ + if (less(n_ptr->bclink.acked, msg_bcast_ack(msg))) { + if (node_is_up(n_ptr) && n_ptr->bclink.supported) + bclink_acknowledge(n_ptr, msg_bcast_ack(msg)); + } + + crs = l_ptr->first_out; + while ((crs != l_ptr->next_out) && + less_eq(msg_seqno(buf_msg(crs)), ackd)) { + struct sk_buff *next = crs->next; + + buf_discard(crs); + crs = next; + released++; + } + if (released) { + l_ptr->first_out = crs; + l_ptr->out_queue_size -= released; + } + if (unlikely(l_ptr->next_out)) + link_push_queue(l_ptr); + if (unlikely(!list_empty(&l_ptr->waiting_ports))) + link_wakeup_ports(l_ptr, 0); + if (unlikely(++l_ptr->unacked_window >= TIPC_MIN_LINK_WIN)) { + l_ptr->stats.sent_acks++; + link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + } + +protocol_check: + if (likely(link_working_working(l_ptr))) { + if (likely(seq_no == mod(l_ptr->next_in_no))) { + l_ptr->next_in_no++; + if (unlikely(l_ptr->oldest_deferred_in)) + head = link_insert_deferred_queue(l_ptr, + head); + if (likely(msg_is_dest(msg, tipc_own_addr))) { +deliver: + if (likely(msg_isdata(msg))) { + node_unlock(n_ptr); + port_recv_msg(buf); + continue; + } + switch (msg_user(msg)) { + case MSG_BUNDLER: + l_ptr->stats.recv_bundles++; + l_ptr->stats.recv_bundled += + msg_msgcnt(msg); + node_unlock(n_ptr); + link_recv_bundle(buf); + continue; + case ROUTE_DISTRIBUTOR: + node_unlock(n_ptr); + cluster_recv_routing_table(buf); + continue; + case NAME_DISTRIBUTOR: + node_unlock(n_ptr); + named_recv(buf); + continue; + case CONN_MANAGER: + node_unlock(n_ptr); + port_recv_proto_msg(buf); + continue; + case MSG_FRAGMENTER: + l_ptr->stats.recv_fragments++; + if (link_recv_fragment( + &l_ptr->defragm_buf, + &buf, &msg)) { + l_ptr->stats.recv_fragmented++; + goto deliver; + } + break; + case CHANGEOVER_PROTOCOL: + type = msg_type(msg); + if (link_recv_changeover_msg( + &l_ptr, &buf)) { + msg = buf_msg(buf); + seq_no = msg_seqno(msg); + TIPC_SKB_CB(buf)->handle + = b_ptr; + if (type == ORIGINAL_MSG) + goto deliver; + goto protocol_check; + } + break; + } + } + node_unlock(n_ptr); + net_route_msg(buf); + continue; + } + link_handle_out_of_seq_msg(l_ptr, buf); + head = link_insert_deferred_queue(l_ptr, head); + node_unlock(n_ptr); + continue; + } + + if (msg_user(msg) == LINK_PROTOCOL) { + link_recv_proto_msg(l_ptr, buf); + head = link_insert_deferred_queue(l_ptr, head); + node_unlock(n_ptr); + continue; + } + msg_dbg(msg,"NSEQ<REC<"); + link_state_event(l_ptr, TRAFFIC_MSG_EVT); + + if (link_working_working(l_ptr)) { + /* Re-insert in front of queue */ + msg_dbg(msg,"RECV-REINS:"); + buf->next = head; + head = buf; + node_unlock(n_ptr); + continue; + } + node_unlock(n_ptr); +cont: + buf_discard(buf); + } + read_unlock_bh(&net_lock); +} + +/* + * link_defer_buf(): Sort a received out-of-sequence packet + * into the deferred reception queue. + * Returns the increase of the queue length,i.e. 0 or 1 + */ + +u32 link_defer_pkt(struct sk_buff **head, + struct sk_buff **tail, + struct sk_buff *buf) +{ + struct sk_buff *prev = 0; + struct sk_buff *crs = *head; + u32 seq_no = msg_seqno(buf_msg(buf)); + + buf->next = NULL; + + /* Empty queue ? */ + if (*head == NULL) { + *head = *tail = buf; + return 1; + } + + /* Last ? */ + if (less(msg_seqno(buf_msg(*tail)), seq_no)) { + (*tail)->next = buf; + *tail = buf; + return 1; + } + + /* Scan through queue and sort it in */ + do { + struct tipc_msg *msg = buf_msg(crs); + + if (less(seq_no, msg_seqno(msg))) { + buf->next = crs; + if (prev) + prev->next = buf; + else + *head = buf; + return 1; + } + if (seq_no == msg_seqno(msg)) { + break; + } + prev = crs; + crs = crs->next; + } + while (crs); + + /* Message is a duplicate of an existing message */ + + buf_discard(buf); + return 0; +} + +/** + * link_handle_out_of_seq_msg - handle arrival of out-of-sequence packet + */ + +static void link_handle_out_of_seq_msg(struct link *l_ptr, + struct sk_buff *buf) +{ + u32 seq_no = msg_seqno(buf_msg(buf)); + + if (likely(msg_user(buf_msg(buf)) == LINK_PROTOCOL)) { + link_recv_proto_msg(l_ptr, buf); + return; + } + + dbg("rx OOS msg: seq_no %u, expecting %u (%u)\n", + seq_no, mod(l_ptr->next_in_no), l_ptr->next_in_no); + + /* Record OOS packet arrival (force mismatch on next timeout) */ + + l_ptr->checkpoint--; + + /* + * Discard packet if a duplicate; otherwise add it to deferred queue + * and notify peer of gap as per protocol specification + */ + + if (less(seq_no, mod(l_ptr->next_in_no))) { + l_ptr->stats.duplicates++; + buf_discard(buf); + return; + } + + if (link_defer_pkt(&l_ptr->oldest_deferred_in, + &l_ptr->newest_deferred_in, buf)) { + l_ptr->deferred_inqueue_sz++; + l_ptr->stats.deferred_recv++; + if ((l_ptr->deferred_inqueue_sz % 16) == 1) + link_send_proto_msg(l_ptr, STATE_MSG, 0, 0, 0, 0, 0); + } else + l_ptr->stats.duplicates++; +} + +/* + * Send protocol message to the other endpoint. + */ +void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int probe_msg, + u32 gap, u32 tolerance, u32 priority, u32 ack_mtu) +{ + struct sk_buff *buf = 0; + struct tipc_msg *msg = l_ptr->pmsg; + u32 msg_size = sizeof(l_ptr->proto_msg); + + if (link_blocked(l_ptr)) + return; + msg_set_type(msg, msg_typ); + msg_set_net_plane(msg, l_ptr->b_ptr->net_plane); + msg_set_bcast_ack(msg, mod(l_ptr->owner->bclink.last_in)); + msg_set_last_bcast(msg, bclink_get_last_sent()); + + if (msg_typ == STATE_MSG) { + u32 next_sent = mod(l_ptr->next_out_no); + + if (!link_is_up(l_ptr)) + return; + if (l_ptr->next_out) + next_sent = msg_seqno(buf_msg(l_ptr->next_out)); + msg_set_next_sent(msg, next_sent); + if (l_ptr->oldest_deferred_in) { + u32 rec = msg_seqno(buf_msg(l_ptr->oldest_deferred_in)); + gap = mod(rec - mod(l_ptr->next_in_no)); + } + msg_set_seq_gap(msg, gap); + if (gap) + l_ptr->stats.sent_nacks++; + msg_set_link_tolerance(msg, tolerance); + msg_set_linkprio(msg, priority); + msg_set_max_pkt(msg, ack_mtu); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); + msg_set_probe(msg, probe_msg != 0); + if (probe_msg) { + u32 mtu = l_ptr->max_pkt; + + if ((mtu < l_ptr->max_pkt_target) && + link_working_working(l_ptr) && + l_ptr->fsm_msg_cnt) { + msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; + if (l_ptr->max_pkt_probes == 10) { + l_ptr->max_pkt_target = (msg_size - 4); + l_ptr->max_pkt_probes = 0; + msg_size = (mtu + (l_ptr->max_pkt_target - mtu)/2 + 2) & ~3; + } + l_ptr->max_pkt_probes++; + } + + l_ptr->stats.sent_probes++; + } + l_ptr->stats.sent_states++; + } else { /* RESET_MSG or ACTIVATE_MSG */ + msg_set_ack(msg, mod(l_ptr->reset_checkpoint - 1)); + msg_set_seq_gap(msg, 0); + msg_set_next_sent(msg, 1); + msg_set_link_tolerance(msg, l_ptr->tolerance); + msg_set_linkprio(msg, l_ptr->priority); + msg_set_max_pkt(msg, l_ptr->max_pkt_target); + } + + if (node_has_redundant_links(l_ptr->owner)) { + msg_set_redundant_link(msg); + } else { + msg_clear_redundant_link(msg); + } + msg_set_linkprio(msg, l_ptr->priority); + + /* Ensure sequence number will not fit : */ + + msg_set_seqno(msg, mod(l_ptr->next_out_no + (0xffff/2))); + + /* Congestion? */ + + if (bearer_congested(l_ptr->b_ptr, l_ptr)) { + if (!l_ptr->proto_msg_queue) { + l_ptr->proto_msg_queue = + buf_acquire(sizeof(l_ptr->proto_msg)); + } + buf = l_ptr->proto_msg_queue; + if (!buf) + return; + memcpy(buf->data, (unchar *)msg, sizeof(l_ptr->proto_msg)); + return; + } + msg_set_timestamp(msg, jiffies_to_msecs(jiffies)); + + /* Message can be sent */ + + msg_dbg(msg, ">>"); + + buf = buf_acquire(msg_size); + if (!buf) + return; + + memcpy(buf->data, (unchar *)msg, sizeof(l_ptr->proto_msg)); + msg_set_size(buf_msg(buf), msg_size); + + if (bearer_send(l_ptr->b_ptr, buf, &l_ptr->media_addr)) { + l_ptr->unacked_window = 0; + buf_discard(buf); + return; + } + + /* New congestion */ + bearer_schedule(l_ptr->b_ptr, l_ptr); + l_ptr->proto_msg_queue = buf; + l_ptr->stats.bearer_congs++; +} + +/* + * Receive protocol message : + * Note that network plane id propagates through the network, and may + * change at any time. The node with lowest address rules + */ + +static void link_recv_proto_msg(struct link *l_ptr, struct sk_buff *buf) +{ + u32 rec_gap = 0; + u32 max_pkt_info; + u32 max_pkt_ack; + u32 msg_tol; + struct tipc_msg *msg = buf_msg(buf); + + dbg("AT(%u):", jiffies_to_msecs(jiffies)); + msg_dbg(msg, "<<"); + if (link_blocked(l_ptr)) + goto exit; + + /* record unnumbered packet arrival (force mismatch on next timeout) */ + + l_ptr->checkpoint--; + + if (l_ptr->b_ptr->net_plane != msg_net_plane(msg)) + if (tipc_own_addr > msg_prevnode(msg)) + l_ptr->b_ptr->net_plane = msg_net_plane(msg); + + l_ptr->owner->permit_changeover = msg_redundant_link(msg); + + switch (msg_type(msg)) { + + case RESET_MSG: + if (!link_working_unknown(l_ptr) && l_ptr->peer_session) { + if (msg_session(msg) == l_ptr->peer_session) { + dbg("Duplicate RESET: %u<->%u\n", + msg_session(msg), l_ptr->peer_session); + break; /* duplicate: ignore */ + } + } + /* fall thru' */ + case ACTIVATE_MSG: + /* Update link settings according other endpoint's values */ + + strcpy((strrchr(l_ptr->name, ':') + 1), (char *)msg_data(msg)); + + if ((msg_tol = msg_link_tolerance(msg)) && + (msg_tol > l_ptr->tolerance)) + link_set_supervision_props(l_ptr, msg_tol); + + if (msg_linkprio(msg) > l_ptr->priority) + l_ptr->priority = msg_linkprio(msg); + + max_pkt_info = msg_max_pkt(msg); + if (max_pkt_info) { + if (max_pkt_info < l_ptr->max_pkt_target) + l_ptr->max_pkt_target = max_pkt_info; + if (l_ptr->max_pkt > l_ptr->max_pkt_target) + l_ptr->max_pkt = l_ptr->max_pkt_target; + } else { + l_ptr->max_pkt = l_ptr->max_pkt_target; + } + l_ptr->owner->bclink.supported = (max_pkt_info != 0); + + link_state_event(l_ptr, msg_type(msg)); + + l_ptr->peer_session = msg_session(msg); + l_ptr->peer_bearer_id = msg_bearer_id(msg); + + /* Synchronize broadcast sequence numbers */ + if (!node_has_redundant_links(l_ptr->owner)) { + l_ptr->owner->bclink.last_in = mod(msg_last_bcast(msg)); + } + break; + case STATE_MSG: + + if ((msg_tol = msg_link_tolerance(msg))) + link_set_supervision_props(l_ptr, msg_tol); + + if (msg_linkprio(msg) && + (msg_linkprio(msg) != l_ptr->priority)) { + warn("Changing prio <%s>: %u->%u\n", + l_ptr->name, l_ptr->priority, msg_linkprio(msg)); + l_ptr->priority = msg_linkprio(msg); + link_reset(l_ptr); /* Enforce change to take effect */ + break; + } + link_state_event(l_ptr, TRAFFIC_MSG_EVT); + l_ptr->stats.recv_states++; + if (link_reset_unknown(l_ptr)) + break; + + if (less_eq(mod(l_ptr->next_in_no), msg_next_sent(msg))) { + rec_gap = mod(msg_next_sent(msg) - + mod(l_ptr->next_in_no)); + } + + max_pkt_ack = msg_max_pkt(msg); + if (max_pkt_ack > l_ptr->max_pkt) { + dbg("Link <%s> updated MTU %u -> %u\n", + l_ptr->name, l_ptr->max_pkt, max_pkt_ack); + l_ptr->max_pkt = max_pkt_ack; + l_ptr->max_pkt_probes = 0; + } + + max_pkt_ack = 0; + if (msg_probe(msg)) { + l_ptr->stats.recv_probes++; + if (msg_size(msg) > sizeof(l_ptr->proto_msg)) { + max_pkt_ack = msg_size(msg); + } + } + + /* Protocol message before retransmits, reduce loss risk */ + + bclink_check_gap(l_ptr->owner, msg_last_bcast(msg)); + + if (rec_gap || (msg_probe(msg))) { + link_send_proto_msg(l_ptr, STATE_MSG, + 0, rec_gap, 0, 0, max_pkt_ack); + } + if (msg_seq_gap(msg)) { + msg_dbg(msg, "With Gap:"); + l_ptr->stats.recv_nacks++; + link_retransmit(l_ptr, l_ptr->first_out, + msg_seq_gap(msg)); + } + break; + default: + msg_dbg(buf_msg(buf), "<DISCARDING UNKNOWN<"); + } +exit: + buf_discard(buf); +} + + +/* + * link_tunnel(): Send one message via a link belonging to + * another bearer. Owner node is locked. + */ +void link_tunnel(struct link *l_ptr, + struct tipc_msg *tunnel_hdr, + struct tipc_msg *msg, + u32 selector) +{ + struct link *tunnel; + struct sk_buff *buf; + u32 length = msg_size(msg); + + tunnel = l_ptr->owner->active_links[selector & 1]; + if (!link_is_up(tunnel)) + return; + msg_set_size(tunnel_hdr, length + INT_H_SIZE); + buf = buf_acquire(length + INT_H_SIZE); + if (!buf) + return; + memcpy(buf->data, (unchar *)tunnel_hdr, INT_H_SIZE); + memcpy(buf->data + INT_H_SIZE, (unchar *)msg, length); + dbg("%c->%c:", l_ptr->b_ptr->net_plane, tunnel->b_ptr->net_plane); + msg_dbg(buf_msg(buf), ">SEND>"); + assert(tunnel); + link_send_buf(tunnel, buf); +} + + + +/* + * changeover(): Send whole message queue via the remaining link + * Owner node is locked. + */ + +void link_changeover(struct link *l_ptr) +{ + u32 msgcount = l_ptr->out_queue_size; + struct sk_buff *crs = l_ptr->first_out; + struct link *tunnel = l_ptr->owner->active_links[0]; + int split_bundles = node_has_redundant_links(l_ptr->owner); + struct tipc_msg tunnel_hdr; + + if (!tunnel) + return; + + if (!l_ptr->owner->permit_changeover) + return; + + msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, + ORIGINAL_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); + msg_set_msgcnt(&tunnel_hdr, msgcount); + if (!l_ptr->first_out) { + struct sk_buff *buf; + + assert(!msgcount); + buf = buf_acquire(INT_H_SIZE); + if (buf) { + memcpy(buf->data, (unchar *)&tunnel_hdr, INT_H_SIZE); + msg_set_size(&tunnel_hdr, INT_H_SIZE); + dbg("%c->%c:", l_ptr->b_ptr->net_plane, + tunnel->b_ptr->net_plane); + msg_dbg(&tunnel_hdr, "EMPTY>SEND>"); + link_send_buf(tunnel, buf); + } else { + warn("Memory squeeze; link changeover failed\n"); + } + return; + } + while (crs) { + struct tipc_msg *msg = buf_msg(crs); + + if ((msg_user(msg) == MSG_BUNDLER) && split_bundles) { + u32 msgcount = msg_msgcnt(msg); + struct tipc_msg *m = msg_get_wrapped(msg); + unchar* pos = (unchar*)m; + + while (msgcount--) { + msg_set_seqno(m,msg_seqno(msg)); + link_tunnel(l_ptr, &tunnel_hdr, m, + msg_link_selector(m)); + pos += align(msg_size(m)); + m = (struct tipc_msg *)pos; + } + } else { + link_tunnel(l_ptr, &tunnel_hdr, msg, + msg_link_selector(msg)); + } + crs = crs->next; + } +} + +void link_send_duplicate(struct link *l_ptr, struct link *tunnel) +{ + struct sk_buff *iter; + struct tipc_msg tunnel_hdr; + + msg_init(&tunnel_hdr, CHANGEOVER_PROTOCOL, + DUPLICATE_MSG, TIPC_OK, INT_H_SIZE, l_ptr->addr); + msg_set_msgcnt(&tunnel_hdr, l_ptr->out_queue_size); + msg_set_bearer_id(&tunnel_hdr, l_ptr->peer_bearer_id); + iter = l_ptr->first_out; + while (iter) { + struct sk_buff *outbuf; + struct tipc_msg *msg = buf_msg(iter); + u32 length = msg_size(msg); + + if (msg_user(msg) == MSG_BUNDLER) + msg_set_type(msg, CLOSED_MSG); + msg_set_ack(msg, mod(l_ptr->next_in_no - 1)); /* Update */ + msg_set_bcast_ack(msg, l_ptr->owner->bclink.last_in); + msg_set_size(&tunnel_hdr, length + INT_H_SIZE); + outbuf = buf_acquire(length + INT_H_SIZE); + if (outbuf == NULL) { + warn("Memory squeeze; buffer duplication failed\n"); + return; + } + memcpy(outbuf->data, (unchar *)&tunnel_hdr, INT_H_SIZE); + memcpy(outbuf->data + INT_H_SIZE, iter->data, length); + dbg("%c->%c:", l_ptr->b_ptr->net_plane, + tunnel->b_ptr->net_plane); + msg_dbg(buf_msg(outbuf), ">SEND>"); + link_send_buf(tunnel, outbuf); + if (!link_is_up(l_ptr)) + return; + iter = iter->next; + } +} + + + +/** + * buf_extract - extracts embedded TIPC message from another message + * @skb: encapsulating message buffer + * @from_pos: offset to extract from + * + * Returns a new message buffer containing an embedded message. The + * encapsulating message itself is left unchanged. + */ + +static struct sk_buff *buf_extract(struct sk_buff *skb, u32 from_pos) +{ + struct tipc_msg *msg = (struct tipc_msg *)(skb->data + from_pos); + u32 size = msg_size(msg); + struct sk_buff *eb; + + eb = buf_acquire(size); + if (eb) + memcpy(eb->data, (unchar *)msg, size); + return eb; +} + +/* + * link_recv_changeover_msg(): Receive tunneled packet sent + * via other link. Node is locked. Return extracted buffer. + */ + +static int link_recv_changeover_msg(struct link **l_ptr, + struct sk_buff **buf) +{ + struct sk_buff *tunnel_buf = *buf; + struct link *dest_link; + struct tipc_msg *msg; + struct tipc_msg *tunnel_msg = buf_msg(tunnel_buf); + u32 msg_typ = msg_type(tunnel_msg); + u32 msg_count = msg_msgcnt(tunnel_msg); + + dest_link = (*l_ptr)->owner->links[msg_bearer_id(tunnel_msg)]; + assert(dest_link != *l_ptr); + if (!dest_link) { + msg_dbg(tunnel_msg, "NOLINK/<REC<"); + goto exit; + } + dbg("%c<-%c:", dest_link->b_ptr->net_plane, + (*l_ptr)->b_ptr->net_plane); + *l_ptr = dest_link; + msg = msg_get_wrapped(tunnel_msg); + + if (msg_typ == DUPLICATE_MSG) { + if (less(msg_seqno(msg), mod(dest_link->next_in_no))) { + msg_dbg(tunnel_msg, "DROP/<REC<"); + goto exit; + } + *buf = buf_extract(tunnel_buf,INT_H_SIZE); + if (*buf == NULL) { + warn("Memory squeeze; failed to extract msg\n"); + goto exit; + } + msg_dbg(tunnel_msg, "TNL<REC<"); + buf_discard(tunnel_buf); + return 1; + } + + /* First original message ?: */ + + if (link_is_up(dest_link)) { + msg_dbg(tunnel_msg, "UP/FIRST/<REC<"); + link_reset(dest_link); + dest_link->exp_msg_count = msg_count; + if (!msg_count) + goto exit; + } else if (dest_link->exp_msg_count == START_CHANGEOVER) { + msg_dbg(tunnel_msg, "BLK/FIRST/<REC<"); + dest_link->exp_msg_count = msg_count; + if (!msg_count) + goto exit; + } + + /* Receive original message */ + + if (dest_link->exp_msg_count == 0) { + msg_dbg(tunnel_msg, "OVERDUE/DROP/<REC<"); + dbg_print_link(dest_link, "LINK:"); + goto exit; + } + dest_link->exp_msg_count--; + if (less(msg_seqno(msg), dest_link->reset_checkpoint)) { + msg_dbg(tunnel_msg, "DROP/DUPL/<REC<"); + goto exit; + } else { + *buf = buf_extract(tunnel_buf, INT_H_SIZE); + if (*buf != NULL) { + msg_dbg(tunnel_msg, "TNL<REC<"); + buf_discard(tunnel_buf); + return 1; + } else { + warn("Memory squeeze; dropped incoming msg\n"); + } + } +exit: + *buf = 0; + buf_discard(tunnel_buf); + return 0; +} + +/* + * Bundler functionality: + */ +void link_recv_bundle(struct sk_buff *buf) +{ + u32 msgcount = msg_msgcnt(buf_msg(buf)); + u32 pos = INT_H_SIZE; + struct sk_buff *obuf; + + msg_dbg(buf_msg(buf), "<BNDL<: "); + while (msgcount--) { + obuf = buf_extract(buf, pos); + if (obuf == NULL) { + char addr_string[16]; + + warn("Buffer allocation failure;\n"); + warn(" incoming message(s) from %s lost\n", + addr_string_fill(addr_string, + msg_orignode(buf_msg(buf)))); + return; + }; + pos += align(msg_size(buf_msg(obuf))); + msg_dbg(buf_msg(obuf), " /"); + net_route_msg(obuf); + } + buf_discard(buf); +} + +/* + * Fragmentation/defragmentation: + */ + + +/* + * link_send_long_buf: Entry for buffers needing fragmentation. + * The buffer is complete, inclusive total message length. + * Returns user data length. + */ +int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf) +{ + struct tipc_msg *inmsg = buf_msg(buf); + struct tipc_msg fragm_hdr; + u32 insize = msg_size(inmsg); + u32 dsz = msg_data_sz(inmsg); + unchar *crs = buf->data; + u32 rest = insize; + u32 pack_sz = link_max_pkt(l_ptr); + u32 fragm_sz = pack_sz - INT_H_SIZE; + u32 fragm_no = 1; + u32 destaddr = msg_destnode(inmsg); + + if (msg_short(inmsg)) + destaddr = l_ptr->addr; + + if (msg_routed(inmsg)) + msg_set_prevnode(inmsg, tipc_own_addr); + + /* Prepare reusable fragment header: */ + + msg_init(&fragm_hdr, MSG_FRAGMENTER, FIRST_FRAGMENT, + TIPC_OK, INT_H_SIZE, destaddr); + msg_set_link_selector(&fragm_hdr, msg_link_selector(inmsg)); + msg_set_long_msgno(&fragm_hdr, mod(l_ptr->long_msg_seq_no++)); + msg_set_fragm_no(&fragm_hdr, fragm_no); + l_ptr->stats.sent_fragmented++; + + /* Chop up message: */ + + while (rest > 0) { + struct sk_buff *fragm; + + if (rest <= fragm_sz) { + fragm_sz = rest; + msg_set_type(&fragm_hdr, LAST_FRAGMENT); + } + fragm = buf_acquire(fragm_sz + INT_H_SIZE); + if (fragm == NULL) { + warn("Memory squeeze; failed to fragment msg\n"); + dsz = -ENOMEM; + goto exit; + } + msg_set_size(&fragm_hdr, fragm_sz + INT_H_SIZE); + memcpy(fragm->data, (unchar *)&fragm_hdr, INT_H_SIZE); + memcpy(fragm->data + INT_H_SIZE, crs, fragm_sz); + + /* Send queued messages first, if any: */ + + l_ptr->stats.sent_fragments++; + link_send_buf(l_ptr, fragm); + if (!link_is_up(l_ptr)) + return dsz; + msg_set_fragm_no(&fragm_hdr, ++fragm_no); + rest -= fragm_sz; + crs += fragm_sz; + msg_set_type(&fragm_hdr, FRAGMENT); + } +exit: + buf_discard(buf); + return dsz; +} + +/* + * A pending message being re-assembled must store certain values + * to handle subsequent fragments correctly. The following functions + * help storing these values in unused, available fields in the + * pending message. This makes dynamic memory allocation unecessary. + */ + +static inline u32 get_long_msg_seqno(struct sk_buff *buf) +{ + return msg_seqno(buf_msg(buf)); +} + +static inline void set_long_msg_seqno(struct sk_buff *buf, u32 seqno) +{ + msg_set_seqno(buf_msg(buf), seqno); +} + +static inline u32 get_fragm_size(struct sk_buff *buf) +{ + return msg_ack(buf_msg(buf)); +} + +static inline void set_fragm_size(struct sk_buff *buf, u32 sz) +{ + msg_set_ack(buf_msg(buf), sz); +} + +static inline u32 get_expected_frags(struct sk_buff *buf) +{ + return msg_bcast_ack(buf_msg(buf)); +} + +static inline void set_expected_frags(struct sk_buff *buf, u32 exp) +{ + msg_set_bcast_ack(buf_msg(buf), exp); +} + +static inline u32 get_timer_cnt(struct sk_buff *buf) +{ + return msg_reroute_cnt(buf_msg(buf)); +} + +static inline void incr_timer_cnt(struct sk_buff *buf) +{ + msg_incr_reroute_cnt(buf_msg(buf)); +} + +/* + * link_recv_fragment(): Called with node lock on. Returns + * the reassembled buffer if message is complete. + */ +int link_recv_fragment(struct sk_buff **pending, struct sk_buff **fb, + struct tipc_msg **m) +{ + struct sk_buff *prev = 0; + struct sk_buff *fbuf = *fb; + struct tipc_msg *fragm = buf_msg(fbuf); + struct sk_buff *pbuf = *pending; + u32 long_msg_seq_no = msg_long_msgno(fragm); + + *fb = 0; + msg_dbg(fragm,"FRG<REC<"); + + /* Is there an incomplete message waiting for this fragment? */ + + while (pbuf && ((msg_seqno(buf_msg(pbuf)) != long_msg_seq_no) + || (msg_orignode(fragm) != msg_orignode(buf_msg(pbuf))))) { + prev = pbuf; + pbuf = pbuf->next; + } + + if (!pbuf && (msg_type(fragm) == FIRST_FRAGMENT)) { + struct tipc_msg *imsg = (struct tipc_msg *)msg_data(fragm); + u32 msg_sz = msg_size(imsg); + u32 fragm_sz = msg_data_sz(fragm); + u32 exp_fragm_cnt = msg_sz/fragm_sz + !!(msg_sz % fragm_sz); + u32 max = TIPC_MAX_USER_MSG_SIZE + LONG_H_SIZE; + if (msg_type(imsg) == TIPC_MCAST_MSG) + max = TIPC_MAX_USER_MSG_SIZE + MCAST_H_SIZE; + if (msg_size(imsg) > max) { + msg_dbg(fragm,"<REC<Oversized: "); + buf_discard(fbuf); + return 0; + } + pbuf = buf_acquire(msg_size(imsg)); + if (pbuf != NULL) { + pbuf->next = *pending; + *pending = pbuf; + memcpy(pbuf->data, (unchar *)imsg, msg_data_sz(fragm)); + + /* Prepare buffer for subsequent fragments. */ + + set_long_msg_seqno(pbuf, long_msg_seq_no); + set_fragm_size(pbuf,fragm_sz); + set_expected_frags(pbuf,exp_fragm_cnt - 1); + } else { + warn("Memory squeeze; got no defragmenting buffer\n"); + } + buf_discard(fbuf); + return 0; + } else if (pbuf && (msg_type(fragm) != FIRST_FRAGMENT)) { + u32 dsz = msg_data_sz(fragm); + u32 fsz = get_fragm_size(pbuf); + u32 crs = ((msg_fragm_no(fragm) - 1) * fsz); + u32 exp_frags = get_expected_frags(pbuf) - 1; + memcpy(pbuf->data + crs, msg_data(fragm), dsz); + buf_discard(fbuf); + + /* Is message complete? */ + + if (exp_frags == 0) { + if (prev) + prev->next = pbuf->next; + else + *pending = pbuf->next; + msg_reset_reroute_cnt(buf_msg(pbuf)); + *fb = pbuf; + *m = buf_msg(pbuf); + return 1; + } + set_expected_frags(pbuf,exp_frags); + return 0; + } + dbg(" Discarding orphan fragment %x\n",fbuf); + msg_dbg(fragm,"ORPHAN:"); + dbg("Pending long buffers:\n"); + dbg_print_buf_chain(*pending); + buf_discard(fbuf); + return 0; +} + +/** + * link_check_defragm_bufs - flush stale incoming message fragments + * @l_ptr: pointer to link + */ + +static void link_check_defragm_bufs(struct link *l_ptr) +{ + struct sk_buff *prev = 0; + struct sk_buff *next = 0; + struct sk_buff *buf = l_ptr->defragm_buf; + + if (!buf) + return; + if (!link_working_working(l_ptr)) + return; + while (buf) { + u32 cnt = get_timer_cnt(buf); + + next = buf->next; + if (cnt < 4) { + incr_timer_cnt(buf); + prev = buf; + } else { + dbg(" Discarding incomplete long buffer\n"); + msg_dbg(buf_msg(buf), "LONG:"); + dbg_print_link(l_ptr, "curr:"); + dbg("Pending long buffers:\n"); + dbg_print_buf_chain(l_ptr->defragm_buf); + if (prev) + prev->next = buf->next; + else + l_ptr->defragm_buf = buf->next; + buf_discard(buf); + } + buf = next; + } +} + + + +static void link_set_supervision_props(struct link *l_ptr, u32 tolerance) +{ + l_ptr->tolerance = tolerance; + l_ptr->continuity_interval = + ((tolerance / 4) > 500) ? 500 : tolerance / 4; + l_ptr->abort_limit = tolerance / (l_ptr->continuity_interval / 4); +} + + +void link_set_queue_limits(struct link *l_ptr, u32 window) +{ + /* Data messages from this node, inclusive FIRST_FRAGM */ + l_ptr->queue_limit[DATA_LOW] = window; + l_ptr->queue_limit[DATA_MEDIUM] = (window / 3) * 4; + l_ptr->queue_limit[DATA_HIGH] = (window / 3) * 5; + l_ptr->queue_limit[DATA_CRITICAL] = (window / 3) * 6; + /* Transiting data messages,inclusive FIRST_FRAGM */ + l_ptr->queue_limit[DATA_LOW + 4] = 300; + l_ptr->queue_limit[DATA_MEDIUM + 4] = 600; + l_ptr->queue_limit[DATA_HIGH + 4] = 900; + l_ptr->queue_limit[DATA_CRITICAL + 4] = 1200; + l_ptr->queue_limit[CONN_MANAGER] = 1200; + l_ptr->queue_limit[ROUTE_DISTRIBUTOR] = 1200; + l_ptr->queue_limit[CHANGEOVER_PROTOCOL] = 2500; + l_ptr->queue_limit[NAME_DISTRIBUTOR] = 3000; + /* FRAGMENT and LAST_FRAGMENT packets */ + l_ptr->queue_limit[MSG_FRAGMENTER] = 4000; +} + +/** + * link_find_link - locate link by name + * @name - ptr to link name string + * @node - ptr to area to be filled with ptr to associated node + * + * Caller must hold 'net_lock' to ensure node and bearer are not deleted; + * this also prevents link deletion. + * + * Returns pointer to link (or 0 if invalid link name). + */ + +static struct link *link_find_link(const char *name, struct node **node) +{ + struct link_name link_name_parts; + struct bearer *b_ptr; + struct link *l_ptr; + + if (!link_name_validate(name, &link_name_parts)) + return 0; + + b_ptr = bearer_find_interface(link_name_parts.if_local); + if (!b_ptr) + return 0; + + *node = node_find(link_name_parts.addr_peer); + if (!*node) + return 0; + + l_ptr = (*node)->links[b_ptr->identity]; + if (!l_ptr || strcmp(l_ptr->name, name)) + return 0; + + return l_ptr; +} + +struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, + u16 cmd) +{ + struct tipc_link_config *args; + u32 new_value; + struct link *l_ptr; + struct node *node; + int res; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_CONFIG)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + args = (struct tipc_link_config *)TLV_DATA(req_tlv_area); + new_value = ntohl(args->value); + + if (!strcmp(args->name, bc_link_name)) { + if ((cmd == TIPC_CMD_SET_LINK_WINDOW) && + (bclink_set_queue_limits(new_value) == 0)) + return cfg_reply_none(); + return cfg_reply_error_string(TIPC_CFG_NOT_SUPPORTED + " (cannot change setting on broadcast link)"); + } + + read_lock_bh(&net_lock); + l_ptr = link_find_link(args->name, &node); + if (!l_ptr) { + read_unlock_bh(&net_lock); + return cfg_reply_error_string("link not found"); + } + + node_lock(node); + res = -EINVAL; + switch (cmd) { + case TIPC_CMD_SET_LINK_TOL: + if ((new_value >= TIPC_MIN_LINK_TOL) && + (new_value <= TIPC_MAX_LINK_TOL)) { + link_set_supervision_props(l_ptr, new_value); + link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, new_value, 0, 0); + res = TIPC_OK; + } + break; + case TIPC_CMD_SET_LINK_PRI: + if (new_value < TIPC_NUM_LINK_PRI) { + l_ptr->priority = new_value; + link_send_proto_msg(l_ptr, STATE_MSG, + 0, 0, 0, new_value, 0); + res = TIPC_OK; + } + break; + case TIPC_CMD_SET_LINK_WINDOW: + if ((new_value >= TIPC_MIN_LINK_WIN) && + (new_value <= TIPC_MAX_LINK_WIN)) { + link_set_queue_limits(l_ptr, new_value); + res = TIPC_OK; + } + break; + } + node_unlock(node); + + read_unlock_bh(&net_lock); + if (res) + return cfg_reply_error_string("cannot change link setting"); + + return cfg_reply_none(); +} + +/** + * link_reset_statistics - reset link statistics + * @l_ptr: pointer to link + */ + +static void link_reset_statistics(struct link *l_ptr) +{ + memset(&l_ptr->stats, 0, sizeof(l_ptr->stats)); + l_ptr->stats.sent_info = l_ptr->next_out_no; + l_ptr->stats.recv_info = l_ptr->next_in_no; +} + +struct sk_buff *link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space) +{ + char *link_name; + struct link *l_ptr; + struct node *node; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + link_name = (char *)TLV_DATA(req_tlv_area); + if (!strcmp(link_name, bc_link_name)) { + if (bclink_reset_stats()) + return cfg_reply_error_string("link not found"); + return cfg_reply_none(); + } + + read_lock_bh(&net_lock); + l_ptr = link_find_link(link_name, &node); + if (!l_ptr) { + read_unlock_bh(&net_lock); + return cfg_reply_error_string("link not found"); + } + + node_lock(node); + link_reset_statistics(l_ptr); + node_unlock(node); + read_unlock_bh(&net_lock); + return cfg_reply_none(); +} + +/** + * percent - convert count to a percentage of total (rounding up or down) + */ + +static u32 percent(u32 count, u32 total) +{ + return (count * 100 + (total / 2)) / total; +} + +/** + * link_stats - print link statistics + * @name: link name + * @buf: print buffer area + * @buf_size: size of print buffer area + * + * Returns length of print buffer data string (or 0 if error) + */ + +static int link_stats(const char *name, char *buf, const u32 buf_size) +{ + struct print_buf pb; + struct link *l_ptr; + struct node *node; + char *status; + u32 profile_total = 0; + + if (!strcmp(name, bc_link_name)) + return bclink_stats(buf, buf_size); + + printbuf_init(&pb, buf, buf_size); + + read_lock_bh(&net_lock); + l_ptr = link_find_link(name, &node); + if (!l_ptr) { + read_unlock_bh(&net_lock); + return 0; + } + node_lock(node); + + if (link_is_active(l_ptr)) + status = "ACTIVE"; + else if (link_is_up(l_ptr)) + status = "STANDBY"; + else + status = "DEFUNCT"; + tipc_printf(&pb, "Link <%s>\n" + " %s MTU:%u Priority:%u Tolerance:%u ms" + " Window:%u packets\n", + l_ptr->name, status, link_max_pkt(l_ptr), + l_ptr->priority, l_ptr->tolerance, l_ptr->queue_limit[0]); + tipc_printf(&pb, " RX packets:%u fragments:%u/%u bundles:%u/%u\n", + l_ptr->next_in_no - l_ptr->stats.recv_info, + l_ptr->stats.recv_fragments, + l_ptr->stats.recv_fragmented, + l_ptr->stats.recv_bundles, + l_ptr->stats.recv_bundled); + tipc_printf(&pb, " TX packets:%u fragments:%u/%u bundles:%u/%u\n", + l_ptr->next_out_no - l_ptr->stats.sent_info, + l_ptr->stats.sent_fragments, + l_ptr->stats.sent_fragmented, + l_ptr->stats.sent_bundles, + l_ptr->stats.sent_bundled); + profile_total = l_ptr->stats.msg_length_counts; + if (!profile_total) + profile_total = 1; + tipc_printf(&pb, " TX profile sample:%u packets average:%u octets\n" + " 0-64:%u%% -256:%u%% -1024:%u%% -4096:%u%% " + "-16354:%u%% -32768:%u%% -66000:%u%%\n", + l_ptr->stats.msg_length_counts, + l_ptr->stats.msg_lengths_total / profile_total, + percent(l_ptr->stats.msg_length_profile[0], profile_total), + percent(l_ptr->stats.msg_length_profile[1], profile_total), + percent(l_ptr->stats.msg_length_profile[2], profile_total), + percent(l_ptr->stats.msg_length_profile[3], profile_total), + percent(l_ptr->stats.msg_length_profile[4], profile_total), + percent(l_ptr->stats.msg_length_profile[5], profile_total), + percent(l_ptr->stats.msg_length_profile[6], profile_total)); + tipc_printf(&pb, " RX states:%u probes:%u naks:%u defs:%u dups:%u\n", + l_ptr->stats.recv_states, + l_ptr->stats.recv_probes, + l_ptr->stats.recv_nacks, + l_ptr->stats.deferred_recv, + l_ptr->stats.duplicates); + tipc_printf(&pb, " TX states:%u probes:%u naks:%u acks:%u dups:%u\n", + l_ptr->stats.sent_states, + l_ptr->stats.sent_probes, + l_ptr->stats.sent_nacks, + l_ptr->stats.sent_acks, + l_ptr->stats.retransmitted); + tipc_printf(&pb, " Congestion bearer:%u link:%u Send queue max:%u avg:%u\n", + l_ptr->stats.bearer_congs, + l_ptr->stats.link_congs, + l_ptr->stats.max_queue_sz, + l_ptr->stats.queue_sz_counts + ? (l_ptr->stats.accu_queue_sz / l_ptr->stats.queue_sz_counts) + : 0); + + node_unlock(node); + read_unlock_bh(&net_lock); + return printbuf_validate(&pb); +} + +#define MAX_LINK_STATS_INFO 2000 + +struct sk_buff *link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space) +{ + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + int str_len; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_LINK_NAME)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + buf = cfg_reply_alloc(TLV_SPACE(MAX_LINK_STATS_INFO)); + if (!buf) + return NULL; + + rep_tlv = (struct tlv_desc *)buf->data; + + str_len = link_stats((char *)TLV_DATA(req_tlv_area), + (char *)TLV_DATA(rep_tlv), MAX_LINK_STATS_INFO); + if (!str_len) { + buf_discard(buf); + return cfg_reply_error_string("link not found"); + } + + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +#if 0 +int link_control(const char *name, u32 op, u32 val) +{ + int res = -EINVAL; + struct link *l_ptr; + u32 bearer_id; + struct node * node; + u32 a; + + a = link_name2addr(name, &bearer_id); + read_lock_bh(&net_lock); + node = node_find(a); + if (node) { + node_lock(node); + l_ptr = node->links[bearer_id]; + if (l_ptr) { + if (op == TIPC_REMOVE_LINK) { + struct bearer *b_ptr = l_ptr->b_ptr; + spin_lock_bh(&b_ptr->publ.lock); + link_delete(l_ptr); + spin_unlock_bh(&b_ptr->publ.lock); + } + if (op == TIPC_CMD_BLOCK_LINK) { + link_reset(l_ptr); + l_ptr->blocked = 1; + } + if (op == TIPC_CMD_UNBLOCK_LINK) { + l_ptr->blocked = 0; + } + res = TIPC_OK; + } + node_unlock(node); + } + read_unlock_bh(&net_lock); + return res; +} +#endif + +/** + * link_get_max_pkt - get maximum packet size to use when sending to destination + * @dest: network address of destination node + * @selector: used to select from set of active links + * + * If no active link can be found, uses default maximum packet size. + */ + +u32 link_get_max_pkt(u32 dest, u32 selector) +{ + struct node *n_ptr; + struct link *l_ptr; + u32 res = MAX_PKT_DEFAULT; + + if (dest == tipc_own_addr) + return MAX_MSG_SIZE; + + read_lock_bh(&net_lock); + n_ptr = node_select(dest, selector); + if (n_ptr) { + node_lock(n_ptr); + l_ptr = n_ptr->active_links[selector & 1]; + if (l_ptr) + res = link_max_pkt(l_ptr); + node_unlock(n_ptr); + } + read_unlock_bh(&net_lock); + return res; +} + +#if 0 +static void link_dump_rec_queue(struct link *l_ptr) +{ + struct sk_buff *crs; + + if (!l_ptr->oldest_deferred_in) { + info("Reception queue empty\n"); + return; + } + info("Contents of Reception queue:\n"); + crs = l_ptr->oldest_deferred_in; + while (crs) { + if (crs->data == (void *)0x0000a3a3) { + info("buffer %x invalid\n", crs); + return; + } + msg_dbg(buf_msg(crs), "In rec queue: \n"); + crs = crs->next; + } +} +#endif + +static void link_dump_send_queue(struct link *l_ptr) +{ + if (l_ptr->next_out) { + info("\nContents of unsent queue:\n"); + dbg_print_buf_chain(l_ptr->next_out); + } + info("\nContents of send queue:\n"); + if (l_ptr->first_out) { + dbg_print_buf_chain(l_ptr->first_out); + } + info("Empty send queue\n"); +} + +static void link_print(struct link *l_ptr, struct print_buf *buf, + const char *str) +{ + tipc_printf(buf, str); + if (link_reset_reset(l_ptr) || link_reset_unknown(l_ptr)) + return; + tipc_printf(buf, "Link %x<%s>:", + l_ptr->addr, l_ptr->b_ptr->publ.name); + tipc_printf(buf, ": NXO(%u):", mod(l_ptr->next_out_no)); + tipc_printf(buf, "NXI(%u):", mod(l_ptr->next_in_no)); + tipc_printf(buf, "SQUE"); + if (l_ptr->first_out) { + tipc_printf(buf, "[%u..", msg_seqno(buf_msg(l_ptr->first_out))); + if (l_ptr->next_out) + tipc_printf(buf, "%u..", + msg_seqno(buf_msg(l_ptr->next_out))); + tipc_printf(buf, "%u]", + msg_seqno(buf_msg + (l_ptr->last_out)), l_ptr->out_queue_size); + if ((mod(msg_seqno(buf_msg(l_ptr->last_out)) - + msg_seqno(buf_msg(l_ptr->first_out))) + != (l_ptr->out_queue_size - 1)) + || (l_ptr->last_out->next != 0)) { + tipc_printf(buf, "\nSend queue inconsistency\n"); + tipc_printf(buf, "first_out= %x ", l_ptr->first_out); + tipc_printf(buf, "next_out= %x ", l_ptr->next_out); + tipc_printf(buf, "last_out= %x ", l_ptr->last_out); + link_dump_send_queue(l_ptr); + } + } else + tipc_printf(buf, "[]"); + tipc_printf(buf, "SQSIZ(%u)", l_ptr->out_queue_size); + if (l_ptr->oldest_deferred_in) { + u32 o = msg_seqno(buf_msg(l_ptr->oldest_deferred_in)); + u32 n = msg_seqno(buf_msg(l_ptr->newest_deferred_in)); + tipc_printf(buf, ":RQUE[%u..%u]", o, n); + if (l_ptr->deferred_inqueue_sz != mod((n + 1) - o)) { + tipc_printf(buf, ":RQSIZ(%u)", + l_ptr->deferred_inqueue_sz); + } + } + if (link_working_unknown(l_ptr)) + tipc_printf(buf, ":WU"); + if (link_reset_reset(l_ptr)) + tipc_printf(buf, ":RR"); + if (link_reset_unknown(l_ptr)) + tipc_printf(buf, ":RU"); + if (link_working_working(l_ptr)) + tipc_printf(buf, ":WW"); + tipc_printf(buf, "\n"); +} + diff --git a/net/tipc/link.h b/net/tipc/link.h new file mode 100644 index 00000000000..c2553f07375 --- /dev/null +++ b/net/tipc/link.h @@ -0,0 +1,296 @@ +/* + * net/tipc/link.h: Include file for TIPC link code + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_LINK_H +#define _TIPC_LINK_H + +#include "dbg.h" +#include "msg.h" +#include "bearer.h" +#include "node.h" + +#define PUSH_FAILED 1 +#define PUSH_FINISHED 2 + +/* + * Link states + */ + +#define WORKING_WORKING 560810u +#define WORKING_UNKNOWN 560811u +#define RESET_UNKNOWN 560812u +#define RESET_RESET 560813u + +/* + * Starting value for maximum packet size negotiation on unicast links + * (unless bearer MTU is less) + */ + +#define MAX_PKT_DEFAULT 1500 + +/** + * struct link - TIPC link data structure + * @addr: network address of link's peer node + * @name: link name character string + * @media_addr: media address to use when sending messages over link + * @timer: link timer + * @owner: pointer to peer node + * @link_list: adjacent links in bearer's list of links + * @started: indicates if link has been started + * @checkpoint: reference point for triggering link continuity checking + * @peer_session: link session # being used by peer end of link + * @peer_bearer_id: bearer id used by link's peer endpoint + * @b_ptr: pointer to bearer used by link + * @tolerance: minimum link continuity loss needed to reset link [in ms] + * @continuity_interval: link continuity testing interval [in ms] + * @abort_limit: # of unacknowledged continuity probes needed to reset link + * @state: current state of link FSM + * @blocked: indicates if link has been administratively blocked + * @fsm_msg_cnt: # of protocol messages link FSM has sent in current state + * @proto_msg: template for control messages generated by link + * @pmsg: convenience pointer to "proto_msg" field + * @priority: current link priority + * @queue_limit: outbound message queue congestion thresholds (indexed by user) + * @exp_msg_count: # of tunnelled messages expected during link changeover + * @reset_checkpoint: seq # of last acknowledged message at time of link reset + * @max_pkt: current maximum packet size for this link + * @max_pkt_target: desired maximum packet size for this link + * @max_pkt_probes: # of probes based on current (max_pkt, max_pkt_target) + * @out_queue_size: # of messages in outbound message queue + * @first_out: ptr to first outbound message in queue + * @last_out: ptr to last outbound message in queue + * @next_out_no: next sequence number to use for outbound messages + * @last_retransmitted: sequence number of most recently retransmitted message + * @stale_count: # of identical retransmit requests made by peer + * @next_in_no: next sequence number to expect for inbound messages + * @deferred_inqueue_sz: # of messages in inbound message queue + * @oldest_deferred_in: ptr to first inbound message in queue + * @newest_deferred_in: ptr to last inbound message in queue + * @unacked_window: # of inbound messages rx'd without ack'ing back to peer + * @proto_msg_queue: ptr to (single) outbound control message + * @retransm_queue_size: number of messages to retransmit + * @retransm_queue_head: sequence number of first message to retransmit + * @next_out: ptr to first unsent outbound message in queue + * @waiting_ports: linked list of ports waiting for link congestion to abate + * @long_msg_seq_no: next identifier to use for outbound fragmented messages + * @defragm_buf: list of partially reassembled inbound message fragments + * @stats: collects statistics regarding link activity + * @print_buf: print buffer used to log link activity + */ + +struct link { + u32 addr; + char name[TIPC_MAX_LINK_NAME]; + struct tipc_media_addr media_addr; + struct timer_list timer; + struct node *owner; + struct list_head link_list; + + /* Management and link supervision data */ + int started; + u32 checkpoint; + u32 peer_session; + u32 peer_bearer_id; + struct bearer *b_ptr; + u32 tolerance; + u32 continuity_interval; + u32 abort_limit; + int state; + int blocked; + u32 fsm_msg_cnt; + struct { + unchar hdr[INT_H_SIZE]; + unchar body[TIPC_MAX_IF_NAME]; + } proto_msg; + struct tipc_msg *pmsg; + u32 priority; + u32 queue_limit[15]; /* queue_limit[0]==window limit */ + + /* Changeover */ + u32 exp_msg_count; + u32 reset_checkpoint; + + /* Max packet negotiation */ + u32 max_pkt; + u32 max_pkt_target; + u32 max_pkt_probes; + + /* Sending */ + u32 out_queue_size; + struct sk_buff *first_out; + struct sk_buff *last_out; + u32 next_out_no; + u32 last_retransmitted; + u32 stale_count; + + /* Reception */ + u32 next_in_no; + u32 deferred_inqueue_sz; + struct sk_buff *oldest_deferred_in; + struct sk_buff *newest_deferred_in; + u32 unacked_window; + + /* Congestion handling */ + struct sk_buff *proto_msg_queue; + u32 retransm_queue_size; + u32 retransm_queue_head; + struct sk_buff *next_out; + struct list_head waiting_ports; + + /* Fragmentation/defragmentation */ + u32 long_msg_seq_no; + struct sk_buff *defragm_buf; + + /* Statistics */ + struct { + u32 sent_info; /* used in counting # sent packets */ + u32 recv_info; /* used in counting # recv'd packets */ + u32 sent_states; + u32 recv_states; + u32 sent_probes; + u32 recv_probes; + u32 sent_nacks; + u32 recv_nacks; + u32 sent_acks; + u32 sent_bundled; + u32 sent_bundles; + u32 recv_bundled; + u32 recv_bundles; + u32 retransmitted; + u32 sent_fragmented; + u32 sent_fragments; + u32 recv_fragmented; + u32 recv_fragments; + u32 link_congs; /* # port sends blocked by congestion */ + u32 bearer_congs; + u32 deferred_recv; + u32 duplicates; + + /* for statistical profiling of send queue size */ + + u32 max_queue_sz; + u32 accu_queue_sz; + u32 queue_sz_counts; + + /* for statistical profiling of message lengths */ + + u32 msg_length_counts; + u32 msg_lengths_total; + u32 msg_length_profile[7]; +#if 0 + u32 sent_tunneled; + u32 recv_tunneled; +#endif + } stats; + + struct print_buf print_buf; +}; + +struct port; + +struct link *link_create(struct bearer *b_ptr, const u32 peer, + const struct tipc_media_addr *media_addr); +void link_delete(struct link *l_ptr); +void link_changeover(struct link *l_ptr); +void link_send_duplicate(struct link *l_ptr, struct link *dest); +void link_reset_fragments(struct link *l_ptr); +int link_is_up(struct link *l_ptr); +int link_is_active(struct link *l_ptr); +void link_start(struct link *l_ptr); +u32 link_push_packet(struct link *l_ptr); +void link_stop(struct link *l_ptr); +struct sk_buff *link_cmd_config(const void *req_tlv_area, int req_tlv_space, u16 cmd); +struct sk_buff *link_cmd_show_stats(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *link_cmd_reset_stats(const void *req_tlv_area, int req_tlv_space); +void link_reset(struct link *l_ptr); +int link_send(struct sk_buff *buf, u32 dest, u32 selector); +int link_send_buf(struct link *l_ptr, struct sk_buff *buf); +u32 link_get_max_pkt(u32 dest,u32 selector); +int link_send_sections_fast(struct port* sender, + struct iovec const *msg_sect, + const u32 num_sect, + u32 destnode); + +int link_send_long_buf(struct link *l_ptr, struct sk_buff *buf); +void link_tunnel(struct link *l_ptr, struct tipc_msg *tnl_hdr, + struct tipc_msg *msg, u32 selector); +void link_recv_bundle(struct sk_buff *buf); +int link_recv_fragment(struct sk_buff **pending, + struct sk_buff **fb, + struct tipc_msg **msg); +void link_send_proto_msg(struct link *l_ptr, u32 msg_typ, int prob, u32 gap, + u32 tolerance, u32 priority, u32 acked_mtu); +void link_push_queue(struct link *l_ptr); +u32 link_defer_pkt(struct sk_buff **head, struct sk_buff **tail, + struct sk_buff *buf); +void link_wakeup_ports(struct link *l_ptr, int all); +void link_set_queue_limits(struct link *l_ptr, u32 window); +void link_retransmit(struct link *l_ptr, struct sk_buff *start, u32 retransmits); + +/* + * Link sequence number manipulation routines (uses modulo 2**16 arithmetic) + */ + +static inline u32 mod(u32 x) +{ + return x & 0xffffu; +} + +static inline int between(u32 lower, u32 upper, u32 n) +{ + if ((lower < n) && (n < upper)) + return 1; + if ((upper < lower) && ((n > lower) || (n < upper))) + return 1; + return 0; +} + +static inline int less_eq(u32 left, u32 right) +{ + return (mod(right - left) < 32768u); +} + +static inline int less(u32 left, u32 right) +{ + return (less_eq(left, right) && (mod(right) != mod(left))); +} + +static inline u32 lesser(u32 left, u32 right) +{ + return less_eq(left, right) ? left : right; +} + +#endif diff --git a/net/tipc/msg.c b/net/tipc/msg.c new file mode 100644 index 00000000000..03dbc55cb04 --- /dev/null +++ b/net/tipc/msg.c @@ -0,0 +1,334 @@ +/* + * net/tipc/msg.c: TIPC message header routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "addr.h" +#include "dbg.h" +#include "msg.h" +#include "bearer.h" + + +void msg_set_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) +{ + memcpy(&((int *)m)[5], a, sizeof(*a)); +} + +void msg_get_media_addr(struct tipc_msg *m, struct tipc_media_addr *a) +{ + memcpy(a, &((int*)m)[5], sizeof(*a)); +} + + +void msg_print(struct print_buf *buf, struct tipc_msg *msg, const char *str) +{ + u32 usr = msg_user(msg); + tipc_printf(buf, str); + + switch (usr) { + case MSG_BUNDLER: + tipc_printf(buf, "BNDL::"); + tipc_printf(buf, "MSGS(%u):", msg_msgcnt(msg)); + break; + case BCAST_PROTOCOL: + tipc_printf(buf, "BCASTP::"); + break; + case MSG_FRAGMENTER: + tipc_printf(buf, "FRAGM::"); + switch (msg_type(msg)) { + case FIRST_FRAGMENT: + tipc_printf(buf, "FIRST:"); + break; + case FRAGMENT: + tipc_printf(buf, "BODY:"); + break; + case LAST_FRAGMENT: + tipc_printf(buf, "LAST:"); + break; + default: + tipc_printf(buf, "UNKNOWN:%x",msg_type(msg)); + + } + tipc_printf(buf, "NO(%u/%u):",msg_long_msgno(msg), + msg_fragm_no(msg)); + break; + case DATA_LOW: + case DATA_MEDIUM: + case DATA_HIGH: + case DATA_CRITICAL: + tipc_printf(buf, "DAT%u:", msg_user(msg)); + if (msg_short(msg)) { + tipc_printf(buf, "CON:"); + break; + } + switch (msg_type(msg)) { + case TIPC_CONN_MSG: + tipc_printf(buf, "CON:"); + break; + case TIPC_MCAST_MSG: + tipc_printf(buf, "MCST:"); + break; + case TIPC_NAMED_MSG: + tipc_printf(buf, "NAM:"); + break; + case TIPC_DIRECT_MSG: + tipc_printf(buf, "DIR:"); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE %u",msg_type(msg)); + } + if (msg_routed(msg) && !msg_non_seq(msg)) + tipc_printf(buf, "ROUT:"); + if (msg_reroute_cnt(msg)) + tipc_printf(buf, "REROUTED(%u):", + msg_reroute_cnt(msg)); + break; + case NAME_DISTRIBUTOR: + tipc_printf(buf, "NMD::"); + switch (msg_type(msg)) { + case PUBLICATION: + tipc_printf(buf, "PUBL(%u):", (msg_size(msg) - msg_hdr_sz(msg)) / 20); /* Items */ + break; + case WITHDRAWAL: + tipc_printf(buf, "WDRW:"); + break; + default: + tipc_printf(buf, "UNKNOWN:%x",msg_type(msg)); + } + if (msg_routed(msg)) + tipc_printf(buf, "ROUT:"); + if (msg_reroute_cnt(msg)) + tipc_printf(buf, "REROUTED(%u):", + msg_reroute_cnt(msg)); + break; + case CONN_MANAGER: + tipc_printf(buf, "CONN_MNG:"); + switch (msg_type(msg)) { + case CONN_PROBE: + tipc_printf(buf, "PROBE:"); + break; + case CONN_PROBE_REPLY: + tipc_printf(buf, "PROBE_REPLY:"); + break; + case CONN_ACK: + tipc_printf(buf, "CONN_ACK:"); + tipc_printf(buf, "ACK(%u):",msg_msgcnt(msg)); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); + } + if (msg_routed(msg)) + tipc_printf(buf, "ROUT:"); + if (msg_reroute_cnt(msg)) + tipc_printf(buf, "REROUTED(%u):",msg_reroute_cnt(msg)); + break; + case LINK_PROTOCOL: + tipc_printf(buf, "PROT:TIM(%u):",msg_timestamp(msg)); + switch (msg_type(msg)) { + case STATE_MSG: + tipc_printf(buf, "STATE:"); + tipc_printf(buf, "%s:",msg_probe(msg) ? "PRB" :""); + tipc_printf(buf, "NXS(%u):",msg_next_sent(msg)); + tipc_printf(buf, "GAP(%u):",msg_seq_gap(msg)); + tipc_printf(buf, "LSTBC(%u):",msg_last_bcast(msg)); + break; + case RESET_MSG: + tipc_printf(buf, "RESET:"); + if (msg_size(msg) != msg_hdr_sz(msg)) + tipc_printf(buf, "BEAR:%s:",msg_data(msg)); + break; + case ACTIVATE_MSG: + tipc_printf(buf, "ACTIVATE:"); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); + } + tipc_printf(buf, "PLANE(%c):",msg_net_plane(msg)); + tipc_printf(buf, "SESS(%u):",msg_session(msg)); + break; + case CHANGEOVER_PROTOCOL: + tipc_printf(buf, "TUNL:"); + switch (msg_type(msg)) { + case DUPLICATE_MSG: + tipc_printf(buf, "DUPL:"); + break; + case ORIGINAL_MSG: + tipc_printf(buf, "ORIG:"); + tipc_printf(buf, "EXP(%u)",msg_msgcnt(msg)); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); + } + break; + case ROUTE_DISTRIBUTOR: + tipc_printf(buf, "ROUTING_MNG:"); + switch (msg_type(msg)) { + case EXT_ROUTING_TABLE: + tipc_printf(buf, "EXT_TBL:"); + tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); + break; + case LOCAL_ROUTING_TABLE: + tipc_printf(buf, "LOCAL_TBL:"); + tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); + break; + case SLAVE_ROUTING_TABLE: + tipc_printf(buf, "DP_TBL:"); + tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); + break; + case ROUTE_ADDITION: + tipc_printf(buf, "ADD:"); + tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); + break; + case ROUTE_REMOVAL: + tipc_printf(buf, "REMOVE:"); + tipc_printf(buf, "TO:%x:",msg_remote_node(msg)); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE:%x",msg_type(msg)); + } + break; + case LINK_CONFIG: + tipc_printf(buf, "CFG:"); + switch (msg_type(msg)) { + case DSC_REQ_MSG: + tipc_printf(buf, "DSC_REQ:"); + break; + case DSC_RESP_MSG: + tipc_printf(buf, "DSC_RESP:"); + break; + default: + tipc_printf(buf, "UNKNOWN TYPE:%x:",msg_type(msg)); + break; + } + break; + default: + tipc_printf(buf, "UNKNOWN USER:"); + } + + switch (usr) { + case CONN_MANAGER: + case NAME_DISTRIBUTOR: + case DATA_LOW: + case DATA_MEDIUM: + case DATA_HIGH: + case DATA_CRITICAL: + if (msg_short(msg)) + break; /* No error */ + switch (msg_errcode(msg)) { + case TIPC_OK: + break; + case TIPC_ERR_NO_NAME: + tipc_printf(buf, "NO_NAME:"); + break; + case TIPC_ERR_NO_PORT: + tipc_printf(buf, "NO_PORT:"); + break; + case TIPC_ERR_NO_NODE: + tipc_printf(buf, "NO_PROC:"); + break; + case TIPC_ERR_OVERLOAD: + tipc_printf(buf, "OVERLOAD:"); + break; + case TIPC_CONN_SHUTDOWN: + tipc_printf(buf, "SHUTDOWN:"); + break; + default: + tipc_printf(buf, "UNKNOWN ERROR(%x):", + msg_errcode(msg)); + } + default:{} + } + + tipc_printf(buf, "HZ(%u):", msg_hdr_sz(msg)); + tipc_printf(buf, "SZ(%u):", msg_size(msg)); + tipc_printf(buf, "SQNO(%u):", msg_seqno(msg)); + + if (msg_non_seq(msg)) + tipc_printf(buf, "NOSEQ:"); + else { + tipc_printf(buf, "ACK(%u):", msg_ack(msg)); + } + tipc_printf(buf, "BACK(%u):", msg_bcast_ack(msg)); + tipc_printf(buf, "PRND(%x)", msg_prevnode(msg)); + + if (msg_isdata(msg)) { + if (msg_named(msg)) { + tipc_printf(buf, "NTYP(%u):", msg_nametype(msg)); + tipc_printf(buf, "NINST(%u)", msg_nameinst(msg)); + } + } + + if ((usr != LINK_PROTOCOL) && (usr != LINK_CONFIG) && + (usr != MSG_BUNDLER)) { + if (!msg_short(msg)) { + tipc_printf(buf, ":ORIG(%x:%u):", + msg_orignode(msg), msg_origport(msg)); + tipc_printf(buf, ":DEST(%x:%u):", + msg_destnode(msg), msg_destport(msg)); + } else { + tipc_printf(buf, ":OPRT(%u):", msg_origport(msg)); + tipc_printf(buf, ":DPRT(%u):", msg_destport(msg)); + } + if (msg_routed(msg) && !msg_non_seq(msg)) + tipc_printf(buf, ":TSEQN(%u)", msg_transp_seqno(msg)); + } + if (msg_user(msg) == NAME_DISTRIBUTOR) { + tipc_printf(buf, ":ONOD(%x):", msg_orignode(msg)); + tipc_printf(buf, ":DNOD(%x):", msg_destnode(msg)); + if (msg_routed(msg)) { + tipc_printf(buf, ":CSEQN(%u)", msg_transp_seqno(msg)); + } + } + + if (msg_user(msg) == LINK_CONFIG) { + u32* raw = (u32*)msg; + struct tipc_media_addr* orig = (struct tipc_media_addr*)&raw[5]; + tipc_printf(buf, ":REQL(%u):", msg_req_links(msg)); + tipc_printf(buf, ":DDOM(%x):", msg_dest_domain(msg)); + tipc_printf(buf, ":NETID(%u):", msg_bc_netid(msg)); + media_addr_printf(buf, orig); + } + if (msg_user(msg) == BCAST_PROTOCOL) { + tipc_printf(buf, "BCNACK:AFTER(%u):", msg_bcgap_after(msg)); + tipc_printf(buf, "TO(%u):", msg_bcgap_to(msg)); + } + tipc_printf(buf, "\n"); + if ((usr == CHANGEOVER_PROTOCOL) && (msg_msgcnt(msg))) { + msg_print(buf,msg_get_wrapped(msg)," /"); + } + if ((usr == MSG_FRAGMENTER) && (msg_type(msg) == FIRST_FRAGMENT)) { + msg_print(buf,msg_get_wrapped(msg)," /"); + } +} diff --git a/net/tipc/msg.h b/net/tipc/msg.h new file mode 100644 index 00000000000..662c81862a0 --- /dev/null +++ b/net/tipc/msg.h @@ -0,0 +1,818 @@ +/* + * net/tipc/msg.h: Include file for TIPC message header routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_MSG_H +#define _TIPC_MSG_H + +#include <net/tipc/tipc_msg.h> + +#define TIPC_VERSION 2 +#define DATA_LOW TIPC_LOW_IMPORTANCE +#define DATA_MEDIUM TIPC_MEDIUM_IMPORTANCE +#define DATA_HIGH TIPC_HIGH_IMPORTANCE +#define DATA_CRITICAL TIPC_CRITICAL_IMPORTANCE +#define SHORT_H_SIZE 24 /* Connected,in cluster */ +#define DIR_MSG_H_SIZE 32 /* Directly addressed messages */ +#define CONN_MSG_H_SIZE 36 /* Routed connected msgs*/ +#define LONG_H_SIZE 40 /* Named Messages */ +#define MCAST_H_SIZE 44 /* Multicast messages */ +#define MAX_H_SIZE 60 /* Inclusive full options */ +#define MAX_MSG_SIZE (MAX_H_SIZE + TIPC_MAX_USER_MSG_SIZE) +#define LINK_CONFIG 13 + + +/* + TIPC user data message header format, version 2 + + - Fundamental definitions available to privileged TIPC users + are located in tipc_msg.h. + - Remaining definitions available to TIPC internal users appear below. +*/ + + +static inline void msg_set_word(struct tipc_msg *m, u32 w, u32 val) +{ + m->hdr[w] = htonl(val); +} + +static inline void msg_set_bits(struct tipc_msg *m, u32 w, + u32 pos, u32 mask, u32 val) +{ + u32 word = msg_word(m,w) & ~(mask << pos); + msg_set_word(m, w, (word |= (val << pos))); +} + +/* + * Word 0 + */ + +static inline u32 msg_version(struct tipc_msg *m) +{ + return msg_bits(m, 0, 29, 7); +} + +static inline void msg_set_version(struct tipc_msg *m) +{ + msg_set_bits(m, 0, 29, 0xf, TIPC_VERSION); +} + +static inline u32 msg_user(struct tipc_msg *m) +{ + return msg_bits(m, 0, 25, 0xf); +} + +static inline u32 msg_isdata(struct tipc_msg *m) +{ + return (msg_user(m) <= DATA_CRITICAL); +} + +static inline void msg_set_user(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 0, 25, 0xf, n); +} + +static inline void msg_set_importance(struct tipc_msg *m, u32 i) +{ + msg_set_user(m, i); +} + +static inline void msg_set_hdr_sz(struct tipc_msg *m,u32 n) +{ + msg_set_bits(m, 0, 21, 0xf, n>>2); +} + +static inline int msg_non_seq(struct tipc_msg *m) +{ + return msg_bits(m, 0, 20, 1); +} + +static inline void msg_set_non_seq(struct tipc_msg *m) +{ + msg_set_bits(m, 0, 20, 1, 1); +} + +static inline int msg_dest_droppable(struct tipc_msg *m) +{ + return msg_bits(m, 0, 19, 1); +} + +static inline void msg_set_dest_droppable(struct tipc_msg *m, u32 d) +{ + msg_set_bits(m, 0, 19, 1, d); +} + +static inline int msg_src_droppable(struct tipc_msg *m) +{ + return msg_bits(m, 0, 18, 1); +} + +static inline void msg_set_src_droppable(struct tipc_msg *m, u32 d) +{ + msg_set_bits(m, 0, 18, 1, d); +} + +static inline void msg_set_size(struct tipc_msg *m, u32 sz) +{ + m->hdr[0] = htonl((msg_word(m, 0) & ~0x1ffff) | sz); +} + + +/* + * Word 1 + */ + +static inline void msg_set_type(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 29, 0x7, n); +} + +static inline void msg_set_errcode(struct tipc_msg *m, u32 err) +{ + msg_set_bits(m, 1, 25, 0xf, err); +} + +static inline u32 msg_reroute_cnt(struct tipc_msg *m) +{ + return msg_bits(m, 1, 21, 0xf); +} + +static inline void msg_incr_reroute_cnt(struct tipc_msg *m) +{ + msg_set_bits(m, 1, 21, 0xf, msg_reroute_cnt(m) + 1); +} + +static inline void msg_reset_reroute_cnt(struct tipc_msg *m) +{ + msg_set_bits(m, 1, 21, 0xf, 0); +} + +static inline u32 msg_lookup_scope(struct tipc_msg *m) +{ + return msg_bits(m, 1, 19, 0x3); +} + +static inline void msg_set_lookup_scope(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 19, 0x3, n); +} + +static inline void msg_set_options(struct tipc_msg *m, const char *opt, u32 sz) +{ + u32 hsz = msg_hdr_sz(m); + char *to = (char *)&m->hdr[hsz/4]; + + if ((hsz < DIR_MSG_H_SIZE) || ((hsz + sz) > MAX_H_SIZE)) + return; + msg_set_bits(m, 1, 16, 0x7, (hsz - 28)/4); + msg_set_hdr_sz(m, hsz + sz); + memcpy(to, opt, sz); +} + +static inline u32 msg_bcast_ack(struct tipc_msg *m) +{ + return msg_bits(m, 1, 0, 0xffff); +} + +static inline void msg_set_bcast_ack(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 0, 0xffff, n); +} + + +/* + * Word 2 + */ + +static inline u32 msg_ack(struct tipc_msg *m) +{ + return msg_bits(m, 2, 16, 0xffff); +} + +static inline void msg_set_ack(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 16, 0xffff, n); +} + +static inline u32 msg_seqno(struct tipc_msg *m) +{ + return msg_bits(m, 2, 0, 0xffff); +} + +static inline void msg_set_seqno(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 0, 0xffff, n); +} + + +/* + * Words 3-10 + */ + + +static inline void msg_set_prevnode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 3, a); +} + +static inline void msg_set_origport(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 4, p); +} + +static inline void msg_set_destport(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 5, p); +} + +static inline void msg_set_mc_netid(struct tipc_msg *m, u32 p) +{ + msg_set_word(m, 5, p); +} + +static inline void msg_set_orignode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 6, a); +} + +static inline void msg_set_destnode(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, 7, a); +} + +static inline int msg_is_dest(struct tipc_msg *m, u32 d) +{ + return(msg_short(m) || (msg_destnode(m) == d)); +} + +static inline u32 msg_routed(struct tipc_msg *m) +{ + if (likely(msg_short(m))) + return 0; + return(msg_destnode(m) ^ msg_orignode(m)) >> 11; +} + +static inline void msg_set_nametype(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 8, n); +} + +static inline u32 msg_transp_seqno(struct tipc_msg *m) +{ + return msg_word(m, 8); +} + +static inline void msg_set_timestamp(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 8, n); +} + +static inline u32 msg_timestamp(struct tipc_msg *m) +{ + return msg_word(m, 8); +} + +static inline void msg_set_transp_seqno(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 8, n); +} + +static inline void msg_set_namelower(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 9, n); +} + +static inline void msg_set_nameinst(struct tipc_msg *m, u32 n) +{ + msg_set_namelower(m, n); +} + +static inline void msg_set_nameupper(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 10, n); +} + +static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) +{ + return (struct tipc_msg *)msg_data(m); +} + +static inline void msg_expand(struct tipc_msg *m, u32 destnode) +{ + if (!msg_short(m)) + return; + msg_set_hdr_sz(m, LONG_H_SIZE); + msg_set_orignode(m, msg_prevnode(m)); + msg_set_destnode(m, destnode); + memset(&m->hdr[8], 0, 12); +} + + + +/* + TIPC internal message header format, version 2 + + 1 0 9 8 7 6 5 4|3 2 1 0 9 8 7 6|5 4 3 2 1 0 9 8|7 6 5 4 3 2 1 0 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w0:|vers |msg usr|hdr sz |n|resrv| packet size | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w1:|m typ|rsv=0| sequence gap | broadcast ack no | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w2:| link level ack no/bc_gap_from | seq no / bcast_gap_to | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w3:| previous node | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w4:| next sent broadcast/fragm no | next sent pkt/ fragm msg no | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w5:| session no |rsv=0|r|berid|link prio|netpl|p| + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w6:| originating node | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w7:| destination node | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w8:| transport sequence number | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + w9:| msg count / bcast tag | link tolerance | + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + \ \ + / User Specific Data / + \ \ + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + + NB: CONN_MANAGER use data message format. LINK_CONFIG has own format. +*/ + +/* + * Internal users + */ + +#define BCAST_PROTOCOL 5 +#define MSG_BUNDLER 6 +#define LINK_PROTOCOL 7 +#define CONN_MANAGER 8 +#define ROUTE_DISTRIBUTOR 9 +#define CHANGEOVER_PROTOCOL 10 +#define NAME_DISTRIBUTOR 11 +#define MSG_FRAGMENTER 12 +#define LINK_CONFIG 13 +#define INT_H_SIZE 40 +#define DSC_H_SIZE 40 + +/* + * Connection management protocol messages + */ + +#define CONN_PROBE 0 +#define CONN_PROBE_REPLY 1 +#define CONN_ACK 2 + +/* + * Name distributor messages + */ + +#define PUBLICATION 0 +#define WITHDRAWAL 1 + + +/* + * Word 1 + */ + +static inline u32 msg_seq_gap(struct tipc_msg *m) +{ + return msg_bits(m, 1, 16, 0xff); +} + +static inline void msg_set_seq_gap(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 16, 0xff, n); +} + +static inline u32 msg_req_links(struct tipc_msg *m) +{ + return msg_bits(m, 1, 16, 0xfff); +} + +static inline void msg_set_req_links(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 1, 16, 0xfff, n); +} + + +/* + * Word 2 + */ + +static inline u32 msg_dest_domain(struct tipc_msg *m) +{ + return msg_word(m, 2); +} + +static inline void msg_set_dest_domain(struct tipc_msg *m, u32 n) +{ + msg_set_word(m, 2, n); +} + +static inline u32 msg_bcgap_after(struct tipc_msg *m) +{ + return msg_bits(m, 2, 16, 0xffff); +} + +static inline void msg_set_bcgap_after(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 16, 0xffff, n); +} + +static inline u32 msg_bcgap_to(struct tipc_msg *m) +{ + return msg_bits(m, 2, 0, 0xffff); +} + +static inline void msg_set_bcgap_to(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 2, 0, 0xffff, n); +} + + +/* + * Word 4 + */ + +static inline u32 msg_last_bcast(struct tipc_msg *m) +{ + return msg_bits(m, 4, 16, 0xffff); +} + +static inline void msg_set_last_bcast(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 16, 0xffff, n); +} + + +static inline u32 msg_fragm_no(struct tipc_msg *m) +{ + return msg_bits(m, 4, 16, 0xffff); +} + +static inline void msg_set_fragm_no(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 16, 0xffff, n); +} + + +static inline u32 msg_next_sent(struct tipc_msg *m) +{ + return msg_bits(m, 4, 0, 0xffff); +} + +static inline void msg_set_next_sent(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 0xffff, n); +} + + +static inline u32 msg_long_msgno(struct tipc_msg *m) +{ + return msg_bits(m, 4, 0, 0xffff); +} + +static inline void msg_set_long_msgno(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 0xffff, n); +} + +static inline u32 msg_bc_netid(struct tipc_msg *m) +{ + return msg_word(m, 4); +} + +static inline void msg_set_bc_netid(struct tipc_msg *m, u32 id) +{ + msg_set_word(m, 4, id); +} + +static inline u32 msg_link_selector(struct tipc_msg *m) +{ + return msg_bits(m, 4, 0, 1); +} + +static inline void msg_set_link_selector(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 4, 0, 1, (n & 1)); +} + +/* + * Word 5 + */ + +static inline u32 msg_session(struct tipc_msg *m) +{ + return msg_bits(m, 5, 16, 0xffff); +} + +static inline void msg_set_session(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 16, 0xffff, n); +} + +static inline u32 msg_probe(struct tipc_msg *m) +{ + return msg_bits(m, 5, 0, 1); +} + +static inline void msg_set_probe(struct tipc_msg *m, u32 val) +{ + msg_set_bits(m, 5, 0, 1, (val & 1)); +} + +static inline char msg_net_plane(struct tipc_msg *m) +{ + return msg_bits(m, 5, 1, 7) + 'A'; +} + +static inline void msg_set_net_plane(struct tipc_msg *m, char n) +{ + msg_set_bits(m, 5, 1, 7, (n - 'A')); +} + +static inline u32 msg_linkprio(struct tipc_msg *m) +{ + return msg_bits(m, 5, 4, 0x1f); +} + +static inline void msg_set_linkprio(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 4, 0x1f, n); +} + +static inline u32 msg_bearer_id(struct tipc_msg *m) +{ + return msg_bits(m, 5, 9, 0x7); +} + +static inline void msg_set_bearer_id(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 5, 9, 0x7, n); +} + +static inline u32 msg_redundant_link(struct tipc_msg *m) +{ + return msg_bits(m, 5, 12, 0x1); +} + +static inline void msg_set_redundant_link(struct tipc_msg *m) +{ + msg_set_bits(m, 5, 12, 0x1, 1); +} + +static inline void msg_clear_redundant_link(struct tipc_msg *m) +{ + msg_set_bits(m, 5, 12, 0x1, 0); +} + + +/* + * Word 9 + */ + +static inline u32 msg_msgcnt(struct tipc_msg *m) +{ + return msg_bits(m, 9, 16, 0xffff); +} + +static inline void msg_set_msgcnt(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, n); +} + +static inline u32 msg_bcast_tag(struct tipc_msg *m) +{ + return msg_bits(m, 9, 16, 0xffff); +} + +static inline void msg_set_bcast_tag(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, n); +} + +static inline u32 msg_max_pkt(struct tipc_msg *m) +{ + return (msg_bits(m, 9, 16, 0xffff) * 4); +} + +static inline void msg_set_max_pkt(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 16, 0xffff, (n / 4)); +} + +static inline u32 msg_link_tolerance(struct tipc_msg *m) +{ + return msg_bits(m, 9, 0, 0xffff); +} + +static inline void msg_set_link_tolerance(struct tipc_msg *m, u32 n) +{ + msg_set_bits(m, 9, 0, 0xffff, n); +} + +/* + * Routing table message data + */ + + +static inline u32 msg_remote_node(struct tipc_msg *m) +{ + return msg_word(m, msg_hdr_sz(m)/4); +} + +static inline void msg_set_remote_node(struct tipc_msg *m, u32 a) +{ + msg_set_word(m, msg_hdr_sz(m)/4, a); +} + +static inline int msg_dataoctet(struct tipc_msg *m, u32 pos) +{ + return(msg_data(m)[pos + 4] != 0); +} + +static inline void msg_set_dataoctet(struct tipc_msg *m, u32 pos) +{ + msg_data(m)[pos + 4] = 1; +} + +/* + * Segmentation message types + */ + +#define FIRST_FRAGMENT 0 +#define FRAGMENT 1 +#define LAST_FRAGMENT 2 + +/* + * Link management protocol message types + */ + +#define STATE_MSG 0 +#define RESET_MSG 1 +#define ACTIVATE_MSG 2 + +/* + * Changeover tunnel message types + */ +#define DUPLICATE_MSG 0 +#define ORIGINAL_MSG 1 + +/* + * Routing table message types + */ +#define EXT_ROUTING_TABLE 0 +#define LOCAL_ROUTING_TABLE 1 +#define SLAVE_ROUTING_TABLE 2 +#define ROUTE_ADDITION 3 +#define ROUTE_REMOVAL 4 + +/* + * Config protocol message types + */ + +#define DSC_REQ_MSG 0 +#define DSC_RESP_MSG 1 + +static inline u32 msg_tot_importance(struct tipc_msg *m) +{ + if (likely(msg_isdata(m))) { + if (likely(msg_orignode(m) == tipc_own_addr)) + return msg_importance(m); + return msg_importance(m) + 4; + } + if ((msg_user(m) == MSG_FRAGMENTER) && + (msg_type(m) == FIRST_FRAGMENT)) + return msg_importance(msg_get_wrapped(m)); + return msg_importance(m); +} + + +static inline void msg_init(struct tipc_msg *m, u32 user, u32 type, + u32 err, u32 hsize, u32 destnode) +{ + memset(m, 0, hsize); + msg_set_version(m); + msg_set_user(m, user); + msg_set_hdr_sz(m, hsize); + msg_set_size(m, hsize); + msg_set_prevnode(m, tipc_own_addr); + msg_set_type(m, type); + msg_set_errcode(m, err); + if (!msg_short(m)) { + msg_set_orignode(m, tipc_own_addr); + msg_set_destnode(m, destnode); + } +} + +/** + * msg_calc_data_size - determine total data size for message + */ + +static inline int msg_calc_data_size(struct iovec const *msg_sect, u32 num_sect) +{ + int dsz = 0; + int i; + + for (i = 0; i < num_sect; i++) + dsz += msg_sect[i].iov_len; + return dsz; +} + +/** + * msg_build - create message using specified header and data + * + * Note: Caller must not hold any locks in case copy_from_user() is interrupted! + * + * Returns message data size or errno + */ + +static inline int msg_build(struct tipc_msg *hdr, + struct iovec const *msg_sect, u32 num_sect, + int max_size, int usrmem, struct sk_buff** buf) +{ + int dsz, sz, hsz, pos, res, cnt; + + dsz = msg_calc_data_size(msg_sect, num_sect); + if (unlikely(dsz > TIPC_MAX_USER_MSG_SIZE)) { + *buf = NULL; + return -EINVAL; + } + + pos = hsz = msg_hdr_sz(hdr); + sz = hsz + dsz; + msg_set_size(hdr, sz); + if (unlikely(sz > max_size)) { + *buf = NULL; + return dsz; + } + + *buf = buf_acquire(sz); + if (!(*buf)) + return -ENOMEM; + memcpy((*buf)->data, (unchar *)hdr, hsz); + for (res = 1, cnt = 0; res && (cnt < num_sect); cnt++) { + if (likely(usrmem)) + res = !copy_from_user((*buf)->data + pos, + msg_sect[cnt].iov_base, + msg_sect[cnt].iov_len); + else + memcpy((*buf)->data + pos, msg_sect[cnt].iov_base, + msg_sect[cnt].iov_len); + pos += msg_sect[cnt].iov_len; + } + if (likely(res)) + return dsz; + + buf_discard(*buf); + *buf = NULL; + return -EFAULT; +} + + +struct tipc_media_addr; + +extern void msg_set_media_addr(struct tipc_msg *m, + struct tipc_media_addr *a); + +extern void msg_get_media_addr(struct tipc_msg *m, + struct tipc_media_addr *a); + + +#endif diff --git a/net/tipc/name_distr.c b/net/tipc/name_distr.c new file mode 100644 index 00000000000..41cbaf1a4a7 --- /dev/null +++ b/net/tipc/name_distr.c @@ -0,0 +1,309 @@ +/* + * net/tipc/name_distr.c: TIPC name distribution code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "cluster.h" +#include "dbg.h" +#include "link.h" +#include "msg.h" +#include "name_distr.h" + +#undef DBG_OUTPUT +#define DBG_OUTPUT NULL + +#define ITEM_SIZE sizeof(struct distr_item) + +/** + * struct distr_item - publication info distributed to other nodes + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @ref: publishing port reference + * @key: publication key + * + * ===> All fields are stored in network byte order. <=== + * + * First 3 fields identify (name or) name sequence being published. + * Reference field uniquely identifies port that published name sequence. + * Key field uniquely identifies publication, in the event a port has + * multiple publications of the same name sequence. + * + * Note: There is no field that identifies the publishing node because it is + * the same for all items contained within a publication message. + */ + +struct distr_item { + u32 type; + u32 lower; + u32 upper; + u32 ref; + u32 key; +}; + +/** + * List of externally visible publications by this node -- + * that is, all publications having scope > TIPC_NODE_SCOPE. + */ + +static LIST_HEAD(publ_root); +static u32 publ_cnt = 0; + +/** + * publ_to_item - add publication info to a publication message + */ + +static void publ_to_item(struct distr_item *i, struct publication *p) +{ + i->type = htonl(p->type); + i->lower = htonl(p->lower); + i->upper = htonl(p->upper); + i->ref = htonl(p->ref); + i->key = htonl(p->key); + dbg("publ_to_item: %u, %u, %u\n", p->type, p->lower, p->upper); +} + +/** + * named_prepare_buf - allocate & initialize a publication message + */ + +static struct sk_buff *named_prepare_buf(u32 type, u32 size, u32 dest) +{ + struct sk_buff *buf = buf_acquire(LONG_H_SIZE + size); + struct tipc_msg *msg; + + if (buf != NULL) { + msg = buf_msg(buf); + msg_init(msg, NAME_DISTRIBUTOR, type, TIPC_OK, + LONG_H_SIZE, dest); + msg_set_size(msg, LONG_H_SIZE + size); + } + return buf; +} + +/** + * named_publish - tell other nodes about a new publication by this node + */ + +void named_publish(struct publication *publ) +{ + struct sk_buff *buf; + struct distr_item *item; + + list_add(&publ->local_list, &publ_root); + publ_cnt++; + + buf = named_prepare_buf(PUBLICATION, ITEM_SIZE, 0); + if (!buf) { + warn("Memory squeeze; failed to distribute publication\n"); + return; + } + + item = (struct distr_item *)msg_data(buf_msg(buf)); + publ_to_item(item, publ); + dbg("named_withdraw: broadcasting publish msg\n"); + cluster_broadcast(buf); +} + +/** + * named_withdraw - tell other nodes about a withdrawn publication by this node + */ + +void named_withdraw(struct publication *publ) +{ + struct sk_buff *buf; + struct distr_item *item; + + list_del(&publ->local_list); + publ_cnt--; + + buf = named_prepare_buf(WITHDRAWAL, ITEM_SIZE, 0); + if (!buf) { + warn("Memory squeeze; failed to distribute withdrawal\n"); + return; + } + + item = (struct distr_item *)msg_data(buf_msg(buf)); + publ_to_item(item, publ); + dbg("named_withdraw: broadcasting withdraw msg\n"); + cluster_broadcast(buf); +} + +/** + * named_node_up - tell specified node about all publications by this node + */ + +void named_node_up(unsigned long node) +{ + struct publication *publ; + struct distr_item *item = 0; + struct sk_buff *buf = 0; + u32 left = 0; + u32 rest; + u32 max_item_buf; + + assert(in_own_cluster(node)); + read_lock_bh(&nametbl_lock); + max_item_buf = TIPC_MAX_USER_MSG_SIZE / ITEM_SIZE; + max_item_buf *= ITEM_SIZE; + rest = publ_cnt * ITEM_SIZE; + + list_for_each_entry(publ, &publ_root, local_list) { + if (!buf) { + left = (rest <= max_item_buf) ? rest : max_item_buf; + rest -= left; + buf = named_prepare_buf(PUBLICATION, left, node); + if (buf == NULL) { + warn("Memory Squeeze; could not send publication\n"); + goto exit; + } + item = (struct distr_item *)msg_data(buf_msg(buf)); + } + publ_to_item(item, publ); + item++; + left -= ITEM_SIZE; + if (!left) { + msg_set_link_selector(buf_msg(buf), node); + dbg("named_node_up: sending publish msg to " + "<%u.%u.%u>\n", tipc_zone(node), + tipc_cluster(node), tipc_node(node)); + link_send(buf, node, node); + buf = 0; + } + } +exit: + read_unlock_bh(&nametbl_lock); +} + +/** + * node_is_down - remove publication associated with a failed node + * + * Invoked for each publication issued by a newly failed node. + * Removes publication structure from name table & deletes it. + * In rare cases the link may have come back up again when this + * function is called, and we have two items representing the same + * publication. Nudge this item's key to distinguish it from the other. + * (Note: Publication's node subscription is already unsubscribed.) + */ + +static void node_is_down(struct publication *publ) +{ + struct publication *p; + write_lock_bh(&nametbl_lock); + dbg("node_is_down: withdrawing %u, %u, %u\n", + publ->type, publ->lower, publ->upper); + publ->key += 1222345; + p = nametbl_remove_publ(publ->type, publ->lower, + publ->node, publ->ref, publ->key); + assert(p == publ); + write_unlock_bh(&nametbl_lock); + if (publ) + kfree(publ); +} + +/** + * named_recv - process name table update message sent by another node + */ + +void named_recv(struct sk_buff *buf) +{ + struct publication *publ; + struct tipc_msg *msg = buf_msg(buf); + struct distr_item *item = (struct distr_item *)msg_data(msg); + u32 count = msg_data_sz(msg) / ITEM_SIZE; + + write_lock_bh(&nametbl_lock); + while (count--) { + if (msg_type(msg) == PUBLICATION) { + dbg("named_recv: got publication for %u, %u, %u\n", + ntohl(item->type), ntohl(item->lower), + ntohl(item->upper)); + publ = nametbl_insert_publ(ntohl(item->type), + ntohl(item->lower), + ntohl(item->upper), + TIPC_CLUSTER_SCOPE, + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); + if (publ) { + nodesub_subscribe(&publ->subscr, + msg_orignode(msg), + publ, + (net_ev_handler)node_is_down); + } + } else if (msg_type(msg) == WITHDRAWAL) { + dbg("named_recv: got withdrawl for %u, %u, %u\n", + ntohl(item->type), ntohl(item->lower), + ntohl(item->upper)); + publ = nametbl_remove_publ(ntohl(item->type), + ntohl(item->lower), + msg_orignode(msg), + ntohl(item->ref), + ntohl(item->key)); + + if (publ) { + nodesub_unsubscribe(&publ->subscr); + kfree(publ); + } + } else { + warn("named_recv: unknown msg\n"); + } + item++; + } + write_unlock_bh(&nametbl_lock); + buf_discard(buf); +} + +/** + * named_reinit - re-initialize local publication list + * + * This routine is called whenever TIPC networking is (re)enabled. + * All existing publications by this node that have "cluster" or "zone" scope + * are updated to reflect the node's current network address. + * (If the node's address is unchanged, the update loop terminates immediately.) + */ + +void named_reinit(void) +{ + struct publication *publ; + + write_lock_bh(&nametbl_lock); + list_for_each_entry(publ, &publ_root, local_list) { + if (publ->node == tipc_own_addr) + break; + publ->node = tipc_own_addr; + } + write_unlock_bh(&nametbl_lock); +} diff --git a/net/tipc/name_distr.h b/net/tipc/name_distr.h new file mode 100644 index 00000000000..a04bdeac84e --- /dev/null +++ b/net/tipc/name_distr.h @@ -0,0 +1,48 @@ +/* + * net/tipc/name_distr.h: Include file for TIPC name distribution code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NAME_DISTR_H +#define _TIPC_NAME_DISTR_H + +#include "name_table.h" + +void named_publish(struct publication *publ); +void named_withdraw(struct publication *publ); +void named_node_up(unsigned long node); +void named_recv(struct sk_buff *buf); +void named_reinit(void); + +#endif diff --git a/net/tipc/name_table.c b/net/tipc/name_table.c new file mode 100644 index 00000000000..972c83eb83b --- /dev/null +++ b/net/tipc/name_table.c @@ -0,0 +1,1079 @@ +/* + * net/tipc/name_table.c: TIPC name table code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "dbg.h" +#include "name_table.h" +#include "name_distr.h" +#include "addr.h" +#include "node_subscr.h" +#include "subscr.h" +#include "port.h" +#include "cluster.h" +#include "bcast.h" + +int tipc_nametbl_size = 1024; /* must be a power of 2 */ + +/** + * struct sub_seq - container for all published instances of a name sequence + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @node_list: circular list of matching publications with >= node scope + * @cluster_list: circular list of matching publications with >= cluster scope + * @zone_list: circular list of matching publications with >= zone scope + */ + +struct sub_seq { + u32 lower; + u32 upper; + struct publication *node_list; + struct publication *cluster_list; + struct publication *zone_list; +}; + +/** + * struct name_seq - container for all published instances of a name type + * @type: 32 bit 'type' value for name sequence + * @sseq: pointer to dynamically-sized array of sub-sequences of this 'type'; + * sub-sequences are sorted in ascending order + * @alloc: number of sub-sequences currently in array + * @first_free: upper bound of highest sub-sequence + 1 + * @ns_list: links to adjacent name sequences in hash chain + * @subscriptions: list of subscriptions for this 'type' + * @lock: spinlock controlling access to name sequence structure + */ + +struct name_seq { + u32 type; + struct sub_seq *sseqs; + u32 alloc; + u32 first_free; + struct hlist_node ns_list; + struct list_head subscriptions; + spinlock_t lock; +}; + +/** + * struct name_table - table containing all existing port name publications + * @types: pointer to fixed-sized array of name sequence lists, + * accessed via hashing on 'type'; name sequence lists are *not* sorted + * @local_publ_count: number of publications issued by this node + */ + +struct name_table { + struct hlist_head *types; + u32 local_publ_count; +}; + +struct name_table table = { NULL } ; +static atomic_t rsv_publ_ok = ATOMIC_INIT(0); +rwlock_t nametbl_lock = RW_LOCK_UNLOCKED; + + +static inline int hash(int x) +{ + return(x & (tipc_nametbl_size - 1)); +} + +/** + * publ_create - create a publication structure + */ + +static struct publication *publ_create(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port_ref, + u32 key) +{ + struct publication *publ = + (struct publication *)kmalloc(sizeof(*publ), GFP_ATOMIC); + if (publ == NULL) { + warn("Memory squeeze; failed to create publication\n"); + return 0; + } + + memset(publ, 0, sizeof(*publ)); + publ->type = type; + publ->lower = lower; + publ->upper = upper; + publ->scope = scope; + publ->node = node; + publ->ref = port_ref; + publ->key = key; + INIT_LIST_HEAD(&publ->local_list); + INIT_LIST_HEAD(&publ->pport_list); + INIT_LIST_HEAD(&publ->subscr.nodesub_list); + return publ; +} + +/** + * subseq_alloc - allocate a specified number of sub-sequence structures + */ + +struct sub_seq *subseq_alloc(u32 cnt) +{ + u32 sz = cnt * sizeof(struct sub_seq); + struct sub_seq *sseq = (struct sub_seq *)kmalloc(sz, GFP_ATOMIC); + + if (sseq) + memset(sseq, 0, sz); + return sseq; +} + +/** + * nameseq_create - create a name sequence structure for the specified 'type' + * + * Allocates a single sub-sequence structure and sets it to all 0's. + */ + +struct name_seq *nameseq_create(u32 type, struct hlist_head *seq_head) +{ + struct name_seq *nseq = + (struct name_seq *)kmalloc(sizeof(*nseq), GFP_ATOMIC); + struct sub_seq *sseq = subseq_alloc(1); + + if (!nseq || !sseq) { + warn("Memory squeeze; failed to create name sequence\n"); + kfree(nseq); + kfree(sseq); + return 0; + } + + memset(nseq, 0, sizeof(*nseq)); + nseq->lock = SPIN_LOCK_UNLOCKED; + nseq->type = type; + nseq->sseqs = sseq; + dbg("nameseq_create() nseq = %x type %u, ssseqs %x, ff: %u\n", + nseq, type, nseq->sseqs, nseq->first_free); + nseq->alloc = 1; + INIT_HLIST_NODE(&nseq->ns_list); + INIT_LIST_HEAD(&nseq->subscriptions); + hlist_add_head(&nseq->ns_list, seq_head); + return nseq; +} + +/** + * nameseq_find_subseq - find sub-sequence (if any) matching a name instance + * + * Very time-critical, so binary searches through sub-sequence array. + */ + +static inline struct sub_seq *nameseq_find_subseq(struct name_seq *nseq, + u32 instance) +{ + struct sub_seq *sseqs = nseq->sseqs; + int low = 0; + int high = nseq->first_free - 1; + int mid; + + while (low <= high) { + mid = (low + high) / 2; + if (instance < sseqs[mid].lower) + high = mid - 1; + else if (instance > sseqs[mid].upper) + low = mid + 1; + else + return &sseqs[mid]; + } + return 0; +} + +/** + * nameseq_locate_subseq - determine position of name instance in sub-sequence + * + * Returns index in sub-sequence array of the entry that contains the specified + * instance value; if no entry contains that value, returns the position + * where a new entry for it would be inserted in the array. + * + * Note: Similar to binary search code for locating a sub-sequence. + */ + +static u32 nameseq_locate_subseq(struct name_seq *nseq, u32 instance) +{ + struct sub_seq *sseqs = nseq->sseqs; + int low = 0; + int high = nseq->first_free - 1; + int mid; + + while (low <= high) { + mid = (low + high) / 2; + if (instance < sseqs[mid].lower) + high = mid - 1; + else if (instance > sseqs[mid].upper) + low = mid + 1; + else + return mid; + } + return low; +} + +/** + * nameseq_insert_publ - + */ + +struct publication *nameseq_insert_publ(struct name_seq *nseq, + u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port, u32 key) +{ + struct subscription *s; + struct subscription *st; + struct publication *publ; + struct sub_seq *sseq; + int created_subseq = 0; + + assert(nseq->first_free <= nseq->alloc); + sseq = nameseq_find_subseq(nseq, lower); + dbg("nameseq_ins: for seq %x,<%u,%u>, found sseq %x\n", + nseq, type, lower, sseq); + if (sseq) { + + /* Lower end overlaps existing entry => need an exact match */ + + if ((sseq->lower != lower) || (sseq->upper != upper)) { + warn("Overlapping publ <%u,%u,%u>\n", type, lower, upper); + return 0; + } + } else { + u32 inspos; + struct sub_seq *freesseq; + + /* Find where lower end should be inserted */ + + inspos = nameseq_locate_subseq(nseq, lower); + + /* Fail if upper end overlaps into an existing entry */ + + if ((inspos < nseq->first_free) && + (upper >= nseq->sseqs[inspos].lower)) { + warn("Overlapping publ <%u,%u,%u>\n", type, lower, upper); + return 0; + } + + /* Ensure there is space for new sub-sequence */ + + if (nseq->first_free == nseq->alloc) { + struct sub_seq *sseqs = nseq->sseqs; + nseq->sseqs = subseq_alloc(nseq->alloc * 2); + if (nseq->sseqs != NULL) { + memcpy(nseq->sseqs, sseqs, + nseq->alloc * sizeof (struct sub_seq)); + kfree(sseqs); + dbg("Allocated %u sseqs\n", nseq->alloc); + nseq->alloc *= 2; + } else { + warn("Memory squeeze; failed to create sub-sequence\n"); + return 0; + } + } + dbg("Have %u sseqs for type %u\n", nseq->alloc, type); + + /* Insert new sub-sequence */ + + dbg("ins in pos %u, ff = %u\n", inspos, nseq->first_free); + sseq = &nseq->sseqs[inspos]; + freesseq = &nseq->sseqs[nseq->first_free]; + memmove(sseq + 1, sseq, (freesseq - sseq) * sizeof (*sseq)); + memset(sseq, 0, sizeof (*sseq)); + nseq->first_free++; + sseq->lower = lower; + sseq->upper = upper; + created_subseq = 1; + } + dbg("inserting (%u %u %u) from %x:%u into sseq %x(%u,%u) of seq %x\n", + type, lower, upper, node, port, sseq, + sseq->lower, sseq->upper, nseq); + + /* Insert a publication: */ + + publ = publ_create(type, lower, upper, scope, node, port, key); + if (!publ) + return 0; + dbg("inserting publ %x, node=%x publ->node=%x, subscr->node=%x\n", + publ, node, publ->node, publ->subscr.node); + + if (!sseq->zone_list) + sseq->zone_list = publ->zone_list_next = publ; + else { + publ->zone_list_next = sseq->zone_list->zone_list_next; + sseq->zone_list->zone_list_next = publ; + } + + if (in_own_cluster(node)) { + if (!sseq->cluster_list) + sseq->cluster_list = publ->cluster_list_next = publ; + else { + publ->cluster_list_next = + sseq->cluster_list->cluster_list_next; + sseq->cluster_list->cluster_list_next = publ; + } + } + + if (node == tipc_own_addr) { + if (!sseq->node_list) + sseq->node_list = publ->node_list_next = publ; + else { + publ->node_list_next = sseq->node_list->node_list_next; + sseq->node_list->node_list_next = publ; + } + } + + /* + * Any subscriptions waiting for notification? + */ + list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { + dbg("calling report_overlap()\n"); + subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_PUBLISHED, + publ->ref, + publ->node, + created_subseq); + } + return publ; +} + +/** + * nameseq_remove_publ - + */ + +struct publication *nameseq_remove_publ(struct name_seq *nseq, u32 inst, + u32 node, u32 ref, u32 key) +{ + struct publication *publ; + struct publication *prev; + struct sub_seq *sseq = nameseq_find_subseq(nseq, inst); + struct sub_seq *free; + struct subscription *s, *st; + int removed_subseq = 0; + + assert(nseq); + + if (!sseq) { + int i; + + warn("Withdraw unknown <%u,%u>?\n", nseq->type, inst); + assert(nseq->sseqs); + dbg("Dumping subseqs %x for %x, alloc = %u,ff=%u\n", + nseq->sseqs, nseq, nseq->alloc, + nseq->first_free); + for (i = 0; i < nseq->first_free; i++) { + dbg("Subseq %u(%x): lower = %u,upper = %u\n", + i, &nseq->sseqs[i], nseq->sseqs[i].lower, + nseq->sseqs[i].upper); + } + return 0; + } + dbg("nameseq_remove: seq: %x, sseq %x, <%u,%u> key %u\n", + nseq, sseq, nseq->type, inst, key); + + prev = sseq->zone_list; + publ = sseq->zone_list->zone_list_next; + while ((publ->key != key) || (publ->ref != ref) || + (publ->node && (publ->node != node))) { + prev = publ; + publ = publ->zone_list_next; + assert(prev != sseq->zone_list); + } + if (publ != sseq->zone_list) + prev->zone_list_next = publ->zone_list_next; + else if (publ->zone_list_next != publ) { + prev->zone_list_next = publ->zone_list_next; + sseq->zone_list = publ->zone_list_next; + } else { + sseq->zone_list = 0; + } + + if (in_own_cluster(node)) { + prev = sseq->cluster_list; + publ = sseq->cluster_list->cluster_list_next; + while ((publ->key != key) || (publ->ref != ref) || + (publ->node && (publ->node != node))) { + prev = publ; + publ = publ->cluster_list_next; + assert(prev != sseq->cluster_list); + } + if (publ != sseq->cluster_list) + prev->cluster_list_next = publ->cluster_list_next; + else if (publ->cluster_list_next != publ) { + prev->cluster_list_next = publ->cluster_list_next; + sseq->cluster_list = publ->cluster_list_next; + } else { + sseq->cluster_list = 0; + } + } + + if (node == tipc_own_addr) { + prev = sseq->node_list; + publ = sseq->node_list->node_list_next; + while ((publ->key != key) || (publ->ref != ref) || + (publ->node && (publ->node != node))) { + prev = publ; + publ = publ->node_list_next; + assert(prev != sseq->node_list); + } + if (publ != sseq->node_list) + prev->node_list_next = publ->node_list_next; + else if (publ->node_list_next != publ) { + prev->node_list_next = publ->node_list_next; + sseq->node_list = publ->node_list_next; + } else { + sseq->node_list = 0; + } + } + assert(!publ->node || (publ->node == node)); + assert(publ->ref == ref); + assert(publ->key == key); + + /* + * Contract subseq list if no more publications: + */ + if (!sseq->node_list && !sseq->cluster_list && !sseq->zone_list) { + free = &nseq->sseqs[nseq->first_free--]; + memmove(sseq, sseq + 1, (free - (sseq + 1)) * sizeof (*sseq)); + removed_subseq = 1; + } + + /* + * Any subscriptions waiting ? + */ + list_for_each_entry_safe(s, st, &nseq->subscriptions, nameseq_list) { + subscr_report_overlap(s, + publ->lower, + publ->upper, + TIPC_WITHDRAWN, + publ->ref, + publ->node, + removed_subseq); + } + return publ; +} + +/** + * nameseq_subscribe: attach a subscription, and issue + * the prescribed number of events if there is any sub- + * sequence overlapping with the requested sequence + */ + +void nameseq_subscribe(struct name_seq *nseq, struct subscription *s) +{ + struct sub_seq *sseq = nseq->sseqs; + + list_add(&s->nameseq_list, &nseq->subscriptions); + + if (!sseq) + return; + + while (sseq != &nseq->sseqs[nseq->first_free]) { + struct publication *zl = sseq->zone_list; + if (zl && subscr_overlap(s,sseq->lower,sseq->upper)) { + struct publication *crs = zl; + int must_report = 1; + + do { + subscr_report_overlap(s, + sseq->lower, + sseq->upper, + TIPC_PUBLISHED, + crs->ref, + crs->node, + must_report); + must_report = 0; + crs = crs->zone_list_next; + } while (crs != zl); + } + sseq++; + } +} + +static struct name_seq *nametbl_find_seq(u32 type) +{ + struct hlist_head *seq_head; + struct hlist_node *seq_node; + struct name_seq *ns; + + dbg("find_seq %u,(%u,0x%x) table = %p, hash[type] = %u\n", + type, ntohl(type), type, table.types, hash(type)); + + seq_head = &table.types[hash(type)]; + hlist_for_each_entry(ns, seq_node, seq_head, ns_list) { + if (ns->type == type) { + dbg("found %x\n", ns); + return ns; + } + } + + return 0; +}; + +struct publication *nametbl_insert_publ(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 port, u32 key) +{ + struct name_seq *seq = nametbl_find_seq(type); + + dbg("ins_publ: <%u,%x,%x> found %x\n", type, lower, upper, seq); + if (lower > upper) { + warn("Failed to publish illegal <%u,%u,%u>\n", + type, lower, upper); + return 0; + } + + dbg("Publishing <%u,%u,%u> from %x\n", type, lower, upper, node); + if (!seq) { + seq = nameseq_create(type, &table.types[hash(type)]); + dbg("nametbl_insert_publ: created %x\n", seq); + } + if (!seq) + return 0; + + assert(seq->type == type); + return nameseq_insert_publ(seq, type, lower, upper, + scope, node, port, key); +} + +struct publication *nametbl_remove_publ(u32 type, u32 lower, + u32 node, u32 ref, u32 key) +{ + struct publication *publ; + struct name_seq *seq = nametbl_find_seq(type); + + if (!seq) + return 0; + + dbg("Withdrawing <%u,%u> from %x\n", type, lower, node); + publ = nameseq_remove_publ(seq, lower, node, ref, key); + + if (!seq->first_free && list_empty(&seq->subscriptions)) { + hlist_del_init(&seq->ns_list); + kfree(seq->sseqs); + kfree(seq); + } + return publ; +} + +/* + * nametbl_translate(): Translate tipc_name -> tipc_portid. + * Very time-critical. + * + * Note: on entry 'destnode' is the search domain used during translation; + * on exit it passes back the node address of the matching port (if any) + */ + +u32 nametbl_translate(u32 type, u32 instance, u32 *destnode) +{ + struct sub_seq *sseq; + struct publication *publ = 0; + struct name_seq *seq; + u32 ref; + + if (!in_scope(*destnode, tipc_own_addr)) + return 0; + + read_lock_bh(&nametbl_lock); + seq = nametbl_find_seq(type); + if (unlikely(!seq)) + goto not_found; + sseq = nameseq_find_subseq(seq, instance); + if (unlikely(!sseq)) + goto not_found; + spin_lock_bh(&seq->lock); + + /* Closest-First Algorithm: */ + if (likely(!*destnode)) { + publ = sseq->node_list; + if (publ) { + sseq->node_list = publ->node_list_next; +found: + ref = publ->ref; + *destnode = publ->node; + spin_unlock_bh(&seq->lock); + read_unlock_bh(&nametbl_lock); + return ref; + } + publ = sseq->cluster_list; + if (publ) { + sseq->cluster_list = publ->cluster_list_next; + goto found; + } + publ = sseq->zone_list; + if (publ) { + sseq->zone_list = publ->zone_list_next; + goto found; + } + } + + /* Round-Robin Algorithm: */ + else if (*destnode == tipc_own_addr) { + publ = sseq->node_list; + if (publ) { + sseq->node_list = publ->node_list_next; + goto found; + } + } else if (in_own_cluster(*destnode)) { + publ = sseq->cluster_list; + if (publ) { + sseq->cluster_list = publ->cluster_list_next; + goto found; + } + } else { + publ = sseq->zone_list; + if (publ) { + sseq->zone_list = publ->zone_list_next; + goto found; + } + } + spin_unlock_bh(&seq->lock); +not_found: + *destnode = 0; + read_unlock_bh(&nametbl_lock); + return 0; +} + +/** + * nametbl_mc_translate - find multicast destinations + * + * Creates list of all local ports that overlap the given multicast address; + * also determines if any off-node ports overlap. + * + * Note: Publications with a scope narrower than 'limit' are ignored. + * (i.e. local node-scope publications mustn't receive messages arriving + * from another node, even if the multcast link brought it here) + * + * Returns non-zero if any off-node ports overlap + */ + +int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, + struct port_list *dports) +{ + struct name_seq *seq; + struct sub_seq *sseq; + struct sub_seq *sseq_stop; + int res = 0; + + read_lock_bh(&nametbl_lock); + seq = nametbl_find_seq(type); + if (!seq) + goto exit; + + spin_lock_bh(&seq->lock); + + sseq = seq->sseqs + nameseq_locate_subseq(seq, lower); + sseq_stop = seq->sseqs + seq->first_free; + for (; sseq != sseq_stop; sseq++) { + struct publication *publ; + + if (sseq->lower > upper) + break; + publ = sseq->cluster_list; + if (publ && (publ->scope <= limit)) + do { + if (publ->node == tipc_own_addr) + port_list_add(dports, publ->ref); + else + res = 1; + publ = publ->cluster_list_next; + } while (publ != sseq->cluster_list); + } + + spin_unlock_bh(&seq->lock); +exit: + read_unlock_bh(&nametbl_lock); + return res; +} + +/** + * nametbl_publish_rsv - publish port name using a reserved name type + */ + +int nametbl_publish_rsv(u32 ref, unsigned int scope, + struct tipc_name_seq const *seq) +{ + int res; + + atomic_inc(&rsv_publ_ok); + res = tipc_publish(ref, scope, seq); + atomic_dec(&rsv_publ_ok); + return res; +} + +/** + * nametbl_publish - add name publication to network name tables + */ + +struct publication *nametbl_publish(u32 type, u32 lower, u32 upper, + u32 scope, u32 port_ref, u32 key) +{ + struct publication *publ; + + if (table.local_publ_count >= tipc_max_publications) { + warn("Failed publish: max %u local publication\n", + tipc_max_publications); + return 0; + } + if ((type < TIPC_RESERVED_TYPES) && !atomic_read(&rsv_publ_ok)) { + warn("Failed to publish reserved name <%u,%u,%u>\n", + type, lower, upper); + return 0; + } + + write_lock_bh(&nametbl_lock); + table.local_publ_count++; + publ = nametbl_insert_publ(type, lower, upper, scope, + tipc_own_addr, port_ref, key); + if (publ && (scope != TIPC_NODE_SCOPE)) { + named_publish(publ); + } + write_unlock_bh(&nametbl_lock); + return publ; +} + +/** + * nametbl_withdraw - withdraw name publication from network name tables + */ + +int nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key) +{ + struct publication *publ; + + dbg("nametbl_withdraw:<%d,%d,%d>\n", type, lower, key); + write_lock_bh(&nametbl_lock); + publ = nametbl_remove_publ(type, lower, tipc_own_addr, ref, key); + if (publ) { + table.local_publ_count--; + if (publ->scope != TIPC_NODE_SCOPE) + named_withdraw(publ); + write_unlock_bh(&nametbl_lock); + list_del_init(&publ->pport_list); + kfree(publ); + return 1; + } + write_unlock_bh(&nametbl_lock); + return 0; +} + +/** + * nametbl_subscribe - add a subscription object to the name table + */ + +void +nametbl_subscribe(struct subscription *s) +{ + u32 type = s->seq.type; + struct name_seq *seq; + + write_lock_bh(&nametbl_lock); + seq = nametbl_find_seq(type); + if (!seq) { + seq = nameseq_create(type, &table.types[hash(type)]); + } + if (seq){ + spin_lock_bh(&seq->lock); + dbg("nametbl_subscribe:found %x for <%u,%u,%u>\n", + seq, type, s->seq.lower, s->seq.upper); + assert(seq->type == type); + nameseq_subscribe(seq, s); + spin_unlock_bh(&seq->lock); + } + write_unlock_bh(&nametbl_lock); +} + +/** + * nametbl_unsubscribe - remove a subscription object from name table + */ + +void +nametbl_unsubscribe(struct subscription *s) +{ + struct name_seq *seq; + + write_lock_bh(&nametbl_lock); + seq = nametbl_find_seq(s->seq.type); + if (seq != NULL){ + spin_lock_bh(&seq->lock); + list_del_init(&s->nameseq_list); + spin_unlock_bh(&seq->lock); + if ((seq->first_free == 0) && list_empty(&seq->subscriptions)) { + hlist_del_init(&seq->ns_list); + kfree(seq->sseqs); + kfree(seq); + } + } + write_unlock_bh(&nametbl_lock); +} + + +/** + * subseq_list: print specified sub-sequence contents into the given buffer + */ + +static void subseq_list(struct sub_seq *sseq, struct print_buf *buf, u32 depth, + u32 index) +{ + char portIdStr[27]; + char *scopeStr; + struct publication *publ = sseq->zone_list; + + tipc_printf(buf, "%-10u %-10u ", sseq->lower, sseq->upper); + + if (depth == 2 || !publ) { + tipc_printf(buf, "\n"); + return; + } + + do { + sprintf (portIdStr, "<%u.%u.%u:%u>", + tipc_zone(publ->node), tipc_cluster(publ->node), + tipc_node(publ->node), publ->ref); + tipc_printf(buf, "%-26s ", portIdStr); + if (depth > 3) { + if (publ->node != tipc_own_addr) + scopeStr = ""; + else if (publ->scope == TIPC_NODE_SCOPE) + scopeStr = "node"; + else if (publ->scope == TIPC_CLUSTER_SCOPE) + scopeStr = "cluster"; + else + scopeStr = "zone"; + tipc_printf(buf, "%-10u %s", publ->key, scopeStr); + } + + publ = publ->zone_list_next; + if (publ == sseq->zone_list) + break; + + tipc_printf(buf, "\n%33s", " "); + } while (1); + + tipc_printf(buf, "\n"); +} + +/** + * nameseq_list: print specified name sequence contents into the given buffer + */ + +static void nameseq_list(struct name_seq *seq, struct print_buf *buf, u32 depth, + u32 type, u32 lowbound, u32 upbound, u32 index) +{ + struct sub_seq *sseq; + char typearea[11]; + + sprintf(typearea, "%-10u", seq->type); + + if (depth == 1) { + tipc_printf(buf, "%s\n", typearea); + return; + } + + for (sseq = seq->sseqs; sseq != &seq->sseqs[seq->first_free]; sseq++) { + if ((lowbound <= sseq->upper) && (upbound >= sseq->lower)) { + tipc_printf(buf, "%s ", typearea); + subseq_list(sseq, buf, depth, index); + sprintf(typearea, "%10s", " "); + } + } +} + +/** + * nametbl_header - print name table header into the given buffer + */ + +static void nametbl_header(struct print_buf *buf, u32 depth) +{ + tipc_printf(buf, "Type "); + + if (depth > 1) + tipc_printf(buf, "Lower Upper "); + if (depth > 2) + tipc_printf(buf, "Port Identity "); + if (depth > 3) + tipc_printf(buf, "Publication"); + + tipc_printf(buf, "\n-----------"); + + if (depth > 1) + tipc_printf(buf, "--------------------- "); + if (depth > 2) + tipc_printf(buf, "-------------------------- "); + if (depth > 3) + tipc_printf(buf, "------------------"); + + tipc_printf(buf, "\n"); +} + +/** + * nametbl_list - print specified name table contents into the given buffer + */ + +static void nametbl_list(struct print_buf *buf, u32 depth_info, + u32 type, u32 lowbound, u32 upbound) +{ + struct hlist_head *seq_head; + struct hlist_node *seq_node; + struct name_seq *seq; + int all_types; + u32 depth; + u32 i; + + all_types = (depth_info & TIPC_NTQ_ALLTYPES); + depth = (depth_info & ~TIPC_NTQ_ALLTYPES); + + if (depth == 0) + return; + + if (all_types) { + /* display all entries in name table to specified depth */ + nametbl_header(buf, depth); + lowbound = 0; + upbound = ~0; + for (i = 0; i < tipc_nametbl_size; i++) { + seq_head = &table.types[i]; + hlist_for_each_entry(seq, seq_node, seq_head, ns_list) { + nameseq_list(seq, buf, depth, seq->type, + lowbound, upbound, i); + } + } + } else { + /* display only the sequence that matches the specified type */ + if (upbound < lowbound) { + tipc_printf(buf, "invalid name sequence specified\n"); + return; + } + nametbl_header(buf, depth); + i = hash(type); + seq_head = &table.types[i]; + hlist_for_each_entry(seq, seq_node, seq_head, ns_list) { + if (seq->type == type) { + nameseq_list(seq, buf, depth, type, + lowbound, upbound, i); + break; + } + } + } +} + +void nametbl_print(struct print_buf *buf, const char *str) +{ + tipc_printf(buf, str); + read_lock_bh(&nametbl_lock); + nametbl_list(buf, 0, 0, 0, 0); + read_unlock_bh(&nametbl_lock); +} + +#define MAX_NAME_TBL_QUERY 32768 + +struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space) +{ + struct sk_buff *buf; + struct tipc_name_table_query *argv; + struct tlv_desc *rep_tlv; + struct print_buf b; + int str_len; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NAME_TBL_QUERY)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + buf = cfg_reply_alloc(TLV_SPACE(MAX_NAME_TBL_QUERY)); + if (!buf) + return NULL; + + rep_tlv = (struct tlv_desc *)buf->data; + printbuf_init(&b, TLV_DATA(rep_tlv), MAX_NAME_TBL_QUERY); + argv = (struct tipc_name_table_query *)TLV_DATA(req_tlv_area); + read_lock_bh(&nametbl_lock); + nametbl_list(&b, ntohl(argv->depth), ntohl(argv->type), + ntohl(argv->lowbound), ntohl(argv->upbound)); + read_unlock_bh(&nametbl_lock); + str_len = printbuf_validate(&b); + + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +void nametbl_dump(void) +{ + nametbl_list(CONS, 0, 0, 0, 0); +} + +int nametbl_init(void) +{ + int array_size = sizeof(struct hlist_head) * tipc_nametbl_size; + + table.types = (struct hlist_head *)kmalloc(array_size, GFP_ATOMIC); + if (!table.types) + return -ENOMEM; + + write_lock_bh(&nametbl_lock); + memset(table.types, 0, array_size); + table.local_publ_count = 0; + write_unlock_bh(&nametbl_lock); + return 0; +} + +void nametbl_stop(void) +{ + struct hlist_head *seq_head; + struct hlist_node *seq_node; + struct hlist_node *tmp; + struct name_seq *seq; + u32 i; + + if (!table.types) + return; + + write_lock_bh(&nametbl_lock); + for (i = 0; i < tipc_nametbl_size; i++) { + seq_head = &table.types[i]; + hlist_for_each_entry_safe(seq, seq_node, tmp, seq_head, ns_list) { + struct sub_seq *sseq = seq->sseqs; + + for (; sseq != &seq->sseqs[seq->first_free]; sseq++) { + struct publication *publ = sseq->zone_list; + assert(publ); + do { + struct publication *next = + publ->zone_list_next; + kfree(publ); + publ = next; + } + while (publ != sseq->zone_list); + } + } + } + kfree(table.types); + table.types = NULL; + write_unlock_bh(&nametbl_lock); +} diff --git a/net/tipc/name_table.h b/net/tipc/name_table.h new file mode 100644 index 00000000000..f82693384f6 --- /dev/null +++ b/net/tipc/name_table.h @@ -0,0 +1,108 @@ +/* + * net/tipc/name_table.h: Include file for TIPC name table code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NAME_TABLE_H +#define _TIPC_NAME_TABLE_H + +#include "node_subscr.h" + +struct subscription; +struct port_list; + +/* + * TIPC name types reserved for internal TIPC use (both current and planned) + */ + +#define TIPC_ZM_SRV 3 /* zone master service name type */ + + +/** + * struct publication - info about a published (name or) name sequence + * @type: name sequence type + * @lower: name sequence lower bound + * @upper: name sequence upper bound + * @scope: scope of publication + * @node: network address of publishing port's node + * @ref: publishing port + * @key: publication key + * @subscr: subscription to "node down" event (for off-node publications only) + * @local_list: adjacent entries in list of publications made by this node + * @pport_list: adjacent entries in list of publications made by this port + * @node_list: next matching name seq publication with >= node scope + * @cluster_list: next matching name seq publication with >= cluster scope + * @zone_list: next matching name seq publication with >= zone scope + * + * Note that the node list, cluster list, and zone list are circular lists. + */ + +struct publication { + u32 type; + u32 lower; + u32 upper; + u32 scope; + u32 node; + u32 ref; + u32 key; + struct node_subscr subscr; + struct list_head local_list; + struct list_head pport_list; + struct publication *node_list_next; + struct publication *cluster_list_next; + struct publication *zone_list_next; +}; + + +extern rwlock_t nametbl_lock; + +struct sk_buff *nametbl_get(const void *req_tlv_area, int req_tlv_space); +u32 nametbl_translate(u32 type, u32 instance, u32 *node); +int nametbl_mc_translate(u32 type, u32 lower, u32 upper, u32 limit, + struct port_list *dports); +int nametbl_publish_rsv(u32 ref, unsigned int scope, + struct tipc_name_seq const *seq); +struct publication *nametbl_publish(u32 type, u32 lower, u32 upper, + u32 scope, u32 port_ref, u32 key); +int nametbl_withdraw(u32 type, u32 lower, u32 ref, u32 key); +struct publication *nametbl_insert_publ(u32 type, u32 lower, u32 upper, + u32 scope, u32 node, u32 ref, u32 key); +struct publication *nametbl_remove_publ(u32 type, u32 lower, + u32 node, u32 ref, u32 key); +void nametbl_subscribe(struct subscription *s); +void nametbl_unsubscribe(struct subscription *s); +int nametbl_init(void); +void nametbl_stop(void); + +#endif diff --git a/net/tipc/net.c b/net/tipc/net.c new file mode 100644 index 00000000000..6826b493c1d --- /dev/null +++ b/net/tipc/net.c @@ -0,0 +1,311 @@ +/* + * net/tipc/net.c: TIPC network routing code + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "bearer.h" +#include "net.h" +#include "zone.h" +#include "addr.h" +#include "name_table.h" +#include "name_distr.h" +#include "subscr.h" +#include "link.h" +#include "msg.h" +#include "port.h" +#include "bcast.h" +#include "discover.h" +#include "config.h" + +/* + * The TIPC locking policy is designed to ensure a very fine locking + * granularity, permitting complete parallel access to individual + * port and node/link instances. The code consists of three major + * locking domains, each protected with their own disjunct set of locks. + * + * 1: The routing hierarchy. + * Comprises the structures 'zone', 'cluster', 'node', 'link' + * and 'bearer'. The whole hierarchy is protected by a big + * read/write lock, net_lock, to enssure that nothing is added + * or removed while code is accessing any of these structures. + * This layer must not be called from the two others while they + * hold any of their own locks. + * Neither must it itself do any upcalls to the other two before + * it has released net_lock and other protective locks. + * + * Within the net_lock domain there are two sub-domains;'node' and + * 'bearer', where local write operations are permitted, + * provided that those are protected by individual spin_locks + * per instance. Code holding net_lock(read) and a node spin_lock + * is permitted to poke around in both the node itself and its + * subordinate links. I.e, it can update link counters and queues, + * change link state, send protocol messages, and alter the + * "active_links" array in the node; but it can _not_ remove a link + * or a node from the overall structure. + * Correspondingly, individual bearers may change status within a + * net_lock(read), protected by an individual spin_lock ber bearer + * instance, but it needs net_lock(write) to remove/add any bearers. + * + * + * 2: The transport level of the protocol. + * This consists of the structures port, (and its user level + * representations, such as user_port and tipc_sock), reference and + * tipc_user (port.c, reg.c, socket.c). + * + * This layer has four different locks: + * - The tipc_port spin_lock. This is protecting each port instance + * from parallel data access and removal. Since we can not place + * this lock in the port itself, it has been placed in the + * corresponding reference table entry, which has the same life + * cycle as the module. This entry is difficult to access from + * outside the TIPC core, however, so a pointer to the lock has + * been added in the port instance, -to be used for unlocking + * only. + * - A read/write lock to protect the reference table itself (teg.c). + * (Nobody is using read-only access to this, so it can just as + * well be changed to a spin_lock) + * - A spin lock to protect the registry of kernel/driver users (reg.c) + * - A global spin_lock (port_lock), which only task is to ensure + * consistency where more than one port is involved in an operation, + * i.e., whe a port is part of a linked list of ports. + * There are two such lists; 'port_list', which is used for management, + * and 'wait_list', which is used to queue ports during congestion. + * + * 3: The name table (name_table.c, name_distr.c, subscription.c) + * - There is one big read/write-lock (nametbl_lock) protecting the + * overall name table structure. Nothing must be added/removed to + * this structure without holding write access to it. + * - There is one local spin_lock per sub_sequence, which can be seen + * as a sub-domain to the nametbl_lock domain. It is used only + * for translation operations, and is needed because a translation + * steps the root of the 'publication' linked list between each lookup. + * This is always used within the scope of a nametbl_lock(read). + * - A local spin_lock protecting the queue of subscriber events. +*/ + +rwlock_t net_lock = RW_LOCK_UNLOCKED; +struct network net = { 0 }; + +struct node *net_select_remote_node(u32 addr, u32 ref) +{ + return zone_select_remote_node(net.zones[tipc_zone(addr)], addr, ref); +} + +u32 net_select_router(u32 addr, u32 ref) +{ + return zone_select_router(net.zones[tipc_zone(addr)], addr, ref); +} + + +u32 net_next_node(u32 a) +{ + if (net.zones[tipc_zone(a)]) + return zone_next_node(a); + return 0; +} + +void net_remove_as_router(u32 router) +{ + u32 z_num; + + for (z_num = 1; z_num <= tipc_max_zones; z_num++) { + if (!net.zones[z_num]) + continue; + zone_remove_as_router(net.zones[z_num], router); + } +} + +void net_send_external_routes(u32 dest) +{ + u32 z_num; + + for (z_num = 1; z_num <= tipc_max_zones; z_num++) { + if (net.zones[z_num]) + zone_send_external_routes(net.zones[z_num], dest); + } +} + +int net_init(void) +{ + u32 sz = sizeof(struct _zone *) * (tipc_max_zones + 1); + + memset(&net, 0, sizeof(net)); + net.zones = (struct _zone **)kmalloc(sz, GFP_ATOMIC); + if (!net.zones) { + return -ENOMEM; + } + memset(net.zones, 0, sz); + return TIPC_OK; +} + +void net_stop(void) +{ + u32 z_num; + + if (!net.zones) + return; + + for (z_num = 1; z_num <= tipc_max_zones; z_num++) { + zone_delete(net.zones[z_num]); + } + kfree(net.zones); + net.zones = 0; +} + +static void net_route_named_msg(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + u32 dnode; + u32 dport; + + if (!msg_named(msg)) { + msg_dbg(msg, "net->drop_nam:"); + buf_discard(buf); + return; + } + + dnode = addr_domain(msg_lookup_scope(msg)); + dport = nametbl_translate(msg_nametype(msg), msg_nameinst(msg), &dnode); + dbg("net->lookup<%u,%u>-><%u,%x>\n", + msg_nametype(msg), msg_nameinst(msg), dport, dnode); + if (dport) { + msg_set_destnode(msg, dnode); + msg_set_destport(msg, dport); + net_route_msg(buf); + return; + } + msg_dbg(msg, "net->rej:NO NAME: "); + tipc_reject_msg(buf, TIPC_ERR_NO_NAME); +} + +void net_route_msg(struct sk_buff *buf) +{ + struct tipc_msg *msg; + u32 dnode; + + if (!buf) + return; + msg = buf_msg(buf); + + msg_incr_reroute_cnt(msg); + if (msg_reroute_cnt(msg) > 6) { + if (msg_errcode(msg)) { + msg_dbg(msg, "NET>DISC>:"); + buf_discard(buf); + } else { + msg_dbg(msg, "NET>REJ>:"); + tipc_reject_msg(buf, msg_destport(msg) ? + TIPC_ERR_NO_PORT : TIPC_ERR_NO_NAME); + } + return; + } + + msg_dbg(msg, "net->rout: "); + + /* Handle message for this node */ + dnode = msg_short(msg) ? tipc_own_addr : msg_destnode(msg); + if (in_scope(dnode, tipc_own_addr)) { + if (msg_isdata(msg)) { + if (msg_mcast(msg)) + port_recv_mcast(buf, NULL); + else if (msg_destport(msg)) + port_recv_msg(buf); + else + net_route_named_msg(buf); + return; + } + switch (msg_user(msg)) { + case ROUTE_DISTRIBUTOR: + cluster_recv_routing_table(buf); + break; + case NAME_DISTRIBUTOR: + named_recv(buf); + break; + case CONN_MANAGER: + port_recv_proto_msg(buf); + break; + default: + msg_dbg(msg,"DROP/NET/<REC<"); + buf_discard(buf); + } + return; + } + + /* Handle message for another node */ + msg_dbg(msg, "NET>SEND>: "); + link_send(buf, dnode, msg_link_selector(msg)); +} + +int tipc_start_net(void) +{ + char addr_string[16]; + int res; + + if (tipc_mode != TIPC_NODE_MODE) + return -ENOPROTOOPT; + + tipc_mode = TIPC_NET_MODE; + named_reinit(); + port_reinit(); + + if ((res = bearer_init()) || + (res = net_init()) || + (res = cluster_init()) || + (res = bclink_init())) { + return res; + } + subscr_stop(); + cfg_stop(); + k_signal((Handler)subscr_start, 0); + k_signal((Handler)cfg_init, 0); + info("Started in network mode\n"); + info("Own node address %s, network identity %u\n", + addr_string_fill(addr_string, tipc_own_addr), tipc_net_id); + return TIPC_OK; +} + +void tipc_stop_net(void) +{ + if (tipc_mode != TIPC_NET_MODE) + return; + write_lock_bh(&net_lock); + bearer_stop(); + tipc_mode = TIPC_NODE_MODE; + bclink_stop(); + net_stop(); + write_unlock_bh(&net_lock); + info("Left network mode \n"); +} + diff --git a/net/tipc/net.h b/net/tipc/net.h new file mode 100644 index 00000000000..948c6d42102 --- /dev/null +++ b/net/tipc/net.h @@ -0,0 +1,66 @@ +/* + * net/tipc/net.h: Include file for TIPC network routing code + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NET_H +#define _TIPC_NET_H + +struct _zone; + +/** + * struct network - TIPC network structure + * @zones: array of pointers to all zones within network + */ + +struct network { + struct _zone **zones; +}; + + +extern struct network net; +extern rwlock_t net_lock; + +int net_init(void); +void net_stop(void); +void net_remove_as_router(u32 router); +void net_send_external_routes(u32 dest); +void net_route_msg(struct sk_buff *buf); +struct node *net_select_remote_node(u32 addr, u32 ref); +u32 net_select_router(u32 addr, u32 ref); + +int tipc_start_net(void); +void tipc_stop_net(void); + +#endif diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c new file mode 100644 index 00000000000..19b3f402253 --- /dev/null +++ b/net/tipc/netlink.c @@ -0,0 +1,112 @@ +/* + * net/tipc/netlink.c: TIPC configuration handling + * + * Copyright (c) 2005-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include <net/genetlink.h> + +static int handle_cmd(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *rep_buf; + struct nlmsghdr *rep_nlh; + struct nlmsghdr *req_nlh = info->nlhdr; + struct tipc_genlmsghdr *req_userhdr = info->userhdr; + int hdr_space = NLMSG_SPACE(GENL_HDRLEN + TIPC_GENL_HDRLEN); + + if ((req_userhdr->cmd & 0xC000) && (!capable(CAP_NET_ADMIN))) + rep_buf = cfg_reply_error_string(TIPC_CFG_NOT_NET_ADMIN); + else + rep_buf = cfg_do_cmd(req_userhdr->dest, + req_userhdr->cmd, + NLMSG_DATA(req_nlh) + GENL_HDRLEN + TIPC_GENL_HDRLEN, + NLMSG_PAYLOAD(req_nlh, GENL_HDRLEN + TIPC_GENL_HDRLEN), + hdr_space); + + if (rep_buf) { + skb_push(rep_buf, hdr_space); + rep_nlh = (struct nlmsghdr *)rep_buf->data; + memcpy(rep_nlh, req_nlh, hdr_space); + rep_nlh->nlmsg_len = rep_buf->len; + genlmsg_unicast(rep_buf, req_nlh->nlmsg_pid); + } + + return 0; +} + +static struct genl_family family = { + .id = GENL_ID_GENERATE, + .name = TIPC_GENL_NAME, + .version = TIPC_GENL_VERSION, + .hdrsize = TIPC_GENL_HDRLEN, + .maxattr = 0, +}; + +static struct genl_ops ops = { + .cmd = TIPC_GENL_CMD, + .doit = handle_cmd, +}; + +static int family_registered = 0; + +int netlink_start(void) +{ + + + if (genl_register_family(&family)) + goto err; + + family_registered = 1; + + if (genl_register_ops(&family, &ops)) + goto err_unregister; + + return 0; + + err_unregister: + genl_unregister_family(&family); + family_registered = 0; + err: + err("Failed to register netlink interface\n"); + return -EFAULT; +} + +void netlink_stop(void) +{ + if (family_registered) { + genl_unregister_family(&family); + family_registered = 0; + } +} diff --git a/net/tipc/node.c b/net/tipc/node.c new file mode 100644 index 00000000000..05688d01138 --- /dev/null +++ b/net/tipc/node.c @@ -0,0 +1,679 @@ +/* + * net/tipc/node.c: TIPC node management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "node.h" +#include "cluster.h" +#include "net.h" +#include "addr.h" +#include "node_subscr.h" +#include "link.h" +#include "port.h" +#include "bearer.h" +#include "name_distr.h" +#include "net.h" + +void node_print(struct print_buf *buf, struct node *n_ptr, char *str); +static void node_lost_contact(struct node *n_ptr); +static void node_established_contact(struct node *n_ptr); + +struct node *nodes = NULL; /* sorted list of nodes within cluster */ + +u32 tipc_own_tag = 0; + +struct node *node_create(u32 addr) +{ + struct cluster *c_ptr; + struct node *n_ptr; + struct node **curr_node; + + n_ptr = kmalloc(sizeof(*n_ptr),GFP_ATOMIC); + if (n_ptr != NULL) { + memset(n_ptr, 0, sizeof(*n_ptr)); + n_ptr->addr = addr; + n_ptr->lock = SPIN_LOCK_UNLOCKED; + INIT_LIST_HEAD(&n_ptr->nsub); + + c_ptr = cluster_find(addr); + if (c_ptr == NULL) + c_ptr = cluster_create(addr); + if (c_ptr != NULL) { + n_ptr->owner = c_ptr; + cluster_attach_node(c_ptr, n_ptr); + n_ptr->last_router = -1; + + /* Insert node into ordered list */ + for (curr_node = &nodes; *curr_node; + curr_node = &(*curr_node)->next) { + if (addr < (*curr_node)->addr) { + n_ptr->next = *curr_node; + break; + } + } + (*curr_node) = n_ptr; + } else { + kfree(n_ptr); + n_ptr = NULL; + } + } + return n_ptr; +} + +void node_delete(struct node *n_ptr) +{ + if (!n_ptr) + return; + +#if 0 + /* Not needed because links are already deleted via bearer_stop() */ + + u32 l_num; + + for (l_num = 0; l_num < MAX_BEARERS; l_num++) { + link_delete(n_ptr->links[l_num]); + } +#endif + + dbg("node %x deleted\n", n_ptr->addr); + kfree(n_ptr); +} + + +/** + * node_link_up - handle addition of link + * + * Link becomes active (alone or shared) or standby, depending on its priority. + */ + +void node_link_up(struct node *n_ptr, struct link *l_ptr) +{ + struct link **active = &n_ptr->active_links[0]; + + info("Established link <%s> on network plane %c\n", + l_ptr->name, l_ptr->b_ptr->net_plane); + + if (!active[0]) { + dbg(" link %x into %x/%x\n", l_ptr, &active[0], &active[1]); + active[0] = active[1] = l_ptr; + node_established_contact(n_ptr); + return; + } + if (l_ptr->priority < active[0]->priority) { + info("Link is standby\n"); + return; + } + link_send_duplicate(active[0], l_ptr); + if (l_ptr->priority == active[0]->priority) { + active[0] = l_ptr; + return; + } + info("Link <%s> on network plane %c becomes standby\n", + active[0]->name, active[0]->b_ptr->net_plane); + active[0] = active[1] = l_ptr; +} + +/** + * node_select_active_links - select active link + */ + +static void node_select_active_links(struct node *n_ptr) +{ + struct link **active = &n_ptr->active_links[0]; + u32 i; + u32 highest_prio = 0; + + active[0] = active[1] = 0; + + for (i = 0; i < MAX_BEARERS; i++) { + struct link *l_ptr = n_ptr->links[i]; + + if (!l_ptr || !link_is_up(l_ptr) || + (l_ptr->priority < highest_prio)) + continue; + + if (l_ptr->priority > highest_prio) { + highest_prio = l_ptr->priority; + active[0] = active[1] = l_ptr; + } else { + active[1] = l_ptr; + } + } +} + +/** + * node_link_down - handle loss of link + */ + +void node_link_down(struct node *n_ptr, struct link *l_ptr) +{ + struct link **active; + + if (!link_is_active(l_ptr)) { + info("Lost standby link <%s> on network plane %c\n", + l_ptr->name, l_ptr->b_ptr->net_plane); + return; + } + info("Lost link <%s> on network plane %c\n", + l_ptr->name, l_ptr->b_ptr->net_plane); + + active = &n_ptr->active_links[0]; + if (active[0] == l_ptr) + active[0] = active[1]; + if (active[1] == l_ptr) + active[1] = active[0]; + if (active[0] == l_ptr) + node_select_active_links(n_ptr); + if (node_is_up(n_ptr)) + link_changeover(l_ptr); + else + node_lost_contact(n_ptr); +} + +int node_has_active_links(struct node *n_ptr) +{ + return (n_ptr && + ((n_ptr->active_links[0]) || (n_ptr->active_links[1]))); +} + +int node_has_redundant_links(struct node *n_ptr) +{ + return (node_has_active_links(n_ptr) && + (n_ptr->active_links[0] != n_ptr->active_links[1])); +} + +int node_has_active_routes(struct node *n_ptr) +{ + return (n_ptr && (n_ptr->last_router >= 0)); +} + +int node_is_up(struct node *n_ptr) +{ + return (node_has_active_links(n_ptr) || node_has_active_routes(n_ptr)); +} + +struct node *node_attach_link(struct link *l_ptr) +{ + struct node *n_ptr = node_find(l_ptr->addr); + + if (!n_ptr) + n_ptr = node_create(l_ptr->addr); + if (n_ptr) { + u32 bearer_id = l_ptr->b_ptr->identity; + char addr_string[16]; + + assert(bearer_id < MAX_BEARERS); + if (n_ptr->link_cnt >= 2) { + char addr_string[16]; + + err("Attempt to create third link to %s\n", + addr_string_fill(addr_string, n_ptr->addr)); + return 0; + } + + if (!n_ptr->links[bearer_id]) { + n_ptr->links[bearer_id] = l_ptr; + net.zones[tipc_zone(l_ptr->addr)]->links++; + n_ptr->link_cnt++; + return n_ptr; + } + err("Attempt to establish second link on <%s> to <%s> \n", + l_ptr->b_ptr->publ.name, + addr_string_fill(addr_string, l_ptr->addr)); + } + return 0; +} + +void node_detach_link(struct node *n_ptr, struct link *l_ptr) +{ + n_ptr->links[l_ptr->b_ptr->identity] = 0; + net.zones[tipc_zone(l_ptr->addr)]->links--; + n_ptr->link_cnt--; +} + +/* + * Routing table management - five cases to handle: + * + * 1: A link towards a zone/cluster external node comes up. + * => Send a multicast message updating routing tables of all + * system nodes within own cluster that the new destination + * can be reached via this node. + * (node.establishedContact()=>cluster.multicastNewRoute()) + * + * 2: A link towards a slave node comes up. + * => Send a multicast message updating routing tables of all + * system nodes within own cluster that the new destination + * can be reached via this node. + * (node.establishedContact()=>cluster.multicastNewRoute()) + * => Send a message to the slave node about existence + * of all system nodes within cluster: + * (node.establishedContact()=>cluster.sendLocalRoutes()) + * + * 3: A new cluster local system node becomes available. + * => Send message(s) to this particular node containing + * information about all cluster external and slave + * nodes which can be reached via this node. + * (node.establishedContact()==>network.sendExternalRoutes()) + * (node.establishedContact()==>network.sendSlaveRoutes()) + * => Send messages to all directly connected slave nodes + * containing information about the existence of the new node + * (node.establishedContact()=>cluster.multicastNewRoute()) + * + * 4: The link towards a zone/cluster external node or slave + * node goes down. + * => Send a multcast message updating routing tables of all + * nodes within cluster that the new destination can not any + * longer be reached via this node. + * (node.lostAllLinks()=>cluster.bcastLostRoute()) + * + * 5: A cluster local system node becomes unavailable. + * => Remove all references to this node from the local + * routing tables. Note: This is a completely node + * local operation. + * (node.lostAllLinks()=>network.removeAsRouter()) + * => Send messages to all directly connected slave nodes + * containing information about loss of the node + * (node.establishedContact()=>cluster.multicastLostRoute()) + * + */ + +static void node_established_contact(struct node *n_ptr) +{ + struct cluster *c_ptr; + + dbg("node_established_contact:-> %x\n", n_ptr->addr); + if (!node_has_active_routes(n_ptr)) { + k_signal((Handler)named_node_up, n_ptr->addr); + } + + /* Syncronize broadcast acks */ + n_ptr->bclink.acked = bclink_get_last_sent(); + + if (is_slave(tipc_own_addr)) + return; + if (!in_own_cluster(n_ptr->addr)) { + /* Usage case 1 (see above) */ + c_ptr = cluster_find(tipc_own_addr); + if (!c_ptr) + c_ptr = cluster_create(tipc_own_addr); + if (c_ptr) + cluster_bcast_new_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); + return; + } + + c_ptr = n_ptr->owner; + if (is_slave(n_ptr->addr)) { + /* Usage case 2 (see above) */ + cluster_bcast_new_route(c_ptr, n_ptr->addr, 1, tipc_max_nodes); + cluster_send_local_routes(c_ptr, n_ptr->addr); + return; + } + + if (n_ptr->bclink.supported) { + nmap_add(&cluster_bcast_nodes, n_ptr->addr); + if (n_ptr->addr < tipc_own_addr) + tipc_own_tag++; + } + + /* Case 3 (see above) */ + net_send_external_routes(n_ptr->addr); + cluster_send_slave_routes(c_ptr, n_ptr->addr); + cluster_bcast_new_route(c_ptr, n_ptr->addr, LOWEST_SLAVE, + highest_allowed_slave); +} + +static void node_lost_contact(struct node *n_ptr) +{ + struct cluster *c_ptr; + struct node_subscr *ns, *tns; + char addr_string[16]; + u32 i; + + /* Clean up broadcast reception remains */ + n_ptr->bclink.gap_after = n_ptr->bclink.gap_to = 0; + while (n_ptr->bclink.deferred_head) { + struct sk_buff* buf = n_ptr->bclink.deferred_head; + n_ptr->bclink.deferred_head = buf->next; + buf_discard(buf); + } + if (n_ptr->bclink.defragm) { + buf_discard(n_ptr->bclink.defragm); + n_ptr->bclink.defragm = NULL; + } + if (in_own_cluster(n_ptr->addr) && n_ptr->bclink.supported) { + bclink_acknowledge(n_ptr, mod(n_ptr->bclink.acked + 10000)); + } + + /* Update routing tables */ + if (is_slave(tipc_own_addr)) { + net_remove_as_router(n_ptr->addr); + } else { + if (!in_own_cluster(n_ptr->addr)) { + /* Case 4 (see above) */ + c_ptr = cluster_find(tipc_own_addr); + cluster_bcast_lost_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); + } else { + /* Case 5 (see above) */ + c_ptr = cluster_find(n_ptr->addr); + if (is_slave(n_ptr->addr)) { + cluster_bcast_lost_route(c_ptr, n_ptr->addr, 1, + tipc_max_nodes); + } else { + if (n_ptr->bclink.supported) { + nmap_remove(&cluster_bcast_nodes, + n_ptr->addr); + if (n_ptr->addr < tipc_own_addr) + tipc_own_tag--; + } + net_remove_as_router(n_ptr->addr); + cluster_bcast_lost_route(c_ptr, n_ptr->addr, + LOWEST_SLAVE, + highest_allowed_slave); + } + } + } + if (node_has_active_routes(n_ptr)) + return; + + info("Lost contact with %s\n", + addr_string_fill(addr_string, n_ptr->addr)); + + /* Abort link changeover */ + for (i = 0; i < MAX_BEARERS; i++) { + struct link *l_ptr = n_ptr->links[i]; + if (!l_ptr) + continue; + l_ptr->reset_checkpoint = l_ptr->next_in_no; + l_ptr->exp_msg_count = 0; + link_reset_fragments(l_ptr); + } + + /* Notify subscribers */ + list_for_each_entry_safe(ns, tns, &n_ptr->nsub, nodesub_list) { + ns->node = 0; + list_del_init(&ns->nodesub_list); + k_signal((Handler)ns->handle_node_down, + (unsigned long)ns->usr_handle); + } +} + +/** + * node_select_next_hop - find the next-hop node for a message + * + * Called by when cluster local lookup has failed. + */ + +struct node *node_select_next_hop(u32 addr, u32 selector) +{ + struct node *n_ptr; + u32 router_addr; + + if (!addr_domain_valid(addr)) + return 0; + + /* Look for direct link to destination processsor */ + n_ptr = node_find(addr); + if (n_ptr && node_has_active_links(n_ptr)) + return n_ptr; + + /* Cluster local system nodes *must* have direct links */ + if (!is_slave(addr) && in_own_cluster(addr)) + return 0; + + /* Look for cluster local router with direct link to node */ + router_addr = node_select_router(n_ptr, selector); + if (router_addr) + return node_select(router_addr, selector); + + /* Slave nodes can only be accessed within own cluster via a + known router with direct link -- if no router was found,give up */ + if (is_slave(addr)) + return 0; + + /* Inter zone/cluster -- find any direct link to remote cluster */ + addr = tipc_addr(tipc_zone(addr), tipc_cluster(addr), 0); + n_ptr = net_select_remote_node(addr, selector); + if (n_ptr && node_has_active_links(n_ptr)) + return n_ptr; + + /* Last resort -- look for any router to anywhere in remote zone */ + router_addr = net_select_router(addr, selector); + if (router_addr) + return node_select(router_addr, selector); + + return 0; +} + +/** + * node_select_router - select router to reach specified node + * + * Uses a deterministic and fair algorithm for selecting router node. + */ + +u32 node_select_router(struct node *n_ptr, u32 ref) +{ + u32 ulim; + u32 mask; + u32 start; + u32 r; + + if (!n_ptr) + return 0; + + if (n_ptr->last_router < 0) + return 0; + ulim = ((n_ptr->last_router + 1) * 32) - 1; + + /* Start entry must be random */ + mask = tipc_max_nodes; + while (mask > ulim) + mask >>= 1; + start = ref & mask; + r = start; + + /* Lookup upwards with wrap-around */ + do { + if (((n_ptr->routers[r / 32]) >> (r % 32)) & 1) + break; + } while (++r <= ulim); + if (r > ulim) { + r = 1; + do { + if (((n_ptr->routers[r / 32]) >> (r % 32)) & 1) + break; + } while (++r < start); + assert(r != start); + } + assert(r && (r <= ulim)); + return tipc_addr(own_zone(), own_cluster(), r); +} + +void node_add_router(struct node *n_ptr, u32 router) +{ + u32 r_num = tipc_node(router); + + n_ptr->routers[r_num / 32] = + ((1 << (r_num % 32)) | n_ptr->routers[r_num / 32]); + n_ptr->last_router = tipc_max_nodes / 32; + while ((--n_ptr->last_router >= 0) && + !n_ptr->routers[n_ptr->last_router]); +} + +void node_remove_router(struct node *n_ptr, u32 router) +{ + u32 r_num = tipc_node(router); + + if (n_ptr->last_router < 0) + return; /* No routes */ + + n_ptr->routers[r_num / 32] = + ((~(1 << (r_num % 32))) & (n_ptr->routers[r_num / 32])); + n_ptr->last_router = tipc_max_nodes / 32; + while ((--n_ptr->last_router >= 0) && + !n_ptr->routers[n_ptr->last_router]); + + if (!node_is_up(n_ptr)) + node_lost_contact(n_ptr); +} + +#if 0 +void node_print(struct print_buf *buf, struct node *n_ptr, char *str) +{ + u32 i; + + tipc_printf(buf, "\n\n%s", str); + for (i = 0; i < MAX_BEARERS; i++) { + if (!n_ptr->links[i]) + continue; + tipc_printf(buf, "Links[%u]: %x, ", i, n_ptr->links[i]); + } + tipc_printf(buf, "Active links: [%x,%x]\n", + n_ptr->active_links[0], n_ptr->active_links[1]); +} +#endif + +u32 tipc_available_nodes(const u32 domain) +{ + struct node *n_ptr; + u32 cnt = 0; + + for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + if (!in_scope(domain, n_ptr->addr)) + continue; + if (node_is_up(n_ptr)) + cnt++; + } + return cnt; +} + +struct sk_buff *node_get_nodes(const void *req_tlv_area, int req_tlv_space) +{ + u32 domain; + struct sk_buff *buf; + struct node *n_ptr; + struct tipc_node_info node_info; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + domain = *(u32 *)TLV_DATA(req_tlv_area); + domain = ntohl(domain); + if (!addr_domain_valid(domain)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); + + if (!nodes) + return cfg_reply_none(); + + /* For now, get space for all other nodes + (will need to modify this when slave nodes are supported */ + + buf = cfg_reply_alloc(TLV_SPACE(sizeof(node_info)) * + (tipc_max_nodes - 1)); + if (!buf) + return NULL; + + /* Add TLVs for all nodes in scope */ + + for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + if (!in_scope(domain, n_ptr->addr)) + continue; + node_info.addr = htonl(n_ptr->addr); + node_info.up = htonl(node_is_up(n_ptr)); + cfg_append_tlv(buf, TIPC_TLV_NODE_INFO, + &node_info, sizeof(node_info)); + } + + return buf; +} + +struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space) +{ + u32 domain; + struct sk_buff *buf; + struct node *n_ptr; + struct tipc_link_info link_info; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_NET_ADDR)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + domain = *(u32 *)TLV_DATA(req_tlv_area); + domain = ntohl(domain); + if (!addr_domain_valid(domain)) + return cfg_reply_error_string(TIPC_CFG_INVALID_VALUE + " (network address)"); + + if (!nodes) + return cfg_reply_none(); + + /* For now, get space for 2 links to all other nodes + bcast link + (will need to modify this when slave nodes are supported */ + + buf = cfg_reply_alloc(TLV_SPACE(sizeof(link_info)) * + (2 * (tipc_max_nodes - 1) + 1)); + if (!buf) + return NULL; + + /* Add TLV for broadcast link */ + + link_info.dest = tipc_own_addr & 0xfffff00; + link_info.dest = htonl(link_info.dest); + link_info.up = htonl(1); + sprintf(link_info.str, bc_link_name); + cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, &link_info, sizeof(link_info)); + + /* Add TLVs for any other links in scope */ + + for (n_ptr = nodes; n_ptr; n_ptr = n_ptr->next) { + u32 i; + + if (!in_scope(domain, n_ptr->addr)) + continue; + for (i = 0; i < MAX_BEARERS; i++) { + if (!n_ptr->links[i]) + continue; + link_info.dest = htonl(n_ptr->addr); + link_info.up = htonl(link_is_up(n_ptr->links[i])); + strcpy(link_info.str, n_ptr->links[i]->name); + cfg_append_tlv(buf, TIPC_TLV_LINK_INFO, + &link_info, sizeof(link_info)); + } + } + + return buf; +} diff --git a/net/tipc/node.h b/net/tipc/node.h new file mode 100644 index 00000000000..b39442badcc --- /dev/null +++ b/net/tipc/node.h @@ -0,0 +1,144 @@ +/* + * net/tipc/node.h: Include file for TIPC node management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NODE_H +#define _TIPC_NODE_H + +#include "node_subscr.h" +#include "addr.h" +#include "cluster.h" +#include "bearer.h" + +/** + * struct node - TIPC node structure + * @addr: network address of node + * @lock: spinlock governing access to structure + * @owner: pointer to cluster that node belongs to + * @next: pointer to next node in sorted list of cluster's nodes + * @nsub: list of "node down" subscriptions monitoring node + * @active_links: pointers to active links to node + * @links: pointers to all links to node + * @link_cnt: number of links to node + * @permit_changeover: non-zero if node has redundant links to this system + * @routers: bitmap (used for multicluster communication) + * @last_router: (used for multicluster communication) + * @bclink: broadcast-related info + * @supported: non-zero if node supports TIPC b'cast capability + * @acked: sequence # of last outbound b'cast message acknowledged by node + * @last_in: sequence # of last in-sequence b'cast message received from node + * @gap_after: sequence # of last message not requiring a NAK request + * @gap_to: sequence # of last message requiring a NAK request + * @nack_sync: counter that determines when NAK requests should be sent + * @deferred_head: oldest OOS b'cast message received from node + * @deferred_tail: newest OOS b'cast message received from node + * @defragm: list of partially reassembled b'cast message fragments from node + */ + +struct node { + u32 addr; + spinlock_t lock; + struct cluster *owner; + struct node *next; + struct list_head nsub; + struct link *active_links[2]; + struct link *links[MAX_BEARERS]; + int link_cnt; + int permit_changeover; + u32 routers[512/32]; + int last_router; + struct { + int supported; + u32 acked; + u32 last_in; + u32 gap_after; + u32 gap_to; + u32 nack_sync; + struct sk_buff *deferred_head; + struct sk_buff *deferred_tail; + struct sk_buff *defragm; + } bclink; +}; + +extern struct node *nodes; +extern u32 tipc_own_tag; + +struct node *node_create(u32 addr); +void node_delete(struct node *n_ptr); +struct node *node_attach_link(struct link *l_ptr); +void node_detach_link(struct node *n_ptr, struct link *l_ptr); +void node_link_down(struct node *n_ptr, struct link *l_ptr); +void node_link_up(struct node *n_ptr, struct link *l_ptr); +int node_has_active_links(struct node *n_ptr); +int node_has_redundant_links(struct node *n_ptr); +u32 node_select_router(struct node *n_ptr, u32 ref); +struct node *node_select_next_hop(u32 addr, u32 selector); +int node_is_up(struct node *n_ptr); +void node_add_router(struct node *n_ptr, u32 router); +void node_remove_router(struct node *n_ptr, u32 router); +struct sk_buff *node_get_links(const void *req_tlv_area, int req_tlv_space); +struct sk_buff *node_get_nodes(const void *req_tlv_area, int req_tlv_space); + +static inline struct node *node_find(u32 addr) +{ + if (likely(in_own_cluster(addr))) + return local_nodes[tipc_node(addr)]; + else if (addr_domain_valid(addr)) { + struct cluster *c_ptr = cluster_find(addr); + + if (c_ptr) + return c_ptr->nodes[tipc_node(addr)]; + } + return 0; +} + +static inline struct node *node_select(u32 addr, u32 selector) +{ + if (likely(in_own_cluster(addr))) + return local_nodes[tipc_node(addr)]; + return node_select_next_hop(addr, selector); +} + +static inline void node_lock(struct node *n_ptr) +{ + spin_lock_bh(&n_ptr->lock); +} + +static inline void node_unlock(struct node *n_ptr) +{ + spin_unlock_bh(&n_ptr->lock); +} + +#endif diff --git a/net/tipc/node_subscr.c b/net/tipc/node_subscr.c new file mode 100644 index 00000000000..79375927916 --- /dev/null +++ b/net/tipc/node_subscr.c @@ -0,0 +1,79 @@ +/* + * net/tipc/node_subscr.c: TIPC "node down" subscription handling + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "node_subscr.h" +#include "node.h" +#include "addr.h" + +/** + * nodesub_subscribe - create "node down" subscription for specified node + */ + +void nodesub_subscribe(struct node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down) +{ + node_sub->node = 0; + if (addr == tipc_own_addr) + return; + if (!addr_node_valid(addr)) { + warn("node_subscr with illegal %x\n", addr); + return; + } + + node_sub->handle_node_down = handle_down; + node_sub->usr_handle = usr_handle; + node_sub->node = node_find(addr); + assert(node_sub->node); + node_lock(node_sub->node); + list_add_tail(&node_sub->nodesub_list, &node_sub->node->nsub); + node_unlock(node_sub->node); +} + +/** + * nodesub_unsubscribe - cancel "node down" subscription (if any) + */ + +void nodesub_unsubscribe(struct node_subscr *node_sub) +{ + if (!node_sub->node) + return; + + node_lock(node_sub->node); + list_del_init(&node_sub->nodesub_list); + node_unlock(node_sub->node); +} diff --git a/net/tipc/node_subscr.h b/net/tipc/node_subscr.h new file mode 100644 index 00000000000..a3b87ac4859 --- /dev/null +++ b/net/tipc/node_subscr.h @@ -0,0 +1,63 @@ +/* + * net/tipc/node_subscr.h: Include file for TIPC "node down" subscription handling + * + * Copyright (c) 1995-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_NODE_SUBSCR_H +#define _TIPC_NODE_SUBSCR_H + +#include "addr.h" + +typedef void (*net_ev_handler) (void *usr_handle); + +/** + * struct node_subscr - "node down" subscription entry + * @node: ptr to node structure of interest (or NULL, if none) + * @handle_node_down: routine to invoke when node fails + * @usr_handle: argument to pass to routine when node fails + * @nodesub_list: adjacent entries in list of subscriptions for the node + */ + +struct node_subscr { + struct node *node; + net_ev_handler handle_node_down; + void *usr_handle; + struct list_head nodesub_list; +}; + +void nodesub_subscribe(struct node_subscr *node_sub, u32 addr, + void *usr_handle, net_ev_handler handle_down); +void nodesub_unsubscribe(struct node_subscr *node_sub); + +#endif diff --git a/net/tipc/port.c b/net/tipc/port.c new file mode 100644 index 00000000000..66caca7abe9 --- /dev/null +++ b/net/tipc/port.c @@ -0,0 +1,1708 @@ +/* + * net/tipc/port.c: TIPC port code + * + * Copyright (c) 1992-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "config.h" +#include "dbg.h" +#include "port.h" +#include "addr.h" +#include "link.h" +#include "node.h" +#include "port.h" +#include "name_table.h" +#include "user_reg.h" +#include "msg.h" +#include "bcast.h" + +/* Connection management: */ +#define PROBING_INTERVAL 3600000 /* [ms] => 1 h */ +#define CONFIRMED 0 +#define PROBING 1 + +#define MAX_REJECT_SIZE 1024 + +static struct sk_buff *msg_queue_head = 0; +static struct sk_buff *msg_queue_tail = 0; + +spinlock_t port_list_lock = SPIN_LOCK_UNLOCKED; +static spinlock_t queue_lock = SPIN_LOCK_UNLOCKED; + +LIST_HEAD(ports); +static void port_handle_node_down(unsigned long ref); +static struct sk_buff* port_build_self_abort_msg(struct port *,u32 err); +static struct sk_buff* port_build_peer_abort_msg(struct port *,u32 err); +static void port_timeout(unsigned long ref); + + +static inline u32 port_peernode(struct port *p_ptr) +{ + return msg_destnode(&p_ptr->publ.phdr); +} + +static inline u32 port_peerport(struct port *p_ptr) +{ + return msg_destport(&p_ptr->publ.phdr); +} + +static inline u32 port_out_seqno(struct port *p_ptr) +{ + return msg_transp_seqno(&p_ptr->publ.phdr); +} + +static inline void port_set_out_seqno(struct port *p_ptr, u32 seqno) +{ + msg_set_transp_seqno(&p_ptr->publ.phdr,seqno); +} + +static inline void port_incr_out_seqno(struct port *p_ptr) +{ + struct tipc_msg *m = &p_ptr->publ.phdr; + + if (likely(!msg_routed(m))) + return; + msg_set_transp_seqno(m, (msg_transp_seqno(m) + 1)); +} + +/** + * tipc_multicast - send a multicast message to local and remote destinations + */ + +int tipc_multicast(u32 ref, struct tipc_name_seq const *seq, u32 domain, + u32 num_sect, struct iovec const *msg_sect) +{ + struct tipc_msg *hdr; + struct sk_buff *buf; + struct sk_buff *ibuf = NULL; + struct port_list dports = {0, NULL, }; + struct port *oport = port_deref(ref); + int ext_targets; + int res; + + if (unlikely(!oport)) + return -EINVAL; + + /* Create multicast message */ + + hdr = &oport->publ.phdr; + msg_set_type(hdr, TIPC_MCAST_MSG); + msg_set_nametype(hdr, seq->type); + msg_set_namelower(hdr, seq->lower); + msg_set_nameupper(hdr, seq->upper); + msg_set_hdr_sz(hdr, MCAST_H_SIZE); + res = msg_build(hdr, msg_sect, num_sect, MAX_MSG_SIZE, + !oport->user_port, &buf); + if (unlikely(!buf)) + return res; + + /* Figure out where to send multicast message */ + + ext_targets = nametbl_mc_translate(seq->type, seq->lower, seq->upper, + TIPC_NODE_SCOPE, &dports); + + /* Send message to destinations (duplicate it only if necessary) */ + + if (ext_targets) { + if (dports.count != 0) { + ibuf = skb_copy(buf, GFP_ATOMIC); + if (ibuf == NULL) { + port_list_free(&dports); + buf_discard(buf); + return -ENOMEM; + } + } + res = bclink_send_msg(buf); + if ((res < 0) && (dports.count != 0)) { + buf_discard(ibuf); + } + } else { + ibuf = buf; + } + + if (res >= 0) { + if (ibuf) + port_recv_mcast(ibuf, &dports); + } else { + port_list_free(&dports); + } + return res; +} + +/** + * port_recv_mcast - deliver multicast message to all destination ports + * + * If there is no port list, perform a lookup to create one + */ + +void port_recv_mcast(struct sk_buff *buf, struct port_list *dp) +{ + struct tipc_msg* msg; + struct port_list dports = {0, NULL, }; + struct port_list *item = dp; + int cnt = 0; + + assert(buf); + msg = buf_msg(buf); + + /* Create destination port list, if one wasn't supplied */ + + if (dp == NULL) { + nametbl_mc_translate(msg_nametype(msg), + msg_namelower(msg), + msg_nameupper(msg), + TIPC_CLUSTER_SCOPE, + &dports); + item = dp = &dports; + } + + /* Deliver a copy of message to each destination port */ + + if (dp->count != 0) { + if (dp->count == 1) { + msg_set_destport(msg, dp->ports[0]); + port_recv_msg(buf); + port_list_free(dp); + return; + } + for (; cnt < dp->count; cnt++) { + int index = cnt % PLSIZE; + struct sk_buff *b = skb_clone(buf, GFP_ATOMIC); + + if (b == NULL) { + warn("Buffer allocation failure\n"); + msg_dbg(msg, "LOST:"); + goto exit; + } + if ((index == 0) && (cnt != 0)) { + item = item->next; + } + msg_set_destport(buf_msg(b),item->ports[index]); + port_recv_msg(b); + } + } +exit: + buf_discard(buf); + port_list_free(dp); +} + +/** + * tipc_createport_raw - create a native TIPC port + * + * Returns local port reference + */ + +u32 tipc_createport_raw(void *usr_handle, + u32 (*dispatcher)(struct tipc_port *, struct sk_buff *), + void (*wakeup)(struct tipc_port *), + const u32 importance) +{ + struct port *p_ptr; + struct tipc_msg *msg; + u32 ref; + + p_ptr = kmalloc(sizeof(*p_ptr), GFP_ATOMIC); + if (p_ptr == NULL) { + warn("Memory squeeze; failed to create port\n"); + return 0; + } + memset(p_ptr, 0, sizeof(*p_ptr)); + ref = ref_acquire(p_ptr, &p_ptr->publ.lock); + if (!ref) { + warn("Reference Table Exhausted\n"); + kfree(p_ptr); + return 0; + } + + port_lock(ref); + p_ptr->publ.ref = ref; + msg = &p_ptr->publ.phdr; + msg_init(msg, DATA_LOW, TIPC_NAMED_MSG, TIPC_OK, LONG_H_SIZE, 0); + msg_set_orignode(msg, tipc_own_addr); + msg_set_prevnode(msg, tipc_own_addr); + msg_set_origport(msg, ref); + msg_set_importance(msg,importance); + p_ptr->last_in_seqno = 41; + p_ptr->sent = 1; + p_ptr->publ.usr_handle = usr_handle; + INIT_LIST_HEAD(&p_ptr->wait_list); + INIT_LIST_HEAD(&p_ptr->subscription.nodesub_list); + p_ptr->congested_link = 0; + p_ptr->max_pkt = MAX_PKT_DEFAULT; + p_ptr->dispatcher = dispatcher; + p_ptr->wakeup = wakeup; + p_ptr->user_port = 0; + k_init_timer(&p_ptr->timer, (Handler)port_timeout, ref); + spin_lock_bh(&port_list_lock); + INIT_LIST_HEAD(&p_ptr->publications); + INIT_LIST_HEAD(&p_ptr->port_list); + list_add_tail(&p_ptr->port_list, &ports); + spin_unlock_bh(&port_list_lock); + port_unlock(p_ptr); + return ref; +} + +int tipc_deleteport(u32 ref) +{ + struct port *p_ptr; + struct sk_buff *buf = 0; + + tipc_withdraw(ref, 0, 0); + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + + ref_discard(ref); + port_unlock(p_ptr); + + k_cancel_timer(&p_ptr->timer); + if (p_ptr->publ.connected) { + buf = port_build_peer_abort_msg(p_ptr, TIPC_ERR_NO_PORT); + nodesub_unsubscribe(&p_ptr->subscription); + } + if (p_ptr->user_port) { + reg_remove_port(p_ptr->user_port); + kfree(p_ptr->user_port); + } + + spin_lock_bh(&port_list_lock); + list_del(&p_ptr->port_list); + list_del(&p_ptr->wait_list); + spin_unlock_bh(&port_list_lock); + k_term_timer(&p_ptr->timer); + kfree(p_ptr); + dbg("Deleted port %u\n", ref); + net_route_msg(buf); + return TIPC_OK; +} + +/** + * tipc_get_port() - return port associated with 'ref' + * + * Note: Port is not locked. + */ + +struct tipc_port *tipc_get_port(const u32 ref) +{ + return (struct tipc_port *)ref_deref(ref); +} + +/** + * tipc_get_handle - return user handle associated to port 'ref' + */ + +void *tipc_get_handle(const u32 ref) +{ + struct port *p_ptr; + void * handle; + + p_ptr = port_lock(ref); + if (!p_ptr) + return 0; + handle = p_ptr->publ.usr_handle; + port_unlock(p_ptr); + return handle; +} + +static inline int port_unreliable(struct port *p_ptr) +{ + return msg_src_droppable(&p_ptr->publ.phdr); +} + +int tipc_portunreliable(u32 ref, unsigned int *isunreliable) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + *isunreliable = port_unreliable(p_ptr); + spin_unlock_bh(p_ptr->publ.lock); + return TIPC_OK; +} + +int tipc_set_portunreliable(u32 ref, unsigned int isunreliable) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + msg_set_src_droppable(&p_ptr->publ.phdr, (isunreliable != 0)); + port_unlock(p_ptr); + return TIPC_OK; +} + +static inline int port_unreturnable(struct port *p_ptr) +{ + return msg_dest_droppable(&p_ptr->publ.phdr); +} + +int tipc_portunreturnable(u32 ref, unsigned int *isunrejectable) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + *isunrejectable = port_unreturnable(p_ptr); + spin_unlock_bh(p_ptr->publ.lock); + return TIPC_OK; +} + +int tipc_set_portunreturnable(u32 ref, unsigned int isunrejectable) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + msg_set_dest_droppable(&p_ptr->publ.phdr, (isunrejectable != 0)); + port_unlock(p_ptr); + return TIPC_OK; +} + +/* + * port_build_proto_msg(): build a port level protocol + * or a connection abortion message. Called with + * tipc_port lock on. + */ +static struct sk_buff *port_build_proto_msg(u32 destport, u32 destnode, + u32 origport, u32 orignode, + u32 usr, u32 type, u32 err, + u32 seqno, u32 ack) +{ + struct sk_buff *buf; + struct tipc_msg *msg; + + buf = buf_acquire(LONG_H_SIZE); + if (buf) { + msg = buf_msg(buf); + msg_init(msg, usr, type, err, LONG_H_SIZE, destnode); + msg_set_destport(msg, destport); + msg_set_origport(msg, origport); + msg_set_destnode(msg, destnode); + msg_set_orignode(msg, orignode); + msg_set_transp_seqno(msg, seqno); + msg_set_msgcnt(msg, ack); + msg_dbg(msg, "PORT>SEND>:"); + } + return buf; +} + +int tipc_set_msg_option(struct tipc_port *tp_ptr, const char *opt, const u32 sz) +{ + msg_expand(&tp_ptr->phdr, msg_destnode(&tp_ptr->phdr)); + msg_set_options(&tp_ptr->phdr, opt, sz); + return TIPC_OK; +} + +int tipc_reject_msg(struct sk_buff *buf, u32 err) +{ + struct tipc_msg *msg = buf_msg(buf); + struct sk_buff *rbuf; + struct tipc_msg *rmsg; + int hdr_sz; + u32 imp = msg_importance(msg); + u32 data_sz = msg_data_sz(msg); + + if (data_sz > MAX_REJECT_SIZE) + data_sz = MAX_REJECT_SIZE; + if (msg_connected(msg) && (imp < TIPC_CRITICAL_IMPORTANCE)) + imp++; + msg_dbg(msg, "port->rej: "); + + /* discard rejected message if it shouldn't be returned to sender */ + if (msg_errcode(msg) || msg_dest_droppable(msg)) { + buf_discard(buf); + return data_sz; + } + + /* construct rejected message */ + if (msg_mcast(msg)) + hdr_sz = MCAST_H_SIZE; + else + hdr_sz = LONG_H_SIZE; + rbuf = buf_acquire(data_sz + hdr_sz); + if (rbuf == NULL) { + buf_discard(buf); + return data_sz; + } + rmsg = buf_msg(rbuf); + msg_init(rmsg, imp, msg_type(msg), err, hdr_sz, msg_orignode(msg)); + msg_set_destport(rmsg, msg_origport(msg)); + msg_set_prevnode(rmsg, tipc_own_addr); + msg_set_origport(rmsg, msg_destport(msg)); + if (msg_short(msg)) + msg_set_orignode(rmsg, tipc_own_addr); + else + msg_set_orignode(rmsg, msg_destnode(msg)); + msg_set_size(rmsg, data_sz + hdr_sz); + msg_set_nametype(rmsg, msg_nametype(msg)); + msg_set_nameinst(rmsg, msg_nameinst(msg)); + memcpy(rbuf->data + hdr_sz, msg_data(msg), data_sz); + + /* send self-abort message when rejecting on a connected port */ + if (msg_connected(msg)) { + struct sk_buff *abuf = 0; + struct port *p_ptr = port_lock(msg_destport(msg)); + + if (p_ptr) { + if (p_ptr->publ.connected) + abuf = port_build_self_abort_msg(p_ptr, err); + port_unlock(p_ptr); + } + net_route_msg(abuf); + } + + /* send rejected message */ + buf_discard(buf); + net_route_msg(rbuf); + return data_sz; +} + +int port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, + struct iovec const *msg_sect, u32 num_sect, + int err) +{ + struct sk_buff *buf; + int res; + + res = msg_build(hdr, msg_sect, num_sect, MAX_MSG_SIZE, + !p_ptr->user_port, &buf); + if (!buf) + return res; + + return tipc_reject_msg(buf, err); +} + +static void port_timeout(unsigned long ref) +{ + struct port *p_ptr = port_lock(ref); + struct sk_buff *buf = 0; + + if (!p_ptr || !p_ptr->publ.connected) + return; + + /* Last probe answered ? */ + if (p_ptr->probing_state == PROBING) { + buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_PORT); + } else { + buf = port_build_proto_msg(port_peerport(p_ptr), + port_peernode(p_ptr), + p_ptr->publ.ref, + tipc_own_addr, + CONN_MANAGER, + CONN_PROBE, + TIPC_OK, + port_out_seqno(p_ptr), + 0); + port_incr_out_seqno(p_ptr); + p_ptr->probing_state = PROBING; + k_start_timer(&p_ptr->timer, p_ptr->probing_interval); + } + port_unlock(p_ptr); + net_route_msg(buf); +} + + +static void port_handle_node_down(unsigned long ref) +{ + struct port *p_ptr = port_lock(ref); + struct sk_buff* buf = 0; + + if (!p_ptr) + return; + buf = port_build_self_abort_msg(p_ptr, TIPC_ERR_NO_NODE); + port_unlock(p_ptr); + net_route_msg(buf); +} + + +static struct sk_buff *port_build_self_abort_msg(struct port *p_ptr, u32 err) +{ + u32 imp = msg_importance(&p_ptr->publ.phdr); + + if (!p_ptr->publ.connected) + return 0; + if (imp < TIPC_CRITICAL_IMPORTANCE) + imp++; + return port_build_proto_msg(p_ptr->publ.ref, + tipc_own_addr, + port_peerport(p_ptr), + port_peernode(p_ptr), + imp, + TIPC_CONN_MSG, + err, + p_ptr->last_in_seqno + 1, + 0); +} + + +static struct sk_buff *port_build_peer_abort_msg(struct port *p_ptr, u32 err) +{ + u32 imp = msg_importance(&p_ptr->publ.phdr); + + if (!p_ptr->publ.connected) + return 0; + if (imp < TIPC_CRITICAL_IMPORTANCE) + imp++; + return port_build_proto_msg(port_peerport(p_ptr), + port_peernode(p_ptr), + p_ptr->publ.ref, + tipc_own_addr, + imp, + TIPC_CONN_MSG, + err, + port_out_seqno(p_ptr), + 0); +} + +void port_recv_proto_msg(struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct port *p_ptr = port_lock(msg_destport(msg)); + u32 err = TIPC_OK; + struct sk_buff *r_buf = 0; + struct sk_buff *abort_buf = 0; + + msg_dbg(msg, "PORT<RECV<:"); + + if (!p_ptr) { + err = TIPC_ERR_NO_PORT; + } else if (p_ptr->publ.connected) { + if (port_peernode(p_ptr) != msg_orignode(msg)) + err = TIPC_ERR_NO_PORT; + if (port_peerport(p_ptr) != msg_origport(msg)) + err = TIPC_ERR_NO_PORT; + if (!err && msg_routed(msg)) { + u32 seqno = msg_transp_seqno(msg); + u32 myno = ++p_ptr->last_in_seqno; + if (seqno != myno) { + err = TIPC_ERR_NO_PORT; + abort_buf = port_build_self_abort_msg(p_ptr, err); + } + } + if (msg_type(msg) == CONN_ACK) { + int wakeup = port_congested(p_ptr) && + p_ptr->publ.congested && + p_ptr->wakeup; + p_ptr->acked += msg_msgcnt(msg); + if (port_congested(p_ptr)) + goto exit; + p_ptr->publ.congested = 0; + if (!wakeup) + goto exit; + p_ptr->wakeup(&p_ptr->publ); + goto exit; + } + } else if (p_ptr->publ.published) { + err = TIPC_ERR_NO_PORT; + } + if (err) { + r_buf = port_build_proto_msg(msg_origport(msg), + msg_orignode(msg), + msg_destport(msg), + tipc_own_addr, + DATA_HIGH, + TIPC_CONN_MSG, + err, + 0, + 0); + goto exit; + } + + /* All is fine */ + if (msg_type(msg) == CONN_PROBE) { + r_buf = port_build_proto_msg(msg_origport(msg), + msg_orignode(msg), + msg_destport(msg), + tipc_own_addr, + CONN_MANAGER, + CONN_PROBE_REPLY, + TIPC_OK, + port_out_seqno(p_ptr), + 0); + } + p_ptr->probing_state = CONFIRMED; + port_incr_out_seqno(p_ptr); +exit: + if (p_ptr) + port_unlock(p_ptr); + net_route_msg(r_buf); + net_route_msg(abort_buf); + buf_discard(buf); +} + +static void port_print(struct port *p_ptr, struct print_buf *buf, int full_id) +{ + struct publication *publ; + + if (full_id) + tipc_printf(buf, "<%u.%u.%u:%u>:", + tipc_zone(tipc_own_addr), tipc_cluster(tipc_own_addr), + tipc_node(tipc_own_addr), p_ptr->publ.ref); + else + tipc_printf(buf, "%-10u:", p_ptr->publ.ref); + + if (p_ptr->publ.connected) { + u32 dport = port_peerport(p_ptr); + u32 destnode = port_peernode(p_ptr); + + tipc_printf(buf, " connected to <%u.%u.%u:%u>", + tipc_zone(destnode), tipc_cluster(destnode), + tipc_node(destnode), dport); + if (p_ptr->publ.conn_type != 0) + tipc_printf(buf, " via {%u,%u}", + p_ptr->publ.conn_type, + p_ptr->publ.conn_instance); + } + else if (p_ptr->publ.published) { + tipc_printf(buf, " bound to"); + list_for_each_entry(publ, &p_ptr->publications, pport_list) { + if (publ->lower == publ->upper) + tipc_printf(buf, " {%u,%u}", publ->type, + publ->lower); + else + tipc_printf(buf, " {%u,%u,%u}", publ->type, + publ->lower, publ->upper); + } + } + tipc_printf(buf, "\n"); +} + +#define MAX_PORT_QUERY 32768 + +struct sk_buff *port_get_ports(void) +{ + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + struct print_buf pb; + struct port *p_ptr; + int str_len; + + buf = cfg_reply_alloc(TLV_SPACE(MAX_PORT_QUERY)); + if (!buf) + return NULL; + rep_tlv = (struct tlv_desc *)buf->data; + + printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_QUERY); + spin_lock_bh(&port_list_lock); + list_for_each_entry(p_ptr, &ports, port_list) { + spin_lock_bh(p_ptr->publ.lock); + port_print(p_ptr, &pb, 0); + spin_unlock_bh(p_ptr->publ.lock); + } + spin_unlock_bh(&port_list_lock); + str_len = printbuf_validate(&pb); + + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +#if 0 + +#define MAX_PORT_STATS 2000 + +struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space) +{ + u32 ref; + struct port *p_ptr; + struct sk_buff *buf; + struct tlv_desc *rep_tlv; + struct print_buf pb; + int str_len; + + if (!TLV_CHECK(req_tlv_area, req_tlv_space, TIPC_TLV_PORT_REF)) + return cfg_reply_error_string(TIPC_CFG_TLV_ERROR); + + ref = *(u32 *)TLV_DATA(req_tlv_area); + ref = ntohl(ref); + + p_ptr = port_lock(ref); + if (!p_ptr) + return cfg_reply_error_string("port not found"); + + buf = cfg_reply_alloc(TLV_SPACE(MAX_PORT_STATS)); + if (!buf) { + port_unlock(p_ptr); + return NULL; + } + rep_tlv = (struct tlv_desc *)buf->data; + + printbuf_init(&pb, TLV_DATA(rep_tlv), MAX_PORT_STATS); + port_print(p_ptr, &pb, 1); + /* NEED TO FILL IN ADDITIONAL PORT STATISTICS HERE */ + port_unlock(p_ptr); + str_len = printbuf_validate(&pb); + + skb_put(buf, TLV_SPACE(str_len)); + TLV_SET(rep_tlv, TIPC_TLV_ULTRA_STRING, NULL, str_len); + + return buf; +} + +#endif + +void port_reinit(void) +{ + struct port *p_ptr; + struct tipc_msg *msg; + + spin_lock_bh(&port_list_lock); + list_for_each_entry(p_ptr, &ports, port_list) { + msg = &p_ptr->publ.phdr; + if (msg_orignode(msg) == tipc_own_addr) + break; + msg_set_orignode(msg, tipc_own_addr); + } + spin_unlock_bh(&port_list_lock); +} + + +/* + * port_dispatcher_sigh(): Signal handler for messages destinated + * to the tipc_port interface. + */ + +static void port_dispatcher_sigh(void *dummy) +{ + struct sk_buff *buf; + + spin_lock_bh(&queue_lock); + buf = msg_queue_head; + msg_queue_head = 0; + spin_unlock_bh(&queue_lock); + + while (buf) { + struct port *p_ptr; + struct user_port *up_ptr; + struct tipc_portid orig; + struct tipc_name_seq dseq; + void *usr_handle; + int connected; + int published; + + struct sk_buff *next = buf->next; + struct tipc_msg *msg = buf_msg(buf); + u32 dref = msg_destport(msg); + + p_ptr = port_lock(dref); + if (!p_ptr) { + /* Port deleted while msg in queue */ + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + buf = next; + continue; + } + orig.ref = msg_origport(msg); + orig.node = msg_orignode(msg); + up_ptr = p_ptr->user_port; + usr_handle = up_ptr->usr_handle; + connected = p_ptr->publ.connected; + published = p_ptr->publ.published; + + if (unlikely(msg_errcode(msg))) + goto err; + + switch (msg_type(msg)) { + + case TIPC_CONN_MSG:{ + tipc_conn_msg_event cb = up_ptr->conn_msg_cb; + u32 peer_port = port_peerport(p_ptr); + u32 peer_node = port_peernode(p_ptr); + + spin_unlock_bh(p_ptr->publ.lock); + if (unlikely(!connected)) { + if (unlikely(published)) + goto reject; + tipc_connect2port(dref,&orig); + } + if (unlikely(msg_origport(msg) != peer_port)) + goto reject; + if (unlikely(msg_orignode(msg) != peer_node)) + goto reject; + if (unlikely(!cb)) + goto reject; + if (unlikely(++p_ptr->publ.conn_unacked >= + TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(dref, + p_ptr->publ.conn_unacked); + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg)); + break; + } + case TIPC_DIRECT_MSG:{ + tipc_msg_event cb = up_ptr->msg_cb; + + spin_unlock_bh(p_ptr->publ.lock); + if (unlikely(connected)) + goto reject; + if (unlikely(!cb)) + goto reject; + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg), msg_importance(msg), + &orig); + break; + } + case TIPC_NAMED_MSG:{ + tipc_named_msg_event cb = up_ptr->named_msg_cb; + + spin_unlock_bh(p_ptr->publ.lock); + if (unlikely(connected)) + goto reject; + if (unlikely(!cb)) + goto reject; + if (unlikely(!published)) + goto reject; + dseq.type = msg_nametype(msg); + dseq.lower = msg_nameinst(msg); + dseq.upper = dseq.lower; + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg), msg_importance(msg), + &orig, &dseq); + break; + } + } + if (buf) + buf_discard(buf); + buf = next; + continue; +err: + switch (msg_type(msg)) { + + case TIPC_CONN_MSG:{ + tipc_conn_shutdown_event cb = + up_ptr->conn_err_cb; + u32 peer_port = port_peerport(p_ptr); + u32 peer_node = port_peernode(p_ptr); + + spin_unlock_bh(p_ptr->publ.lock); + if (!connected || !cb) + break; + if (msg_origport(msg) != peer_port) + break; + if (msg_orignode(msg) != peer_node) + break; + tipc_disconnect(dref); + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg), msg_errcode(msg)); + break; + } + case TIPC_DIRECT_MSG:{ + tipc_msg_err_event cb = up_ptr->err_cb; + + spin_unlock_bh(p_ptr->publ.lock); + if (connected || !cb) + break; + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg), msg_errcode(msg), &orig); + break; + } + case TIPC_NAMED_MSG:{ + tipc_named_msg_err_event cb = + up_ptr->named_err_cb; + + spin_unlock_bh(p_ptr->publ.lock); + if (connected || !cb) + break; + dseq.type = msg_nametype(msg); + dseq.lower = msg_nameinst(msg); + dseq.upper = dseq.lower; + skb_pull(buf, msg_hdr_sz(msg)); + cb(usr_handle, dref, &buf, msg_data(msg), + msg_data_sz(msg), msg_errcode(msg), &dseq); + break; + } + } + if (buf) + buf_discard(buf); + buf = next; + continue; +reject: + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + buf = next; + } +} + +/* + * port_dispatcher(): Dispatcher for messages destinated + * to the tipc_port interface. Called with port locked. + */ + +static u32 port_dispatcher(struct tipc_port *dummy, struct sk_buff *buf) +{ + buf->next = NULL; + spin_lock_bh(&queue_lock); + if (msg_queue_head) { + msg_queue_tail->next = buf; + msg_queue_tail = buf; + } else { + msg_queue_tail = msg_queue_head = buf; + k_signal((Handler)port_dispatcher_sigh, 0); + } + spin_unlock_bh(&queue_lock); + return TIPC_OK; +} + +/* + * Wake up port after congestion: Called with port locked, + * + */ + +static void port_wakeup_sh(unsigned long ref) +{ + struct port *p_ptr; + struct user_port *up_ptr; + tipc_continue_event cb = 0; + void *uh = 0; + + p_ptr = port_lock(ref); + if (p_ptr) { + up_ptr = p_ptr->user_port; + if (up_ptr) { + cb = up_ptr->continue_event_cb; + uh = up_ptr->usr_handle; + } + port_unlock(p_ptr); + } + if (cb) + cb(uh, ref); +} + + +static void port_wakeup(struct tipc_port *p_ptr) +{ + k_signal((Handler)port_wakeup_sh, p_ptr->ref); +} + +void tipc_acknowledge(u32 ref, u32 ack) +{ + struct port *p_ptr; + struct sk_buff *buf = 0; + + p_ptr = port_lock(ref); + if (!p_ptr) + return; + if (p_ptr->publ.connected) { + p_ptr->publ.conn_unacked -= ack; + buf = port_build_proto_msg(port_peerport(p_ptr), + port_peernode(p_ptr), + ref, + tipc_own_addr, + CONN_MANAGER, + CONN_ACK, + TIPC_OK, + port_out_seqno(p_ptr), + ack); + } + port_unlock(p_ptr); + net_route_msg(buf); +} + +/* + * tipc_createport(): user level call. Will add port to + * registry if non-zero user_ref. + */ + +int tipc_createport(u32 user_ref, + void *usr_handle, + unsigned int importance, + tipc_msg_err_event error_cb, + tipc_named_msg_err_event named_error_cb, + tipc_conn_shutdown_event conn_error_cb, + tipc_msg_event msg_cb, + tipc_named_msg_event named_msg_cb, + tipc_conn_msg_event conn_msg_cb, + tipc_continue_event continue_event_cb,/* May be zero */ + u32 *portref) +{ + struct user_port *up_ptr; + struct port *p_ptr; + u32 ref; + + up_ptr = (struct user_port *)kmalloc(sizeof(*up_ptr), GFP_ATOMIC); + if (up_ptr == NULL) { + return -ENOMEM; + } + ref = tipc_createport_raw(0, port_dispatcher, port_wakeup, importance); + p_ptr = port_lock(ref); + if (!p_ptr) { + kfree(up_ptr); + return -ENOMEM; + } + + p_ptr->user_port = up_ptr; + up_ptr->user_ref = user_ref; + up_ptr->usr_handle = usr_handle; + up_ptr->ref = p_ptr->publ.ref; + up_ptr->err_cb = error_cb; + up_ptr->named_err_cb = named_error_cb; + up_ptr->conn_err_cb = conn_error_cb; + up_ptr->msg_cb = msg_cb; + up_ptr->named_msg_cb = named_msg_cb; + up_ptr->conn_msg_cb = conn_msg_cb; + up_ptr->continue_event_cb = continue_event_cb; + INIT_LIST_HEAD(&up_ptr->uport_list); + reg_add_port(up_ptr); + *portref = p_ptr->publ.ref; + dbg(" tipc_createport: %x with ref %u\n", p_ptr, p_ptr->publ.ref); + port_unlock(p_ptr); + return TIPC_OK; +} + +int tipc_ownidentity(u32 ref, struct tipc_portid *id) +{ + id->ref = ref; + id->node = tipc_own_addr; + return TIPC_OK; +} + +int tipc_portimportance(u32 ref, unsigned int *importance) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + *importance = (unsigned int)msg_importance(&p_ptr->publ.phdr); + spin_unlock_bh(p_ptr->publ.lock); + return TIPC_OK; +} + +int tipc_set_portimportance(u32 ref, unsigned int imp) +{ + struct port *p_ptr; + + if (imp > TIPC_CRITICAL_IMPORTANCE) + return -EINVAL; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + msg_set_importance(&p_ptr->publ.phdr, (u32)imp); + spin_unlock_bh(p_ptr->publ.lock); + return TIPC_OK; +} + + +int tipc_publish(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) +{ + struct port *p_ptr; + struct publication *publ; + u32 key; + int res = -EINVAL; + + p_ptr = port_lock(ref); + dbg("tipc_publ %u, p_ptr = %x, conn = %x, scope = %x, " + "lower = %u, upper = %u\n", + ref, p_ptr, p_ptr->publ.connected, scope, seq->lower, seq->upper); + if (!p_ptr) + return -EINVAL; + if (p_ptr->publ.connected) + goto exit; + if (seq->lower > seq->upper) + goto exit; + if ((scope < TIPC_ZONE_SCOPE) || (scope > TIPC_NODE_SCOPE)) + goto exit; + key = ref + p_ptr->pub_count + 1; + if (key == ref) { + res = -EADDRINUSE; + goto exit; + } + publ = nametbl_publish(seq->type, seq->lower, seq->upper, + scope, p_ptr->publ.ref, key); + if (publ) { + list_add(&publ->pport_list, &p_ptr->publications); + p_ptr->pub_count++; + p_ptr->publ.published = 1; + res = TIPC_OK; + } +exit: + port_unlock(p_ptr); + return res; +} + +int tipc_withdraw(u32 ref, unsigned int scope, struct tipc_name_seq const *seq) +{ + struct port *p_ptr; + struct publication *publ; + struct publication *tpubl; + int res = -EINVAL; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + if (!p_ptr->publ.published) + goto exit; + if (!seq) { + list_for_each_entry_safe(publ, tpubl, + &p_ptr->publications, pport_list) { + nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + } + res = TIPC_OK; + } else { + list_for_each_entry_safe(publ, tpubl, + &p_ptr->publications, pport_list) { + if (publ->scope != scope) + continue; + if (publ->type != seq->type) + continue; + if (publ->lower != seq->lower) + continue; + if (publ->upper != seq->upper) + break; + nametbl_withdraw(publ->type, publ->lower, + publ->ref, publ->key); + res = TIPC_OK; + break; + } + } + if (list_empty(&p_ptr->publications)) + p_ptr->publ.published = 0; +exit: + port_unlock(p_ptr); + return res; +} + +int tipc_connect2port(u32 ref, struct tipc_portid const *peer) +{ + struct port *p_ptr; + struct tipc_msg *msg; + int res = -EINVAL; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + if (p_ptr->publ.published || p_ptr->publ.connected) + goto exit; + if (!peer->ref) + goto exit; + + msg = &p_ptr->publ.phdr; + msg_set_destnode(msg, peer->node); + msg_set_destport(msg, peer->ref); + msg_set_orignode(msg, tipc_own_addr); + msg_set_origport(msg, p_ptr->publ.ref); + msg_set_transp_seqno(msg, 42); + msg_set_type(msg, TIPC_CONN_MSG); + if (!may_route(peer->node)) + msg_set_hdr_sz(msg, SHORT_H_SIZE); + else + msg_set_hdr_sz(msg, LONG_H_SIZE); + + p_ptr->probing_interval = PROBING_INTERVAL; + p_ptr->probing_state = CONFIRMED; + p_ptr->publ.connected = 1; + k_start_timer(&p_ptr->timer, p_ptr->probing_interval); + + nodesub_subscribe(&p_ptr->subscription,peer->node, + (void *)(unsigned long)ref, + (net_ev_handler)port_handle_node_down); + res = TIPC_OK; +exit: + port_unlock(p_ptr); + p_ptr->max_pkt = link_get_max_pkt(peer->node, ref); + return res; +} + +/* + * tipc_disconnect(): Disconnect port form peer. + * This is a node local operation. + */ + +int tipc_disconnect(u32 ref) +{ + struct port *p_ptr; + int res = -ENOTCONN; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + if (p_ptr->publ.connected) { + p_ptr->publ.connected = 0; + /* let timer expire on it's own to avoid deadlock! */ + nodesub_unsubscribe(&p_ptr->subscription); + res = TIPC_OK; + } + port_unlock(p_ptr); + return res; +} + +/* + * tipc_shutdown(): Send a SHUTDOWN msg to peer and disconnect + */ +int tipc_shutdown(u32 ref) +{ + struct port *p_ptr; + struct sk_buff *buf = 0; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + + if (p_ptr->publ.connected) { + u32 imp = msg_importance(&p_ptr->publ.phdr); + if (imp < TIPC_CRITICAL_IMPORTANCE) + imp++; + buf = port_build_proto_msg(port_peerport(p_ptr), + port_peernode(p_ptr), + ref, + tipc_own_addr, + imp, + TIPC_CONN_MSG, + TIPC_CONN_SHUTDOWN, + port_out_seqno(p_ptr), + 0); + } + port_unlock(p_ptr); + net_route_msg(buf); + return tipc_disconnect(ref); +} + +int tipc_isconnected(u32 ref, int *isconnected) +{ + struct port *p_ptr; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + *isconnected = p_ptr->publ.connected; + port_unlock(p_ptr); + return TIPC_OK; +} + +int tipc_peer(u32 ref, struct tipc_portid *peer) +{ + struct port *p_ptr; + int res; + + p_ptr = port_lock(ref); + if (!p_ptr) + return -EINVAL; + if (p_ptr->publ.connected) { + peer->ref = port_peerport(p_ptr); + peer->node = port_peernode(p_ptr); + res = TIPC_OK; + } else + res = -ENOTCONN; + port_unlock(p_ptr); + return res; +} + +int tipc_ref_valid(u32 ref) +{ + /* Works irrespective of type */ + return !!ref_deref(ref); +} + + +/* + * port_recv_sections(): Concatenate and deliver sectioned + * message for this node. + */ + +int port_recv_sections(struct port *sender, unsigned int num_sect, + struct iovec const *msg_sect) +{ + struct sk_buff *buf; + int res; + + res = msg_build(&sender->publ.phdr, msg_sect, num_sect, + MAX_MSG_SIZE, !sender->user_port, &buf); + if (likely(buf)) + port_recv_msg(buf); + return res; +} + +/** + * tipc_send - send message sections on connection + */ + +int tipc_send(u32 ref, unsigned int num_sect, struct iovec const *msg_sect) +{ + struct port *p_ptr; + u32 destnode; + int res; + + p_ptr = port_deref(ref); + if (!p_ptr || !p_ptr->publ.connected) + return -EINVAL; + + p_ptr->publ.congested = 1; + if (!port_congested(p_ptr)) { + destnode = port_peernode(p_ptr); + if (likely(destnode != tipc_own_addr)) + res = link_send_sections_fast(p_ptr, msg_sect, num_sect, + destnode); + else + res = port_recv_sections(p_ptr, num_sect, msg_sect); + + if (likely(res != -ELINKCONG)) { + port_incr_out_seqno(p_ptr); + p_ptr->publ.congested = 0; + p_ptr->sent++; + return res; + } + } + if (port_unreliable(p_ptr)) { + p_ptr->publ.congested = 0; + /* Just calculate msg length and return */ + return msg_calc_data_size(msg_sect, num_sect); + } + return -ELINKCONG; +} + +/** + * tipc_send_buf - send message buffer on connection + */ + +int tipc_send_buf(u32 ref, struct sk_buff *buf, unsigned int dsz) +{ + struct port *p_ptr; + struct tipc_msg *msg; + u32 destnode; + u32 hsz; + u32 sz; + u32 res; + + p_ptr = port_deref(ref); + if (!p_ptr || !p_ptr->publ.connected) + return -EINVAL; + + msg = &p_ptr->publ.phdr; + hsz = msg_hdr_sz(msg); + sz = hsz + dsz; + msg_set_size(msg, sz); + if (skb_cow(buf, hsz)) + return -ENOMEM; + + skb_push(buf, hsz); + memcpy(buf->data, (unchar *)msg, hsz); + destnode = msg_destnode(msg); + p_ptr->publ.congested = 1; + if (!port_congested(p_ptr)) { + if (likely(destnode != tipc_own_addr)) + res = tipc_send_buf_fast(buf, destnode); + else { + port_recv_msg(buf); + res = sz; + } + if (likely(res != -ELINKCONG)) { + port_incr_out_seqno(p_ptr); + p_ptr->sent++; + p_ptr->publ.congested = 0; + return res; + } + } + if (port_unreliable(p_ptr)) { + p_ptr->publ.congested = 0; + return dsz; + } + return -ELINKCONG; +} + +/** + * tipc_forward2name - forward message sections to port name + */ + +int tipc_forward2name(u32 ref, + struct tipc_name const *name, + u32 domain, + u32 num_sect, + struct iovec const *msg_sect, + struct tipc_portid const *orig, + unsigned int importance) +{ + struct port *p_ptr; + struct tipc_msg *msg; + u32 destnode = domain; + u32 destport = 0; + int res; + + p_ptr = port_deref(ref); + if (!p_ptr || p_ptr->publ.connected) + return -EINVAL; + + msg = &p_ptr->publ.phdr; + msg_set_type(msg, TIPC_NAMED_MSG); + msg_set_orignode(msg, orig->node); + msg_set_origport(msg, orig->ref); + msg_set_hdr_sz(msg, LONG_H_SIZE); + msg_set_nametype(msg, name->type); + msg_set_nameinst(msg, name->instance); + msg_set_lookup_scope(msg, addr_scope(domain)); + if (importance <= TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(msg,importance); + destport = nametbl_translate(name->type, name->instance, &destnode); + msg_set_destnode(msg, destnode); + msg_set_destport(msg, destport); + + if (likely(destport || destnode)) { + p_ptr->sent++; + if (likely(destnode == tipc_own_addr)) + return port_recv_sections(p_ptr, num_sect, msg_sect); + res = link_send_sections_fast(p_ptr, msg_sect, num_sect, + destnode); + if (likely(res != -ELINKCONG)) + return res; + if (port_unreliable(p_ptr)) { + /* Just calculate msg length and return */ + return msg_calc_data_size(msg_sect, num_sect); + } + return -ELINKCONG; + } + return port_reject_sections(p_ptr, msg, msg_sect, num_sect, + TIPC_ERR_NO_NAME); +} + +/** + * tipc_send2name - send message sections to port name + */ + +int tipc_send2name(u32 ref, + struct tipc_name const *name, + unsigned int domain, + unsigned int num_sect, + struct iovec const *msg_sect) +{ + struct tipc_portid orig; + + orig.ref = ref; + orig.node = tipc_own_addr; + return tipc_forward2name(ref, name, domain, num_sect, msg_sect, &orig, + TIPC_PORT_IMPORTANCE); +} + +/** + * tipc_forward_buf2name - forward message buffer to port name + */ + +int tipc_forward_buf2name(u32 ref, + struct tipc_name const *name, + u32 domain, + struct sk_buff *buf, + unsigned int dsz, + struct tipc_portid const *orig, + unsigned int importance) +{ + struct port *p_ptr; + struct tipc_msg *msg; + u32 destnode = domain; + u32 destport = 0; + int res; + + p_ptr = (struct port *)ref_deref(ref); + if (!p_ptr || p_ptr->publ.connected) + return -EINVAL; + + msg = &p_ptr->publ.phdr; + if (importance <= TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(msg, importance); + msg_set_type(msg, TIPC_NAMED_MSG); + msg_set_orignode(msg, orig->node); + msg_set_origport(msg, orig->ref); + msg_set_nametype(msg, name->type); + msg_set_nameinst(msg, name->instance); + msg_set_lookup_scope(msg, addr_scope(domain)); + msg_set_hdr_sz(msg, LONG_H_SIZE); + msg_set_size(msg, LONG_H_SIZE + dsz); + destport = nametbl_translate(name->type, name->instance, &destnode); + msg_set_destnode(msg, destnode); + msg_set_destport(msg, destport); + msg_dbg(msg, "forw2name ==> "); + if (skb_cow(buf, LONG_H_SIZE)) + return -ENOMEM; + skb_push(buf, LONG_H_SIZE); + memcpy(buf->data, (unchar *)msg, LONG_H_SIZE); + msg_dbg(buf_msg(buf),"PREP:"); + if (likely(destport || destnode)) { + p_ptr->sent++; + if (destnode == tipc_own_addr) + return port_recv_msg(buf); + res = tipc_send_buf_fast(buf, destnode); + if (likely(res != -ELINKCONG)) + return res; + if (port_unreliable(p_ptr)) + return dsz; + return -ELINKCONG; + } + return tipc_reject_msg(buf, TIPC_ERR_NO_NAME); +} + +/** + * tipc_send_buf2name - send message buffer to port name + */ + +int tipc_send_buf2name(u32 ref, + struct tipc_name const *dest, + u32 domain, + struct sk_buff *buf, + unsigned int dsz) +{ + struct tipc_portid orig; + + orig.ref = ref; + orig.node = tipc_own_addr; + return tipc_forward_buf2name(ref, dest, domain, buf, dsz, &orig, + TIPC_PORT_IMPORTANCE); +} + +/** + * tipc_forward2port - forward message sections to port identity + */ + +int tipc_forward2port(u32 ref, + struct tipc_portid const *dest, + unsigned int num_sect, + struct iovec const *msg_sect, + struct tipc_portid const *orig, + unsigned int importance) +{ + struct port *p_ptr; + struct tipc_msg *msg; + int res; + + p_ptr = port_deref(ref); + if (!p_ptr || p_ptr->publ.connected) + return -EINVAL; + + msg = &p_ptr->publ.phdr; + msg_set_type(msg, TIPC_DIRECT_MSG); + msg_set_orignode(msg, orig->node); + msg_set_origport(msg, orig->ref); + msg_set_destnode(msg, dest->node); + msg_set_destport(msg, dest->ref); + msg_set_hdr_sz(msg, DIR_MSG_H_SIZE); + if (importance <= TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(msg, importance); + p_ptr->sent++; + if (dest->node == tipc_own_addr) + return port_recv_sections(p_ptr, num_sect, msg_sect); + res = link_send_sections_fast(p_ptr, msg_sect, num_sect, dest->node); + if (likely(res != -ELINKCONG)) + return res; + if (port_unreliable(p_ptr)) { + /* Just calculate msg length and return */ + return msg_calc_data_size(msg_sect, num_sect); + } + return -ELINKCONG; +} + +/** + * tipc_send2port - send message sections to port identity + */ + +int tipc_send2port(u32 ref, + struct tipc_portid const *dest, + unsigned int num_sect, + struct iovec const *msg_sect) +{ + struct tipc_portid orig; + + orig.ref = ref; + orig.node = tipc_own_addr; + return tipc_forward2port(ref, dest, num_sect, msg_sect, &orig, + TIPC_PORT_IMPORTANCE); +} + +/** + * tipc_forward_buf2port - forward message buffer to port identity + */ +int tipc_forward_buf2port(u32 ref, + struct tipc_portid const *dest, + struct sk_buff *buf, + unsigned int dsz, + struct tipc_portid const *orig, + unsigned int importance) +{ + struct port *p_ptr; + struct tipc_msg *msg; + int res; + + p_ptr = (struct port *)ref_deref(ref); + if (!p_ptr || p_ptr->publ.connected) + return -EINVAL; + + msg = &p_ptr->publ.phdr; + msg_set_type(msg, TIPC_DIRECT_MSG); + msg_set_orignode(msg, orig->node); + msg_set_origport(msg, orig->ref); + msg_set_destnode(msg, dest->node); + msg_set_destport(msg, dest->ref); + msg_set_hdr_sz(msg, DIR_MSG_H_SIZE); + if (importance <= TIPC_CRITICAL_IMPORTANCE) + msg_set_importance(msg, importance); + msg_set_size(msg, DIR_MSG_H_SIZE + dsz); + if (skb_cow(buf, DIR_MSG_H_SIZE)) + return -ENOMEM; + + skb_push(buf, DIR_MSG_H_SIZE); + memcpy(buf->data, (unchar *)msg, DIR_MSG_H_SIZE); + msg_dbg(msg, "buf2port: "); + p_ptr->sent++; + if (dest->node == tipc_own_addr) + return port_recv_msg(buf); + res = tipc_send_buf_fast(buf, dest->node); + if (likely(res != -ELINKCONG)) + return res; + if (port_unreliable(p_ptr)) + return dsz; + return -ELINKCONG; +} + +/** + * tipc_send_buf2port - send message buffer to port identity + */ + +int tipc_send_buf2port(u32 ref, + struct tipc_portid const *dest, + struct sk_buff *buf, + unsigned int dsz) +{ + struct tipc_portid orig; + + orig.ref = ref; + orig.node = tipc_own_addr; + return tipc_forward_buf2port(ref, dest, buf, dsz, &orig, + TIPC_PORT_IMPORTANCE); +} + diff --git a/net/tipc/port.h b/net/tipc/port.h new file mode 100644 index 00000000000..e829a99d3b7 --- /dev/null +++ b/net/tipc/port.h @@ -0,0 +1,209 @@ +/* + * net/tipc/port.h: Include file for TIPC port code + * + * Copyright (c) 1994-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_PORT_H +#define _TIPC_PORT_H + +#include <net/tipc/tipc_port.h> +#include "ref.h" +#include "net.h" +#include "msg.h" +#include "dbg.h" +#include "node_subscr.h" + +/** + * struct user_port - TIPC user port (used with native API) + * @user_ref: id of user who created user port + * @usr_handle: user-specified field + * @ref: object reference to associated TIPC port + * <various callback routines> + * @uport_list: adjacent user ports in list of ports held by user + */ + +struct user_port { + u32 user_ref; + void *usr_handle; + u32 ref; + tipc_msg_err_event err_cb; + tipc_named_msg_err_event named_err_cb; + tipc_conn_shutdown_event conn_err_cb; + tipc_msg_event msg_cb; + tipc_named_msg_event named_msg_cb; + tipc_conn_msg_event conn_msg_cb; + tipc_continue_event continue_event_cb; + struct list_head uport_list; +}; + +/** + * struct port - TIPC port structure + * @publ: TIPC port info available to privileged users + * @port_list: adjacent ports in TIPC's global list of ports + * @dispatcher: ptr to routine which handles received messages + * @wakeup: ptr to routine to call when port is no longer congested + * @user_port: ptr to user port associated with port (if any) + * @wait_list: adjacent ports in list of ports waiting on link congestion + * @congested_link: ptr to congested link port is waiting on + * @waiting_pkts: + * @sent: + * @acked: + * @publications: list of publications for port + * @pub_count: total # of publications port has made during its lifetime + * @max_pkt: maximum packet size "hint" used when building messages sent by port + * @probing_state: + * @probing_interval: + * @last_in_seqno: + * @timer_ref: + * @subscription: "node down" subscription used to terminate failed connections + */ + +struct port { + struct tipc_port publ; + struct list_head port_list; + u32 (*dispatcher)(struct tipc_port *, struct sk_buff *); + void (*wakeup)(struct tipc_port *); + struct user_port *user_port; + struct list_head wait_list; + struct link *congested_link; + u32 waiting_pkts; + u32 sent; + u32 acked; + struct list_head publications; + u32 pub_count; + u32 max_pkt; + u32 probing_state; + u32 probing_interval; + u32 last_in_seqno; + struct timer_list timer; + struct node_subscr subscription; +}; + +extern spinlock_t port_list_lock; +struct port_list; + +int port_recv_sections(struct port *p_ptr, u32 num_sect, + struct iovec const *msg_sect); +int port_reject_sections(struct port *p_ptr, struct tipc_msg *hdr, + struct iovec const *msg_sect, u32 num_sect, + int err); +struct sk_buff *port_get_ports(void); +struct sk_buff *port_show_stats(const void *req_tlv_area, int req_tlv_space); +void port_recv_proto_msg(struct sk_buff *buf); +void port_recv_mcast(struct sk_buff *buf, struct port_list *dp); +void port_reinit(void); + +/** + * port_lock - lock port instance referred to and return its pointer + */ + +static inline struct port *port_lock(u32 ref) +{ + return (struct port *)ref_lock(ref); +} + +/** + * port_unlock - unlock a port instance + * + * Can use pointer instead of ref_unlock() since port is already locked. + */ + +static inline void port_unlock(struct port *p_ptr) +{ + spin_unlock_bh(p_ptr->publ.lock); +} + +static inline struct port* port_deref(u32 ref) +{ + return (struct port *)ref_deref(ref); +} + +static inline u32 peer_port(struct port *p_ptr) +{ + return msg_destport(&p_ptr->publ.phdr); +} + +static inline u32 peer_node(struct port *p_ptr) +{ + return msg_destnode(&p_ptr->publ.phdr); +} + +static inline int port_congested(struct port *p_ptr) +{ + return((p_ptr->sent - p_ptr->acked) >= (TIPC_FLOW_CONTROL_WIN * 2)); +} + +/** + * port_recv_msg - receive message from lower layer and deliver to port user + */ + +static inline int port_recv_msg(struct sk_buff *buf) +{ + struct port *p_ptr; + struct tipc_msg *msg = buf_msg(buf); + u32 destport = msg_destport(msg); + u32 dsz = msg_data_sz(msg); + u32 err; + + /* forward unresolved named message */ + if (unlikely(!destport)) { + net_route_msg(buf); + return dsz; + } + + /* validate destination & pass to port, otherwise reject message */ + p_ptr = port_lock(destport); + if (likely(p_ptr)) { + if (likely(p_ptr->publ.connected)) { + if ((unlikely(msg_origport(msg) != peer_port(p_ptr))) || + (unlikely(msg_orignode(msg) != peer_node(p_ptr))) || + (unlikely(!msg_connected(msg)))) { + err = TIPC_ERR_NO_PORT; + port_unlock(p_ptr); + goto reject; + } + } + err = p_ptr->dispatcher(&p_ptr->publ, buf); + port_unlock(p_ptr); + if (likely(!err)) + return dsz; + } else { + err = TIPC_ERR_NO_PORT; + } +reject: + dbg("port->rejecting, err = %x..\n",err); + return tipc_reject_msg(buf, err); +} + +#endif diff --git a/net/tipc/ref.c b/net/tipc/ref.c new file mode 100644 index 00000000000..944093fe246 --- /dev/null +++ b/net/tipc/ref.c @@ -0,0 +1,189 @@ +/* + * net/tipc/ref.c: TIPC object registry code + * + * Copyright (c) 1991-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "ref.h" +#include "port.h" +#include "subscr.h" +#include "name_distr.h" +#include "name_table.h" +#include "config.h" +#include "discover.h" +#include "bearer.h" +#include "node.h" +#include "bcast.h" + +/* + * Object reference table consists of 2**N entries. + * + * A used entry has object ptr != 0, reference == XXXX|own index + * (XXXX changes each time entry is acquired) + * A free entry has object ptr == 0, reference == YYYY|next free index + * (YYYY is one more than last used XXXX) + * + * Free list is initially chained from entry (2**N)-1 to entry 1. + * Entry 0 is not used to allow index 0 to indicate the end of the free list. + * + * Note: Any accidental reference of the form XXXX|0--0 won't match entry 0 + * because entry 0's reference field has the form XXXX|1--1. + */ + +struct ref_table ref_table = { 0 }; + +rwlock_t reftbl_lock = RW_LOCK_UNLOCKED; + +/** + * ref_table_init - create reference table for objects + */ + +int ref_table_init(u32 requested_size, u32 start) +{ + struct reference *table; + u32 sz = 1 << 4; + u32 index_mask; + int i; + + while (sz < requested_size) { + sz <<= 1; + } + table = (struct reference *)vmalloc(sz * sizeof(struct reference)); + if (table == NULL) + return -ENOMEM; + + write_lock_bh(&reftbl_lock); + index_mask = sz - 1; + for (i = sz - 1; i >= 0; i--) { + table[i].object = 0; + table[i].lock = SPIN_LOCK_UNLOCKED; + table[i].data.next_plus_upper = (start & ~index_mask) + i - 1; + } + ref_table.entries = table; + ref_table.index_mask = index_mask; + ref_table.first_free = sz - 1; + ref_table.last_free = 1; + write_unlock_bh(&reftbl_lock); + return TIPC_OK; +} + +/** + * ref_table_stop - destroy reference table for objects + */ + +void ref_table_stop(void) +{ + if (!ref_table.entries) + return; + + vfree(ref_table.entries); + ref_table.entries = 0; +} + +/** + * ref_acquire - create reference to an object + * + * Return a unique reference value which can be translated back to the pointer + * 'object' at a later time. Also, pass back a pointer to the lock protecting + * the object, but without locking it. + */ + +u32 ref_acquire(void *object, spinlock_t **lock) +{ + struct reference *entry; + u32 index; + u32 index_mask; + u32 next_plus_upper; + u32 reference = 0; + + assert(ref_table.entries && object); + + write_lock_bh(&reftbl_lock); + if (ref_table.first_free) { + index = ref_table.first_free; + entry = &(ref_table.entries[index]); + index_mask = ref_table.index_mask; + /* take lock in case a previous user of entry still holds it */ + spin_lock_bh(&entry->lock); + next_plus_upper = entry->data.next_plus_upper; + ref_table.first_free = next_plus_upper & index_mask; + reference = (next_plus_upper & ~index_mask) + index; + entry->data.reference = reference; + entry->object = object; + if (lock != 0) + *lock = &entry->lock; + spin_unlock_bh(&entry->lock); + } + write_unlock_bh(&reftbl_lock); + return reference; +} + +/** + * ref_discard - invalidate references to an object + * + * Disallow future references to an object and free up the entry for re-use. + * Note: The entry's spin_lock may still be busy after discard + */ + +void ref_discard(u32 ref) +{ + struct reference *entry; + u32 index; + u32 index_mask; + + assert(ref_table.entries); + assert(ref != 0); + + write_lock_bh(&reftbl_lock); + index_mask = ref_table.index_mask; + index = ref & index_mask; + entry = &(ref_table.entries[index]); + assert(entry->object != 0); + assert(entry->data.reference == ref); + + /* mark entry as unused */ + entry->object = 0; + if (ref_table.first_free == 0) + ref_table.first_free = index; + else + /* next_plus_upper is always XXXX|0--0 for last free entry */ + ref_table.entries[ref_table.last_free].data.next_plus_upper + |= index; + ref_table.last_free = index; + + /* increment upper bits of entry to invalidate subsequent references */ + entry->data.next_plus_upper = (ref & ~index_mask) + (index_mask + 1); + write_unlock_bh(&reftbl_lock); +} + diff --git a/net/tipc/ref.h b/net/tipc/ref.h new file mode 100644 index 00000000000..429cde57228 --- /dev/null +++ b/net/tipc/ref.h @@ -0,0 +1,131 @@ +/* + * net/tipc/ref.h: Include file for TIPC object registry code + * + * Copyright (c) 1991-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_REF_H +#define _TIPC_REF_H + +/** + * struct reference - TIPC object reference entry + * @object: pointer to object associated with reference entry + * @lock: spinlock controlling access to object + * @data: reference value associated with object (or link to next unused entry) + */ + +struct reference { + void *object; + spinlock_t lock; + union { + u32 next_plus_upper; + u32 reference; + } data; +}; + +/** + * struct ref_table - table of TIPC object reference entries + * @entries: pointer to array of reference entries + * @index_mask: bitmask for array index portion of reference values + * @first_free: array index of first unused object reference entry + * @last_free: array index of last unused object reference entry + */ + +struct ref_table { + struct reference *entries; + u32 index_mask; + u32 first_free; + u32 last_free; +}; + +extern struct ref_table ref_table; + +int ref_table_init(u32 requested_size, u32 start); +void ref_table_stop(void); + +u32 ref_acquire(void *object, spinlock_t **lock); +void ref_discard(u32 ref); + + +/** + * ref_lock - lock referenced object and return pointer to it + */ + +static inline void *ref_lock(u32 ref) +{ + if (likely(ref_table.entries)) { + struct reference *r = + &ref_table.entries[ref & ref_table.index_mask]; + + spin_lock_bh(&r->lock); + if (likely(r->data.reference == ref)) + return r->object; + spin_unlock_bh(&r->lock); + } + return 0; +} + +/** + * ref_unlock - unlock referenced object + */ + +static inline void ref_unlock(u32 ref) +{ + if (likely(ref_table.entries)) { + struct reference *r = + &ref_table.entries[ref & ref_table.index_mask]; + + if (likely(r->data.reference == ref)) + spin_unlock_bh(&r->lock); + else + err("ref_unlock() invoked using obsolete reference\n"); + } +} + +/** + * ref_deref - return pointer referenced object (without locking it) + */ + +static inline void *ref_deref(u32 ref) +{ + if (likely(ref_table.entries)) { + struct reference *r = + &ref_table.entries[ref & ref_table.index_mask]; + + if (likely(r->data.reference == ref)) + return r->object; + } + return 0; +} + +#endif diff --git a/net/tipc/socket.c b/net/tipc/socket.c new file mode 100644 index 00000000000..d21f8c0cd25 --- /dev/null +++ b/net/tipc/socket.c @@ -0,0 +1,1726 @@ +/* + * net/tipc/socket.c: TIPC socket API + * + * Copyright (c) 2001-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include <linux/module.h> +#include <linux/types.h> +#include <linux/net.h> +#include <linux/socket.h> +#include <linux/errno.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/poll.h> +#include <linux/version.h> +#include <linux/fcntl.h> +#include <linux/version.h> +#include <asm/semaphore.h> +#include <asm/string.h> +#include <asm/atomic.h> +#include <net/sock.h> + +#include <linux/tipc.h> +#include <linux/tipc_config.h> +#include <net/tipc/tipc_msg.h> +#include <net/tipc/tipc_port.h> + +#include "core.h" + +#define SS_LISTENING -1 /* socket is listening */ +#define SS_READY -2 /* socket is connectionless */ + +#define OVERLOAD_LIMIT_BASE 5000 + +struct tipc_sock { + struct sock sk; + struct tipc_port *p; + struct semaphore sem; +}; + +#define tipc_sk(sk) ((struct tipc_sock*)sk) + +static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf); +static void wakeupdispatch(struct tipc_port *tport); + +static struct proto_ops packet_ops; +static struct proto_ops stream_ops; +static struct proto_ops msg_ops; + +static struct proto tipc_proto; + +static int sockets_enabled = 0; + +static atomic_t tipc_queue_size = ATOMIC_INIT(0); + + +/* + * sock_lock(): Lock a port/socket pair. lock_sock() can + * not be used here, since the same lock must protect ports + * with non-socket interfaces. + * See net.c for description of locking policy. + */ +static inline void sock_lock(struct tipc_sock* tsock) +{ + spin_lock_bh(tsock->p->lock); +} + +/* + * sock_unlock(): Unlock a port/socket pair + */ +static inline void sock_unlock(struct tipc_sock* tsock) +{ + spin_unlock_bh(tsock->p->lock); +} + +/** + * pollmask - determine the current set of poll() events for a socket + * @sock: socket structure + * + * TIPC sets the returned events as follows: + * a) POLLRDNORM and POLLIN are set if the socket's receive queue is non-empty + * or if a connection-oriented socket is does not have an active connection + * (i.e. a read operation will not block). + * b) POLLOUT is set except when a socket's connection has been terminated + * (i.e. a write operation will not block). + * c) POLLHUP is set when a socket's connection has been terminated. + * + * IMPORTANT: The fact that a read or write operation will not block does NOT + * imply that the operation will succeed! + * + * Returns pollmask value + */ + +static inline u32 pollmask(struct socket *sock) +{ + u32 mask; + + if ((skb_queue_len(&sock->sk->sk_receive_queue) != 0) || + (sock->state == SS_UNCONNECTED) || + (sock->state == SS_DISCONNECTING)) + mask = (POLLRDNORM | POLLIN); + else + mask = 0; + + if (sock->state == SS_DISCONNECTING) + mask |= POLLHUP; + else + mask |= POLLOUT; + + return mask; +} + + +/** + * advance_queue - discard first buffer in queue + * @tsock: TIPC socket + */ + +static inline void advance_queue(struct tipc_sock *tsock) +{ + sock_lock(tsock); + buf_discard(skb_dequeue(&tsock->sk.sk_receive_queue)); + sock_unlock(tsock); + atomic_dec(&tipc_queue_size); +} + +/** + * tipc_create - create a TIPC socket + * @sock: pre-allocated socket structure + * @protocol: protocol indicator (must be 0) + * + * This routine creates and attaches a 'struct sock' to the 'struct socket', + * then create and attaches a TIPC port to the 'struct sock' part. + * + * Returns 0 on success, errno otherwise + */ +static int tipc_create(struct socket *sock, int protocol) +{ + struct tipc_sock *tsock; + struct tipc_port *port; + struct sock *sk; + u32 ref; + + if ((sock->type != SOCK_STREAM) && + (sock->type != SOCK_SEQPACKET) && + (sock->type != SOCK_DGRAM) && + (sock->type != SOCK_RDM)) + return -EPROTOTYPE; + + if (unlikely(protocol != 0)) + return -EPROTONOSUPPORT; + + ref = tipc_createport_raw(0, &dispatch, &wakeupdispatch, TIPC_LOW_IMPORTANCE); + if (unlikely(!ref)) + return -ENOMEM; + + sock->state = SS_UNCONNECTED; + + switch (sock->type) { + case SOCK_STREAM: + sock->ops = &stream_ops; + break; + case SOCK_SEQPACKET: + sock->ops = &packet_ops; + break; + case SOCK_DGRAM: + tipc_set_portunreliable(ref, 1); + /* fall through */ + case SOCK_RDM: + tipc_set_portunreturnable(ref, 1); + sock->ops = &msg_ops; + sock->state = SS_READY; + break; + } + + sk = sk_alloc(AF_TIPC, GFP_KERNEL, &tipc_proto, 1); + if (!sk) { + tipc_deleteport(ref); + return -ENOMEM; + } + + sock_init_data(sock, sk); + init_waitqueue_head(sk->sk_sleep); + sk->sk_rcvtimeo = 8 * HZ; /* default connect timeout = 8s */ + + tsock = tipc_sk(sk); + port = tipc_get_port(ref); + + tsock->p = port; + port->usr_handle = tsock; + + init_MUTEX(&tsock->sem); + + dbg("sock_create: %x\n",tsock); + + atomic_inc(&tipc_user_count); + + return 0; +} + +/** + * release - destroy a TIPC socket + * @sock: socket to destroy + * + * This routine cleans up any messages that are still queued on the socket. + * For DGRAM and RDM socket types, all queued messages are rejected. + * For SEQPACKET and STREAM socket types, the first message is rejected + * and any others are discarded. (If the first message on a STREAM socket + * is partially-read, it is discarded and the next one is rejected instead.) + * + * NOTE: Rejected messages are not necessarily returned to the sender! They + * are returned or discarded according to the "destination droppable" setting + * specified for the message by the sender. + * + * Returns 0 on success, errno otherwise + */ + +static int release(struct socket *sock) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + int res = TIPC_OK; + struct sk_buff *buf; + + dbg("sock_delete: %x\n",tsock); + if (!tsock) + return 0; + down_interruptible(&tsock->sem); + if (!sock->sk) { + up(&tsock->sem); + return 0; + } + + /* Reject unreceived messages, unless no longer connected */ + + while (sock->state != SS_DISCONNECTING) { + sock_lock(tsock); + buf = skb_dequeue(&sk->sk_receive_queue); + if (!buf) + tsock->p->usr_handle = 0; + sock_unlock(tsock); + if (!buf) + break; + if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) + buf_discard(buf); + else + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + atomic_dec(&tipc_queue_size); + } + + /* Delete TIPC port */ + + res = tipc_deleteport(tsock->p->ref); + sock->sk = NULL; + + /* Discard any remaining messages */ + + while ((buf = skb_dequeue(&sk->sk_receive_queue))) { + buf_discard(buf); + atomic_dec(&tipc_queue_size); + } + + up(&tsock->sem); + + sock_put(sk); + + atomic_dec(&tipc_user_count); + return res; +} + +/** + * bind - associate or disassocate TIPC name(s) with a socket + * @sock: socket structure + * @uaddr: socket address describing name(s) and desired operation + * @uaddr_len: size of socket address data structure + * + * Name and name sequence binding is indicated using a positive scope value; + * a negative scope value unbinds the specified name. Specifying no name + * (i.e. a socket address length of 0) unbinds all names from the socket. + * + * Returns 0 on success, errno otherwise + */ + +static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + int res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + if (unlikely(!uaddr_len)) { + res = tipc_withdraw(tsock->p->ref, 0, 0); + goto exit; + } + + if (uaddr_len < sizeof(struct sockaddr_tipc)) { + res = -EINVAL; + goto exit; + } + + if (addr->family != AF_TIPC) { + res = -EAFNOSUPPORT; + goto exit; + } + if (addr->addrtype == TIPC_ADDR_NAME) + addr->addr.nameseq.upper = addr->addr.nameseq.lower; + else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { + res = -EAFNOSUPPORT; + goto exit; + } + + if (addr->scope > 0) + res = tipc_publish(tsock->p->ref, addr->scope, + &addr->addr.nameseq); + else + res = tipc_withdraw(tsock->p->ref, -addr->scope, + &addr->addr.nameseq); +exit: + up(&tsock->sem); + return res; +} + +/** + * get_name - get port ID of socket or peer socket + * @sock: socket structure + * @uaddr: area for returned socket address + * @uaddr_len: area for returned length of socket address + * @peer: 0 to obtain socket name, 1 to obtain peer socket name + * + * Returns 0 on success, errno otherwise + */ + +static int get_name(struct socket *sock, struct sockaddr *uaddr, + int *uaddr_len, int peer) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + u32 res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + *uaddr_len = sizeof(*addr); + addr->addrtype = TIPC_ADDR_ID; + addr->family = AF_TIPC; + addr->scope = 0; + if (peer) + res = tipc_peer(tsock->p->ref, &addr->addr.id); + else + res = tipc_ownidentity(tsock->p->ref, &addr->addr.id); + addr->addr.name.domain = 0; + + up(&tsock->sem); + return res; +} + +/** + * poll - read and possibly block on pollmask + * @file: file structure associated with the socket + * @sock: socket for which to calculate the poll bits + * @wait: ??? + * + * Returns the pollmask + */ + +static unsigned int poll(struct file *file, struct socket *sock, + poll_table *wait) +{ + poll_wait(file, sock->sk->sk_sleep, wait); + /* NEED LOCK HERE? */ + return pollmask(sock); +} + +/** + * dest_name_check - verify user is permitted to send to specified port name + * @dest: destination address + * @m: descriptor for message to be sent + * + * Prevents restricted configuration commands from being issued by + * unauthorized users. + * + * Returns 0 if permission is granted, otherwise errno + */ + +static inline int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) +{ + struct tipc_cfg_msg_hdr hdr; + + if (likely(dest->addr.name.name.type >= TIPC_RESERVED_TYPES)) + return 0; + if (likely(dest->addr.name.name.type == TIPC_TOP_SRV)) + return 0; + + if (likely(dest->addr.name.name.type != TIPC_CFG_SRV)) + return -EACCES; + + if (copy_from_user(&hdr, m->msg_iov[0].iov_base, sizeof(hdr))) + return -EFAULT; + if ((ntohs(hdr.tcm_type) & 0xC000) & (!capable(CAP_NET_ADMIN))) + return -EACCES; + + return 0; +} + +/** + * send_msg - send message in connectionless manner + * @iocb: (unused) + * @sock: socket structure + * @m: message to send + * @total_len: (unused) + * + * Message must have an destination specified explicitly. + * Used for SOCK_RDM and SOCK_DGRAM messages, + * and for 'SYN' messages on SOCK_SEQPACKET and SOCK_STREAM connections. + * (Note: 'SYN+' is prohibited on SOCK_STREAM.) + * + * Returns the number of bytes sent on success, or errno otherwise + */ + +static int send_msg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; + struct sk_buff *buf; + int needs_conn; + int res = -EINVAL; + + if (unlikely(!dest)) + return -EDESTADDRREQ; + if (unlikely(dest->family != AF_TIPC)) + return -EINVAL; + + needs_conn = (sock->state != SS_READY); + if (unlikely(needs_conn)) { + if (sock->state == SS_LISTENING) + return -EPIPE; + if (sock->state != SS_UNCONNECTED) + return -EISCONN; + if ((tsock->p->published) || + ((sock->type == SOCK_STREAM) && (total_len != 0))) + return -EOPNOTSUPP; + } + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + if (needs_conn) { + + /* Abort any pending connection attempts (very unlikely) */ + + while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + atomic_dec(&tipc_queue_size); + } + + sock->state = SS_CONNECTING; + } + + do { + if (dest->addrtype == TIPC_ADDR_NAME) { + if ((res = dest_name_check(dest, m))) + goto exit; + res = tipc_send2name(tsock->p->ref, + &dest->addr.name.name, + dest->addr.name.domain, + m->msg_iovlen, + m->msg_iov); + } + else if (dest->addrtype == TIPC_ADDR_ID) { + res = tipc_send2port(tsock->p->ref, + &dest->addr.id, + m->msg_iovlen, + m->msg_iov); + } + else if (dest->addrtype == TIPC_ADDR_MCAST) { + if (needs_conn) { + res = -EOPNOTSUPP; + goto exit; + } + if ((res = dest_name_check(dest, m))) + goto exit; + res = tipc_multicast(tsock->p->ref, + &dest->addr.nameseq, + 0, + m->msg_iovlen, + m->msg_iov); + } + if (likely(res != -ELINKCONG)) { +exit: + up(&tsock->sem); + return res; + } + if (m->msg_flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + if (wait_event_interruptible(*sock->sk->sk_sleep, + !tsock->p->congested)) { + res = -ERESTARTSYS; + goto exit; + } + } while (1); +} + +/** + * send_packet - send a connection-oriented message + * @iocb: (unused) + * @sock: socket structure + * @m: message to send + * @total_len: (unused) + * + * Used for SOCK_SEQPACKET messages and SOCK_STREAM data. + * + * Returns the number of bytes sent on success, or errno otherwise + */ + +static int send_packet(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; + int res; + + /* Handle implied connection establishment */ + + if (unlikely(dest)) + return send_msg(iocb, sock, m, total_len); + + if (down_interruptible(&tsock->sem)) { + return -ERESTARTSYS; + } + + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + res = -EPIPE; + else + res = -ENOTCONN; + goto exit; + } + + do { + res = tipc_send(tsock->p->ref, m->msg_iovlen, m->msg_iov); + if (likely(res != -ELINKCONG)) { +exit: + up(&tsock->sem); + return res; + } + if (m->msg_flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + if (wait_event_interruptible(*sock->sk->sk_sleep, + !tsock->p->congested)) { + res = -ERESTARTSYS; + goto exit; + } + } while (1); +} + +/** + * send_stream - send stream-oriented data + * @iocb: (unused) + * @sock: socket structure + * @m: data to send + * @total_len: total length of data to be sent + * + * Used for SOCK_STREAM data. + * + * Returns the number of bytes sent on success, or errno otherwise + */ + + +static int send_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t total_len) +{ + struct msghdr my_msg; + struct iovec my_iov; + struct iovec *curr_iov; + int curr_iovlen; + char __user *curr_start; + int curr_left; + int bytes_to_send; + int res; + + if (likely(total_len <= TIPC_MAX_USER_MSG_SIZE)) + return send_packet(iocb, sock, m, total_len); + + /* Can only send large data streams if already connected */ + + if (unlikely(sock->state != SS_CONNECTED)) { + if (sock->state == SS_DISCONNECTING) + return -EPIPE; + else + return -ENOTCONN; + } + + /* + * Send each iovec entry using one or more messages + * + * Note: This algorithm is good for the most likely case + * (i.e. one large iovec entry), but could be improved to pass sets + * of small iovec entries into send_packet(). + */ + + my_msg = *m; + curr_iov = my_msg.msg_iov; + curr_iovlen = my_msg.msg_iovlen; + my_msg.msg_iov = &my_iov; + my_msg.msg_iovlen = 1; + + while (curr_iovlen--) { + curr_start = curr_iov->iov_base; + curr_left = curr_iov->iov_len; + + while (curr_left) { + bytes_to_send = (curr_left < TIPC_MAX_USER_MSG_SIZE) + ? curr_left : TIPC_MAX_USER_MSG_SIZE; + my_iov.iov_base = curr_start; + my_iov.iov_len = bytes_to_send; + if ((res = send_packet(iocb, sock, &my_msg, 0)) < 0) + return res; + curr_left -= bytes_to_send; + curr_start += bytes_to_send; + } + + curr_iov++; + } + + return total_len; +} + +/** + * auto_connect - complete connection setup to a remote port + * @sock: socket structure + * @tsock: TIPC-specific socket structure + * @msg: peer's response message + * + * Returns 0 on success, errno otherwise + */ + +static int auto_connect(struct socket *sock, struct tipc_sock *tsock, + struct tipc_msg *msg) +{ + struct tipc_portid peer; + + if (msg_errcode(msg)) { + sock->state = SS_DISCONNECTING; + return -ECONNREFUSED; + } + + peer.ref = msg_origport(msg); + peer.node = msg_orignode(msg); + tipc_connect2port(tsock->p->ref, &peer); + tipc_set_portimportance(tsock->p->ref, msg_importance(msg)); + sock->state = SS_CONNECTED; + return 0; +} + +/** + * set_orig_addr - capture sender's address for received message + * @m: descriptor for message info + * @msg: received message header + * + * Note: Address is not captured if not requested by receiver. + */ + +static inline void set_orig_addr(struct msghdr *m, struct tipc_msg *msg) +{ + struct sockaddr_tipc *addr = (struct sockaddr_tipc *)m->msg_name; + + if (addr) { + addr->family = AF_TIPC; + addr->addrtype = TIPC_ADDR_ID; + addr->addr.id.ref = msg_origport(msg); + addr->addr.id.node = msg_orignode(msg); + addr->addr.name.domain = 0; /* could leave uninitialized */ + addr->scope = 0; /* could leave uninitialized */ + m->msg_namelen = sizeof(struct sockaddr_tipc); + } +} + +/** + * anc_data_recv - optionally capture ancillary data for received message + * @m: descriptor for message info + * @msg: received message header + * @tport: TIPC port associated with message + * + * Note: Ancillary data is not captured if not requested by receiver. + * + * Returns 0 if successful, otherwise errno + */ + +static inline int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, + struct tipc_port *tport) +{ + u32 anc_data[3]; + u32 err; + u32 dest_type; + int res; + + if (likely(m->msg_controllen == 0)) + return 0; + + /* Optionally capture errored message object(s) */ + + err = msg ? msg_errcode(msg) : 0; + if (unlikely(err)) { + anc_data[0] = err; + anc_data[1] = msg_data_sz(msg); + if ((res = put_cmsg(m, SOL_SOCKET, TIPC_ERRINFO, 8, anc_data))) + return res; + if (anc_data[1] && + (res = put_cmsg(m, SOL_SOCKET, TIPC_RETDATA, anc_data[1], + msg_data(msg)))) + return res; + } + + /* Optionally capture message destination object */ + + dest_type = msg ? msg_type(msg) : TIPC_DIRECT_MSG; + switch (dest_type) { + case TIPC_NAMED_MSG: + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_namelower(msg); + break; + case TIPC_MCAST_MSG: + anc_data[0] = msg_nametype(msg); + anc_data[1] = msg_namelower(msg); + anc_data[2] = msg_nameupper(msg); + break; + case TIPC_CONN_MSG: + anc_data[0] = tport->conn_type; + anc_data[1] = tport->conn_instance; + anc_data[2] = tport->conn_instance; + break; + default: + anc_data[0] = 0; + } + if (anc_data[0] && + (res = put_cmsg(m, SOL_SOCKET, TIPC_DESTNAME, 12, anc_data))) + return res; + + return 0; +} + +/** + * recv_msg - receive packet-oriented message + * @iocb: (unused) + * @m: descriptor for message info + * @buf_len: total size of user buffer area + * @flags: receive flags + * + * Used for SOCK_DGRAM, SOCK_RDM, and SOCK_SEQPACKET messages. + * If the complete message doesn't fit in user area, truncate it. + * + * Returns size of returned message data, errno otherwise + */ + +static int recv_msg(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + struct tipc_msg *msg; + unsigned int q_len; + unsigned int sz; + u32 err; + int res; + + /* Currently doesn't support receiving into multiple iovec entries */ + + if (m->msg_iovlen != 1) + return -EOPNOTSUPP; + + /* Catch invalid receive attempts */ + + if (unlikely(!buf_len)) + return -EINVAL; + + if (sock->type == SOCK_SEQPACKET) { + if (unlikely(sock->state == SS_UNCONNECTED)) + return -ENOTCONN; + if (unlikely((sock->state == SS_DISCONNECTING) && + (skb_queue_len(&sock->sk->sk_receive_queue) == 0))) + return -ENOTCONN; + } + + /* Look for a message in receive queue; wait if necessary */ + + if (unlikely(down_interruptible(&tsock->sem))) + return -ERESTARTSYS; + +restart: + if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && + (flags & MSG_DONTWAIT))) { + res = -EWOULDBLOCK; + goto exit; + } + + if ((res = wait_event_interruptible( + *sock->sk->sk_sleep, + ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || + (sock->state == SS_DISCONNECTING))) )) { + goto exit; + } + + /* Catch attempt to receive on an already terminated connection */ + /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + + if (!q_len) { + res = -ENOTCONN; + goto exit; + } + + /* Get access to first message in receive queue */ + + buf = skb_peek(&sock->sk->sk_receive_queue); + msg = buf_msg(buf); + sz = msg_data_sz(msg); + err = msg_errcode(msg); + + /* Complete connection setup for an implied connect */ + + if (unlikely(sock->state == SS_CONNECTING)) { + if ((res = auto_connect(sock, tsock, msg))) + goto exit; + } + + /* Discard an empty non-errored message & try again */ + + if ((!sz) && (!err)) { + advance_queue(tsock); + goto restart; + } + + /* Capture sender's address (optional) */ + + set_orig_addr(m, msg); + + /* Capture ancillary data (optional) */ + + if ((res = anc_data_recv(m, msg, tsock->p))) + goto exit; + + /* Capture message data (if valid) & compute return value (always) */ + + if (!err) { + if (unlikely(buf_len < sz)) { + sz = buf_len; + m->msg_flags |= MSG_TRUNC; + } + if (unlikely(copy_to_user(m->msg_iov->iov_base, msg_data(msg), + sz))) { + res = -EFAULT; + goto exit; + } + res = sz; + } else { + if ((sock->state == SS_READY) || + ((err == TIPC_CONN_SHUTDOWN) || m->msg_control)) + res = 0; + else + res = -ECONNRESET; + } + + /* Consume received message (optional) */ + + if (likely(!(flags & MSG_PEEK))) { + if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); + advance_queue(tsock); + } +exit: + up(&tsock->sem); + return res; +} + +/** + * recv_stream - receive stream-oriented data + * @iocb: (unused) + * @m: descriptor for message info + * @buf_len: total size of user buffer area + * @flags: receive flags + * + * Used for SOCK_STREAM messages only. If not enough data is available + * will optionally wait for more; never truncates data. + * + * Returns size of returned message data, errno otherwise + */ + +static int recv_stream(struct kiocb *iocb, struct socket *sock, + struct msghdr *m, size_t buf_len, int flags) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + struct tipc_msg *msg; + unsigned int q_len; + unsigned int sz; + int sz_to_copy; + int sz_copied = 0; + int needed; + char *crs = m->msg_iov->iov_base; + unsigned char *buf_crs; + u32 err; + int res; + + /* Currently doesn't support receiving into multiple iovec entries */ + + if (m->msg_iovlen != 1) + return -EOPNOTSUPP; + + /* Catch invalid receive attempts */ + + if (unlikely(!buf_len)) + return -EINVAL; + + if (unlikely(sock->state == SS_DISCONNECTING)) { + if (skb_queue_len(&sock->sk->sk_receive_queue) == 0) + return -ENOTCONN; + } else if (unlikely(sock->state != SS_CONNECTED)) + return -ENOTCONN; + + /* Look for a message in receive queue; wait if necessary */ + + if (unlikely(down_interruptible(&tsock->sem))) + return -ERESTARTSYS; + +restart: + if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && + (flags & MSG_DONTWAIT))) { + res = (sz_copied == 0) ? -EWOULDBLOCK : 0; + goto exit; + } + + if ((res = wait_event_interruptible( + *sock->sk->sk_sleep, + ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || + (sock->state == SS_DISCONNECTING))) )) { + goto exit; + } + + /* Catch attempt to receive on an already terminated connection */ + /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + + if (!q_len) { + res = -ENOTCONN; + goto exit; + } + + /* Get access to first message in receive queue */ + + buf = skb_peek(&sock->sk->sk_receive_queue); + msg = buf_msg(buf); + sz = msg_data_sz(msg); + err = msg_errcode(msg); + + /* Discard an empty non-errored message & try again */ + + if ((!sz) && (!err)) { + advance_queue(tsock); + goto restart; + } + + /* Optionally capture sender's address & ancillary data of first msg */ + + if (sz_copied == 0) { + set_orig_addr(m, msg); + if ((res = anc_data_recv(m, msg, tsock->p))) + goto exit; + } + + /* Capture message data (if valid) & compute return value (always) */ + + if (!err) { + buf_crs = (unsigned char *)(TIPC_SKB_CB(buf)->handle); + sz = buf->tail - buf_crs; + + needed = (buf_len - sz_copied); + sz_to_copy = (sz <= needed) ? sz : needed; + if (unlikely(copy_to_user(crs, buf_crs, sz_to_copy))) { + res = -EFAULT; + goto exit; + } + sz_copied += sz_to_copy; + + if (sz_to_copy < sz) { + if (!(flags & MSG_PEEK)) + TIPC_SKB_CB(buf)->handle = buf_crs + sz_to_copy; + goto exit; + } + + crs += sz_to_copy; + } else { + if (sz_copied != 0) + goto exit; /* can't add error msg to valid data */ + + if ((err == TIPC_CONN_SHUTDOWN) || m->msg_control) + res = 0; + else + res = -ECONNRESET; + } + + /* Consume received message (optional) */ + + if (likely(!(flags & MSG_PEEK))) { + if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); + advance_queue(tsock); + } + + /* Loop around if more data is required */ + + if ((sz_copied < buf_len) /* didn't get all requested data */ + && (flags & MSG_WAITALL) /* ... and need to wait for more */ + && (!(flags & MSG_PEEK)) /* ... and aren't just peeking at data */ + && (!err) /* ... and haven't reached a FIN */ + ) + goto restart; + +exit: + up(&tsock->sem); + return res ? res : sz_copied; +} + +/** + * queue_overloaded - test if queue overload condition exists + * @queue_size: current size of queue + * @base: nominal maximum size of queue + * @msg: message to be added to queue + * + * Returns 1 if queue is currently overloaded, 0 otherwise + */ + +static int queue_overloaded(u32 queue_size, u32 base, struct tipc_msg *msg) +{ + u32 threshold; + u32 imp = msg_importance(msg); + + if (imp == TIPC_LOW_IMPORTANCE) + threshold = base; + else if (imp == TIPC_MEDIUM_IMPORTANCE) + threshold = base * 2; + else if (imp == TIPC_HIGH_IMPORTANCE) + threshold = base * 100; + else + return 0; + + if (msg_connected(msg)) + threshold *= 4; + + return (queue_size > threshold); +} + +/** + * async_disconnect - wrapper function used to disconnect port + * @portref: TIPC port reference (passed as pointer-sized value) + */ + +static void async_disconnect(unsigned long portref) +{ + tipc_disconnect((u32)portref); +} + +/** + * dispatch - handle arriving message + * @tport: TIPC port that received message + * @buf: message + * + * Called with port locked. Must not take socket lock to avoid deadlock risk. + * + * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + */ + +static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) +{ + struct tipc_msg *msg = buf_msg(buf); + struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; + struct socket *sock; + u32 recv_q_len; + + /* Reject message if socket is closing */ + + if (!tsock) + return TIPC_ERR_NO_PORT; + + /* Reject message if it is wrong sort of message for socket */ + + /* + * WOULD IT BE BETTER TO JUST DISCARD THESE MESSAGES INSTEAD? + * "NO PORT" ISN'T REALLY THE RIGHT ERROR CODE, AND THERE MAY + * BE SECURITY IMPLICATIONS INHERENT IN REJECTING INVALID TRAFFIC + */ + sock = tsock->sk.sk_socket; + if (sock->state == SS_READY) { + if (msg_connected(msg)) { + msg_dbg(msg, "dispatch filter 1\n"); + return TIPC_ERR_NO_PORT; + } + } else { + if (msg_mcast(msg)) { + msg_dbg(msg, "dispatch filter 2\n"); + return TIPC_ERR_NO_PORT; + } + if (sock->state == SS_CONNECTED) { + if (!msg_connected(msg)) { + msg_dbg(msg, "dispatch filter 3\n"); + return TIPC_ERR_NO_PORT; + } + } + else if (sock->state == SS_CONNECTING) { + if (!msg_connected(msg) && (msg_errcode(msg) == 0)) { + msg_dbg(msg, "dispatch filter 4\n"); + return TIPC_ERR_NO_PORT; + } + } + else if (sock->state == SS_LISTENING) { + if (msg_connected(msg) || msg_errcode(msg)) { + msg_dbg(msg, "dispatch filter 5\n"); + return TIPC_ERR_NO_PORT; + } + } + else if (sock->state == SS_DISCONNECTING) { + msg_dbg(msg, "dispatch filter 6\n"); + return TIPC_ERR_NO_PORT; + } + else /* (sock->state == SS_UNCONNECTED) */ { + if (msg_connected(msg) || msg_errcode(msg)) { + msg_dbg(msg, "dispatch filter 7\n"); + return TIPC_ERR_NO_PORT; + } + } + } + + /* Reject message if there isn't room to queue it */ + + if (unlikely((u32)atomic_read(&tipc_queue_size) > + OVERLOAD_LIMIT_BASE)) { + if (queue_overloaded(atomic_read(&tipc_queue_size), + OVERLOAD_LIMIT_BASE, msg)) + return TIPC_ERR_OVERLOAD; + } + recv_q_len = skb_queue_len(&tsock->sk.sk_receive_queue); + if (unlikely(recv_q_len > (OVERLOAD_LIMIT_BASE / 2))) { + if (queue_overloaded(recv_q_len, + OVERLOAD_LIMIT_BASE / 2, msg)) + return TIPC_ERR_OVERLOAD; + } + + /* Initiate connection termination for an incoming 'FIN' */ + + if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) { + sock->state = SS_DISCONNECTING; + /* Note: Use signal since port lock is already taken! */ + k_signal((Handler)async_disconnect, tport->ref); + } + + /* Enqueue message (finally!) */ + + msg_dbg(msg,"<DISP<: "); + TIPC_SKB_CB(buf)->handle = msg_data(msg); + atomic_inc(&tipc_queue_size); + skb_queue_tail(&sock->sk->sk_receive_queue, buf); + + wake_up_interruptible(sock->sk->sk_sleep); + return TIPC_OK; +} + +/** + * wakeupdispatch - wake up port after congestion + * @tport: port to wakeup + * + * Called with port lock on. + */ + +static void wakeupdispatch(struct tipc_port *tport) +{ + struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; + + wake_up_interruptible(tsock->sk.sk_sleep); +} + +/** + * connect - establish a connection to another TIPC port + * @sock: socket structure + * @dest: socket address for destination port + * @destlen: size of socket address data structure + * @flags: (unused) + * + * Returns 0 on success, errno otherwise + */ + +static int connect(struct socket *sock, struct sockaddr *dest, int destlen, + int flags) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; + struct msghdr m = {0,}; + struct sk_buff *buf; + struct tipc_msg *msg; + int res; + + /* For now, TIPC does not allow use of connect() with DGRAM or RDM types */ + + if (sock->state == SS_READY) + return -EOPNOTSUPP; + + /* MOVE THE REST OF THIS ERROR CHECKING TO send_msg()? */ + if (sock->state == SS_LISTENING) + return -EOPNOTSUPP; + if (sock->state == SS_CONNECTING) + return -EALREADY; + if (sock->state != SS_UNCONNECTED) + return -EISCONN; + + if ((dst->family != AF_TIPC) || + ((dst->addrtype != TIPC_ADDR_NAME) && (dst->addrtype != TIPC_ADDR_ID))) + return -EINVAL; + + /* Send a 'SYN-' to destination */ + + m.msg_name = dest; + if ((res = send_msg(0, sock, &m, 0)) < 0) { + sock->state = SS_DISCONNECTING; + return res; + } + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + /* Wait for destination's 'ACK' response */ + + res = wait_event_interruptible_timeout(*sock->sk->sk_sleep, + skb_queue_len(&sock->sk->sk_receive_queue), + sock->sk->sk_rcvtimeo); + buf = skb_peek(&sock->sk->sk_receive_queue); + if (res > 0) { + msg = buf_msg(buf); + res = auto_connect(sock, tsock, msg); + if (!res) { + if (dst->addrtype == TIPC_ADDR_NAME) { + tsock->p->conn_type = dst->addr.name.name.type; + tsock->p->conn_instance = dst->addr.name.name.instance; + } + if (!msg_data_sz(msg)) + advance_queue(tsock); + } + } else { + if (res == 0) { + res = -ETIMEDOUT; + } else + { /* leave "res" unchanged */ } + sock->state = SS_DISCONNECTING; + } + + up(&tsock->sem); + return res; +} + +/** + * listen - allow socket to listen for incoming connections + * @sock: socket structure + * @len: (unused) + * + * Returns 0 on success, errno otherwise + */ + +static int listen(struct socket *sock, int len) +{ + /* REQUIRES SOCKET LOCKING OF SOME SORT? */ + + if (sock->state == SS_READY) + return -EOPNOTSUPP; + if (sock->state != SS_UNCONNECTED) + return -EINVAL; + sock->state = SS_LISTENING; + return 0; +} + +/** + * accept - wait for connection request + * @sock: listening socket + * @newsock: new socket that is to be connected + * @flags: file-related flags associated with socket + * + * Returns 0 on success, errno otherwise + */ + +static int accept(struct socket *sock, struct socket *newsock, int flags) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + int res = -EFAULT; + + if (sock->state == SS_READY) + return -EOPNOTSUPP; + if (sock->state != SS_LISTENING) + return -EINVAL; + + if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && + (flags & O_NONBLOCK))) + return -EWOULDBLOCK; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + if (wait_event_interruptible(*sock->sk->sk_sleep, + skb_queue_len(&sock->sk->sk_receive_queue))) { + res = -ERESTARTSYS; + goto exit; + } + buf = skb_peek(&sock->sk->sk_receive_queue); + + res = tipc_create(newsock, 0); + if (!res) { + struct tipc_sock *new_tsock = tipc_sk(newsock->sk); + struct tipc_portid id; + struct tipc_msg *msg = buf_msg(buf); + u32 new_ref = new_tsock->p->ref; + + id.ref = msg_origport(msg); + id.node = msg_orignode(msg); + tipc_connect2port(new_ref, &id); + newsock->state = SS_CONNECTED; + + tipc_set_portimportance(new_ref, msg_importance(msg)); + if (msg_named(msg)) { + new_tsock->p->conn_type = msg_nametype(msg); + new_tsock->p->conn_instance = msg_nameinst(msg); + } + + /* + * Respond to 'SYN-' by discarding it & returning 'ACK'-. + * Respond to 'SYN+' by queuing it on new socket. + */ + + msg_dbg(msg,"<ACC<: "); + if (!msg_data_sz(msg)) { + struct msghdr m = {0,}; + + send_packet(0, newsock, &m, 0); + advance_queue(tsock); + } else { + sock_lock(tsock); + skb_dequeue(&sock->sk->sk_receive_queue); + sock_unlock(tsock); + skb_queue_head(&newsock->sk->sk_receive_queue, buf); + } + } +exit: + up(&tsock->sem); + return res; +} + +/** + * shutdown - shutdown socket connection + * @sock: socket structure + * @how: direction to close (always treated as read + write) + * + * Terminates connection (if necessary), then purges socket's receive queue. + * + * Returns 0 on success, errno otherwise + */ + +static int shutdown(struct socket *sock, int how) +{ + struct tipc_sock* tsock = tipc_sk(sock->sk); + struct sk_buff *buf; + int res; + + /* Could return -EINVAL for an invalid "how", but why bother? */ + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + sock_lock(tsock); + + switch (sock->state) { + case SS_CONNECTED: + + /* Send 'FIN+' or 'FIN-' message to peer */ + + sock_unlock(tsock); +restart: + if ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { + atomic_dec(&tipc_queue_size); + if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) { + buf_discard(buf); + goto restart; + } + tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN); + } + else { + tipc_shutdown(tsock->p->ref); + } + sock_lock(tsock); + + /* fall through */ + + case SS_DISCONNECTING: + + /* Discard any unreceived messages */ + + while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { + atomic_dec(&tipc_queue_size); + buf_discard(buf); + } + tsock->p->conn_unacked = 0; + + /* fall through */ + + case SS_CONNECTING: + sock->state = SS_DISCONNECTING; + res = 0; + break; + + default: + res = -ENOTCONN; + } + + sock_unlock(tsock); + + up(&tsock->sem); + return res; +} + +/** + * setsockopt - set socket option + * @sock: socket structure + * @lvl: option level + * @opt: option identifier + * @ov: pointer to new option value + * @ol: length of option value + * + * For stream sockets only, accepts and ignores all IPPROTO_TCP options + * (to ease compatibility). + * + * Returns 0 on success, errno otherwise + */ + +static int setsockopt(struct socket *sock, int lvl, int opt, char *ov, int ol) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + u32 value; + int res; + + if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) + return 0; + if (lvl != SOL_TIPC) + return -ENOPROTOOPT; + if (ol < sizeof(value)) + return -EINVAL; + if ((res = get_user(value, (u32 *)ov))) + return res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + switch (opt) { + case TIPC_IMPORTANCE: + res = tipc_set_portimportance(tsock->p->ref, value); + break; + case TIPC_SRC_DROPPABLE: + if (sock->type != SOCK_STREAM) + res = tipc_set_portunreliable(tsock->p->ref, value); + else + res = -ENOPROTOOPT; + break; + case TIPC_DEST_DROPPABLE: + res = tipc_set_portunreturnable(tsock->p->ref, value); + break; + case TIPC_CONN_TIMEOUT: + sock->sk->sk_rcvtimeo = (value * HZ / 1000); + break; + default: + res = -EINVAL; + } + + up(&tsock->sem); + return res; +} + +/** + * getsockopt - get socket option + * @sock: socket structure + * @lvl: option level + * @opt: option identifier + * @ov: receptacle for option value + * @ol: receptacle for length of option value + * + * For stream sockets only, returns 0 length result for all IPPROTO_TCP options + * (to ease compatibility). + * + * Returns 0 on success, errno otherwise + */ + +static int getsockopt(struct socket *sock, int lvl, int opt, char *ov, int *ol) +{ + struct tipc_sock *tsock = tipc_sk(sock->sk); + int len; + u32 value; + int res; + + if ((lvl == IPPROTO_TCP) && (sock->type == SOCK_STREAM)) + return put_user(0, ol); + if (lvl != SOL_TIPC) + return -ENOPROTOOPT; + if ((res = get_user(len, ol))) + return res; + + if (down_interruptible(&tsock->sem)) + return -ERESTARTSYS; + + switch (opt) { + case TIPC_IMPORTANCE: + res = tipc_portimportance(tsock->p->ref, &value); + break; + case TIPC_SRC_DROPPABLE: + res = tipc_portunreliable(tsock->p->ref, &value); + break; + case TIPC_DEST_DROPPABLE: + res = tipc_portunreturnable(tsock->p->ref, &value); + break; + case TIPC_CONN_TIMEOUT: + value = (sock->sk->sk_rcvtimeo * 1000) / HZ; + break; + default: + res = -EINVAL; + } + + if (res) { + /* "get" failed */ + } + else if (len < sizeof(value)) { + res = -EINVAL; + } + else if ((res = copy_to_user(ov, &value, sizeof(value)))) { + /* couldn't return value */ + } + else { + res = put_user(sizeof(value), ol); + } + + up(&tsock->sem); + return res; +} + +/** + * Placeholders for non-implemented functionality + * + * Returns error code (POSIX-compliant where defined) + */ + +static int ioctl(struct socket *s, u32 cmd, unsigned long arg) +{ + return -EINVAL; +} + +static int no_mmap(struct file *file, struct socket *sock, + struct vm_area_struct *vma) +{ + return -EINVAL; +} +static ssize_t no_sendpage(struct socket *sock, struct page *page, + int offset, size_t size, int flags) +{ + return -EINVAL; +} + +static int no_skpair(struct socket *s1, struct socket *s2) +{ + return -EOPNOTSUPP; +} + +/** + * Protocol switches for the various types of TIPC sockets + */ + +static struct proto_ops msg_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = release, + .bind = bind, + .connect = connect, + .socketpair = no_skpair, + .accept = accept, + .getname = get_name, + .poll = poll, + .ioctl = ioctl, + .listen = listen, + .shutdown = shutdown, + .setsockopt = setsockopt, + .getsockopt = getsockopt, + .sendmsg = send_msg, + .recvmsg = recv_msg, + .mmap = no_mmap, + .sendpage = no_sendpage +}; + +static struct proto_ops packet_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = release, + .bind = bind, + .connect = connect, + .socketpair = no_skpair, + .accept = accept, + .getname = get_name, + .poll = poll, + .ioctl = ioctl, + .listen = listen, + .shutdown = shutdown, + .setsockopt = setsockopt, + .getsockopt = getsockopt, + .sendmsg = send_packet, + .recvmsg = recv_msg, + .mmap = no_mmap, + .sendpage = no_sendpage +}; + +static struct proto_ops stream_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .release = release, + .bind = bind, + .connect = connect, + .socketpair = no_skpair, + .accept = accept, + .getname = get_name, + .poll = poll, + .ioctl = ioctl, + .listen = listen, + .shutdown = shutdown, + .setsockopt = setsockopt, + .getsockopt = getsockopt, + .sendmsg = send_stream, + .recvmsg = recv_stream, + .mmap = no_mmap, + .sendpage = no_sendpage +}; + +static struct net_proto_family tipc_family_ops = { + .owner = THIS_MODULE, + .family = AF_TIPC, + .create = tipc_create +}; + +static struct proto tipc_proto = { + .name = "TIPC", + .owner = THIS_MODULE, + .obj_size = sizeof(struct tipc_sock) +}; + +/** + * socket_init - initialize TIPC socket interface + * + * Returns 0 on success, errno otherwise + */ +int socket_init(void) +{ + int res; + + res = proto_register(&tipc_proto, 1); + if (res) { + err("Failed to register TIPC protocol type\n"); + goto out; + } + + res = sock_register(&tipc_family_ops); + if (res) { + err("Failed to register TIPC socket type\n"); + proto_unregister(&tipc_proto); + goto out; + } + + sockets_enabled = 1; + out: + return res; +} + +/** + * sock_stop - stop TIPC socket interface + */ +void socket_stop(void) +{ + if (!sockets_enabled) + return; + + sockets_enabled = 0; + sock_unregister(tipc_family_ops.family); + proto_unregister(&tipc_proto); +} + diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c new file mode 100644 index 00000000000..80e219ba527 --- /dev/null +++ b/net/tipc/subscr.c @@ -0,0 +1,527 @@ +/* + * net/tipc/subscr.c: TIPC subscription service + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "dbg.h" +#include "subscr.h" +#include "name_table.h" +#include "ref.h" + +/** + * struct subscriber - TIPC network topology subscriber + * @ref: object reference to subscriber object itself + * @lock: pointer to spinlock controlling access to subscriber object + * @subscriber_list: adjacent subscribers in top. server's list of subscribers + * @subscription_list: list of subscription objects for this subscriber + * @port_ref: object reference to port used to communicate with subscriber + * @swap: indicates if subscriber uses opposite endianness in its messages + */ + +struct subscriber { + u32 ref; + spinlock_t *lock; + struct list_head subscriber_list; + struct list_head subscription_list; + u32 port_ref; + int swap; +}; + +/** + * struct top_srv - TIPC network topology subscription service + * @user_ref: TIPC userid of subscription service + * @setup_port: reference to TIPC port that handles subscription requests + * @subscription_count: number of active subscriptions (not subscribers!) + * @subscriber_list: list of ports subscribing to service + * @lock: spinlock govering access to subscriber list + */ + +struct top_srv { + u32 user_ref; + u32 setup_port; + atomic_t subscription_count; + struct list_head subscriber_list; + spinlock_t lock; +}; + +static struct top_srv topsrv = { 0 }; + +/** + * htohl - convert value to endianness used by destination + * @in: value to convert + * @swap: non-zero if endianness must be reversed + * + * Returns converted value + */ + +static inline u32 htohl(u32 in, int swap) +{ + char *c = (char *)∈ + + return swap ? ((c[3] << 3) + (c[2] << 2) + (c[1] << 1) + c[0]) : in; +} + +/** + * subscr_send_event - send a message containing a tipc_event to the subscriber + */ + +static void subscr_send_event(struct subscription *sub, + u32 found_lower, + u32 found_upper, + u32 event, + u32 port_ref, + u32 node) +{ + struct iovec msg_sect; + + msg_sect.iov_base = (void *)&sub->evt; + msg_sect.iov_len = sizeof(struct tipc_event); + + sub->evt.event = htohl(event, sub->owner->swap); + sub->evt.found_lower = htohl(found_lower, sub->owner->swap); + sub->evt.found_upper = htohl(found_upper, sub->owner->swap); + sub->evt.port.ref = htohl(port_ref, sub->owner->swap); + sub->evt.port.node = htohl(node, sub->owner->swap); + tipc_send(sub->owner->port_ref, 1, &msg_sect); +} + +/** + * subscr_overlap - test for subscription overlap with the given values + * + * Returns 1 if there is overlap, otherwise 0. + */ + +int subscr_overlap(struct subscription *sub, + u32 found_lower, + u32 found_upper) + +{ + if (found_lower < sub->seq.lower) + found_lower = sub->seq.lower; + if (found_upper > sub->seq.upper) + found_upper = sub->seq.upper; + if (found_lower > found_upper) + return 0; + return 1; +} + +/** + * subscr_report_overlap - issue event if there is subscription overlap + * + * Protected by nameseq.lock in name_table.c + */ + +void subscr_report_overlap(struct subscription *sub, + u32 found_lower, + u32 found_upper, + u32 event, + u32 port_ref, + u32 node, + int must) +{ + dbg("Rep overlap %u:%u,%u<->%u,%u\n", sub->seq.type, sub->seq.lower, + sub->seq.upper, found_lower, found_upper); + if (!subscr_overlap(sub, found_lower, found_upper)) + return; + if (!must && (sub->filter != TIPC_SUB_PORTS)) + return; + subscr_send_event(sub, found_lower, found_upper, event, port_ref, node); +} + +/** + * subscr_timeout - subscription timeout has occurred + */ + +static void subscr_timeout(struct subscription *sub) +{ + struct subscriber *subscriber; + u32 subscriber_ref; + + /* Validate subscriber reference (in case subscriber is terminating) */ + + subscriber_ref = sub->owner->ref; + subscriber = (struct subscriber *)ref_lock(subscriber_ref); + if (subscriber == NULL) + return; + + /* Unlink subscription from name table */ + + nametbl_unsubscribe(sub); + + /* Notify subscriber of timeout, then unlink subscription */ + + subscr_send_event(sub, + sub->evt.s.seq.lower, + sub->evt.s.seq.upper, + TIPC_SUBSCR_TIMEOUT, + 0, + 0); + list_del(&sub->subscription_list); + + /* Now destroy subscription */ + + ref_unlock(subscriber_ref); + k_term_timer(&sub->timer); + kfree(sub); + atomic_dec(&topsrv.subscription_count); +} + +/** + * subscr_terminate - terminate communication with a subscriber + * + * Called with subscriber locked. Routine must temporarily release this lock + * to enable subscription timeout routine(s) to finish without deadlocking; + * the lock is then reclaimed to allow caller to release it upon return. + * (This should work even in the unlikely event some other thread creates + * a new object reference in the interim that uses this lock; this routine will + * simply wait for it to be released, then claim it.) + */ + +static void subscr_terminate(struct subscriber *subscriber) +{ + struct subscription *sub; + struct subscription *sub_temp; + + /* Invalidate subscriber reference */ + + ref_discard(subscriber->ref); + spin_unlock_bh(subscriber->lock); + + /* Destroy any existing subscriptions for subscriber */ + + list_for_each_entry_safe(sub, sub_temp, &subscriber->subscription_list, + subscription_list) { + if (sub->timeout != TIPC_WAIT_FOREVER) { + k_cancel_timer(&sub->timer); + k_term_timer(&sub->timer); + } + nametbl_unsubscribe(sub); + list_del(&sub->subscription_list); + dbg("Term: Removed sub %u,%u,%u from subscriber %x list\n", + sub->seq.type, sub->seq.lower, sub->seq.upper, subscriber); + kfree(sub); + atomic_dec(&topsrv.subscription_count); + } + + /* Sever connection to subscriber */ + + tipc_shutdown(subscriber->port_ref); + tipc_deleteport(subscriber->port_ref); + + /* Remove subscriber from topology server's subscriber list */ + + spin_lock_bh(&topsrv.lock); + list_del(&subscriber->subscriber_list); + spin_unlock_bh(&topsrv.lock); + + /* Now destroy subscriber */ + + spin_lock_bh(subscriber->lock); + kfree(subscriber); +} + +/** + * subscr_subscribe - create subscription for subscriber + * + * Called with subscriber locked + */ + +static void subscr_subscribe(struct tipc_subscr *s, + struct subscriber *subscriber) +{ + struct subscription *sub; + + /* Refuse subscription if global limit exceeded */ + + if (atomic_read(&topsrv.subscription_count) >= tipc_max_subscriptions) { + warn("Failed: max %u subscriptions\n", tipc_max_subscriptions); + subscr_terminate(subscriber); + return; + } + + /* Allocate subscription object */ + + sub = kmalloc(sizeof(*sub), GFP_ATOMIC); + if (sub == NULL) { + warn("Memory squeeze; ignoring subscription\n"); + subscr_terminate(subscriber); + return; + } + + /* Determine/update subscriber's endianness */ + + if ((s->filter == TIPC_SUB_PORTS) || (s->filter == TIPC_SUB_SERVICE)) + subscriber->swap = 0; + else + subscriber->swap = 1; + + /* Initialize subscription object */ + + memset(sub, 0, sizeof(*sub)); + sub->seq.type = htohl(s->seq.type, subscriber->swap); + sub->seq.lower = htohl(s->seq.lower, subscriber->swap); + sub->seq.upper = htohl(s->seq.upper, subscriber->swap); + sub->timeout = htohl(s->timeout, subscriber->swap); + sub->filter = htohl(s->filter, subscriber->swap); + if ((((sub->filter != TIPC_SUB_PORTS) + && (sub->filter != TIPC_SUB_SERVICE))) + || (sub->seq.lower > sub->seq.upper)) { + warn("Rejecting illegal subscription %u,%u,%u\n", + sub->seq.type, sub->seq.lower, sub->seq.upper); + kfree(sub); + subscr_terminate(subscriber); + return; + } + memcpy(&sub->evt.s, s, sizeof(struct tipc_subscr)); + INIT_LIST_HEAD(&sub->subscription_list); + INIT_LIST_HEAD(&sub->nameseq_list); + list_add(&sub->subscription_list, &subscriber->subscription_list); + atomic_inc(&topsrv.subscription_count); + if (sub->timeout != TIPC_WAIT_FOREVER) { + k_init_timer(&sub->timer, + (Handler)subscr_timeout, (unsigned long)sub); + k_start_timer(&sub->timer, sub->timeout); + } + sub->owner = subscriber; + nametbl_subscribe(sub); +} + +/** + * subscr_conn_shutdown_event - handle termination request from subscriber + */ + +static void subscr_conn_shutdown_event(void *usr_handle, + u32 portref, + struct sk_buff **buf, + unsigned char const *data, + unsigned int size, + int reason) +{ + struct subscriber *subscriber; + spinlock_t *subscriber_lock; + + subscriber = ref_lock((u32)(unsigned long)usr_handle); + if (subscriber == NULL) + return; + + subscriber_lock = subscriber->lock; + subscr_terminate(subscriber); + spin_unlock_bh(subscriber_lock); +} + +/** + * subscr_conn_msg_event - handle new subscription request from subscriber + */ + +static void subscr_conn_msg_event(void *usr_handle, + u32 port_ref, + struct sk_buff **buf, + const unchar *data, + u32 size) +{ + struct subscriber *subscriber; + spinlock_t *subscriber_lock; + + subscriber = ref_lock((u32)(unsigned long)usr_handle); + if (subscriber == NULL) + return; + + subscriber_lock = subscriber->lock; + if (size != sizeof(struct tipc_subscr)) + subscr_terminate(subscriber); + else + subscr_subscribe((struct tipc_subscr *)data, subscriber); + + spin_unlock_bh(subscriber_lock); +} + +/** + * subscr_named_msg_event - handle request to establish a new subscriber + */ + +static void subscr_named_msg_event(void *usr_handle, + u32 port_ref, + struct sk_buff **buf, + const unchar *data, + u32 size, + u32 importance, + struct tipc_portid const *orig, + struct tipc_name_seq const *dest) +{ + struct subscriber *subscriber; + struct iovec msg_sect = {0, 0}; + spinlock_t *subscriber_lock; + + dbg("subscr_named_msg_event: orig = %x own = %x,\n", + orig->node, tipc_own_addr); + if (size && (size != sizeof(struct tipc_subscr))) { + warn("Received tipc_subscr of invalid size\n"); + return; + } + + /* Create subscriber object */ + + subscriber = kmalloc(sizeof(struct subscriber), GFP_ATOMIC); + if (subscriber == NULL) { + warn("Memory squeeze; ignoring subscriber setup\n"); + return; + } + memset(subscriber, 0, sizeof(struct subscriber)); + INIT_LIST_HEAD(&subscriber->subscription_list); + INIT_LIST_HEAD(&subscriber->subscriber_list); + subscriber->ref = ref_acquire(subscriber, &subscriber->lock); + if (subscriber->ref == 0) { + warn("Failed to acquire subscriber reference\n"); + kfree(subscriber); + return; + } + + /* Establish a connection to subscriber */ + + tipc_createport(topsrv.user_ref, + (void *)(unsigned long)subscriber->ref, + importance, + 0, + 0, + subscr_conn_shutdown_event, + 0, + 0, + subscr_conn_msg_event, + 0, + &subscriber->port_ref); + if (subscriber->port_ref == 0) { + warn("Memory squeeze; failed to create subscription port\n"); + ref_discard(subscriber->ref); + kfree(subscriber); + return; + } + tipc_connect2port(subscriber->port_ref, orig); + + + /* Add subscriber to topology server's subscriber list */ + + ref_lock(subscriber->ref); + spin_lock_bh(&topsrv.lock); + list_add(&subscriber->subscriber_list, &topsrv.subscriber_list); + spin_unlock_bh(&topsrv.lock); + + /* + * Subscribe now if message contains a subscription, + * otherwise send an empty response to complete connection handshaking + */ + + subscriber_lock = subscriber->lock; + if (size) + subscr_subscribe((struct tipc_subscr *)data, subscriber); + else + tipc_send(subscriber->port_ref, 1, &msg_sect); + + spin_unlock_bh(subscriber_lock); +} + +int subscr_start(void) +{ + struct tipc_name_seq seq = {TIPC_TOP_SRV, TIPC_TOP_SRV, TIPC_TOP_SRV}; + int res = -1; + + memset(&topsrv, 0, sizeof (topsrv)); + topsrv.lock = SPIN_LOCK_UNLOCKED; + INIT_LIST_HEAD(&topsrv.subscriber_list); + + spin_lock_bh(&topsrv.lock); + res = tipc_attach(&topsrv.user_ref, 0, 0); + if (res) { + spin_unlock_bh(&topsrv.lock); + return res; + } + + res = tipc_createport(topsrv.user_ref, + 0, + TIPC_CRITICAL_IMPORTANCE, + 0, + 0, + 0, + 0, + subscr_named_msg_event, + 0, + 0, + &topsrv.setup_port); + if (res) + goto failed; + + res = nametbl_publish_rsv(topsrv.setup_port, TIPC_NODE_SCOPE, &seq); + if (res) + goto failed; + + spin_unlock_bh(&topsrv.lock); + return 0; + +failed: + err("Failed to create subscription service\n"); + tipc_detach(topsrv.user_ref); + topsrv.user_ref = 0; + spin_unlock_bh(&topsrv.lock); + return res; +} + +void subscr_stop(void) +{ + struct subscriber *subscriber; + struct subscriber *subscriber_temp; + spinlock_t *subscriber_lock; + + if (topsrv.user_ref) { + tipc_deleteport(topsrv.setup_port); + list_for_each_entry_safe(subscriber, subscriber_temp, + &topsrv.subscriber_list, + subscriber_list) { + ref_lock(subscriber->ref); + subscriber_lock = subscriber->lock; + subscr_terminate(subscriber); + spin_unlock_bh(subscriber_lock); + } + tipc_detach(topsrv.user_ref); + topsrv.user_ref = 0; + } +} + + +int tipc_ispublished(struct tipc_name const *name) +{ + u32 domain = 0; + + return(nametbl_translate(name->type, name->instance,&domain) != 0); +} + diff --git a/net/tipc/subscr.h b/net/tipc/subscr.h new file mode 100644 index 00000000000..ccff4efcb75 --- /dev/null +++ b/net/tipc/subscr.h @@ -0,0 +1,80 @@ +/* + * net/tipc/subscr.h: Include file for TIPC subscription service + * + * Copyright (c) 2003-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_SUBSCR_H +#define _TIPC_SUBSCR_H + +/** + * struct subscription - TIPC network topology subscription object + * @seq: name sequence associated with subscription + * @timeout: duration of subscription (in ms) + * @filter: event filtering to be done for subscription + * @evt: template for events generated by subscription + * @subscription_list: adjacent subscriptions in subscriber's subscription list + * @nameseq_list: adjacent subscriptions in name sequence's subscription list + * @timer_ref: reference to timer governing subscription duration (may be NULL) + * @owner: pointer to subscriber object associated with this subscription + */ + +struct subscription { + struct tipc_name_seq seq; + u32 timeout; + u32 filter; + struct tipc_event evt; + struct list_head subscription_list; + struct list_head nameseq_list; + struct timer_list timer; + struct subscriber *owner; +}; + +int subscr_overlap(struct subscription * sub, + u32 found_lower, + u32 found_upper); + +void subscr_report_overlap(struct subscription * sub, + u32 found_lower, + u32 found_upper, + u32 event, + u32 port_ref, + u32 node, + int must_report); + +int subscr_start(void); + +void subscr_stop(void); + + +#endif diff --git a/net/tipc/user_reg.c b/net/tipc/user_reg.c new file mode 100644 index 00000000000..35ec7dc8211 --- /dev/null +++ b/net/tipc/user_reg.c @@ -0,0 +1,265 @@ +/* + * net/tipc/user_reg.c: TIPC user registry code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2004-2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "user_reg.h" + +/* + * TIPC user registry keeps track of users of the tipc_port interface. + * + * The registry utilizes an array of "TIPC user" entries; + * a user's ID is the index of their associated array entry. + * Array entry 0 is not used, so userid 0 is not valid; + * TIPC sometimes uses this value to denote an anonymous user. + * The list of free entries is initially chained from last entry to entry 1. + */ + +/** + * struct tipc_user - registered TIPC user info + * @next: index of next free registry entry (or -1 for an allocated entry) + * @callback: ptr to routine to call when TIPC mode changes (NULL if none) + * @usr_handle: user-defined value passed to callback routine + * @ports: list of user ports owned by the user + */ + +struct tipc_user { + int next; + tipc_mode_event callback; + void *usr_handle; + struct list_head ports; +}; + +#define MAX_USERID 64 +#define USER_LIST_SIZE ((MAX_USERID + 1) * sizeof(struct tipc_user)) + +static struct tipc_user *users = 0; +static u32 next_free_user = MAX_USERID + 1; +static spinlock_t reg_lock = SPIN_LOCK_UNLOCKED; + +/** + * reg_init - create TIPC user registry (but don't activate it) + * + * If registry has been pre-initialized it is left "as is". + * NOTE: This routine may be called when TIPC is inactive. + */ + +static int reg_init(void) +{ + u32 i; + + spin_lock_bh(®_lock); + if (!users) { + users = (struct tipc_user *)kmalloc(USER_LIST_SIZE, GFP_ATOMIC); + if (users) { + memset(users, 0, USER_LIST_SIZE); + for (i = 1; i <= MAX_USERID; i++) { + users[i].next = i - 1; + } + next_free_user = MAX_USERID; + } + } + spin_unlock_bh(®_lock); + return users ? TIPC_OK : -ENOMEM; +} + +/** + * reg_callback - inform TIPC user about current operating mode + */ + +static void reg_callback(struct tipc_user *user_ptr) +{ + tipc_mode_event cb; + void *arg; + + spin_lock_bh(®_lock); + cb = user_ptr->callback; + arg = user_ptr->usr_handle; + spin_unlock_bh(®_lock); + + if (cb) + cb(arg, tipc_mode, tipc_own_addr); +} + +/** + * reg_start - activate TIPC user registry + */ + +int reg_start(void) +{ + u32 u; + int res; + + if ((res = reg_init())) + return res; + + for (u = 1; u <= MAX_USERID; u++) { + if (users[u].callback) + k_signal((Handler)reg_callback, + (unsigned long)&users[u]); + } + return TIPC_OK; +} + +/** + * reg_stop - shut down & delete TIPC user registry + */ + +void reg_stop(void) +{ + int id; + + if (!users) + return; + + for (id = 1; id <= MAX_USERID; id++) { + if (users[id].callback) + reg_callback(&users[id]); + } + kfree(users); + users = 0; +} + +/** + * tipc_attach - register a TIPC user + * + * NOTE: This routine may be called when TIPC is inactive. + */ + +int tipc_attach(u32 *userid, tipc_mode_event cb, void *usr_handle) +{ + struct tipc_user *user_ptr; + + if ((tipc_mode == TIPC_NOT_RUNNING) && !cb) + return -ENOPROTOOPT; + if (!users) + reg_init(); + + spin_lock_bh(®_lock); + if (!next_free_user) { + spin_unlock_bh(®_lock); + return -EBUSY; + } + user_ptr = &users[next_free_user]; + *userid = next_free_user; + next_free_user = user_ptr->next; + user_ptr->next = -1; + spin_unlock_bh(®_lock); + + user_ptr->callback = cb; + user_ptr->usr_handle = usr_handle; + INIT_LIST_HEAD(&user_ptr->ports); + atomic_inc(&tipc_user_count); + + if (cb && (tipc_mode != TIPC_NOT_RUNNING)) + k_signal((Handler)reg_callback, (unsigned long)user_ptr); + return TIPC_OK; +} + +/** + * tipc_detach - deregister a TIPC user + */ + +void tipc_detach(u32 userid) +{ + struct tipc_user *user_ptr; + struct list_head ports_temp; + struct user_port *up_ptr, *temp_up_ptr; + + if ((userid == 0) || (userid > MAX_USERID)) + return; + + spin_lock_bh(®_lock); + if ((!users) || (users[userid].next >= 0)) { + spin_unlock_bh(®_lock); + return; + } + + user_ptr = &users[userid]; + user_ptr->callback = NULL; + INIT_LIST_HEAD(&ports_temp); + list_splice(&user_ptr->ports, &ports_temp); + user_ptr->next = next_free_user; + next_free_user = userid; + spin_unlock_bh(®_lock); + + atomic_dec(&tipc_user_count); + + list_for_each_entry_safe(up_ptr, temp_up_ptr, &ports_temp, uport_list) { + tipc_deleteport(up_ptr->ref); + } +} + +/** + * reg_add_port - register a user's driver port + */ + +int reg_add_port(struct user_port *up_ptr) +{ + struct tipc_user *user_ptr; + + if (up_ptr->user_ref == 0) + return TIPC_OK; + if (up_ptr->user_ref > MAX_USERID) + return -EINVAL; + if ((tipc_mode == TIPC_NOT_RUNNING) || !users ) + return -ENOPROTOOPT; + + spin_lock_bh(®_lock); + user_ptr = &users[up_ptr->user_ref]; + list_add(&up_ptr->uport_list, &user_ptr->ports); + spin_unlock_bh(®_lock); + return TIPC_OK; +} + +/** + * reg_remove_port - deregister a user's driver port + */ + +int reg_remove_port(struct user_port *up_ptr) +{ + if (up_ptr->user_ref == 0) + return TIPC_OK; + if (up_ptr->user_ref > MAX_USERID) + return -EINVAL; + if (!users ) + return -ENOPROTOOPT; + + spin_lock_bh(®_lock); + list_del_init(&up_ptr->uport_list); + spin_unlock_bh(®_lock); + return TIPC_OK; +} + diff --git a/net/tipc/user_reg.h b/net/tipc/user_reg.h new file mode 100644 index 00000000000..122ca9be367 --- /dev/null +++ b/net/tipc/user_reg.h @@ -0,0 +1,48 @@ +/* + * net/tipc/user_reg.h: Include file for TIPC user registry code + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_USER_REG_H +#define _TIPC_USER_REG_H + +#include "port.h" + +int reg_start(void); +void reg_stop(void); + +int reg_add_port(struct user_port *up_ptr); +int reg_remove_port(struct user_port *up_ptr); + +#endif diff --git a/net/tipc/zone.c b/net/tipc/zone.c new file mode 100644 index 00000000000..4eaef662d56 --- /dev/null +++ b/net/tipc/zone.c @@ -0,0 +1,169 @@ +/* + * net/tipc/zone.c: TIPC zone management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#include "core.h" +#include "zone.h" +#include "net.h" +#include "addr.h" +#include "node_subscr.h" +#include "cluster.h" +#include "node.h" + +struct _zone *zone_create(u32 addr) +{ + struct _zone *z_ptr = 0; + u32 z_num; + + if (!addr_domain_valid(addr)) + return 0; + + z_ptr = (struct _zone *)kmalloc(sizeof(*z_ptr), GFP_ATOMIC); + if (z_ptr != NULL) { + memset(z_ptr, 0, sizeof(*z_ptr)); + z_num = tipc_zone(addr); + z_ptr->addr = tipc_addr(z_num, 0, 0); + net.zones[z_num] = z_ptr; + } + return z_ptr; +} + +void zone_delete(struct _zone *z_ptr) +{ + u32 c_num; + + if (!z_ptr) + return; + for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { + cluster_delete(z_ptr->clusters[c_num]); + } + kfree(z_ptr); +} + +void zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr) +{ + u32 c_num = tipc_cluster(c_ptr->addr); + + assert(c_ptr->addr); + assert(c_num <= tipc_max_clusters); + assert(z_ptr->clusters[c_num] == 0); + z_ptr->clusters[c_num] = c_ptr; +} + +void zone_remove_as_router(struct _zone *z_ptr, u32 router) +{ + u32 c_num; + + for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { + if (z_ptr->clusters[c_num]) { + cluster_remove_as_router(z_ptr->clusters[c_num], + router); + } + } +} + +void zone_send_external_routes(struct _zone *z_ptr, u32 dest) +{ + u32 c_num; + + for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { + if (z_ptr->clusters[c_num]) { + if (in_own_cluster(z_ptr->addr)) + continue; + cluster_send_ext_routes(z_ptr->clusters[c_num], dest); + } + } +} + +struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref) +{ + struct cluster *c_ptr; + struct node *n_ptr; + u32 c_num; + + if (!z_ptr) + return 0; + c_ptr = z_ptr->clusters[tipc_cluster(addr)]; + if (!c_ptr) + return 0; + n_ptr = cluster_select_node(c_ptr, ref); + if (n_ptr) + return n_ptr; + + /* Links to any other clusters within this zone ? */ + for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { + c_ptr = z_ptr->clusters[c_num]; + if (!c_ptr) + return 0; + n_ptr = cluster_select_node(c_ptr, ref); + if (n_ptr) + return n_ptr; + } + return 0; +} + +u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref) +{ + struct cluster *c_ptr; + u32 c_num; + u32 router; + + if (!z_ptr) + return 0; + c_ptr = z_ptr->clusters[tipc_cluster(addr)]; + router = c_ptr ? cluster_select_router(c_ptr, ref) : 0; + if (router) + return router; + + /* Links to any other clusters within the zone? */ + for (c_num = 1; c_num <= tipc_max_clusters; c_num++) { + c_ptr = z_ptr->clusters[c_num]; + router = c_ptr ? cluster_select_router(c_ptr, ref) : 0; + if (router) + return router; + } + return 0; +} + + +u32 zone_next_node(u32 addr) +{ + struct cluster *c_ptr = cluster_find(addr); + + if (c_ptr) + return cluster_next_node(c_ptr, addr); + return 0; +} + diff --git a/net/tipc/zone.h b/net/tipc/zone.h new file mode 100644 index 00000000000..4326f78d829 --- /dev/null +++ b/net/tipc/zone.h @@ -0,0 +1,71 @@ +/* + * net/tipc/zone.h: Include file for TIPC zone management routines + * + * Copyright (c) 2000-2006, Ericsson AB + * Copyright (c) 2005, Wind River Systems + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. Neither the names of the copyright holders nor the names of its + * contributors may be used to endorse or promote products derived from + * this software without specific prior written permission. + * + * Alternatively, this software may be distributed under the terms of the + * GNU General Public License ("GPL") version 2 as published by the Free + * Software Foundation. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef _TIPC_ZONE_H +#define _TIPC_ZONE_H + +#include "node_subscr.h" +#include "net.h" + + +/** + * struct _zone - TIPC zone structure + * @addr: network address of zone + * @clusters: array of pointers to all clusters within zone + * @links: (used for inter-zone communication) + */ + +struct _zone { + u32 addr; + struct cluster *clusters[2]; /* currently limited to just 1 cluster */ + u32 links; +}; + +struct node *zone_select_remote_node(struct _zone *z_ptr, u32 addr, u32 ref); +u32 zone_select_router(struct _zone *z_ptr, u32 addr, u32 ref); +void zone_remove_as_router(struct _zone *z_ptr, u32 router); +void zone_send_external_routes(struct _zone *z_ptr, u32 dest); +struct _zone *zone_create(u32 addr); +void zone_delete(struct _zone *z_ptr); +void zone_attach_cluster(struct _zone *z_ptr, struct cluster *c_ptr); +u32 zone_next_node(u32 addr); + +static inline struct _zone *zone_find(u32 addr) +{ + return net.zones[tipc_zone(addr)]; +} + +#endif |